Lymphoma

Pathology, Diagnosis, and Treatment

Second Edition
Lymphoma
Pathology, Diagnosis, and Treatment
Second Edition
Edited by:
Robert Marcus
Department of Haematology, Kings College Hospital, Denmark Hill, London, UK

John W. Sweetenham
Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, USA

Michael E. Williams
Division of Hematology/Oncology, University of Virginia School of Medicine, Charlottesville, VA, USA
University Printing House, Cambridge CB2 8BS, United Kingdom

Published in the United States of America by Cambridge University Press, New York
Cambridge University Press is part of the University of Cambridge.
It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107010598
© Cambridge University Press 2014
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2014
Printed and bound in Spain by Graphos SA, Arte sobre papel
A catalog record for this publication is available from the British Library
ISBN 978-1-107-01059-8 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
.................................................................................................
Every effort has been made in preparing this book to provide accurate and
up-to-date information which is in accord with accepted standards and practice
at the time of publication. Although case histories are drawn from actual cases,
every effort has been made to disguise the identities of the individuals involved.
Nevertheless, the authors, editors, and publishers can make no warranties that the
information contained herein is totally free from error, not least because clinical
standards are constantly changing through research and regulation. The authors,
editors, and publishers therefore disclaim all liability for direct or consequential
damages resulting from the use of material contained in this book. Readers
are strongly advised to pay careful attention to information provided by the
manufacturer of any drugs or equipment that they plan to use.
Contents
Preface to second edition page vii
Preface to first edition ix
List of contributors x

1 Epidemiology 1 11 Burkitt and lymphoblastic lymphoma:

Eve Roman and Alexandra G. Smith clinical therapy and outcome 172
Nirali N. Shah, Alan S. Wayne, and Wyndham
2 Prognostic factors for lymphomas 11
H. Wilson
Guillaume Cartron and Philippe Solal-Céligny
12 Therapy of diffuse large B-cell
3 Imaging 32
lymphoma 191
Heok K. Cheow and Ashley S. Shaw
John W. Sweetenham
4 Clinical trials in lymphoma 45
13 Central nervous system lymphomas 208
Thomas M. Habermann and Matthew Maurer
Andrés J. M. Ferreri and Lisa M. DeAngelis
5 Hodgkin lymphoma 61
14 T-cell non-Hodgkin lymphoma 226
Stephanie Sasse and Andreas Engert
Sheetal M. Kircher, Beverly P. Nelson, Steven
6 Follicular lymphoma 87 T. Rosen, and Andrew M. Evens
Kristian Bowles, Daniel Hodson, and Robert 15 Primary cutaneous lymphoma 254
Marcus
Sean Whittaker
7 MALT and other marginal zone
16 Lymphoma in the immunosuppressed 273
lymphomas 104
Michele Spina, Robert Marcus, Shireen Kassam,
Emanuele Zucca and Francesco Bertoni
and Umberto Tirelli
8 Small lymphocytic lymphoma/chronic
17 Atypical lymphoproliferative, histiocytic,
lymphocytic leukemia 121
and dendritic cell disorders 286
Peter Hillmen
Matthew S. Davids and Jennifer R. Brown
9 Waldenström’s macroglobulinemia/
lymphoplasmacytic lymphoma 138
Steven P. Treon and Giampaolo Merlini
10 Mantle cell lymphoma 155 Index 299
Martin Dreyling and Michael E. Williams

v
Preface to second edition
Since the first edition we have seen a number of impor-
tant changes in the diagnosis, staging and therapy for
many lymphoma subtypes. We are beginning to
observe how molecular and cytogenetic characteristics
can profoundly influence prognosis and that such tools
should now be as much a part of our diagnostic arma-
mentarium as histology and immunophenotyping.
In the near future we expect that “next generation”
sequencing will also have a considerable impact not
only on our understanding of the pathogenesis of
lymphoma, but also on our selection of therapy. We
can also envisage how such techniques will themselves
also lead to less toxic and more targeted therapies.
We have seen the steady increase in the incorpo-
ration of PET and PET CT into staging, risk-adapted
therapy, and reassessment such that the presence of a
PET/CT scanner and radiologic expertise should now
be an essential in every major Lymphoma centre. The
promise and precautions for this important tool are
summarized within this edition.
In terms of therapy, the steady progress in both the
use of monoclonal antibodies and new treatment regi-
mens is reflected in the updated chapters. Lymphoma
specialists are also beginning to see the incorporation
of agents that block intra-cellular signaling pathways
into clinical practice, notably ibrutinib in CLL/SLL
and mantle cell lymphoma, and idelalisib in indolent
B-cell lymphoproliferative disorders. These specifi-
cally targeted therapies give us hope that, in the very
near future, the management of lymphoma will be
both more effective and less toxic.
In this new edition we have also endeavored to
rectify some of the omissions in the previous edition;
in particular, we have now additional chapters on
Lymphoplasmacytoid Lymphoma and Atypical
Lymphoproliferative disorders that make the book
more comprehensive.
Once again we have incorporated biology,
pathogenesis, histopathology, and therapy into each
disease based chapter, we hope the “joins” do not
show too much and are most grateful indeed to
Drs. Ott, Rosenwald, and Wotherspoon for allowing
their contributions on molecular biology and
histopathology, respectively, to be separated and
distributed as before.
Our thanks are due too to all the authors and to
colleagues at CUP for their hard work and forbearance
in the preparation of this new edition.
Robert Marcus
John Sweetenham
Michael Williams
vii
Preface to first edition
Why publish a book on lymphoma in 2007? Surely
there are sufficient reviews, meetings and published
educational symposia to make such a work redundant.
Furthermore, doesn’t instant access to online informa-
tion make any work in print out of date before it
appears?
The editors and authors of this work think not. We
firmly believe that there is still a place for a clear
summary of the diagnosis, staging and therapy of
lymphoma in a single volume that reflect the advances
in these areas which have taken place over the past five
years. The problem with the plethora of information
now available is that it is rarely set in a framework of
understanding of the major challenges which still
face those involved with the diagnosis and therapy of
non-Hodgkin’s and Hodgkin’s lymphomas. We, and
our patients, are confronted with many facts but
little judgment. This work is our modest attempt to
rectify this.
Accordingly the layout of the book should enable
the reader to gain an understanding of patterns of
disease, methods of staging, principles of new
approaches to therapy, and interpretation of clinical
trials and prognostic markers by reading the first part
of the book. In the second part the reader will find
separate succinct yet comprehensive reviews of the
individual disease entities which make up the spectrum
of diseases comprising the lymphomas. Here we have
integrated pathology, molecular biology and therapy
for each subtype into a single chapter; the reader will
not need to flick backwards and forwards to gain
a comprehensive understanding of, say, follicular or
diffuse large B-cell lymphoma. Such an integration
has posed significant editorial challenges, and has
been made possible by a willingness on the part of
Dr Wotherspoon, Dr Rosenwald and co-workers to
accept that their contributions on histopathology and
molecular cytogenetics would be divided and distrib-
uted among the relevant chapters. The editors are most
grateful for their support.
Each chapter in followed by a select bibliography
rather than an exhaustive list of references. We feel
that these date very quickly, take up disproportionate
amounts of space and are better found by internet
searches or perusal of current journals.
This book is intended for senior trainees and
fellows in hematology and oncology, together with
more experienced practitioners who regularly treat
these disorders. It is not intended for those who may
feel they could have written the chapters themselves.
It is also not a book where the reader will find detailed
descriptions of rarities seen once in a professional
lifetime.
The appearance of this volume comes at an oppor-
tune time: we have seen over the past five years a
profound understanding of the pathology of lym-
phoma, the use of increasingly sophisticated imaging
techniques, and the integration of monoclonal anti-
bodies into standard therapy for NHL. Our hope is
that these radical changes in the way we diagnose and
treat lymphoma have been reflected in the book and
will stand the reader in good stead even when newer
data become available.
The editors express their sincere and heartfelt
thanks to all the authors, who have, in the main,
responded promptly to our comments and recommen-
dations, and to all those at Cambridge University
Press who have helped with this project: Richard
Barling, who set the wheels of this vehicle in motion,
and especially Betty Fulford and Deborah Russell, who
kept it moving to its final destination whenever it
threatened to stall.
ix
 Contributors

Francesco Bertoni Andrew M. Evens

Lymphoma & Genomics Research Program, Division of Hematology/Oncology, Tufts University
Institute of Oncology Research (IOR), and School of Medicine, and Tufts Cancer Center Boston,
Oncology Institute of Southern Switzerland (IOSI), MA, USA
Bellinzona, Switzerland
Andrés J. M. Ferreri
Kristian Bowles Unit of Lymphoid Malignancies, Medical Oncology
Norwich Medical School, UK Unit, Department of Oncology, San Raffaele Scientific
Institute, Milan, Italy
Jennifer R. Brown
Harvard Medical School, Dana-Farber Cancer Thomas M. Habermann
Institute, Boston, MA, USA Mayo Clinic College of Medicine, Rochester, MN, USA

Guillaume Cartron Peter Hillmen

Department of Haematology University
Hospital, UMR CNRS 5235 Montpellier,
France
Heok K. Cheow
Department of Radiology, Addenbrooke’s Hospital,
Cambridge, UK
Matthew S. Davids
Harvard Medical School, Dana-Farber Cancer
Institute, Boston, MA, USA
Lisa M. DeAngelis
Department of Neurology, Memorial Sloan-Kettering
Cancer Center, New York, NY, USA
Martin Dreyling
University Hospital Grosshadern, Department of
Internal Medicine III, Ludwig-Maximilians-
University, Munich, Germany
Andreas Engert
First Department of Internal Medicine, Trial Secretary
GHSG, University Hospital Cologne, Cologne,
Germany
Department of Haematology, Leeds General
Infirmary, Leeds, UK
Shireen Kassam
Department of Haematology, Kings College Hospital,
Denmark Hill, London, UK
Sheetal M. Kircher
Division of Hematology/Oncology, Department of
Medicine, Northwestern University Feinberg School
of Medicine, Robert H. Lurie Comprehensive Cancer
Center, Chicago, IL, USA
Robert Marcus
Department of Haematology, Kings College Hospital,
Denmark Hill, London, UK
Matthew Maurer
Mayo Clinic College of Medicine, Rochester, MN, USA
Giampaolo Merlini
Department of Biochemistry at the University of
Pavia, and Biotechnology Research Laboratories,
University Hospital Policlinico San Matteo, Pavia,
Italy

x

Beverly P. Nelson
Department of Pathology, Northwestern
University Feinberg School of Medicine, Robert
H. Lurie Comprehensive Cancer Center, Chicago,
IL, USA
German Ott
Division of Hematology, Oncology and Blood &
Marrow Transplantation, Children’s Hospital Los
Angeles, Keck School of Medicine, University of
Southern California, Los Angeles, CA, USA
Eve Roman
Epidemiology & Cancer Statistics Group,
Department of Health Sciences, University of York,
York, UK
Steven T. Rosen
Robert H. Lurie Comprehensive Cancer Center,
Northwestern University Feinberg School of
Medicine, Chicago, IL, USA
Andreas Rosenwald
Institute of Pathology, University of Wurzburg,
Wurzburg, Germany
Stephanie Sasse
First Department of Internal Medicine, University
Hospital Cologne, Cologne, Germany
Nirali N. Shah
Pediatric Oncology Branch, Center
for Cancer Research, National Cancer
Institute, National Institutes of Health, Bethesda,
MD, USA
Ashley S. Shaw
Department of Radiology, Addenbrooke’s Hospital,
Cambridge, UK
Alexandra G. Smith
Epidemiology & Cancer Statistics Group,
Department of Health Sciences, University of York,
York, UK
Philippe Solal-Céligny
Institut de Cancérologie de l’Ouest, Nantes-Angers,
France
List of contributors
Michele Spina
Division of Medical Oncology A, National Cancer
Institute, Aviano, Italy
John W. Sweetenham
Huntsman Cancer Institute, University of Utah, 1950
Circle of Hope, Salt Lake City, UT, USA
Umberto Tirelli
Division of Medical Oncology A, National Cancer
Institute, Aviano, Italy
Steven P. Treon
Bing Center for Waldenstrom’s Macroglobulinemia,
Dana-Farber Cancer Institute, Harvard Medical
School, Boston, MA, USA
Alan S. Wayne
Division of Hematology, Oncology and Blood &
Marrow Transplantation, Children’s Hospital
Los Angeles, Keck School of Medicine,
University of Southern California, Los Angeles,
CA, USA
Sean Whittaker
St Johns Institute of Dermatology, Guys and St
Thomas’ NHS Foundation Trust, and Division of
Genetics and Molecular Medicine, Kings College
London, UK
Michael E. Williams
Hematology/Oncology Division, University of
Virginia Health System, Charlottesville, VA, USA
Wyndham H. Wilson
Lymphoma Therapeutics Section, Metabolism
Branch, Center for Cancer Research, National Cancer
Institute, National Institutes of Health,
Bethesda, MD, USA
Andrew Wotherspoon
Department of Histopathology,
Royal Marsden Hospital, Fulham Road,
London, UK
Emanuele Zucca
Oncology Institute of Southern Switzerland (IOSI),
Ospedale San Giovanni, Bellinzona, Switzerland
xi
 Chapter
1
Epidemiology
Eve Roman and Alexandra G. Smith
Introduction
Epidemiology is the basic quantitative science of
public health and, as such, is concerned with the dis-
tribution, determinants, treatment, management, and
potential control of disease. Concentrating on the first
two of these, this chapter reviews the epidemiology of
lymphomas – a heterogeneous group of malignancies
that is estimated to account for around 3–4% of can-
cers worldwide.
Epidemiological reports on lymphomas often
begin, and sometimes end, by stating that little is
known about the causes of the cancers under study.
This is slowly changing, as evidence about the patho-
logical diversity of the various lymphoma subtypes
accumulates. At present, however, the issue of lym-
phoma classification continues to permeate much of
the literature, since, in order to originate and test
hypotheses about pathogenesis, it is vitally important
to describe accurately and understand underlying
descriptive disease patterns. Implicit in this is the
need to use appropriate disease classifications; it is
this requirement that has beleaguered epidemiological
research into the lymphoid malignancies.
In short, the classification of hematological malig-
nancies has changed markedly over recent decades,
and will continue to do so as biological understand-
ing increases and new diagnostic methods and
techniques are developed. One problem for epide-
miological research concerns the use of historical
classifications emanating from the latter half of the
nineteenth century – long before there were any
effective treatments or real understanding of the rela-
tionships between lymphoid malignancies, the nor-
mal bone marrow and immune system, and before
anything was known about the cellular and genetic
basis of malignant transformation. The application
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
of modern disease classifications is, however, now
beginning to discriminate between subtypes, reveal-
ing many features that future etiological hypotheses
will undoubtedly seek to address. Accordingly, the
next few decades promise to be an exciting time for
epidemiological research into lymphoid, as well as
other hematological, malignancies.
Descriptive epidemiology
In 2001 the World Health Organization (WHO) pro-
duced, for the first time, a consensus classification that
defined malignancies of the hematopoietic and lym-
phoid systems in terms of their immunophenotype,
genetic abnormalities, and clinical features. Up until
then, the use of competing classifications had tended
to make meaningful comparison of results both
between and within populations virtually impossible.
For a variety of reasons, although WHO’s 2001 classi-
fication and its successor was adopted into clinical
practice almost uniformly around the world, it did not
have an immediate effect on population-based epide-
miological research. This is because, unlike many other
cancers, hematological neoplasms are diagnosed using
multiple parameters, including a combination of histol-
ogy, cytology, immunophenotyping, cytogenetics,
imaging, and clinical data. This range and depth of
data is difficult for cancer registries and other research-
ers to access routinely, forming a barrier both to com-
plete ascertainment and to the collection of diagnostic
data at the level of detail required to implement the
latest classification systematically. Furthermore, in
practice, even within some of the best defined WHO
categories, there is a need to qualify the final diagnosis
even further using additional clinical and biological
prognostic factors before valid outcome comparisons
can be made between clinical centers.
1
 Chapter 1: Epidemiology
Gathering accurate information about disease bur-
dens and patterns is central to any successful cancer
information strategy. However, whilst cancer registra-
tion has a long history in many countries, particularly
those in the more affluent regions of the world, nearly
80% of the world’s population is not covered by such
systems (www.globocan.iarc.fr/). Furthermore, in
some of these, even enumerating the population at
risk (denominator) through census counts and vital
registration is challenging. Even so, with a view to
characterizing the global burden of disease, the
WHO’s International Agency for Research on Cancer
(IARC) routinely uses the available data to estimate
worldwide cancer incidence and mortality levels.
Misdiagnosis and underenumeration are recog-
nized as particularly problematic for hematological can-
cers. The acute and rapidly fatal presentation of some
patients leads to underenumeration in those countries
with less well-developed health service infrastructures
and the intermittent and non-specific nature of symp-
toms associated with others poses problems even in
countries with well-developed health service systems
and cancer registration processes. In addition,
population-based data continue to be reported in the
broad anatomical-based categories of non-Hodgkin
lymphoma (NHL), Hodgkin lymphoma (HL), mye-
loma, and leukemia since, for reasons outlined above,
the laboratory data required to classify hematological
cancers appropriately are hard for cancer registries to
access in a timely and systematic fashion.
Geographic variation
Of the 12.68 million new cancers estimated to have
occurred around the world in 2008, 6.64 were in men
and 6.04 in women. Combined, hematological malig-
nancies (lymphomas, leukemias, and myelomas) com-
prised 7.5% of the estimated cancers in males and 6.4%
in females, with lymphomas accounting for around
half of all newly diagnosed hematological neoplasms
in both men and women.
Figure 1.1 shows the estimated global regional rates
and numbers of cases of NHL and HL separately for
men and women (www.globocan.iarc.fr/). In general,
estimated lymphoma rates are higher in more econom-
ically developed regions of the world – North America,
Europe, and Australasia – and lower in less affluent
regions. This relationship with affluence is seen for
both NHL and HL, with one or two interesting pockets
2 such as the consistently low overall rates reported
for Japan (www.globocan.iarc.fr/). Nevertheless, less
than half of all NHL and HL diagnoses occurred in
the three regions with the highest rates – Northern
America, Europe, and Oceania (Figure 1.1) – largely
reflecting the underlying world population distribution.
The most striking disparity is seen for Asia, which has
the lowest rates but more than a third of all of the
estimated worldwide cases. To some extent these
broad regional differences may well reflect, at least in
part, underlying lymphoma subtype variations – with a
relatively high proportion of mature T-/natural killer
cell neoplasms being reported for several Asian popula-
tions. These issues are discussed in more depth in the
later section on etiology.
Age and sex
That lymphoid malignancies are, overall, generally
more common in men than in women in all areas of
the world is evident from Figure 1.1, where for both
NHL and HL the age-standardized rates and numbers
are consistently higher for males than for females.
Interestingly, these gender differences appear to be
slightly more pronounced in less developed regions of
the world (http://globocan.iarc.fr/) – the sex rate ratios
(M:F) for HL, for example, range from 1.8 and 1.6,
respectively, in Africa and Asia through to 1.2 and 1.1,
respectively, in North America and Europe. Whether or
not these gradations reflect genuine underlying inci-
dence differences, either generally or within particular
subtypes, or are in fact caused by enumeration biases
cannot be investigated further from the available cancer
registration data.
The NHL and HL age-specific incidence patterns
seen in more economically developed regions of the
world are all broadly similar to the pattern shown in
Figure 1.2, which presents cancer registration data
from the UK. For NHL, the age-specific male and
female rates increase with increasing age, the diver-
gence between the male and female rates becoming
progressively more marked as age increases, but,
despite this, more women than men are diagnosed
over the age of 80 years. This apparent discrepancy
reflects the fact that the UK, like other economically
developed regions of the world, has an aging popula-
tion structure within which more women than men
survive to reach old age.
The relationship with age and sex is quite different
for NHL and HL. In more affluent regions of the world
HL tends to have a bimodal age distribution that is
characterized by early age peak, within which females
are often in excess but have a deeper following trough,
 Chapter 1: Epidemiology

A Male rate Female rate

Male cases Female cases Number of Cases
80000 70000 60000 50000 40000 30000 20000 10000 0 10000 20000 30000 40000 50000 60000 70000 80000

NORTHERN NORTHERN
AMERICA AMERICA
OCEANIA OCEANIA

EUROPE EUROPE

LATIN AMERICA LATIN AMERICA

& CARIBBEAN & CARIBBEAN
AFRICA AFRICA

ASIA ASIA

18 16 14 12 10 8 6 4 2 0 2 4 6 8 10 12 14 16 18
B Age Standardized Rate
Male rate Female rate
Male cases Female cases
Number of Cases
20000 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

NORTHERN NORTHERN
AMERICA AMERICA
EUROPE EUROPE

OCEANIA OCEANIA

LATIN AMERICA LATIN AMERICA

& CARIBBEAN & CARIBBEAN
AFRICA AFRICA

ASIA ASIA

3 2.5 2 1.5 1 0.5 0 0.5 1 1.5 2 2.5 3

Age Standardized Rate

Figure 1.1 Estimated numbers and age-standardized (world population) incidence rates by region for (A) non-Hodgkin lymphoma and
(B) Hodgkin lymphoma. (Source: GLOBOCAN.)

and a late age peak with a pronounced male excess
(Figure 1.2). The earlier age peak is not clearly evident
in less well developed regions of the world. Inspection
of various case-series has revealed that these patterns
are the result of differences in the age and gender
frequencies of the various HL subtypes. Likewise, the
comparatively smooth pattern seen when all NHLs are
combined (Figure 1.2) conceals considerable age and
sex subtype variability.
The diagnostic challenges posed by hematological
malignancies means that subtype data can only be
obtained from specialist registries. Furthermore,
population-based data with clearly defined numera-
tors and denominators (as opposed to hospital-based
case-series) are required for epidemiological research.
One such registry is the UK-based Haematological
Malignancy Research Network (www.HMRN.org),
and descriptive age and sex patterns for the lympho-
mas diagnosed within HMRN are presented in
Figures 1.3 and 1.4. Established in 2004 and covering
a population of 3.6 million with over 2100 diagnoses
each year, HMRN comprises an ongoing population-
based cohort of all newly diagnosed hematological
malignancies (pediatric and adult). Importantly, as a
matter of policy and irrespective of treatment intent,
all diagnoses within HMRN are made and coded to the
latest WHO classification by a single specialist hema-
topathology laboratory.
HMRN patients newly diagnosed with lymphoma
in the 5 years from September 2004 to August 2009 are
proportionately distributed by WHO diagnostic cate-
gory in Figure 1.3, beginning with the B-cell NHLs,
moving clockwise through the T-cell NHLs and ending
with the HLs. Diffuse large B-cell (DLBCL), follicular
(FL), and marginal zone lymphomas (MZL) dominate –
together accounting for more than 70% of the total. The
high proportion of DLBCL and MZL is a common
feature evident in all reported series, including hospital-
based registers in Asia. By contrast, in certain Asian and
other population groups, diagnoses of the more indo-
lent FL appear to occur far less frequently, whereas 3
diagnoses of T-cell lymphomas are more common.
 Chapter 1: Epidemiology

A Figure 1.2 Numbers of new cases and

800 100 age-specific incidence rates by sex for (A)
non-Hodgkin lymphoma and (B) Hodgkin
90 lymphoma, UK in 2007.

Rate per 100,000 population

80
600
70
Number of cases

60
400 50
40
30
200
20
10
0 0
4
10 9
15 4
20 9
25 24
30 9
35 4
40 9
45 44
50 9
55 54
60 59
65 4
70 9
75 4
80 9
4
+
0–
5–
–1
–1

–2
–3
–3

–4

–6
–6
–7
–7
–8
85
–

–
–

Male cases Female cases

Male rates Female rates

B
120 6

100 5

Rate per 100,000 population

Number of cases

80 4

60 3

40 2

20 1

0 0
4
10 9
15 4
20 9
25 4
30 9
35 4
40 9
45 4
50 9
55 4
60 9
65 4
70 9
75 4
80 9
4
+
0–
5–
–1
–1
–2
–2
–3
–3
–4
–4
–5
–5
–6
–6
–7
–7
–8
85

Male cases Female cases

Male rates Female rates
Age at diagnosis

The corresponding HMRN box-and-whisker age
distributions and sex rate ratios (male rate:female rate),
together with their standard errors, are shown in Figure
1.4. Whilst most B-cell NHLs have a median diagnostic
age over 70 years, a significant minority tend to be
diagnosed at younger ages – follicular, Burkitt, and
mediastinal lymphomas in HMRN having median ages
of 65, 52, and 36 years, respectively. Furthermore, some
subtypes have broad age ranges whilst others are narrow;
4 with no patients diagnosed before the age of 48 years and
a median of 74 years, the age distribution of mantle cell
lymphoma shows the least variation. By contrast, with a
median diagnostic age approaching 71 years, but with
several sporadic pediatric cases, DLBCL covers the larg-
est age range – the scatter of outliers at younger ages is
indicative, perhaps, of diagnostic heterogeneity within
this subtype category.
Overall, in agreement with other reports, within
HMRN T-cell NHLs tend to be diagnosed at younger
ages than B-cell neoplasms. Nonetheless, within T-cell
forms of the disease there is considerable subtype
heterogeneity – the tight age band for enteropathy
 Chapter 1: Epidemiology

Figure 1.3 Lymphoma subtype

frequencies. (Haematological Malignancy
Research Network, 2004–2009.)

Systemic marginal zone

Extranodal marginal zone
Follicular
Mantle cell
Diffuse large B-cell
Mediastinal large B-cell
Burkitt
Peripheral T-cell - common; unspecified
Mycosis fungoides
Primary cutaneous CD30 positive T-cell
Anaplastic large cell - T/null type
Angioimmunoblastic T-cell
Enteropathy-type T-cell
Other non-Hodgkin
Lymphocyte predominant nodular Hodgkin
Nodular sclerosis classical Hodgkin
Mixed cellularity classical Hodgkin
Lymphocyte-rich classical Hodgkin

Systemic marginal zone

Extranodal marginal zone
Follicular
Mantle cell
Diffuse large B-cell
Mediastinal large B-cell
Burkitt

Enteropathy-type T-cell
Mycosis fungoides
Primary cutaneous CD30 positive T-cell
Anaplastic large cell-T/null type
Peripheral T-cell lymphoma – common; unspecified
Angioimmunoblastic T-cell

Lymphocyte predominant nodular Hodgkin

Nodular sclerosis classical Hodgkin
Lymphocyte-rich classical Hodgkin
Mixed cellularity classical Hodgkin

0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 5
Age at Diagnosis (years) Sex Rate Ratio

Figure 1.4 Lymphoma box-and-whisker plot age distributions, and sex rate ratios with standard errors (M:F). (Haematological Malignancy
Research Network, 2004–2009.) (Legend below).

1st Quartile 3rd Quartile

Min Median Max

Outliers

type T-cell contrasting with that seen for anaplastic
large T-cell, for example. Lastly, with a pronounced
pediatric component, HL tends to be diagnosed earlier
still, but, again, there are differences between the sub-
types – the median ages within the classical Hodgkin
lymphoma (CHL) category, for example, ranging from
37 years for nodular sclerosis CHL to 60 years for
mixed cellularity CHL. These subtype differences are
largely responsible for the bimodal HL age and sex
distributions shown in Figure 1.2.
That males are far more likely to develop lymphoma
than females is conspicuously clear in Figures 1.1, 1.2,
and 1.4 – but it is also evident that the gender disparity
is much greater for some subtypes, whilst being almost
absent for others. Among B-cell NHLs, roughly equal
numbers of males and females were diagnosed with FL
and with extranodal MZL; however, the sex rate ratio
for all other B-cell lymphomas shows a male bias – the
most striking being for Burkitt lymphoma which, in 5
these and other series, is at least three times as likely to
 Chapter 1: Epidemiology

A 30 Figure 1.5 Age-standardized (US

2000 population) incidence rates for
(A) non-Hodgkin lymphoma and (B)
25 Hodgkin lymphoma. (Source: SEER.)

20
Rate per 100,000

15 Female rate
Male rate

10

0
1970 1975 1980 1985 1990 1995 2000 2005 2010
Year

B 4.5

3.5

3
Rate per 100,000

2.5
Female rate
2 Male rate

1.5

0.5

0
1970 1975 1980 1985 1990 1995 2000 2005 2010
Year

be diagnosed in males than in females. T-cell NHLs and
HLs also exhibit a male bias, but again there is consid-
erable subtype heterogeneity, with angioimmunoblastic
T-cell lymphoma running counter to the trend by being
more common in women.
The wide diversity of descriptive patterns shown in
Figure 1.4 is indicative of different subtype etiologies,
which many new and ongoing epidemiological studies
are aiming to address. This, coupled with the geo-
graphic variations for T- and B-cell subtypes reported
for various case-series, underscore the importance of
using appropriate classifications. Indeed, the contin-
ued application of site-based classification systems for
routine cancer registration severely limits the use of
6 their data in epidemiological studies. Hopefully, in the
future, new insights into lymphoma epidemiology will
emerge as the WHO classification becomes more
widely applied in a population-based context..
Changes over time
Monitoring disease trends over time is a fundamental
activity of descriptive epidemiology, such analyses
often yielding important etiological clues. Indeed,
there are many examples in the field of cancer epi-
demiology where this has been the case, particularly in
relation to the identification of hazardous occupa-
tional and environmental exposures. In this context,
the temporal changes reported for NHL in recent
decades are unquestionably dramatic, as can be seen
from Figure 1.5, which shows the estimated age-
adjusted incidence rates from the Surveillance,

Epidemiology and End Results (SEER) Program in the
United States (www.seer.cancer.gov).
The sharp increase seen for NHL in the USA
between the 1970s and the 1990s was reported in
many countries with population-based cancer regis-
tration systems (both statutory and voluntary), and the
magnitude and global scale of the effect naturally
captured both scientific and public attention at the
time (Figure 1.5). In males, the doubling of the NHL
rate over a comparatively short 20-year period is par-
ticularly striking, and contrasts with the modest fall in
HL seen among males over the same time period.
These registrational trends coincided with a period of
significant change in clinical understanding and con-
sequent marked taxonomic alterations.
During the 1980s the working formulation domi-
nated in North America, whilst Kiel gained ground in
Europe. In 1994, the Revised European–American
Lymphoma (REAL) classification was published,
and this was followed by the WHO classifications of
2001 and 2008. Quantification of the likely impact of
these diagnostic changes on the patterns seen in
Figure 1.5 is almost impossible to achieve. Whilst it
is generally recognized that shifting clinical diagnos-
tic practice almost certainly fuelled the trends, there
remains debate about whether such changes could
have been responsible for all, or only part, of the
dramatic increase in NHL registrations. In this con-
text, for example, the potential etiological role of
human immunodeficiency virus (HIV) and organ
transplantation, as well as other viruses and environ-
mental exposures, attracted considerable attention at
the time – and these factors are discussed in more
detail in the section on causes below.
Whatever the cause, the rate of change in NHL
slowed in the USA towards the end of the twentieth
century, and similar plateauing has now been reported
for other developed regions of the world. Taken at face
value, this would seem to support the notion that
changes in disease detection, diagnostic practice, and
cancer registration procedures are all likely to have
contributed to the increase seen over the 1970s and
1990s. Indeed, a recent report examining incidence
and survival trends across Europe concluded that ‘the
evolving classification and poor standardization of
data collected on hematological malignancies vitiate
the comparisons of disease incidence and survival over
time and across regions’ (Sant et al., 2009).
Importantly, with respect to the examination of
future time-trends, the role of changing diagnostic
Chapter 1: Epidemiology
practice will undoubtedly remain challenging to evalu-
ate since hematological oncology is changing rapidly,
with new approaches to treatment and diagnosis con-
tinually emerging as diverse patient pathways evolve. In
this regard, the use of more reliable and sensitive diag-
nostic techniques continues to lower the threshold of
disease detection – the ability of flow cytometry to detect
small populations of abnormal cells, for example, as well
as the introduction of the polymerase chain reaction
(PCR) allowing the detection of disease at a molecular
level. Hence, proportionately more of the patients diag-
nosed today have indolent or asymptomatic disease, and
it remains possible that the increasing numbers of regis-
trations may be the result, at least in part, of the recog-
nition of new group(s) of patients that would not have
been diagnosed in previous decades.
Etiology
For the reasons discussed above, etiological studies con-
ducted in previous decades have often been hampered
by the need to aggregate their data into the broad group-
ings of NHL or HL, either because primary source
information was recorded in that way or because diag-
nostic standards were inconsistently applied. Hence,
given the underlying variations in pathologies and prog-
nosis, it is perhaps not surprising that epidemiological
reports on lymphomas have often tended to produce
inconsistent and conflicting results. Hopefully, this sit-
uation will improve in the near future as evidence about
the pathological and clinical diversity of lymphoma
subtypes continues to accumulate, not only revealing
descriptive differences such as those outlined in the
previous section, but also other associations.
With respect to the underlying causes of lym-
phoma, it is generally agreed that immune dysregula-
tion plays a pivotal role in lymphogenesis, and most
epidemiological research has concentrated on factors
and exposures that interact with the immune system –
particularly infection, immunosuppression, and auto-
immune disease. These interrelated associations are
discussed in the sections that follow.
Infection
In the context of immunodeficiency (see following sec-
tion), the strong association between human immuno-
deficiency virus/acquired immunodeficiency syndrome
(HIV/AIDS) and several B-cell lymphomas has been
examined in numerous studies – the increased risks
reported range from 10- to 300-fold, the highest 7
 Chapter 1: Epidemiology
risks generally been observed for the more aggressive
NHL subtypes and the lowest risks for the HL subtypes.
Unsurprisingly, the potential role of the HIV/AIDS
epidemic was one of the factors evaluated in the
context of the striking rise of NHL observed in the
1970s–90s discussed above. In this context it is impor-
tant to note that NHL rates had begun to rise before the
onset of the HIV/AIDS epidemic in the 1980s
(Figure 1.5), and in the era of modern therapies the
contribution of HIV/AIDS to the totality of lymphomas
diagnosed in developed countries remains compara-
tively small.
In addition to HIV, a number of other viruses are
either known or thought to impact on lymphoma risk,
albeit by different mechanisms. The most accepted
viral associations are those with human T-cell leuke-
mia virus type 1 (HTLV-1) and the two herpesviruses
Epstein–Barr virus (EBV) and human herpesvirus 8
(HHV-8) – also known as Kaposi’s sarcoma herpesvi-
rus (KSHV). Infection with HTLV-1 is a necessary but
not sufficient cause of adult T-cell leukemia/lym-
phoma (ATLL), with ATLL developing in around 3%
of those infected. ATLL principally occurs in areas
where HTLV-1 is endemic, including Japan, the
Caribbean, and parts of central Africa and, like most
other lymphomas, it is more common in males than
females.
In contrast to the specific nature of the HTLV-1/
ATLL association, the globally ubiquitous EBV fea-
tures in several lymphoma subtypes, including
Burkitt lymphoma and CHL, as well as those lym-
phomas occurring in immunosuppressed individu-
als. EBV is invariably associated with the endemic
form of Burkitt lymphoma that occurs in equatorial
Africa and New Guinea, where, with a median age of
around 7 years and a male to female ratio of 2:1, it is
the commonest pediatric malignancy. However, as
can be seen from Figure 1.4, the median age at onset
of the comparatively rare sporadic form of Burkitt
lymphoma seen in other areas of the world is much
later, and the male predominance is considerably
greater. In contrast to endemic Burkitt’s, EBV is
only associated with around a third of all sporadic
cases, and the role of EBV in these tumors remains a
continuing area of research. Likewise, the signifi-
cance of the well-established association between
EBV and CHL, observed most frequently in children
and the elderly in around a third of all cases in
resource-rich countries (but even more commonly
8 in less affluent parts of the world), remains an area
of current investigation. HHV-8 has been found in
several comparatively rare tumor subtypes, where it
sometimes occurs with EBV. The nature of these
associations, as well as those with a number of other
viruses, including both hepatitis B and C, are topics
of much ongoing research.
Albeit by very different mechanisms, bacterial
infection resulting in chronic inflammation has
been consistently linked with a marked increase in
the risk of some NHL subtypes – one of the best
known associations being that between Helicobacter
pylori and gastric mucosa-associated lymphoid tissue
(MALT) extranodal MZL; the presence of the bacteria
increase lymphoma risk by about sixfold. As with the
viral associations (e.g. HTLV-1 and ATLL), bacterial
infection often occurs many years before MALT lym-
phoma development, the median age of diagnosis of
all extra-marginal zone lymphomas combined in the
HMRN series being just under 70 years with no
obvious sex bias (Figure 1.4). H. pylori infection,
however, generally occurs in childhood – the preva-
lence being highest in developing countries and fall-
ing as socioeconomic status increases. In the future,
the introduction of effective H. pylori treatments,
coupled with increasing affluence in developed
regions of the world, may well lead to a reduction in
incidence of H. pylori-positive gastric MZLs.
Interestingly, with respect to the timing of infection
and subsequent lymphoma development, detailed anal-
ysis of medical records collected during the course of a
UK case-control study revealed significant increases in
the frequency of non-specific infectious illness episodes
more than 10 years before the diagnosis of CHL, but not
of DLBCL and FL. No associations with specific infec-
tions were, however, found, and whether or not the
excess seen for CHL was a consequence of an under-
lying immune abnormality or whether infection played
a causal role could not be determined.
Immunosuppression
In addition to HIV/AIDS, it has been known for some
time that both inherited and acquired immunodefi-
ciency syndromes predispose towards lymphoprolifer-
ative disorders. A few rare inherited disorders of the
immune system, including Wiskott–Aldrich syn-
drome and ataxia telangiectasia, are well known to be
associated with marked increases in the risk of lym-
phoproliferative disorders. Such conditions are, how-
ever, exceedingly rare in the general population, and

the proportion of total lymphoma diagnoses for which
they account is correspondingly small.
In addition to these rare inherited conditions, lym-
phoma risks in organ transplant recipients have been
extensively investigated, and there are many ongoing
patient cohorts. The spectrum of lymphoma subtypes
that occur in immunosuppressed individuals tends to
differ in important respects from those seen in the non-
immunosuppressed, being more aggressive, less respon-
sive to therapy, often EBV-associated, and more likely
to occur at extranodal sites. Indeed, post-transplant
lymphoproliferative disorders (PTLD) are generally
considered separately, the latest WHO classification
distinguishing several different malignant and benign
PTLD disease forms. Immunosuppression is recognized
as the major causal factor in PTLD, the likelihood of
development varying with the patient’s age and type of
transplant – the incidence being lowest following renal
transplant and highest following heart and lung.
Organ transplantation is an increasingly successful
therapy and, as with HIV/AIDS, it is one of the factors
often considered to have contributed to the temporal
increase in NHLs seen at the end of the last century
(Figure 1.5). Nonetheless, organ transplantation
remains comparatively rare and the proportion of
the increase it could account for remains small.
Autoimmune disease
Several strong associations between autoimmune dis-
ease and subsequent lymphoma development have
been described. In particular, links with autoimmune
disease-based chronic inflammation and MALT lym-
phoma are well recognized – two of the strongest and
most well-known relationships being those between
Hashimoto thyroiditis and thyroid lymphoma and
between Sjögren’s syndrome and salivary gland lym-
phoma. Consistent associations have also been
reported for other chronic inflammatory rheumatic
diseases; for example, rheumatoid arthritis and sys-
temic lupus erythematosus with both MZL and
DLBCL, and gastrointestinal inflammatory autoim-
mune conditions, such as celiac disease and Crohn’s
disease, with the T-cell lymphomas. Indeed, evidence
linking lymphoproliferative disorders with autoim-
munity continues to accumulate, the strongest associ-
ations being seen for MZL, DLBCL, and T-cell
lymphoma, and the weakest with the relatively indo-
lent FL. Associations between chronic inflammatory
rheumatic diseases and CHL, but not nodular sclerosis
HL, have also been reported.
Chapter 1: Epidemiology
Taken as a whole, there is broad consensus that
elucidating the mechanisms underpinning the link
between autoimmune disease and the lymphomas
could lead to fundamental insights into the biology of
both groups of disorders, and there is considerable
speculation about potential mechanisms in the litera-
ture. In this regard routine examination of sex-specific
associations could provide valuable information since it
has long been known that most autoimmune conditions
are considerably more common in women than in men,
whereas the reverse is true for the majority of lympho-
proliferative malignancies. Furthermore, there are well-
documented sex-specific variations in immune response
that may well impact on the risk of these diseases.
Other environmental factors
The dramatic increase in lymphoma registrations seen
at the end of the last century (Figure 1.5) resulted
in a massive increase in epidemiological research.
Accordingly, in addition to investigating the explicit
immunological factors discussed above, many other
putative risk factors have been examined, including
genetic predisposition, associations with other comor-
bidities, and a wide range of occupational and other
potentially hazardous environmental exposures.
However, despite concentrated effort, many of the
potentially hazardous associations identified have
been weak or contradictory, with few consistent asso-
ciations emerging. Indeed, at the present time, the lack
of accepted environmental determinant(s) for lym-
phomas sits starkly against the accumulated knowl-
edge about other cancers such as lung, stomach, skin,
bladder, and breast.
With respect to the continued investigation of lym-
phoma determinants, the InterLymph consortium – a
scientific forum for epidemiologic research on lym-
phoma – was established in 2001. InterLymph members
have reviewed and published on several health-related
states/events and environmental exposures – and their
website includes all their reports and provides up-to-
date information on many putative risk factors (http://
epi.grants.cancer.gov/InterLymph/). Thus far, topics
investigated by InterLymph have included self-reported
obesity (rising levels being another potential factor sug-
gested as contributing to temporal trends), smoking
and alcohol (no consistent effects found); sun exposure
and history of atopic infections (marginally reduced
risks found); and family history of hematological can-
cers (increased risks confirmed). Finally, with a view to 9
investigating genetic susceptibility, the majority of
 Chapter 1: Epidemiology
InterLymph studies collected constitutional biological
material from cases and corresponding controls, and
findings here have, perhaps, been more informative.
Thus far polymorphisms in two immune system-
related genes have been identified using the candidate
gene approach, associations being seen for both DLBCL
and MZL, and genome-wide association studies are
currently ongoing.
Further reading
Anderson LA, Gadalla S, Morton LM, et al. Population-
based study of autoimmune conditions and the risk of
specific lymphoid malignancies. Int J Cancer, 2009;
125(2):398–405.
Ansell P, Simpson J, Lightfoot T, et al. Non-Hodgkin
lymphoma and autoimmunity: does gender matter? Int J
Cancer, 2011;129(2):460–466.
Barrans S, Crouch S, Smith A, et al. Rearrangement of MYC
is associated with poor prognosis in patients with diffuse
large B-cell lymphoma treated in the era of rituximab. J
Clin Oncol, 2010;28(20):3360–3365.
Boffetta P, Armstrong B, Linet M, et al. Consortia in cancer
epidemiology: lessons from InterLymph. Cancer
Epidemiol. Biomarkers Prev, 2007;16(2):197–199.
Bosetti C, Levi F, Ferlay J, et al. Incidence and mortality from
non-Hodgkin lymphoma in Europe: the end of an
epidemic? Int J Cancer, 2008;123(8):1917–1923.
Boyle P. International agency for research on cancer. World
Cancer Report 2008. Lyon: IARC Press; 2008.
Crouch S, Simpson J, Ansell P, et al. Illness patterns prior to
diagnosis of lymphoma: analysis of UK medical records.
Cancer Epidemiol [Internet]. 2010, Sept. 8 [cited Sept. 15,
2010]; Available from: http://www.ncbi.nlm.nih.gov/
pubmed/20832384
Dal Maso L, Serraino D, Franceschi S. Epidemiology of
AIDS-related tumours in developed and developing
countries. Eur J Cancer, 2001;37(10):1188–1201.
Ferlay J, Shin H, Bray F, et al. Estimates of worldwide
burden of cancer in 2008: GLOBOCAN 2008. Int J
Cancer [Internet]. 2010, June 17 [cited Aug. 26, 2010];
Available from: http://www.ncbi.nlm.nih.gov/pubmed/
20560135
Jarrett RF. Viruses and lymphoma/leukaemia. J Pathol,
2006;208(2):176–186.
Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA
Cancer J Clin, 2010;60(5):277–300.
Luminari S, Cesaretti M, Marcheselli L, et al. Decreasing
incidence of gastric MALT lymphomas in the era of anti-
Helicobacter pylori interventions: results from a
population-based study on extranodal marginal zone
10 lymphomas. Ann Oncol, 2010;21(4):855–859.
Marcos-Gragera R, Pollán M, Chirlaque MD, et al. Attenuation
of the epidemic increase in non-Hodgkin’s lymphomas in
Spain. Ann Oncol, 2010;21 Suppl 3:iii90–96.
Mozaheb Z, Aledavood A, Farzad F. Distributions of major
sub-types of lymphoid malignancies among adults in
Mashhad, Iran. Cancer Epidemiol [Internet]. 2010, Oct.
29 [cited Jan. 10, 2011]; Available from: http://www.ncbi.
nlm.nih.gov/pubmed/21036690
Orem J, Mbidde EK, Lambert B, de Sanjose S, Weiderpass E.
Burkitt’s lymphoma in Africa, a review of the epidemiology
and etiology. Afr Health Sci, 2007;7(3):166–175.
Relander T, Johnson NA, Farinha P, et al. Prognostic
factors in follicular lymphoma. J Clin Oncol,
2010;28(17):2902–2913.
Sagaert X, Van Cutsem E, De Hertogh G, Geboes K,
Tousseyn T. Gastric MALT lymphoma: a model of
chronic inflammation-induced tumor development. Nat
Rev Gastroenterol Hepatol, 2010;7(6):336–346.
Sant M, Allemani C, Santaquilani M, et al. EUROCARE-4.
Survival of cancer patients diagnosed in 1995–1999.
Results and commentary. Eur J Cancer,
2009;45(6):931–991.
Sant M, Allemani C, Tereanu C, et al. Incidence of
hematologic malignancies in Europe by morphologic
subtype: results of the HAEMACARE project. Blood,
2010;116(19):3724–3734.
Smedby KE, Hjalgrim H, Askling J, et al. Autoimmune
and chronic inflammatory disorders and risk of
non-Hodgkin lymphoma by subtype. J Natl Cancer Inst,
2006;98(1):51–60.
Smith A, Roman E, Howell D, et al. The Haematological
Malignancy Research Network (HMRN): a new
information strategy for population based
epidemiology and health service research. Br J
Haematol, 2010;148(5):739–753.
Swerdlow S. International Agency for Research on Cancer,
World Health Organization. WHO Classification of
Tumours of Haematopoietic and Lymphoid Tissues. 4th
ed. Lyon, France: International Agency for Research on
Cancer, 2008.
Végső G, Hajdu M, Sebestyén A. Lymphoproliferative
disorders after solid organ transplantation–classification,
incidence, risk factors, early detection and treatment
options. Pathol Oncol Res [Internet]. 2010, Dec. 31 [cited
Jan 12, 2011];Available from: http://www.ncbi.nlm.nih.
gov/pubmed/21193979
Westlake S. Cancer incidence and mortality in the United
Kingdom and constituent countries, 2004–06. Health Stat
Q, 2009;43:56–62.
Yoon SO, Suh C, Lee DH, et al. Distribution of lymphoid
neoplasms in the Republic of Korea: analysis of 5318 cases
according to the World Health Organization
classification. Am J Hematol, 2010;85(10):760–764.
 Chapter
2
Prognostic factors for lymphomas
Guillaume Cartron and Philippe Solal-Céligny
Introduction
Lymphomas are tumors that have largely benefited
from the introduction of cytotoxic chemotherapy.
Treatment strategies based on intensive chemotherapy
have also demonstrated improved survival of patients
with relapsed lymphomas. It is necessary to identify
selected populations with adverse prognostic factors
that would justify such a strategy. Prognostic indices
have been developed both to define therapeutic strat-
egies and compare results of clinical trials. Such indi-
ces are usually based on retrospective studies that
highlight methodological problems in prospective
studies in the monoclonal antibody era. The recent
development of technologies analyzing gene expres-
sion profiling also gives us new tools for both more
accurate diagnosis and prognostic implications. We are
also beginning to see the development of indices based
on lymphoma-specific biological risk factors. Positron
emission tomography using 18F-fluorodeoxyglucose
should also improve assessment of tumor response
and introduce new prognostic measures.
Methodology for building prognostic
indices
The goals of a prognostic index (PI) are as follows:
(i) for a single patient at the time of diagnosis, to help
the physician to predict the probable course of the
disease and propose an individualized treatment,
and to give the patient and his or her family
accurate information;
(ii) to compare the results of clinical trials to ascertain
whether groups of patients share the same
prognosis;
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
(iii) to design clinical trials in homogeneous
subgroups of patients.
A good PI must fulfill several qualities.
It must be accurate
(i) It must include patients with the same lymphoma
subtype. This is easily achieved in individual
entities like Hodgkin’s disease. For some entities,
like mantle cell lymphoma, it may require detailed
pathology review and additional
immunohistochemical and cytogenetic studies.
(ii) Patients should ideally represent the whole
spectrum of the lymphoma subtype.
(iii) In subtypes where treatment has a significant
influence on the course of the disease, only
patients optimally treated should be included.
It must be simple
(i) Although molecular biology and/or sophisticated
cytogenetic studies may improve the accuracy of
PIs, these tests are cumbersome, costly, and
available only in limited centers. Using such tests
will limit the usefulness of an index, since widely
used prognostic indices in lymphoma must rely
on routinely performed tests.
(ii) Some tests can require a consensus before
incorporating them in a PI, such as the cell grade
in follicular lymphomas, and BCL-2 positivity in
large B-cell non-Hodgkin lymphomas (NHLs).
(iii) The number of parameters included in an index
must be small enough (i.e. between three and five)
to be easily memorized and to separate the
patients into a limited number (say between three
11
 Chapter 2: Prognostic factors for lymphomas
and five) risk groups. However, the widely
expanding use of “pocket computers” and smart
phones will permit us to use more complex indices
and/or algorithmic programs, e.g. the MIPI
(Mantle Cell Lymphoma International Prognostic
Index).
It must ideally rely on overall survival (OS)
OS is the most convenient endpoint. PIs have been
developed from retrospective analyses of large groups
of patients. In these retrospective studies, determining
the time to progression (TTP) or time to treatment
failure (TTF) may lead to uncertainties and is a poten-
tial source of bias.
However, OS cannot be used in analyses of patients
with some lymphoma subtypes, as in Hodgkin’s dis-
ease or indolent lymphomas where the OS is very
good, events are rare and an analysis would require
many years of follow-up. It is now widely accepted –
although not definitively proved – that significant
improvement in progression-free survival (PFS) is a
good surrogate of an improvement in OS that can be
used in the evaluation of the influence of a new treat-
ment in different prognostic subgroups.
It must be discriminant
To be useful to clinicians, a PI must separate patients
into risk groups with significantly different OS or PFS
when compared by log-rank test and hazard ratio for
either of these endpoints.
Furthermore, except in some common subtypes
such as large cell NHL or Hodgkin’s disease, there
must be an even distribution between risk groups to
allow assessment of a new treatment in an adequate
number of patients. In developing the PI, it may also
be useful in some frequently occurring subtypes to
identify a small group of patients, either with a very
good or a very poor prognosis.
It must be validated
The first step in building a PI is to collect the data from a
large number of patients, spanning the entire spectrum
of the disease. Then, these patients are randomly sepa-
rated into a test group used for building the index and a
second group for internal validation. This method has
been used to create the International Prognostic Index
(IPI) for aggressive NHLs and the Follicular Lymphoma
12 International Prognostic Index (FLIPI). These indices
must then be validated by other groups and/or centers
in other patients who may differ by age, distribution,
and treatment modalities. This external validation is
mandatory before widespread adoption of the PI. A
univariate analysis is the first step. Parameters here are
selected from the literature and will include demo-
graphic data, clinical and biological markers of tumor
burden, and the effect of the lymphoma upon the host.
This univariate analysis may, however, generate bias:
when creating the FLIPI, it appeared that including the
cell subtype of follicular lymphoma would require a
pathology review since the distribution was obviously
center-dependent. Another example is the absence of
inclusion in all indices of comorbidities, since the pres-
ence and the severity of a concomitant disease will
influence prognosis. This is especially true in the eld-
erly. New scores for evaluating these comorbidities and
their severity have been proposed and should be inte-
grated in new clinical PIs.
There is no consensus on the optimal method for
including continuous variables in univariate analyses.
They may be included as a continuous variable –
which is the most accurate but the most complex
option – or dichotomized according to a threshold
change in prognosis beyond a certain point (for
instance, the threshold of 60 years was chosen for the
IPI since it was considered a maximal age for autolo-
gous bone marrow transplantation). The expansion of
computer software in daily use by physicians in their
day-to-day practice will allow more and more use of
continuous variables in PIs. Parameters that signifi-
cantly influence the OS of patients in the univariate
analysis should then be included in the multivariate
analysis. However, if the number of patients is very
high, a selection of parameters has to be made accord-
ing to statistical (P value in the univariate analysis) and
clinical considerations. Parameters that significantly
influence OS in the multivariate analysis are chosen
for building the PI. Detailed statistical considerations
are beyond the scope of this chapter.
Diffuse large B-cell lymphoma
(DLBCL)
Prognostic factors at diagnosis
Clinical factors
Until the end of the 1980s, age and staging according to
the Ann Arbor classification were the most frequent
parameters used to identify “high-risk” or “low-risk”
patients with aggressive NHL. The Ann Arbor

classification was originally developed for Hodgkin’s
disease, which commonly spreads via contiguous
lymph node groups. Because the patterns of spread are
somewhat different between NHL and HL, it is therefore
not surprising that the Ann Arbor classification was less
efficient in identifying prognostic subgroups of patients
with aggressive NHL. Age at diagnosis was also com-
monly identified as a prognostic factor in numerous
studies. However, most publications that included large
numbers of older patients, demonstrated a tendency to
treat elderly patients with lower doses of chemotherapy
to reduce treatment-related toxicity, while others sug-
gested that older patients could benefit from full-dose
therapy. In this context, many investigators attempted to
identify pre-treatment prognostic factors from their own
series of patients with aggressive NHL. Features inde-
pendently associated with outcome were often used to
develop prognostic-factor models in which the individ-
ual patient’s risk correlated to the number of adverse
factors present at diagnosis. Although the clinical fea-
tures used in these models differed, they were thought to
reflect three basic features:
(1) the tumor’s growth and invasive potential: lactate
dehydrogenase (LDH), Ann Arbor stage, tumor
size, bone marrow (BM) involvement, and number
of nodal and extranodal sites;
(2) the patient’s reaction to the tumor: performance
status (PS), B symptoms;
(3) the patient’s likely ability to tolerate intensive
therapy: PS, BM involvement, age.
The models were similarly predictive for outcome in
large series of patients uniformly treated, suggesting
that, although they relied on different parameters, they
identified the same patient subpopulations.
In 1993, institutions and cooperative groups from the
United States, Canada, and Europe participated in the
International Non-Hodgkin’s Lymphoma Prognostic
Factors Project to attempt to develop a predictive
model (the IPI) based on a large cohort of patients with
aggressive NHL. This model was designed to allow com-
parisons of published clinical trials and to identify
patients at high risk for relapse after anthracycline-
based chemotherapy. From a study of 2031 patients
treated with an anthracycline-based chemotherapy
between 1982 and 1987, clinical features independently
predictive for OS and relapse were age, PS, LDH levels,
Ann Arbor stage, and the number of extranodal sites.
These five factors were incorporated into a model
(Table 2.1) and an individual patient’s relative risk of
Chapter 2: Prognostic factors for lymphomas
Table 2.1 Prognostic factors for survival in International
Index Patients.
Relative P
risk value
(A) Patients of all ages
Age (≤60 years versus >60 years) 1.96 <0.001
LDH (≤normal versus >normal) 1.85 <0.001
Performance status (0,1 versus 1.80 <0.001
2–4) 1.47 <0.001
Ann Arbor stage (I/II versus III/IV) 1.48 <0.001
Extranodal involvement (≤1 site
versus >1 site)
(B) Patients ≤60 years of age
Ann Arbor stage (I/II versus III/IV) 2.17 <0.001
LDH (≤normal versus >normal 1.95 <0.001
Performance status (0, 1 versus 1.81 <0.001
2–4)
death was determined by adding the number of adverse
prognostic factors present at diagnosis. It was therefore
possible to identify four groups that had significantly
different predicted 5-year OS: 73% (low-risk group
with either no or one adverse factor); 51% (low–
intermediate risk group with two adverse factors); 43%
(high–intermediate risk group with three adverse
factors); and 26% (high-risk group with four or five
adverse factors). Since age was an independent predictive
factor for outcome and patients younger than 60 years
were often candidates for intensive treatment, an age-
adjusted model was also developed. From 1274 patients
aged 60 years or younger, three adverse factors were
identified: tumor stage, serum concentration of LDH,
and PS. By adding the number of adverse prognostic
factors, this age-adjusted IPI (aaIPI) identified four risk
groups with a predicted 5-year OS of 83% (low-risk
group with no adverse factor), 69% (low–intermediate
risk group with one adverse factor), 46% (high–inter-
mediate risk group with two adverse factors), and 32%
(high-risk group with three adverse factors).
When clinical characteristics of patients above
and below 60 years were compared, it was seen that
equal percentages of younger and older patients pre-
sented with elevated LDH level, advanced stage, poor
performance status PS (2–4), or had multiple extra-
nodal sites. These demonstrated that the worst out-
come of older patients was not because of more
extensive disease. The comparison of complete
remission (CR) and relapse-free survival (RFS) rates
between older and younger patients (Table 2.2) also
indicated that they had similar CR rates but lower 13
 Chapter 2: Prognostic factors for lymphomas

Table 2.2 The International and Age-Adjusted Index.

Risk group Risk factors Distribution CR rate (%) RFS of CR (%) Survival (%)
of cases (%)
2-year 5-year 2-year 5-year
rate rate rate rate
(A) International Index
(patients of all ages)
Low (L) 0,1 35 87 79 70 84 73
Low–intermediate (LI) 2 27 67 66 50 66 51
High–intermediate (HI) 3 22 55 59 49 54 43
High (H) 4,5 16 44 58 40 34 26
(B) Age-adjusted Index
(patients ≤60 years)
Low (L) 0 22 92 88 86 90 83
Low–intermediate (LI) 1 32 78 74 66 79 69
High–intermediate (HI) 2 32 57 62 53 59 46
High (H) 3 14 46 61 58 37 32
(B) Age-adjusted Index
(patients >60 years)
Low (L) 0 18 91 75 46 80 56
Low–intermediate (LI) 1 31 71 64 45 68 44
High–intermediate (HI) 2 35 56 60 41 48 37
High (H) 3 16 36 47 37 31 21

Table 2.3 The Revised International Prognostic Index.

Revised Number Percentage 4-year 4-year

IPI of IPI factors of patients PFS (%) OS (%)
Very 0 10 94 94
good
Good 1,2 45 80 79
Poor 3,4,5 45 53 55

RFS rates that translated into significant age-related
differences in survival. When survival of patients
with different Ann Arbor stages was compared
according to risk groups defined by the IPI, it was
possible to determine in each different stage signifi-
cantly different outcomes, indicating that the IPI was
more predictive than the Ann Arbor classification.
The IPI was derived from a cohort of patients with
aggressive lymphoma, including both T-cell lym-
phoma and different subtypes of aggressive B lym-
phoma. In addition, none of the patients received the
current gold standard treatment of rituximab and
CHOP (cyclophosphamide, doxorubicin, vincristine,
and prednisone)-like chemotherapy. Subsequently, the
predictive value of IPI was reassessed in a retrospective
study in 365 patients with DLBCL who received six to
14 eight cycles of R-CHOP. The authors found that IPI
remained predictive but they identified only two risk
groups: low and low–intermediate groups with 4-year
OS of 82% and 81%, respectively, and high and high–
intermediate groups with 4-year OS of 49% and 59%,
respectively. They proposed to redistribute these pre-
dictive factors into a Revised International Prognostic
Index (R-IPI) and identified three distinct prognostic
groups with a very good (no adverse prognostic factor,
4-year PFS: 94%, OS: 94%), good (1 or 2 adverse
factors, 4-year PFS: 80%, OS: 79%), and poor (3 or
more adverse factors; 4-year PFS: 53%, OS: 55%) out-
come, respectively (Table 2.3).
The IPI or R-IPI are now used to design therapeu-
tic trials and to select and compare treatment strat-
egies, but these clinical variables clearly represent
surrogates for the intrinsic cellular and molecular het-
erogeneity within aggressive lymphoma and highlight
the need for patient-specific and biologically based risk
factors.

Table 2.4 Immunohistochemical markers having prognostic value in diffuse large B-cell NHL.
Immunohistochemical markers
Lineage-associated and immune antigens
CD5
HLA molecules
CD54
CD86
Proliferation and apoptosis markers
Ki67
MIB 1
P53/P21
Cyclin D3
rb
BCL-2
Survivin
Adhesion molecules
CD44
Stage-specific markers
CD10/BCL-6
Histological factors
The updated World Health Organization (WHO)
classification in 2008 defined DLBCL, not otherwise
specified (NOS), as a proliferation of B-lymphocytes
which diffusely efface the normal architecture of a
node or extranodal site. Three common variants have
been recognized: centroblastic, immunoblastic and
anaplastic. Immunoblast-rich tumors have been
found to have a worse prognosis in several studies
and in some studies associated with frequent involve-
ment of the CNS and BM. WHO classification recog-
nizes four DLBCL subtypes, including T-cell/
histiocyte-rich large B-cell lymphoma, primary
DLBCL of the CNS, primary cutaneous DLBCL
(leg-type) and EBV-positive DLBCL of the elderly.
Until recently, T-cell/histiocyte-rich large B-cell lym-
phomas were considered to have a worse prognosis.
However, most of these studies were retrospective,
included small numbers of patients, and these results
were not confirmed by others. The more recently
defined EBV-positive DLBCL of the elderly and pri-
mary cutaneous DLBCL, leg-type, share a similar
poor prognosis.
DLBCL usually express CD45 (leukocyte com-
mon antigen) and pan-B-cell antigens such as
CD19, CD20, and CD79. They can also express
other markers detectable by immunohistochemistry,
including lineage-associated and immune markers,
Chapter 2: Prognostic factors for lymphomas
Prognostic value
Expression associated with extranodal presentation and shorter survival
Defective expression associated with poor prognosis
Expression associated with shorter relapse-free survival
Expression associated with shorter relapse-free survival
High expression associated with shorter survival
High expression associated with shorter survival
P53+/P21− phenotype associated with poor treatment response
Expression associated with shorter survival
High expression associated with better survival
High expression associated with shorter survival
Expression associated with shorter survival
Expression associated with shorter survival in localized nodal disease
CD10+/BCL-6+ phenotype associated with a better survival
cell adhesion molecules, proliferation and apoptosis
markers. Some of these markers have been consid-
ered to have a prognostic impact and are summarized
in Table 2.4. BCL-2 protein expression is found in
30–60% of cases and the consequence of either
t(14;18)(q32;q21) or BCL-2 amplification leads to an
increase of BCL-2 production. In contrast to the lack
of prognostic significance of the presence of t(14;18)
(q32;q21) in DLBCL, numerous studies have demon-
strated an association between BCL-2 expression and
decreased disease-free survival (DFS) or OS.
However, there was considerable variation in criteria
used by these authors to classify BCL-2-positive cases
(ranging from 10% to more than 50% positive cells).
Moreover, in cohorts of patients treated with chemo-
therapy and rituximab, BCL-2 expression did not
correlate with survival, emphasizing the current
ambiguous prognostic value of BCL-2 expression.
CD5 found primarily in T-cells is also expressed by
a subset of normal B-cells. CD5+ DLBCLs are reported
to be associated with an older age, are predominantly
found in women, with frequent involvement of extra-
nodal sites (especially BM and spleen), and inferior
survival compared to CD5– DLBCL.
By immunostaining of CD10, BCL-6, and IRF4/
MUM1, different groups have proposed immunophe-
notypic subdivision of DLBCL into germinal center-like
(GCB) and non-germinal center-like (non-GCB)
15
 Chapter 2: Prognostic factors for lymphomas
subgroups. Cases with CD10 expression by 30% of cells
are regarded as GC type as well as cases that are CD10–,
BCL-6+, and IRF4/MUM1–. All other cases are regarded
as of non-GC type. However, this immunophenotyping
division does not correlate well with gene expression-
based subgrouping of DLBCL (see Gene expression
profiling, below).
Genetic abnormalities
Several recurrent chromosomal abnormalities have
been described in DLBCL, including those involving
BCL-6 (3q27), BCL-2 (18q21), P53 (17p), and REL
(2p14). None of them are pathognomonic of DLBCL.
BCL-6 is located in 3q27 and is involved in chromo-
somal translocations or mutations leading to deregu-
lation of BCL-6 expression. The chromosomal
translocations involving 3q27 are the most frequent
genetic abnormalities and are detected in 30–40% of
DLBCL. Contrary to the predictive value of increased
BCL-6 mRNA and BCL-6 protein expression, the clin-
ical importance of such translocations has been con-
troversial and most studies have failed to demonstrate
a significant effect on survival. Translocation t(14;18)
(q32;q21) is found in up to 30% of DLBCL, and most
early studies have not shown a prognostic impact of
this translocation in de novo DLBCL.
Mutations and deletion of P53 are reported in up to
20% of DLBCL and this has been associated with a
more aggressive clinical course. However, p53 protein
expression is imperfectly correlated with P53 mutation
and it has been suggested that the analysis of p53
with its downstream target p21 may identify a sub-
group of DLBCL with a particularly poor outcome
(p53+/p21–). MYC rearrangement was observed in up
to 10% of cases of DLBCL. The MYC break partner is
IGH (60%) or non-IGH (40%) genes. In some, MYC
rearrangements are associated with a concurrent BCL-
2–IGH translocation and/or BCL-6 breakpoint or both.
These cases are usually characterized by a high prolif-
eration index and have a very poor prognosis.
Gene expression profiling
Studies using gene expression profiling have identified
different patterns of gene expression allowing the sep-
aration of DLBCL subtypes deriving from distinct
stages of lymphocyte ontogeny. The first group,
named GC-like DLBCL, expresses genes characteristic
of normal B-cells in the germinal center, whereas non-
16 GC-like DLBCL consists of cells with the character-
istics of activated B-cells (ABC). The ABC-like tumors
expressed genes found in in vitro activated peripheral
blood B-cells as well as some genes expressed normally
by plasma cells, suggesting their post-GC origin.
Initially, a “type 3” group was defined, including a
collection of DLBCL which cannot be classified as the
GC and ABC groups, but this does not represent a
distinct group. This stratification related to stages of
B-lymphocyte ontogeny has been found to be correlated
with OS independently of IPI, GC-like DLBCL having
significantly better OS rates when compared to those
with ABC-like DLBCL. However, the microarrays used
in these studies were different and there was no overlap
between the genes identified in these studies. The clin-
ical utility of such technologies is also limited by the cost
and the requirement for fresh or optimally cryopre-
served tumor. A simplified 6-gene predictive model
using quantitative real-time polymerase chain reaction
was subsequently developed, demonstrating that the
expression of three genes (LMO2, BCL-6, FN1) corre-
lated with prolonged survival, whereas the expression of
another set of three genes (BCL-2, CCND2, SCYA3) was
associated with a shorter survival. The predictive value
of the gene expression profile was later confirmed using
tissue microarray (TMA), suggesting the potential of
this technology to replace gene expression profiling in
clinical practice. Gene expression profiling was origi-
nally obtained with tumor samples from patients trea-
ted without rituximab. The predictive value of gene
expression profiling has been subsequently demonstra-
ted in a more recent cohort of patients receiving
R-CHOP.
Treatment-related prognostic factors
Although parameters associated directly with response
to treatment cannot be used to define induction ther-
apy, they provide information regarding the chemo-
sensitivity of the tumor. The recent introduction
of positron emission tomography (PET) using
18
F-fluorodeoxyglucose (18F-FDG) has improved our
ability to evaluate the prognostic relevance of
response.
Time to complete remission
The survival of patients with DLBCL is closely related
to achieving CR at the end of treatment, whereas stable
or progressive disease (refractory disease) are both
associated with poor survival. The time required to
obtain CR has also been identified as an important
prognostic factor of OS, with patients achieving CR
early having the best survival. However, evaluation of

Table 2.5 Predictive value of whole-body PET–18F-FDG for treatment evaluation in NHL.
Clinical Sensitivity Specificity Positive
situations predictive value
Early response 79% 92% 90%
Before 84% 83% 84%
transplantation
Post treatment 67% 100% 100%
response by clinical examination and computer
tomography scans is often limited to evaluating
patients with bulky or disseminated tumors who fre-
quently (30–60%) have residual abnormalities of
uncertain significance.
Positron emission tomography (PET) using
18
F-fluorodeoxyglucose (18FDG)
PET–18F-FDG is now considered as the non-invasive
imaging technique of choice for the detection of resid-
ual disease after treatment. The positive predictive
value of PET–18F-FDG for relapse is close to 100%
(Table 2.5) and persistent PET–18F-FDG uptake after
first line therapy is associated with a high risk of
progression. The follow-up time and the small size of
the studies that included both DLBCL and Hodgkin’s
lymphoma are confounding factors, and the prognos-
tic impact of PET–18F-FDG at the end of treatment
needs to be confirmed in future studies. Despite these
limitations, PET–18F-FDG has now been incorporated
into the revised response criteria for DLBCL.
The prognostic value of 18F-FDG uptake has also
been evaluated before autologous stem cell transplanta-
tion (ASCT), and positive PET–18F-FDG scan was also
correlated with PFS and OS. PET–18F-FDG before trans-
plantation seems to be an even stronger prognostic fac-
tor than IPI and its prognostic significance is higher
before rather than after transplantation. Early prediction
of response to therapy could also distinguish those
patients who might benefit from standard therapy
from those for whom intensive strategies may be
more beneficial. A rapid decrease of 18F-FDG uptake
by tumoral tissue has been observed as early as 7 days
after the first administration of chemotherapy in patients
with NHL, and it might therefore be possible to separate
early patients with or without residual 18F-FDG uptake.
Nevertheless, optimal timing to assess response remains
unknown. Some authors found that the predictive value
of PET–18F-FDG was independent of IPI, and
PET–18F-FDG findings obtained after the first cycle
Chapter 2: Prognostic factors for lymphomas
Negative
predictive value
81%
83%
83%
correlated better with PFS than those observed after
completion of chemotherapy. Results of prospective
studies evaluating the impact of interim PET–18F-FDG
in patient management are awaited.
Dose intensity and schedule
Until the advent of rituximab, the CHOP regimen
(cyclophosphamide, doxorubicin, vincristine, and pred-
nisone) was considered as the gold standard for the
treatment of DLBCL. Early phase II studies have high-
lighted the potential role of dose intensity and schedule
of treatment on survival rates. These second and third
generation regimens failed, however, to demonstrate an
impact on survival. Several recent randomized trials have
focused again on increasing frequency and intensity of
CHOP. They demonstrated that CHOP given every 2
weeks (CHOP-14) or an intensified regimen (ACVBP)
could improve OS rates. More intensive approaches
using high-dose chemotherapy followed by ASCT
have also been tested in first line therapy. Results are
often conflicting, but could suggest that intensive early
therapy might benefit age-adjusted IPI high–inter-
mediate (HI)-risk patients. The benefit of adding rit-
uximab to CHOP on survival has been demonstrated
in both elderly patients with advanced DLBCL and in
young low-risk patients (defined by IPI ≤1). The
impact of intensified chemotherapy regimens in the
era of rituximab (R-CHOP-14) has been studied in
three prospective trials; two have failed to demonstrate
any survival advantage compared to R-CHOP-21. The
role of high-dose chemotherapy with rituximab as
part of initial therapy is currently under investigation.
Prognostic factors at relapse
Clinical and treatment-related prognostic factors
Initial therapy of DLBCL with anthracycline-based
chemotherapy cures approximately 40–50% of
patients and therefore 50–60% of patients will either 17
have disease refractory to initial therapy or will relapse
 Chapter 2: Prognostic factors for lymphomas
after a complete response. For these patients, high-
dose chemotherapy followed by ASCT is the most
successful approach, obtaining 40–45% 5-year event-
free survival (EFS) as demonstrated in the PARMA
trial which compared salvage regimen (DHAP) alone
or followed by ASCT. Early studies of high-dose che-
motherapy have demonstrated that chemosensitivity
to initial second line therapy was an overwhelming
predictor of outcome. Other additional adverse factors
identified in many of these studies include a remission
duration shorter than 12 months, an elevated lactate
dehydrogenase (LDH), bulk disease greater than
10 cm, or three or more chemotherapy regimens
before ASCT. The IPI has also been evaluated at the
initiation of second line therapy as a predictor of out-
come. In a retrospective analysis of the PARMA trial,
aaIPI was highly predictive of both response to DHAP
and OS for the entire cohort of patients. For chemo-
sensitive patients randomized to autologous stem cell
transplantation, aaIPI failed, however, to be predictive
of OS. Furthermore, no difference in PFS or OS was
found between the two arms for patients with an aaIPI
score of 0.
Gene expression profiling
There are very few studies including gene expression
profiling at the time of relapse and its predictive value
is under investigation. The importance of phenotype
assessed by TMA in predicting survival in relapse or
refractory DLBCL treated with ASCT has recently
been reported. No difference in survival was observed
for patients with either the GC phenotype or the non-
GC phenotype.
Follicular lymphomas
Follicular lymphomas (FLs) account for 25–30%
of all NHLs. Until 20 years ago, treatment of FLs
was palliative and did not significantly modify the
natural history of the disease. Since the development
of immunotherapies such as interferon alpha, anti-
CD20 monoclonal antibodies, and combinations
of these with chemotherapy, overall survival has
improved. However, these treatments may have
immediate and/or long-term toxicity, may have
untoward effects on quality of life, and have variable
costs.
It has thus become essential for clinicians to have
detailed information on the prognosis of the disease at
18 diagnosis in order to select treatment with the optimal
efficacy/toxicity/cost ratios.
Prognostic factors at diagnosis
Clinical factors
Several retrospective analyses have looked for prog-
nostic factors in FLs. Those that have been reported
as having a significant influence on prognosis are
listed in Table 2.6. The IPI proposed for aggressive
lymphomas has also been tested in FL, where sev-
eral studies have shown that the IPI can separate
patients into groups with significantly different
prognoses.
However, there are many drawbacks in using the
IPI in FLs: (1) the statistical methodology is inad-
equate since some factors that were not significant in
aggressive lymphomas might influence the prognosis
of FL; (2) the IPI is poorly discriminant in FL, with less
than 15% of patients in the high–intermediate risk and
high-risk groups.
This prompted a group of specialists in 1998 to
initiate a collection of data from approximately 5000
patients with FL. After univariate and multivariate
analyses, the Follicular Lymphoma International
Prognostic Index (FLIPI) was devised. It relies on
five parameters (Table 2.7): age, serum LDH, Ann
Arbor stage, hemoglobin level, and number of
nodal sites. From these, three risk groups were
established.
The characteristics of these groups, distribution of
patients in the test group, 5-year and 10-year survivals,
and relative risks of deaths are seen in Table 2.8.
The FLIPI has been validated by analysis of other
series and has been designed with patients treated
with conventional chemotherapy before the era of
anti-CD20 monoclonal antibodies (MoAb). OS can-
not be used for validation in patients treated with
anti-CD20 MoAbs because of the scarcity of events.
The usefulness of the FLIPI has thus been tested on
PFS and has been validated on different series of
patients treated with cyclophosphamide, vincristine,
and prednisone (CVP) followed by rituximab, CVP
combined with rituximab, CHOP combined with
rituximab, or treatment with radiolabeled anti-
CD20 MoAb.
Since the FLIPI did not include some parameters
that were not measured routinely at the time of
the analysis (tumor bulk, serum β2 microglobulin
level) and relied on the analysis of patients initially
treated without anti-CD20 MoAbs, a new PI has
recently been proposed. The FLIPI2 uses PFS as
the endpoint and prospectively included a larger
 Chapter 2: Prognostic factors for lymphomas

Table 2.6 Adverse prognostic factors in follicular lymphomas (from retrospective analyses of the literature).

Characteristics
Demographic Advanced age
Male gender
Pathological Follicular diffuse architecture
Cell type grade 3
Cytogenetic Absence of t(14;18)
Additional cytogenetic abnormalities
6q deletion, 17p, 1p abnormalities
p53 gene mutations
Molecular Overexpression of macrophage and dendritic cell genes
Overexpression of p53
Tumor mass Ann Arbor stage III–IV
Number of nodal sites involved
Massive marrow infiltration
Presence of bulky tumors
Increased serum LDH level
Increased serum β2 microglobulin level
Increased serum CA125
Consequences upon the host Poor performance status
B symptom(s)
Anemia
Hypoalbuminemia

Table 2.7 Prognostic factors for survival in Follicular

Lymphoma International Prognostic Index (FLIPI).
Relative
risk
Age (≤60 years versus >60 years) 2.38
Ann Arbor stage (I/II versus III/IV) 2.00
Hemoglobin level (<120 g/L 1.55
versus ≥120 g/L)
LDH (≤normal versus >normal) 1.50
Number of nodal sites (≤4 sites 1.39
versus >4 sites)
number of prognostic parameters, with the final
index relying on five (Table 2.9): age, serum β2
microglobulin level, tumor bulk, Ann Arbor stage,
and BM involvement. From this, three risk groups
with significantly different PFS have been established
(Table 2.10).
Histological factors
In the 2008 WHO classification of lymphomas, FLs
are separated into three grades according to the cell
type: grades 1 and 2 are small cell and grade 3 is large
cell FL. Grade 3 FLs account for approximately 10%
of FLs and have been separated into two entities:
grade 3a FL (neoplastic follicles with more than 25
large cells per high-power field on a background of
P small cells) and grade 3b (neoplastic follicles in asso-
value ciation with sheets of diffuse large cell infiltration).
<0.001 Grade 3b FLs are also distinctive cytogenetically with
<0.001 a lower incidence of t(14;18) and absence of the
concomitant rearrangement of BCL-2 and BCL-6.
<0.001
Grade 3b FLs are now considered and treated as
DLBCL.
<0.001
0.001
Genetic abnormalities
The t(14;18)(q32;q21) translocation is the hallmark
of FLs, leading to overexpression of the BCL-2 gene.
However, FLs without the t(14;18) probably have the
same prognosis. Most FL cells carry additional chro-
mosome abnormalities, and cell grade and survival
have been demonstrated to be correlated with the
number of abnormalities. Some additional chromo-
some abnormalities have an influence on prognosis,
some being associated with poor (6q deletion,17p, 1p
abnormalities) or good (trisomy 12, +7, +8) long-
term survival.
Some abnormalities have been associated with
histological transformation of FL into high-grade
NHL such as mutation(s) of the P53 gene, overexpres-
sion of C-MYC, and inactivation of the p15 and p16 19
 Chapter 2: Prognostic factors for lymphomas

Table 2.8 Outcome and relative risk of death according to risk group as defined by the FLIPI.

Risk Number Distribution 5-year 10-year Relative

group of factors of patients (%) OS (%) OS (%) risk
Low 0–1 36 90.6 70.7 1.0
Intermediate 2 37 77.6 50.9 2.3
High ≥3 27 52.5 35.5 2.3

Table 2.9 Adverse prognostic factors for progression-free survival in the FLIPI2.

Serum β2 microglobulin > upper limit of normal

Longest diameter of the largest involved node >6 cm
Bone marrow involvement
Hemoglobin level <120 g/L
Age >60 years

Table 2.10 Outcome and relative risk of progression according to risk group as defined by the FLIPI2.

Risk Number Distribution 5-year HR (95% CI)

group of factors of patients (%) PFS (%) (SE)
Low 0 20 79.5 (5.0) 1.0
Intermediate 1–2 53 51.2 (5.7) 3.19 (2.0–5.15)
High 3–5 27 19.8 (13) 5.76 (3.53–9.4)
Adapted from M. Federico et al. J Clin Oncol, 2009; 27: 4555–4562.
SE, Standard error; CI, confidence intervals.

cyclin-dependent kinase inhibitors located on chro-
mosome 9p21. However, these abnormalities have
not been reported consistently and their relative con-
tribution remains unknown.
Gene expression profiling
Studies have now been carried out on the gene
profile of 191 patients with FL. Two signatures were
distinguished, one associated with overexpression of
T-lymphocyte genes and the other with macrophage
and/or dendritic cell genes. The former signature was
associated with a good prognosis while the latter was
associated with a poor prognosis, and survival differ-
ences were highly significant and independent of clin-
ical factors.
The role of the microenvironment in FLs has been
extensively studied and many cells play a role:
20 – macrophages in the follicular and interfollicular
areas have a negative influence on prognosis only in
patients treated with conventional chemotherapy
but not in patients treated with rituximab-based
regimens;
– a high number of CD4-positive T-cells in the
lymphomatous follicles has a negative influence
on outcome. However, subpopulations of
CD4-positive T-cells have opposite effects. FOX
P3-positive Tregs may inhibit FL cell
proliferation and thereby positively influence
prognosis. Programmed cell death 1-positive CD4
T-cells could also have a positive impact on FL
outcome.
However, these results should be cautiously inter-
preted since discrepant results between studies have
been reported.
Treatment-related prognostic factors
Several factors may allow the prediction of response
to treatment or duration of response. The significance

of persistence after treatment of t(14;18)-positive cells
in the blood and/or bone marrow is controversial.
It does not seem to influence the risk of relapse
after radiation therapy for localized disease, while it
was associated with a short PFS after autologous
bone marrow transplantation. Some studies suggest a
negative influence for these cells, others an absence
of prognostic value. According to a few gene profile
analyses, some signatures are associated with an
improved response to CHOP or to rituximab.
Prognostic factors at relapse
Histological transformation into DLBCL is by far the
most important prognostic factor at relapse, and jus-
tifies nodal biopsy whenever possible. Few studies have
tested the prognostic factors at relapse. The FLIPI was
tested and showed different OS according to risk
group. Sensitivity to initial chemotherapy and delay
between first and second episodes also materially
affects prognosis.
Mantle cell lymphomas
Prognostic factors at diagnosis
Clinical factors
Among NHL subtypes, mantle cell lymphomas (MCL)
have the poorest long-term survival, for several
reasons:
(1) The median age is around 65 and few patients can
be treated with intensive therapies.
(2) Although response rates are high after initial
therapy, complete response (CR) rates are low and
most patients have residual disease.
(3) Even in CR patients, the relapse rate is high after a
median duration of response of around 18 months.
(4) After several relapses, most patients have blastic
transformation, which is highly resistant to all
treatments.
(5) Anti-CD20 MoAbs are less active than in other
B-cell NHLs.
Because of the low incidence of the disease and its
heterogeneity in presentation, there has been no con-
sensus on a clinical prognostic index in MCL patients.
Several retrospective analyses have reported a number
of adverse prognostic factors. These are referred to in
Table 2.11. The IPI designed for aggressive NHLs has
also been applied to patients with MCL. IPI is poorly
Chapter 2: Prognostic factors for lymphomas
discriminant since most patients with MCL have
adverse prognostic factors according to the IPI (age
>60, Ann Arbor stage IV disease, more than one extra-
nodal involvement site) and are in the high–inter-
mediate and high-risk groups.
In 2008, The German Low Grade Lymphoma
Study Group and the European Mantle Cell
Network specifically designed a new prognostic
index for overall survival of patients with MCL: the
Mantle Cell Lymphoma International Prognostic
Index (MIPI). Based on the data of 455 patients
with advanced stage MCL, the simplified MIPI is
based on four parameters: age, Eastern Cooperative
Oncology Group (ECOG) performance status, serum
LDH level, and white blood cell count (Table 2.12).
Weighting each parameter with 0 to 3 points allowed
the identification of patients in three risk group (low
risk: 0–3 points; intermediate risk: 4–5 points; high
risk: >5 points) with significantly different OS. This
index has been validated in a few retrospective anal-
yses, especially in patients treated with high-dose
therapy.
Some authors have advocated the addition of Ki 67
expression in addition to the MIPI since high Ki 67
expression is a poor prognostic factor. However, con-
cordance between pathologists in evaluating Ki 67
expression is poor.
Histological factors
MCL initially infiltrates the mantle zone surrounding
normal germinal centers, generating a pattern called
“nodular” MCL. Progressively, the germinal centers
are destroyed by the neoplastic infiltrate, and the
MCL becomes of the “diffuse” type. In both nodular
and diffuse types, the largest neoplastic compartment
is made of small- to medium-size lymphoid cells,
with few large cells. There is no demonstration of a
difference in prognosis between the nodular and dif-
fuse types. MCL is sometimes characterized by
slightly larger cells with a high mitotic index and is
therefore termed “blastoid” variant. This latter var-
iant is associated with a more aggressive clinical
behavior.
Genetic abnormalities
The genetic hallmark of MCL is the chromosomal
translocation t(11;14) (q13;q32) that can be detected
in virtually all cases of MCL. This translocation leads
to overexpression of cyclin D1, which plays a key role 21
 Chapter 2: Prognostic factors for lymphomas

Table 2.11 Clinical prognostic factors in mantle cell lymphomas from retrospective
analyses.

Reference No. of patients Adverse prognostic factors

R. Oinonen et al. 94 Age >60 years
Ann Arbor stage III–IV
Leukemic disease
LDH >normal
H. Samaha et al. 121 Age >70 years
ECOG performance status ≥2
Hemoglobin <120 g/L
Leukemic involvement
D. Weisenburger et al. 68 Age >60 years
Bone marrow involvement
Ann Arbor stage III–IV
B symptoms
Performance status ≥2
High or intermediate IPI risk group
NS Andersen et al. 105 Age >65 years
Hemoglobin <120 g/L
Splenomegaly
E Zucca et al. 65 Performance status ≥ 2,
Increased LDH level,
Increased β2 microglobulin
Age >65 years
High IPI risk group
LH Argatoff et al. 80 Performance status ≥2
AJ Norton et al. 66 Age >70 years
Ann Arbor stage IV
Splenomegaly
Low sodium
Low albumin

Table 2.12 Simplified Mantle Cell Lymphoma Prognostic Index.

Points Age, y ECOG LDH level (ULN) WBC, 109/L

2 <50 0–1 0.67 <6.7
1 50–59 – 0.67–0.99 6.7–9.99
2 60–69 2–4 1.0–1.49 10–12.99
3 ≥70 – ≥1.5 ≥15.0
Abbrevations: ULN, upper limit of normal; WBC, white blood cell.
For each prognostic factor, 0–3 points are given to each patient and points are summed up
to a maximum of 11.

in the control of the G1 phase of the cell cycle. An
elevated level of cyclin D1 expression in MCL cells
accelerates G1/S phase transition and thus tumor cell
proliferation. The level of cyclin D1 expression – as
measured by mRNA translation – is directly correlated
with tumor cell proliferation rate, as measured by Ki
67 expression, and clinical aggressiveness.
Cyclin D1 is a key regulator of the cell cycle, and
22 elevated levels of cyclin D1 in MCL cells promote
additional cytogenetic alterations which may contribute
to proliferation and poor prognosis. Other abnormalities
of the cell cycle regulatory pathway that have been
reported in MCL are associated with an aggressive
behavior:
– deletions of the cyclin D kinase (CDK) inhibitor
gene (CDK p16INK4a) on chromosome 9;
– gene amplification of CDK4;

– decreased expression of P27KIP1 which is an
inhibitor of cyclin D/CDK complexes;
– inactivation of P53, mainly observed in blastoid
MCL types with a high proliferation index.
These genetic alterations share the fact that they all
predominantly target cell cycle regulation and lead to
an uncontrolled proliferation of MCL lymphoma
cells. This has been confirmed by gene profile studies,
which have demonstrated that a “proliferation sig-
nature,” i.e. an overexpression of proliferation-
associated genes, was a strong predictor of poor
survival. This “proliferation signature” probably
reflects the combination of the genetic abnormalities
mentioned above and was found to be associated
with overexpression of genes involved in drug resist-
ance. In conclusion, all MCLs have a poor prognosis
with conventional treatments. As cell cycle distur-
bances are the main factor in the development and
clinical course of MCL, abnormalities of genes and
proteins of the cell cycle will most probably provide
the most useful prognostic information. This contri-
bution will be further improved when the biological
tools can be integrated with new therapeutic
approaches.
Peripheral T-cell lymphomas
Prognostic factors at diagnosis
Clinical factors
The importance of T phenotype has been recognized
by the Revised European American Lymphoma
(REAL) and the subsequent updated WHO classifica-
tion. These classifications have also allowed us to
appreciate fully the wide heterogeneity of peripheral
T-cell lymphoma (PTCL), which accounts for 10–15%
of all NHLs in western countries. The WHO classifi-
cation divides them into predominantly leukemic,
nodal, and extranodal types. PTCL not otherwise
specified (PTCL-NOS) represents the most common
T-NHL in western countries, accounting for 60–70%
of PTCLs. Because of this heterogeneity and the com-
paratively recent histopathological description, retro-
spective analyses of PTCL prognostic factors have to
be analyzed with caution. For example, earlier reports
based on older classifications failed to find clinical
impact of phenotype on outcome, whereas several
studies using the updated classification indicated that
T-phenotype has a negative impact on survival. The
Chapter 2: Prognostic factors for lymphomas
prognostic value of IPI has also been assessed for
PTCL. Most of these studies included patients with
nodal PTCL and found that IPI had predictive value.
Another model has been developed in a large study
and included some IPI factors (age, LDH, PS) in
addition to BM involvement. PTCL-NOS patients
commonly present with unfavorable characteristics,
including B symptoms, elevated LDH, bulky tumor,
poor PS, and extranodal disease. The majority of
patients (50%) thus fall into the unfavorable IPI cat-
egory. IPI failed to correlate with outcome in patients
with angioimmunoblastic T-cell lymphoma (AIL) in a
large retrospective study, whereas male sex, mediasti-
nal lymphadenopathy, and anemia adversely affected
overall survival. IPI has also been evaluated in primary
nasal natural killer cell lymphoma and was found to be
correlated with survival. Application of IPI in the
remaining subtypes has not been evaluated because
of their scarcity, such as hepatosplenic γδ T-cell lym-
phomas that present almost exclusively with high-risk
IPI scores, and it is unlikely that IPI has any clinical
utility in this setting.
Histological factors
Most PTCLs are predominantly nodal NHLs that
include AIL, PTCL-NOS, and systemic anaplastic
large-cell lymphomas (ALCL). Systemic ALCL has a
superior survival compared to other PTCL, whereas
the prognostic impact of PTCL-NOS compared to AIL
remains controversial, with 5-year overall survival
around 30–35%. It has also been demonstrated that
systemic ALCL were heterogeneous according to
expression of ALK protein. ALK-negative cases often
present in elderly patients, with an advanced stage,
elevated serum LDH levels, B-symptoms, and extra-
nodal involvement, and have a prognosis similar to
that of other PTCL, whereas ALK-positive cases have a
5-year survival close to 90%.
Treatment-related prognostic factors
Because of the low incidence and the wide heterogeneity
of PTCL, the evaluation of treatment regimens has been
limited to non-randomized and retrospective studies. In
addition, most of them included different subtypes of
PTCL known to have different outcomes. As in DLBCL,
reaching a CR at the end of treatment probably
improves the clinical outcome. However, optimal treat-
ment is not defined here and it is therefore difficult to
identify prognostic factors related to treatment. 23
 Chapter 2: Prognostic factors for lymphomas
Marginal zone lymphomas
Marginal zone B-cell lymphomas (MZLs) encompass
three subtypes:
1. extranodal MZLs or mucosa-associated lymphoid
tissue (MALT) lymphomas;
2. splenic MZLs with or without villous lymphoid
cells in the blood and/or bone marrow;
3. nodal MZLs with or without “monocytoid” type
lymphoid cells in the blood and/or bone marrow.
All MZLs have a normal counterpart in B-cells of the
“marginal zone” of the secondary follicles which is
adjacent to the mantle zone. Because of the rarity of
this cell type, data on prognostic factors originate from
retrospective analyses of small series with heterogene-
ous treatments. They therefore have limited value in
guiding treatment choices.
MALT lymphomas
The main prognostic factor of MALT lymphomas
is the demonstration of an infectious agent –
Helicobacter pylori in gastric, Chlamydophila psittaci
in adnexal ocular, Borrelia burgdorferi in skin MALT
lymphomas – and its eradication by the appropriate
antibiotic treatment. Such eradication may allow the
cure of the MALT lymphoma. Overall, the prognosis
of gastrointestinal MALT lymphoma is better than
that of other extranodal MALT lymphomas (median
time to progression of 8.9 years versus 2.9 years,
P < 0.01). The presence of a large-cell component
(so-called MALT lymphoma in transformation) is a
rare event but is associated with a poorer prognosis.
Other clinical factors have been shown to adversely
affect the clinical outcome of patients with MALT
lymphomas in retrospective analyses. These factors
include: disseminated disease in some series, poor
performance status, increased serum LDH levels,
bulky tumor, increased serum β2 microglobulin
level, and decreased serum albumin level. A t(11;18)
(q21;q21) translocation has been described in 30% of
gastric MALT lymphomas. It fuses the amino termi-
nal of the API2 gene to the carboxy terminal of the
MALT1 gene and generates a fusion product that
activates Nuclear Factor kappa B (NFκB). This trans-
location is associated with the absence of concomi-
tant H. pylori infection or resistance to H. pylori
eradication, and to therapy with an alkylating agent
24 but not to resistance to rituximab.
Other MZL lymphomas
Studies of prognostic factors in splenic or nodal MZLs
are scarce. The IPI could not be used in these patients
since very few patients are in high-risk groups. A high
lymphocyte count, increased serum LDH and/or β2
microglobulin levels have also been described as adverse
prognostic factors. In 2006, a prognostic index for
splenic MZL was proposed. Three parameters had a
prognostic influence on cause-specific survival: hemo-
globin <120 g/L, serum LDH level above normal level,
and albuminemia <35 g/L. Five-year OS with no
adverse factors was 83%, 72% with one adverse factor,
and 56% for two or more adverse factors.
Burkitt’s lymphoma
Prognostic factors at diagnosis
Whereas Burkitt’s lymphoma represents 30–50% of
pediatric lymphoma, it accounts for only a small per-
centage of adult lymphomas. The evaluation of treat-
ment regimens in adults is also limited to retrospective
and non-randomized studies. Most knowledge in prog-
nostic factors for Burkitt’s lymphoma comes therefore
from large prospective randomized pediatric studies
which have demonstrated very high rates of survival.
Clinical factors
Age is often considered an adverse prognostic factor for
Burkitt’s lymphoma. In studies including children and
adults treated with the same regimen, this was not
confirmed. In other studies using more intensive che-
motherapy, age below 10 years or 15 years has been
demonstrated to be associated with favorable outcome.
When these pediatric regimens were tested in an adult
cohort, inferior results were observed compared to
results obtained in a pediatric population, especially
for advanced stage disease. These results could be in
part because of an increased toxicity of such intensified
conventional chemotherapies, and the exact impact of
age on survival remains debatable
Patients with Burkitt’s lymphoma more frequently
present with extranodal involvement compared to
DLBCL, and extension of such lymphomas is usually
assessed using Murphy’s classification. Advanced stages
III and IV were initially found to be associated with a
decreased survival in both pediatric and adult patients
whereas the use of an adapted and intensified conven-
tional chemotherapy tends to erase the influence of
stage on survival. Serum LDH level as a surrogate of

Table 2.13 Therapeutic groups of Burkitt’s lymphoma according to the Société Française d’Oncologie Pédiatrique (SFOP)
and the Berlin–Frankfurt–Münster (BFM) Group.
SFOP
Low-risk group
Stage I
Abdominal stage II
Intermediate-risk group
Unresected stage I
Non-abdominal stage II
Any stage III or IV or L3ALL
(<70% blasts)
High-risk group
CNS involvement
L3ALL (>70% blasts)
cell proliferation and a tumour mass ≥10 cm have also
been found to have a poorer prognosis. CNS and BM
involvement are recognized to be the most significant
adverse prognostic factors and justify an intensified
approach. Using these prognostic factors, different
therapeutic groups have been defined, allowing the
development of strategies based upon individual risk
(Table 2.13).
Treatment-related prognostic factors
Survival of patients with Burkitt’s lymphoma requires
patients to enter CR. Stable or progressive disease
(refractory disease) is almost always rapidly fatal.
Some other treatment-related prognostic factors have
been demonstrated, especially in children. First, the
absence of complete response after three multiagent
chemotherapy courses correlated with an adverse out-
come, and few patients who achieved a partial
response at this time could be salvaged with high-
dose chemotherapy with ASCT. Second, patients
whose tumors did not respond to the prephase regi-
men (including cyclophosphamide, vincristine, and
prednisone) ultimately failed. The importance of
early intensification of treatment has also been
shown, with significantly lower EFS if the second
course of chemotherapy (COPADM2) started after
day 21. It has also been established that relapse does
not occur after 1 year in CR, and patients without early
relapse or progression can be considered cured. At
relapse, it has been demonstrated that the ability to
cure patients with high-dose chemotherapy followed
by ASCT was closely related to the chemosensitivity of
Chapter 2: Prognostic factors for lymphomas
BFM
Initial complete resection of lymphoma manifestations
No or incomplete resection of lymphoma manifestations with at least one of
the following criteria:
– only extra-abdominal sites
– abdominal site and LDH <500 UI/L
No or incomplete resection of abdominal lymphoma and LDH ≥500 UI/L
BM involvement and/or CNS disease and/or multifocal bone involvment
relapse. Burkitt’s lymphoma refractory to a second
line of chemotherapy has a low probability of cure,
whatever the treatment.
Primary central nervous system
lymphoma
Prognostic factors at diagnosis
Clinical factors
Primary central nervous system lymphomas (PCNSL)
represent 2% of NHL, making it difficult to identify
robust clinical prognostic factors. A primary brain
localization is an adverse prognostic factor compared
to localized non-CNSL. Immunodeficiency, whether
related to inherited disorders, to HIV, or to organ
transplantation, increases the risk of PCNSL. Even if
these patients are not usually included in prospective
trials, they also seem to have a bad prognosis compared
to immunocompetent PCNSL. Age and PS are the only
two universally accepted prognostic factors for PCNSL.
It has been suggested that they are factors influencing
therapeutic choice rather than independent survival
indicators, but critical reviews of published data and a
more recent large study have confirmed that they are
independent variables. The prognostic value of IPI has
been tested in some retrospective studies. The IPI is not
accurate for PCNSL, which are localized and frequently
associated with a poor PS.
It is also likely that other specific prognostic factors
influence survival. The influence of lymphoma local-
ization or number of tumors and of CSF protein level 25
 Chapter 2: Prognostic factors for lymphomas
Table 2.14 Prognostic factors for survival in PCNSL.
Age (≤60 years versus >60 years)
LDH (≤normal versus >normal)
Performance status (0, 1 versus 2–4)
CSP protein level (normal versus elevated)
Deep lesions (yes versus no)
has therefore been retrospectively evaluated with
conflicting results. More recently, predictors of
response and survival were analyzed in an interna-
tional multicenter retrospective series of 378 immu-
nocompetent patients with PCNSL in an attempt to
develop a prognostic index. More than 90% of patients
were treated with radiotherapy alone or with com-
bined radio-chemotherapy. In the multivariate analy-
sis, five factors were identified to have an impact on
survival (Table 2.12): age, PS, LDH serum level, pro-
tein CSF concentration, and involvement of the deep
structures of the brain. Each variable was assigned a
value of either 0 if favorable, or 1 if unfavorable, and
final score was obtained by the addition of the values of
these five variables. The number of adverse features
was significantly correlated to survival, and prognostic
value of this index was also found in patients treated
with high-dose methotrexate (HD-MTX)-based che-
motherapy, with or without radiotherapy.
Histological factors
Over 90% of B-PCNSL are DLBCL, Burkitt’s, and low-
grade NHL, each representing 5%. The T phenotype is
found in less than 5% of PCNSL. The phenotype and
histological subtypes do not seem to influence survival
in PCNSL. BCL-6 and BCL-2 protein expression has
been found to have a prognostic influence in DLBCL. In
a recent study including 83 patients with PCNSL, GCB
and ABC phenotypes were assessed by TMA; most
patients express an ABC phenotype and have survival
identical to DLBCL expressing the same phenotype.
Treatment-related prognostic factors
An important difference from NHL located outside
the CNS is that, in the case of PCNSL, obtaining a
CR does not seem to have a positive prognostic
impact. In a retrospective study, patients with CR
26 and PR did not have statistically different 5-year sur-
vivals. However, such results highlighted the difficulty
Relative risk P value
1.02 <0.001
1.41 0.008
1.64 <0.001
1.71 0.003
1.45 <0.001
of accurate evaluation of these tumors by CT scan and
MRI. The impact of PET–18F-FDG in PCNSL evalua-
tion has not yet been validated.
PCNSL treatment influences the outcome of
patients. Prospective randomized studies failed to
demonstrate an impact of choice of treatment on sur-
vival but retrospective analyses have shown that radio-
therapy or chemotherapy alone give lower survival
rates compared to combined chemo-radiotherapy.
HD-MTX-based chemotherapy has been found to
have a significant impact on survival in several studies
using multivariate analysis. Concomitant intrathecal
chemotherapy did not show any impact..
Hodgkin’s lymphoma
Prognostic factors at diagnosis
Clinical factors
There are numerous studies analyzing the impact of
clinical characteristics at diagnosis on outcome in
patients with Hodgkin’s lymphoma (HL). However,
most of them included few patients, were heterogeneous
and retrospective, and the prognostic factor models
developed should be considered carefully before they
are used as a basis for deciding on treatment for an
individual patient with HL. Moreover, prognostic factors
are closely related to the treatment used and, with the
improvement in outcome for patients with limited-stage
HL, these models are no longer clinically relevant. In
advanced stage HL, prognostic factors can also be helpful
in identifying patients at high risk of relapse and devel-
oping new therapeutic strategies. All prognostic factors
described are related to tumor spread, tumor burden,
patient characteristics, or interaction of tumor and host.
Related to tumor spread and tumor burden
The Ann Arbor classification describes the anatomic
spread of tumor cells via contiguous lymph node
groups and has been demonstrated to be of prognostic

relevance for disease-free survival (DFS) and OS in
many studies testing different therapeutic regimens
in HL. It is therefore possible to separate “early dis-
ease” (i.e. stages I/II) which represents two-thirds of
patients with an estimated 5-year OS close to 90% and
“advanced disease” (i.e. stages III/IV) which represents
one-third of patients, with 5-year OS of around 70%.
The Ann Arbor classification has been used for more
than 30 years to influence therapeutic strategies, but
clinical studies clearly demonstrated differences of
survival among patients with same stage, demonstrat-
ing the need for additional prognostic factors to adapt
the therapeutic strategy to individual risk.
The total body burden count of tumor is an impor-
tant prognostic factor and the presence of bulk disease,
especially in the mediastinum, is associated with a poor
prognosis. Bulky disease is defined as either a single
mass of tumor tissue exceeding 10 cm in the largest
diameter or a mediastinal mass with a ratio of largest
transverse diameter of the mass to the transverse diam-
eter of the thorax at T5–6 of ≥0.35 on a chest radio-
graph. This definition has been included in the Ann
Arbor classification and is often referred to as the
Cotswold modification. Bulky mediastinum, found in
15–20% of patients with early stage disease, correlated
with a reduced DFS after radiotherapy or chemotherapy
alone as well as after combined treatment, whereas in
advanced-stage HL treated with more aggressive thera-
pies, there are less consistent data on the prognostic
significance of mediastinal bulk. The prognostic value
of bulk disease outside the mediastinum is more uncer-
tain. Some have calculated tumor burden according to
both the number and the size of involved nodes. These
authors demonstrated that, in early stages, tumor bur-
den correlated with both DFS and OS. Other specific
sites of HL, such as BM, spleen, or pleural involvement,
were shown to be of prognostic significance. Their
prognostic value is controversial and seems to have
only a little impact on survival. The number of organs
involved has been correlated with OS and DFS by some
but was not found in larger studies. LDH could be
viewed as a surrogate of tumor burden, and an elevated
level of LDH has been found to have a prognostic value
in HL associated with involvement of marrow and at
least two extranodal sites.
Related to patient characteristics
Some patient characteristics, including age and sex,
are important in prognosis. Age is considered as an
Chapter 2: Prognostic factors for lymphomas
adverse prognostic factor in many malignant diseases,
but increased morbidity and mortality in older
patients from other causes has to be taken into account
and results have to be matched to the general popula-
tion. Secondly, increased comorbidity leads to
increased therapy-related mortality. Nevertheless,
when matching patients to the general population in
multivariate analyses, age has been shown to have an
adverse prognostic impact, especially in advanced
stage disease. Gender also seems to influence outcome
in HL. This could be due in part to the association of
female gender with other good parameters, including
localized stage, lack of B symptoms, or nodular scle-
rosis. In a large cohort of patients, multivariate analy-
sis demonstrated that male gender was independently
associated with a reduced OS.
Related to interaction of tumor and host
B-symptoms (fever, weight loss, and night sweats)
have been known to have clinical relevance for
many years and have been included in the Ann
Arbor classification. They are more frequent in
advanced stage disease compared to early stages and
correlate with other biological parameters, including
increased erythrocyte sedimentation rate (ESR) and
decreased serum albumin or hemoglobin. Despite
this, their independent prognostic value has not
been found in multivariate analysis, but the presence
of B-symptoms is used by most cooperative groups to
define prognostic groups and inform therapeutic
strategies. An elevated ESR was shown to have inde-
pendent prognostic significance for DFS and OS for
different stages and several regimens, especially in
early stage disease. The European Organization for
Research and Treatment of Cancer (EORTC) lym-
phoma group defined a group of patients with either
ESR more than 50 without B-symptoms or ESR more
than 30 with B-symptoms who have a worse outcome
and an increased rate of relapse. Anemia has also
been reported to be associated with a decreased sur-
vival in both early and advanced stages. In large
retrospective analyses, lymphopenia lower than
0.6 × 109/L was independently associated with
decreased survival. Many other biological para-
meters easily available in peripheral blood, such as
ferritin, C-reactive protein, β2-microglobulin, ceru-
loplasmin, or thymidine kinase, have been described
to influence outcome in HD, but their significance fell
to insignificant levels when multiparametric analysis
was used. 27
 Chapter 2: Prognostic factors for lymphomas
Prognostic models
From studies conducted during the 1980s a number
of prognostic models have been developed
(Table 2.15). Most of these studies included few
patients with both early and advanced stages, patients
were treated with different approaches, and these
prognostic models have not been validated in other
groups of patients. Furthermore, they failed to iden-
tify a subgroup of advanced HL patients with a
very high risk of relapse when applied in a cohort
of patients treated homogeneously. In 1998,
Hasenclever et al. used the international database on
HL to develop a parametric model for predicting
survival. This model was based on data from 5023
patients who were at various stages of the disease and
who received varying treatments. Six prognostic fac-
tors (sex, age, hemoglobin level, white blood cell and
lymphocyte counts and serum albumin) were found
to influence freedom from progression (FFP) in a
Cox model (Table 2.16). These factors were therefore
incorporated into a model and the individual
patient’s relative risk for relapse and death was deter-
mined by adding the number of adverse prognostic
factors present at diagnosis. It was therefore possible
Table 2.15 Prognostic factor models developed in HL.
Authors Gobbi et al.
Population analyzed
n = 586 stage I–IV
Prognostic factors used in the model
Clinical Age
Sex
Stage
Histology
B-symptoms
Bulky mediastinal
Biological
ESR
Hemoglobin
Albumin
28
to identify six groups of patients (score, 0, 1, 2, 3,
4, ≥5) who had FFP of disease at 5 years ranging
from 84% (for a score 0) to 42% (for score ≥5). This
International Prognosis Score (IPS) was also predic-
tive for OS. However, only 13% of patients included
in this analysis were limited stage and the IPS
failed to identify the lower risk groups reliably.
Furthermore, a score of 4 or more was still associated
with a 47% FFP and 59% OS at 5 years, demonstrat-
ing the limits of IPS in identifying very high-risk
patients.
Analysis of prognostic factors at diagnosis enables
patients to be assigned to different therapeutic
approaches, allowing trial groups to define therapeutic
strategies based upon individual risk. In Europe, the
German Hodgkin Lymphoma Study Group (GHSG)
stratifies patients according to stage and four prognos-
tic factors (Table 2.17), defining three therapeutic
groups (early, intermediate, and advanced stages).
The European Organization for Research and
Treatment of Cancer group (EORTC) defines localized
subdiaphragmatic stage with favorable (no adverse
factor) or unfavorable (one or more adverse factors)
outcome, whereas IPS is used to identify advanced HL
with high risk of relapse (Table 2.18). Most of these
Proctor et al. Strauss et al.
n = 93 stage I–IV n = 185 stage II–IV
Age Age
Stage
Bulky mediastinal Bulky mediastinal
Inguinal nodes
Hemoglobin
Lymphocyte count
LDH
Marrow involvement
 Chapter 2: Prognostic factors for lymphomas

Table 2.16 Prognostic factors for freedom from progression in International Index Patients.

Factors Relative risk P value

Serum albumin <40 g/L 1.49 <0.001
Hemoglobin <105 g/L 1.35 0.006
Male sex 1.35 0.001
Stage IV 1.26 0.011
Age ≥45 years 1.39 0.001
White cell count ≥15 g/L 1.41 0.001
Lymphocyte count <0.6 g/L or <8% of white cell count 1.38 0.002

Table 2.17 Prognostic and therapeutic groups defined by the German Hodgkin’s Lymphoma Study Group.

Stage
Risk factors IA, IB, IIA IIB III, IV
None Early stages
≥3 lymph node areas
High ESR* Early unfavorable stages
Extranodal involvement
Massive mediastinal tumor Advanced stages
* Erythrocyte sedimentation rate ≥50 without or ≥30 with B-symptoms.

Table 2.18 Prognostic and therapeutic groups for subdiaphragmatic stages I–II defined by the European
Organization for Research and Treatment of Cancer Group.

Stage
Risk factors IA, IB IIA, IIB
None Favorable stages
≥4 lymph node areas
High ESR
Unfavorable stages
Age ≥50 years
Massive mediastinal tumor
* Erythrocyte sedimentation rate ≥50 without or ≥30 with B-symptoms.

prognostic scores were established from patients
treated before the last decade, and recent data have
demonstrated that they lose predictive power with
improved treatment, emphasizing the need for new
prognostic factors.
Histological factors
The prognostic value of the classification established
in 1966 by Lukes and Butler has been analyzed in
many studies. They demonstrated that lymphocyte-
depleted subtype was associated with a worse out-
come. The number of patients suffering from this
subtype is very low, and histologic classification of
HL has been reconsidered in the REAL classification,
separating lymphocyte-predominant HL from classi-
cal HL. Grading according to the number of Hodgkin
Reed–Sternberg (HRS) cells or the characteristics of
the background infiltrate may predict prognosis in
some settings, but is not required for routine clinical
purposes. 29
 Chapter 2: Prognostic factors for lymphomas
Some markers, such as Ki 67, P53, Rb, BCL-2
or LMP-1, expressed by HRS have been analyzed and
correlated with outcome. Such studies have yielded
conflicting results; they were retrospective and
included too small a number of patients to support
any definitive conclusion. Some groups have focused
on non-malignant cells and demonstrated a negative
prognostic impact of CD8 granzyme B-positive T-cells
or eosinophilic infiltrates.
Other biological factors
In the past decade, new biological parameters have
been investigated to predict prognosis. CD30 is con-
sistently expressed by HRS cells and soluble CD30 was
found to have independent prognostic significance in
different studies. Soluble CD30 seemed to correlate
with tumor burden and lymphocyte-depleted HL.
VCAM-1 and ICAM-1 are adhesion molecules and
soluble forms could have prognostic significance.
Several cytokines, such as IL-2, IL-6, IL-7, IL-8, IL-
10, IL-12, IL-13, and TNFα, secreted by both HRS cells
and lymphoid cells surrounding HRS cells, have also
been evaluated but their prognostic value remains to
be defined with any certainty.
Treatment-related prognostic factors
Dose intensity and time to complete remission
Survival of patients with HL is closely related
with achieving CR at the end of treatment, and
stable or progressive disease (refractory disease)
are both associated with inferior survival. In a
retrospective analysis, the dose delivered in the
first three cycles of chemotherapy was the main
factor predicting CR and outcome, demonstrating
the importance of dose intensity in HL treatment.
Moreover, early CR, usually evaluated after three or
four cycles, was also shown to be predictive of a
better outcome.
Positron emission tomography (PET) using
18
F-fluorodeoxyglucose (18FDG)
Most studies demonstrating PET–18F-FDG prognostic
value for post-treatment evaluation included both
DLBCL and HL, and in this situation negative predictive
value seemed to be higher in HL than in HD. Early
prognostic value of PET–18F-FDG has also been dem-
onstrated in early response assessment and correlated
30 independently with PFS and OS, especially for
advanced-stage patients. Prospective studies evaluating
the impact of interim PET–18F-FDG in patient manage-
ment are now awaited. However, because ASCT is not
usually used in front line therapy, the predictive value of
PET–18F-FDG before high-dose chemotherapy and
ASCT in HL has not been systematically evaluated.
Prognostic factors at relapse
Many studies have analyzed prognostic factors of
patients with refractory or relapsed HL. The length
of remission after first-line chemotherapy has been
shown to be an important prognostic factor. Using
conventional salvage therapy, virtually no patient
with primary progressive disease (10% of patients)
survives after 8 years, contrasting with a projected
20-year survival of 11% and 22%, respectively, for
patients relapsing early within 12 months of CR
(15% of patients) and for patients relapsing more
than 12 months after CR (15% of patients).
Further reading
Arcaini L, Lazzarino M, Colombo N, et al. Splenic marginal
zone lymphoma: a prognostic model for clinical use.
Blood, 2006;107(12):4643–4649.
Cheson BD, Pfistner B, Juweid ME, et al. Revised response
criteria for malignant lymphoma. J Clin Oncol,
2007;25(5):579–586.
Dave SS, Wright G, Tan B, et al. Prediction of survival in
follicular lymphoma based on molecular features of
tumor-infiltrating immune cells. N Engl J Med,
2004;351:2159–2169.
Federico M, Bellei M, Marcheselli L, et al. Follicular
lymphoma international prognostic index 2: a new
prognostic index for follicular lymphoma developed by the
international follicular lymphoma prognostic factor
project. J Clin Oncol, 2009;27:4555–4562.
Hasenclever D, Diehl V. A prognostic score for advanced
Hodgkin’s disease. International Prognostic Factors
Project on Advanced Hodgkin’s disease. N Engl J Med,
1998;339:1506–1514.
Hoster E, Dreyling M, Klapper W, et al. A new prognostic
index (MIPI) for patients with advanced stage mantle cell
lymphoma. Blood, 2008;111:558–565.
Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in
large-B-cell lymphomas. N Engl J Med,
2008;359:2313–2323.
Patte C, Auperin A, Michon J, et al. The Societe Francaise
d’Oncologie Pediatrique LMB89 protocol: highly
effective multiagent chemotherapy tailored to the tumor
burden and initial response in 561 unselected children
with B-cell lymphomas and L3 leukemia. Blood,
2001;97:3370–3379.

Philip T, Biron P, Maraninchi D, et al. Role of massive
chemotherapy and autologous bone-marrow
transplantation in non-Hodgkin’s malignant lymphoma.
Lancet, 1984;i:391.
Rosenwald A, Wright G, Chan WC, et al. The molecular use
of profiling to predict survival after chemotherapy for
diffuse large-B cell lymphoma. N Engl J Med,
2002;346:1937–1947.
Sehn LH, Berry B, Chhanabhai B, et al. The Revised
International Prognosis Index (R-IPI) is a better
predictor factor of outcome than the standard IPI for
patients with diffuse large B-cell lymphoma treated with
R-CHOP. Blood, 2007;109:1857–1861.
Solal-Celigny P, Roy P, Colombat P, et al. Follicular
lymphoma international prognostic index. Blood,
2004;104:1258–1265.
Spaepen K, Stroobants S, Dupont P, et al. Prognostic value of
positron emission tomography (PET) with fluorine-18
fluorodeoxyglucose ([18F]FDG) after first-line
Chapter 2: Prognostic factors for lymphomas
chemotherapy in non-Hodgkin’s lymphoma: is [18F]
FDG-PET a valid alternative to conventional diagnostic
methods? J Clin Oncol, 2001;19:414–419.
Spaepen K, Stroobants S, Dupont P, et al. Prognostic value of
pretransplantation positron emission tomography using
fluorine 18-fluorodeoxyglucose in patients with
aggressive lymphoma treated with high-dose
chemotherapy and stem cell transplantation. Blood,
2003;102:53–59.
Swerdlow SH, Campo E, Harris NL (eds). WHO
Classification of Tumours of Hematopietic and Lymphoid
Tissues. Lyon: IARC press, 2008.
The International Non-Hodgkin’s Lymphoma Prognostic
Factors Project. A predictive model for aggressive non-
Hodgkin’s lymphoma. N Engl J Med, 1993;329:987–992.
Thieblemont C, Bastion Y, Berger F, et al. Mucosa-associated
lymphoid tissue gastrointestinal and nongastrointestinal
lymphoma behavior: analysis of 108 patients. J Clin
Oncol, 15:1624–1630.
31
 Chapter

3
Imaging
Heok K. Cheow and Ashley S. Shaw

Introduction Staging techniques

The radiologist plays a key role in the management CT has been the mainstay of lymphoma staging in
of patients with lymphoma at several stages, from the neck, chest, abdomen, and pelvis for the past two
initial diagnosis through to the evaluation of decades and continues to form the basis of clinical prac-
response to treatment. The possible diagnosis of tice and the majority of clinical trials. Developments in
lymphoma is often raised first by the imaging CT technology enable detailed three-dimensional images
department. The alert radiologist who identifies an of the whole body to be obtained within a few seconds.
abnormal mediastinum on a chest radiograph may With the addition of intravenous contrast medium,
consider lymphoma in their diagnosis, and this lesions can be identified within, and adjacent to, the
may prompt further investigations. With the solid organs and their nature discerned from their
increasing advent of computed tomography (CT) enhancement patterns (Figures 3.1 and 3.2). However,
for all thoracic and abdominal problems, the imag- despite these advances, the evaluation of lymph nodes
ing findings may again point to a very strong like- with CT is almost exclusively based on size; lymph
lihood of lymphoma being the responsible cause. A nodes measuring 1 cm or greater in their short axis are
referral from a general practitioner about a patient
with a neck lump may lead to an ultrasound exami-
nation. Frequently, tissue for histopathological
examination is obtained by the radiologist using
image-guided techniques, either at ultrasound (US)
or CT. It is important when performing these pro-
cedures that the radiologist endeavors to obtain
good core biopsies with sufficient tissue for full
histopathological evaluation. Fine needle aspiration
is often inadequate for accurate diagnosis and
should be avoided.
Accurate delineation of the extent of disease is
essential both at initial presentation and follow-up
as this is used to guide management decisions. As
lymphoma may involve almost any tissue in the
body, a variety of imaging techniques may be
required to demonstrate it, both anatomically and
functionally, depending upon the tumor biology and
location. Moreover, as treatment regimens develop,
Figure 3.1 Non-enhancing soft tissue mass replacing the right
it is important that the radiologist is able to recog- kidney with para-aortic nodes. Renal biopsy confirmed diffuse large
nize complications of therapy. B-cell lymphoma.

32 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
 Chapter 3: Imaging

Figure 3.2 A rind of soft tissue is seen in the pararenal fat

surrounding the right kidney and invading the cortex, with Figure 3.3 Paravertebral soft tissue mass in the lower thorax
differential enhancement of the functioning kidney and displacing the aorta anteriorly, the so-called “floating aorta.”
lymphomatous tissue.

considered abnormal, whereas those less than 1 cm are

deemed normal. There are a few sites in the body where
nodes smaller than this may be considered suspicious.
These include retrocrural, left gastric, and obturator
regions. Likewise, multiple small nodes coalescing to
form a larger complex are also suspicious for involve-
ment. It has long been recognized that a significant
proportion of enlarged lymph nodes are hyperplastic
and, conversely, that many small nodes are infiltrated
with malignant cells. There are, however, several patterns
of disease seen on CT that are considered characteristic
of lymphoma. A paraspinal mass in direct continuity
with enlarged para-aortic lymph nodes will often prove
to be lymphoma. So too will the appearance of the aorta
“floating” away from the vertebral body within a mass of
a para-aortic nodes (Figure 3.3). Marked mesenteric
nodal enlargement, in the form of a “hamburger” or
“sandwich” (Figure 3.4), encompassing the superior
mesenteric arterial arcade is highly likely to be due to Figure 3.4 Lymph node mass seen around the aorta and
non-Hodgkin’s lymphoma (NHL). The appearance of inferior vena cava, extending into the small bowel mesentery.
The nodes surround the vascular arcade to give the “hamburger”
lymph nodes in other sites also makes a lymphomatous or “sandwich” sign.
process highly likely (internal mammary nodal enlarge-
ment, splenic hilar enlargement, and pericardiophrenic
nodes). Sometimes the root of the mesentery appears tomography (PET)–CT has been shown to be useful in
somewhat “misty” without a frank mesenteric nodal this situation.
enlargement (Figure 3.5) – the differential diagnosis is Given the heterogeneity of the lymphomas, the
wide and includes all possible causes of abnormal fat on paucity of studies with histopathological correlation at
CT. However, the usual cause is either mesenteric pan- multiple nodal sites, and the variable reporting of
niculitis (which some hold to be the harbinger of future results either by individual nodes, disease sites, or over- 33
malignancy) or NHL. Recently positron emission all staging, the literature is frequently confusing and
 Chapter 3: Imaging
Figure 3.5 Ill-defined increase in opacification of the small bowel
mesentery (misty mesentery) in a patient with follicular lymphoma.
Figure 3.7 Marked thickening of the small bowel loops within the
pelvis resulting from small bowel lymphoma.
contradictory. Using a node-by-node analysis with sur-
gical correlation, one study found CT to have a sensi-
tivity of only 20%, correctly staging only 42% of patients
with Hodgkin’s lymphoma (HL). Conversely, other
workers have reported CT to have 92% sensitivity in
staging HL and 75% sensitivity in staging NHL. The
majority of published work lies between these results,
suggesting an accuracy of approximately 70%.
Splenic involvement may be manifest by either
splenomegaly, for which there are many causes, or
the presence of focal lesions (Figure 3.6). Depending
on the size and number of lesions, together with the
technical parameters relating to intravenous enhance-
34 ment, these may easily be overlooked and/or difficult
to characterize with any confidence.
Figure 3.6 Multiple low attenuation lesions throughout the liver
and spleen were demonstrated by CT with small bilateral pleural
effusions. Biopsy demonstrated this to be due to diffuse large B-cell
lymphoma.
Figure 3.8 Lytic lesions are demonstrated throughout the left ilium
with destruction of the cortex and an associated soft tissue mass,
proven to be DLBCL at biopsy.
Extranodal sites of disease can also be difficult to
identify with CT. Gastrointestinal wall thickness is
difficult to estimate, particularly in the non-distended
stomach, and will only be visible when gross
(Figure 3.7). In the stomach, an interrupted rugal
pattern and focal thinning may provide the only clue.
Bone marrow disease will only be evident when a lytic
or sclerotic lesion has developed or when there is
extension into the adjacent soft tissues (Figure 3.8).
Conversely, pulmonary disease is usually readily appa-
rent but can be very difficult to differentiate from
other causes of nodules and consolidation, particularly
infection and drug toxicity (Figures 3.9 and 3.10).

Figure 3.9 Multiple pulmonary nodules bilaterally measuring up to
1 cm in diameter. Biopsy proved these to be DLBCL.
A
Figure 3.11 (A) Sagittal T2-weighted MRI of the thoracic spine
demonstrating an extradural mass posterior to, and compressing,
the spinal cord.
Magnetic resonance imaging (MRI) is primarily
used to image the central nervous system (CNS), soft
tissue masses, and bone marrow. MRI has a greater
sensitivity than CT in identifying CNS lesions and pro-
vides excellent multiplanar images (Figures 3.11 and
3.12), which may help in radiotherapy planning.
Chapter 3: Imaging
Figure 3.10 Consolidation in the right upper lobe with an air
bronchogram demonstrated at CT. Biopsy proved this to be mantle
cell lymphoma.
B
Figure 3.11 (B) Axial images through the lesion demonstrate the
extradural mass extending through the neural foramina bilaterally.
Similarly, imaging focal deposits in the limbs with MRI
can provide additional information as it offers superior
contrast between different soft tissues over CT. MRI of
the bone marrow may be useful in identifying patients
with focal lymphoma deposits, particularly when bone
marrow biopsy is normal. Lymphoma deposits are seen
as low signal on T1 and high signal on fat-saturated
T2-weighted images as the normal fatty marrow is dis-
placed, often with disease spilling out into the extradural 35
or paravertebral spaces. The identification of focal
 Chapter 3: Imaging
A B
Figure 3.12 (A) Axial T1-weighted image following intravenous
gadolinium demonstrates a uniformly enhancing mass within the
pons, extending posteriorly into the right middle cerebellar
peduncle. Image courtesy of Dr J. Cross.
marrow infiltration in patients with normal biopsies is
well described, with some investigators using MRI to
guide a further biopsy. The accuracy of MRI in detecting
bone marrow infiltration is thought to be around 80%,
but there are clearly difficulties in performing histology-
correlated studies. As such, MRI is not routinely used to
assess the marrow in most centers.
Given the limitations of cross-sectional imaging,
the use of functional imaging to identify disease sites is
attractive. Gallium-67 imaging has been used in some
centers; however, it is a time-consuming study, has
poor sensitivity in the liver and spleen, and involves a
high radiation burden. The overall sensitivity for this
study ranges from 31% to 85%. Other tracers, includ-
ing somatostatin receptor imaging with indium-111
labelled octreotide, have been used in specialist cen-
ters, but their use is limited.
PET with 2-[fluorine-18] fluoro-2-deoxy-D-glucose
(FDG), on the other hand, provides functional informa-
tion about the tissues. FDG mimics the action of glu-
cose, being transported into cells and entering the
glycolytic pathway. However, because it is not a sub-
strate for enzyme glucose-6-phosphate isomerase, it
cannot be metabolized and becomes trapped in the
36 cell, thus effectively providing a map of glucose utiliza-
tion. Because of this mode of action, the technique
Figure 3.12 (B) Coronal T1-weighted image following intravenous
gadolinium demonstrates the lesion extending up into the right
midbrain toward the cerebral peduncle. Image courtesy of Dr J. Cross.
requires the patient to be fasted and have normal
blood glucose levels; elevated blood glucose levels stim-
ulate insulin release which drives glucose, and thus
FDG, into the muscles. It has been reported that elevated
plasma glucose may reduce FDG uptake in the tumor,
thus making interpretation difficult. Therefore, fasting
plasma glucose should ideally be <10 mmol/L. Equally,
anxious patients may demonstrate increased uptake
within neck muscles, whilst if patients are cold then
brown fat uptake may be excessive (Figure 3.13).
Steroid administration usually has little effect in onco-
logical PET studies, although surprisingly low uptake of
FDG has been reported in patients with Hodgkin disease
(HD) and diffuse large B-cell lymphoma (DLBCL) who
have received high doses for only 2–3 days prior
to imaging. The intensity of FDG uptake within a
lesion is expressed as the standardized uptake value
(SUV), a semiquantitative measure that broadly repre-
sents the ratio of uptake in the tumor compared to
that which would be expected if the FDG were evenly
distributed throughout the entire body. There have been
a number of studies investigating the use of the SUV in
lymphoma, which demonstrate that an SUV ≤6 often
represents indolent histology, whereas ≥13 usually indi-
cates aggressive histology. However, the significant
degree of overlap in the SUV of aggressive and indolent
histology, coupled with the known intra-patient

Figure 3.13 Coronal PET image demonstrating extensive
FDG uptake in the cervical, supraclavicular, axillary, and
paravertebral regions (arrows). These regions are the typical
distribution of brown fat.
variability of the SUV, means that it has limited value in
grading NHL.
The accuracy of FDG–PET in the initial staging of
high-grade lymphoma has now been established. It has
been consistently shown that the sensitivity, specifi-
city, and accuracy of FDG–PET are superior to CT.
Indeed, all three of these measures are usually reported
to be in the region of 85–98% for both HD and high-
grade NHL. This is reported to result in a change of
stage in anything up to 59% of patients, although other
studies quote 8–40%. In the vast majority of these
cases, the additional information from PET results in
the disease being upstaged. This stage migration will
lead to an apparent improvement in response rates at
each clinical stage. Patients with true “early” stage
disease will be distinguished from those with occult
“advanced” disease, who in turn are likely to have a
better prognosis than those with clinically overt
Chapter 3: Imaging
disease, thus “improving” response and survival rates
in this group too. This new paradigm in lymphoma
staging will need to be taken into account when com-
paring literature from the pre- and post-PET eras. The
clinical impact of PET in initial staging is significant,
with management changed in many of these cases.
Management changes range from modifying radiation
treatment volume to changing from local to systemic
therapy. Clearly, a degree of caution needs to be exer-
cised when there is a discrepancy between CT and PET
findings, which will result in treatment modifications,
as to over-treat or under-treat could lead to significant
morbidity.
The clinical utility of PET imaging in low-grade
lymphoma, however, appears to vary with histolog-
ical subtype and the tissue involved. In patients with
follicular lymphoma, PET is again reported to iden-
tify 40% more abnormal nodal sites than CT.
Conversely, PET identifies less than 60% of the
abnormal nodes on CT in chronic lymphocytic leu-
kemia. The reason behind this discrepancy lies in the
heterogeneity of this disease with variable metabolic
activity. To differentiate tumor from surrounding
normal tissue, the level of FDG uptake has to be
several folds higher than its neighbor. In low-grade
lymphomas, such as mucosa-associated lymphoid
tissue (MALT) lymphoma, the metabolic activity is
relatively low. Identifying such disease is often
fraught with difficulty. On the other hand, FDG–
PET can be used to guide biopsies when high-grade
transformation is suspected. A summary of FDG
uptake seen in the various histological subtypes of
NHL and HL is given in Table 3.1. As 18FDG is not a
tumor-specific tracer, it may give false-positive stud-
ies in a number of conditions, including infection,
drug toxicity, radiotherapy, granulocyte colony-
stimulating factor (G-CSF) therapy, other forms of
malignancy, and physiological activity in normal tis-
sues. Whereas many of these potential pitfalls are
readily identified, others may present diagnostic con-
fusion. In such instances, correlation with anatomical
imaging is often helpful and the current generation of
PET machines is coupled with CT, to produce PET–
CT. This produces two sets of images without moving
the patient, which may be fused to give a complete
structural and functional overview of the whole body.
False-negative PET studies are a reflection of either
tumor biology, lesion size, or a combination of the
two and it is therefore difficult to quantify accurately
the minimum size of lesion that PET will detect. 37
 Chapter 3: Imaging

Table 3.1 FDG uptake in HL and NHL subtypes.

HL
Nodular sclerosis type High
Mixed cellularity type Moderate to high
Lymphocyte-depleted type Moderate to high
Lymphocyte-predominant type Low
B-cell NHL
Diffuse large B-cell lymphoma High
Burkitt lymphoma High
Large cell and anaplastic lymphoma High
Follicular lymphoma (grade 3) Moderate to high
Follicular lymphoma (grade 1–2) Low to moderate
Mantle cell lymphoma Low to moderate
Marginal zone (inc. MALT) lymphoma None to high
Small lymphocytic lymphoma None to low
T-cell lymphoma
Extranodal NK/ T-cell lymphoma High
Peripheral T-cell lymphoma High
Adult T-cell leukemia–lymphoma High
Cutaneous T-cell lymphoma Moderate
Mycosis fungoides and Sézary syndrome Low

However, in high-grade disease it is in the order of a
few millimeters. In this respect, PET–CT may be
particularly useful in evaluating pulmonary disease,
as tiny nodules may be beyond the resolution of PET
alone. As the evidence base grows and PET–CT tech-
nology becomes more widely available, it will become
the imaging modality of choice for staging and man-
aging high-grade lymphoma.
Novel PET tracers are under development and
have been used in early phase clinical studies for
assessing tumor characteristics such as hypoxia, apop-
tosis, and proliferation. One example is 3-deoxy-
3–18F-fluorothymidine (18FLT), a thymidine analog
that demonstrates tumor proliferation. It has been
used in assessing early treatment response in lym-
phoma, but it is still too early to see whether it can
replace 18FDG in lymphoma imaging. Future imaging
may involve an array of tracers tailored to answer
specific clinical questions.
Staging classifications
The Ann Arbor staging classification with Cotswold
revision (Table 3.2) is used to provide the radiologic
staging of HD and most types of NHL. It is important
for the radiologist to understand the impact of staging
on management so that the report is tailored appro-
priately. For example, in HL, early stage asymptomatic
38 supradiaphragmatic disease is currently treated by a
combination of short-course chemotherapy and
involved field radiotherapy, although the use of radia-
tion may alter when results of prospective studies
utilizing the results of FDG–PET become available.
The presence of disease below the diaphragm will
increase the number of planned cycles of chemother-
apy. Additionally, the phrase “bulky lymph node
mass” in a radiological report should be reserved for
those lesions ≥10 cm in diameter, to avoid confusion.
However, in indolent follicular lymphoma, a
watch-and-wait policy may be adopted for many
patients, even with advanced stage disease. There
are a number of radiological indications for chemo-
therapy, including nodal or extranodal mass ≥7 cm in
diameter, ≥3 nodal sites with a diameter ≥3 cm,
splenic enlargement, compression syndrome, pleural
or pericardial effusion, and the radiologist must be
aware of the potential therapeutic implications of the
report.
For childhood NHL and adult patients with
Burkitt’s lymphoma, the St Jude classification
(Table 3.3) is employed. This differs significantly
from the Ann Arbor system in the way that extrano-
dal disease is staged, reflecting the different patterns
of disease and the prognostic implications for these
two groups of patients.
Response assessment
The International Workshop Criteria (IWC), pub-
lished in 1999, have become the widely accepted

Table 3.2 Ann Arbor staging classification and Cotswold
revision.
Stage Area of involvement
I A single lymph node region or a
single lymph localized
involvement of an extralymphatic
site
IE Localized involvement of a single
extralymphatic organ or site
II Two or more lymph node regions
on the same side of the
diaphragm
IIE Localized involvement of a single
extralymphatic organ or site and
of one or more lymph node
regions on the same side of the
diaphragm
III Lymph node regions on both
sides of the diaphragm
IIIE Lymph node regions on both
sides of the diaphragm
accompanied by localized
involvement of an extralymphatic
organ or site
IV Diffuse involvement of one or
more extranodal organs with or
without lymph node involvement
Localized involvement of a single
extralymphatic organ or site with
non-regional lymph node
involvement
Additional qualifiers
A Absence of systemic symptoms
B Presence of systemic symptoms
X Bulky disease, defined as a nodal
mass >10 cm in maximum
diameter or mediastinal mass
>one-third of the internal
diameter of the thorax on a PA
chest radiograph
standard for the assessment of disease response in
NHL. These have enabled comparability of clinical
trials as well as the day-to-day management of
patients. Although based primarily on CT findings,
the IWC take bone marrow biopsy, clinical, and bio-
chemical information into account and are outlined in
Table 3.4. However, there are a number of drawbacks
with the IWC criteria that need to be addressed.
Following treatment for lymphoma, up to 60% of
patients with nodal disease will have a residual mass
on CT. Of these, only a small proportion (up to 20%)
Chapter 3: Imaging
Table 3.3 St Jude staging classification.
I Single tumor (extranodal)
Single anatomic area (nodal)
Excluding mediastinum or abdomen
II Single tumor (extranodal) with regional node
involvement
Two single (extranodal) tumors without regional
lymph node involvement on the same side of the
diaphragm
Two or more nodal areas on the same side of the
diaphragm
Primary gastrointestinal tumor with or without
involvement of associated mesenteric nodes only,
grossly completely resected
III Two single (extranodal) tumors, one on either side of
the diaphragm
Two or more nodal areas, involving above and below
the diaphragm
Primary intrathoracic tumors (mediastinal, pleural,
thymic)
Extensive primary intra-abdominal disease
Paraspinal or epidural tumors irrespective of other sites
IV Any of the above with initial CNS or bone marrow
involvement
Table 3.4 International Workshop Criteria.
Complete response (CR): complete disappearance of all
detectable disease by imaging, with nodes previously >1.5 cm
regressing to <1.5 cm and nodes of 1.0–1.5 cm to <1.0 cm. In
addition, resolution of disease-related symptoms,
normalization of biochemical abnormalities, and normal bone
marrow biopsy.
Complete response unconfirmed (CRu): as for CR, but with a
residual mass >1.5 cm in diameter that has regressed by >75%.
Partial response (PR): at least 50% reduction in the sum of the
product of the greatest diameters (SPD) of the six largest nodes.
There should be neither increase in the size of other nodes nor
any new sites of disease. Hepatic and splenic lesions should also
reduce >50% in the SPD.
Stable disease (SD): where response is less than a PR but there
is not progressive disease.
Progressive disease (PD): more than 50% increase in the
product of the diameters in any previously abnormal node or
the development of new disease sites either during or at the
end of therapy.
Relapsed disease (RD): the appearance of any new disease or
an increase in size of over 50% of residual lesions in patients
who had previously achieved CR or CRu.
will be positive for lymphoma on re-biopsy.
Moreover, if a patient has been imaged too soon
following treatment, there may have been insufficient
time for any significant volume reduction. As
outlined above, whilst CT provides excellent
39
 Chapter 3: Imaging

morphological information, it is impossible to tell

whether this tissue represents residual disease,
inflammation, or fibrosis. Thus, a significant propor-
tion of patients are designated as having either a
partial response (PR) or complete response uncon-
firmed (CRu). Similarly, CT may detect numerous
Figure 3.14 Axial CT, PET, and fused PET/CT images of a patient
lymph nodes measuring <1 cm in diameter, which following treatment for HL demonstrates a large anterior mediastinal
are thus deemed normal and the patient is considered soft tissue mass. This shows low-grade tracer uptake similar to the
to have had a complete response (CR). In the lungs, a adjacent mediastinum (considered PET-negative), which typically
indicates fibrotic scar tissue.
number of changes may occur related to chemother-
apy, radiotherapy, infection, or disease progression.
Bony abnormalities resolve slowly, if at all, whilst
gastrointestinal lesions may not be visible at all and
cannot be reliably measured.
As with initial staging, there is a growing body of
evidence regarding the clinical utility of FDG–PET in
assessing response to therapy. Rapid decreases in the
tumoral uptake of FDG–PET have been demonstra-
ted within 7 days of commencing chemotherapy.
Early response assessment has been studied by a
number of groups, with results of FDG–PET per-
formed after 1–4 cycles of chemotherapy correlated
with clinical follow-up. A recent meta-analysis of
these studies found that interim FDG–PET imaging
in advanced stage HL had an overall sensitivity of
81% and specificity of 97%, with DLBCL slightly
worse at 78% and 87%, respectively. Similarly, these
early and mid-treatment PET studies have been
Figure 3.15 Coronal CT, PET, and fused PET/CT images. Following
shown to be a good predictor of progression-free chemotherapy for DLBCL, there was a slight reduction in volume of
survival and overall survival. In HL, PET-negative the renal mass at CT, but no FDG uptake (arrows) was seen in the
patients have a 5-year progression-free survival of mass, indicating a fibrotic mass (complete metabolic response) in this
patient.
over 90%, compared with 40% in those who are
PET-positive on interim study. However, no pub-
lished study has yet clearly shown that PET can be method remained disease-free at a median 32
used at an early stage to alter treatment with an months’ follow-up. The overall progression-free sur-
improvement in outcome. A number of studies are vival in the groups designated CR by IWC and
currently evaluating a PET response-adapted IWC+PET was 88% versus 91% at 2 years and 74%
approach to determining treatment strategies. versus 80% at 3 years. Crucially, this group contained
The integration of FDG–PET results into IWC is only 17/54 patients using IWC, but 35/54 using
currently being investigated by several groups. In one IWC + PET. These results need to be validated in
retrospective analysis of 54 patients, comparison of the setting of a prospective trial, but it would seem
IWC alone versus IWC including FDG–PET results likely that the IWC criteria will need to be modified to
demonstrated only 61% concordance in response include PET along the lines used by the authors of
classification. The study highlights that the principal this study (Table 3.5) in the near future. Conversely,
advantage of PET in this situation is in patients with a PET imaging may demonstrate residual disease in
residual mass who are classified as either CRu or PR lymph nodes that would have previously been con-
by CT criteria (Figures 3.14 and 3.15). PET enabled sidered normal (Figure 3.16).
reclassification of CRu to either CR or PR, whilst half The timing of this post-treatment study should
of those designated PR were changed to CR. All but ideally be at least 3–4 weeks following completion of
40 one of the ten patients reclassified as CR by this chemotherapy, and 3 months or more following

Table 3.5 IWC + PET description.
CR CR by IWC with a completely negative PET
CRu, PR or SD by IWC with a completely negative PET
and a negative bone marrow biopsy (BMB) if
positive prior to therapy
PD by IWC with a completely negative PET and CT
abnormalities ≥1.5 cm (≥1.0 cm in the lungs) and
negative BMB if positive prior to therapy
CRu CRu by IWC with a completely negative PET but with
an indeterminate BMB
PR CR, CRu, or PR by IWC with a positive PET at the site
of a previously involved node/nodal mass
CR, CRu, PR, or SD by IWC with a positive PET outside
the site of a previously involved node/nodal mass
SD by IWC with a positive PET at the site of a
previously involved node/nodal mass that
regressed to <1.5 cm if previously >1.5 cm, or
<1.0 cm if previously 1.1–1.5 cm
SD SD by IWC with a positive PET at the site of a
previously involved node/nodal mass (i.e. residual
mass)
PD PD by IWC with a positive PET finding corresponding
to the CT abnormality (new lesion, increasing size
of lesion)
PD by IWC with a negative PET and a CT abnormality
(new lesion, increasing size of lesion) of <1.5 cm
(1.0 cm in the lungs)
Figure 3.16 Axial CT and PET images obtained before and after
chemotherapy demonstrate an enlarged right cervical node that has
reduced in size to <1 cm following chemotherapy. By IWC, this
represents complete response (CR). However, the PET study shows
residual high tracer activity seen in this region, indicating active
residual disease.
completion of radiotherapy. The use of PET to eval-
uate residual lesions in indolent disease is less clear and
studies should be interpreted with caution, particu-
larly if PET was not used during initial staging.
Chapter 3: Imaging
Table 3.6 Manifestations of pulmonary drug toxicity in
lymphoma.
Diffuse alveolar Bleomycin, carmustine,
damage cyclophosphamide
Organizing pneumonia Bleomycin, cyclophosphamide
Pulmonary Cyclophosphamide, cytarabine,
hemorrhage amphotericin B
Non-specific interstitial Chlorambucil, carmustine
pneumonia
Complications of therapy
Patients with lymphoma may be treated with an array
of cytotoxic agents and radiation therapy, which can
themselves result in significant morbidity and occa-
sionally death. The lungs appear to be particularly
sensitive to these insults and this can manifest itself
in a number of ways, the more common of which are
listed in Table 3.6.
Diffuse alveolar damage (DAD) is the histopatho-
logical pattern that results from necrosis of type II
pneumocytes and alveolar endothelial cells. The clin-
ical correlate of this is the acute respiratory distress
syndrome (ARDS). In the acute phase (first week after
insult), there is interstitial and alveolar edema and
hyaline membranes. Following this is a reparative
phase, with proliferation of type II pneumocytes and
interstitial fibrosis (usually at 1–2 weeks). The chest
radiograph may demonstrate bilateral focal areas of
consolidation, often in the mid and lower zones. At
this stage, CT often demonstrates a combination of
consolidation, ground glass opacification, and fibrosis
(Figure 3.17). In the chronic phase, fibrosis may
improve, remain stable, or progress. The fibrotic
changes depicted at CT are predominantly peripheral
and subpleural. A number of drugs have been impli-
cated, with cyclophosphamide, carmustine, and bleo-
mycin being the commonest in the context of
lymphoma.
Cyclophosphamide-related injury is not dose-
related and has been reported up to 13 years after
administration. Conversely, carmustine has a clear
dose-related toxicity, with the overall incidence rising
from 20–30% to 50% once a cumulative dose of 1.5 g/
m2 has been exceeded. Bleomycin pulmonary toxicity
is reported to occur in up to 18% of patients, with a
mortality of 4.2%. There appear to be a number of
risk factors associated, including age >40 years, cumu-
lative dose >450 units, and use of granulocyte 41
 Chapter 3: Imaging
Figure 3.17 Extensive ground glass opacification bilaterally with
areas of consolidation in the left lower lobe. A presumptive diagnosis
of bleomycin toxicity was made, with the patient responding to drug
withdrawal and a course of corticosteroids.
colony-stimulating factor. Of note, the same group
observed that omitting bleomycin from the therapeutic
regime had no impact on rates of complete remission,
progression-free survival, and overall survival.
Patients with an organizing pneumonia typically
present with progressive dyspnea, fever, and a dry
cough. The chest radiograph demonstrates bilateral
scattered opacities with a peripheral distribution. At
CT, ill-defined areas of consolidation, nodules, and
plugging of the small airways (“tree-in-bud” appear-
ance) will be evident. Withdrawal of the causative
agent, with or without steroid therapy, typically results
in a good response. Non-specific interstitial pneumo-
nia (NSIP) and pulmonary hemorrhage are less fre-
quently observed manifestations of pulmonary drug
toxicity. NSIP is an inflammatory condition that
results in mild interstitial fibrosis with areas of ground
glass opacification on CT. Diagnosis of NSIP may be
difficult at both histology and radiology, but the prog-
nosis is good. Diffuse pulmonary hemorrhage results
in acute respiratory distress with focal areas of consol-
idation demonstrated on the plain radiograph. CT will
delineate multiple foci of consolidation and/or ground
glass opacification. Diagnosis may be confirmed by
demonstrating an increase in the diffusion capacity
42 for carbon monoxide (DLCO). The outcome is heavily
dependent on the underlying cause.
Figure 3.18 Multiple foci of consolidation, some with a rim of
ground glass opacification typical of angioinvasive aspergillosis.
Radiation therapy has evolved significantly over the
past decade, with the introduction of 3-D conformal
radiotherapy and, more recently, intensity modulated
radiotherapy. This has resulted in a significant reduc-
tion in the volume of “normal” tissue that is irradiated.
As a result, frank radiation pneumonitis is rarely seen
and only minimal paramediastinal fibrosis is seen on
CT. However, as more patients are cured of their initial
disease, the risk of radiation therapy in the development
of a second malignancy is becoming more apparent. In
the context of lymphoma, it is particularly important to
minimize the dose to the breasts of young female
patients with mediastinal disease. To this end, the role
of PET–CT in defining active disease for radiotherapy
planning requires further investigation.
Severely immunocompromised patients are also
susceptible to a variety of infectious agents that carry
a high morbidity and mortality. Angioinvasive asper-
gillosis classically demonstrates variably sized nod-
ules of consolidation with a halo of ground glass
opacification or pleural-based wedge-shaped areas
of consolidation and is elegantly shown on thin-
section CT sections (especially with an HRCT algo-
rithm)(Figure 3.18). This represents infarcted lung
with a zone of hemorrhage peripherally. The patient
will usually be severely neutropenic, but, as the neu-
trophil count increases, a crescent of air may be seen
as the necrotic tissue falls away from the viable tissue.

A B
Figure 3.19 (A) A patient with Hodgkin disease and HIV presented
with acute dyspnea. Initial CT demonstrates multiple foci of ground
glass opacification and bilateral pleural effusions. This was proven to
be PCP following bronchoscopy.
This will eventually resolve to leave either a cavity or
parenchymal scarring. Unfortunately, the CT fea-
tures are non-specific, with candida, cytomegalovirus
(CMV), and herpesvirus among the infections that
can give similar appearances. However, CMV more
typically has bilateral ground glass opacification with
numerous ill-defined small nodules. Patchy consoli-
dation is seen commonly, whilst pleural effusions are
seen in approximately 60% of cases.
Similarly, Pneumocystis carinii pneumonia (PCP)
presents in severely immunocompromised patients
and has a number of radiographic manifestations.
Most commonly, there is a combination of fine reticular
opacification and ill-defined ground glass opacification
that progresses to consolidation over a period of days.
Less commonly, PCP may present as a focal area of
consolidation, mimicking pulmonary lymphoma, pleu-
ral effusions, mediastinal lymphadenopathy or with
miliary nodules (Figure 3.19). Approximately one-
third of patients go on to develop pneumatoceles
(thin-walled cysts), most often in the upper lobes.
These usually resolve spontaneously within a year of
the acute episode, but may rupture to produce a
pneumothorax.
Chapter 3: Imaging
Figure 3.19 (B) Eleven days later, the patient deteriorated
further and CT demonstrated a pneumomediastinum and left
pneumothorax. The ground glass opacification within the lung
parenchyma is now more confluent and the septal thickening is
more evident.
Further reading
Cheson BD, Horning SJ, Coiffier B, et al. Report of an
international workshop to standardize response criteria
for non-Hodgkin’s lymphomas. NCI Sponsored
International Working Group. J Clin Oncol,
1999;17:1244.
Grubnic S, Vinnicombe SJ, Norman AR, Husband JE. MR
evaluation of normal retroperitoneal and pelvic lymph
nodes. Clin Radiol, 2002;57:193–200.
Hoffmann M, Kletter K, Diemling M, et al. Positron
emission tomography with fluorine-18–2-fluoro-2-
deoxy-D-glucose (F18-FDG) does not visualize
extranodal B-cell lymphoma of the mucosa-associated
lymphoid tissue (MALT)-type. Ann Oncol,
1999;10:1185–1189.
Jerusalem G, Beguin Y, Najjar F, et al. Positron emission
tomography (PET) with 18F-fluorodeoxyglucose
(18F-FDG) for the staging of low-grade non-Hodgkin’s
lymphoma (NHL). Ann Oncol, 2001;12:825–830.
Jerusalem G, Hustinx R, Beguin Y, Fillet G. Evaluation
of therapy for lymphoma. Sem Nucl Med,
2005;35:186–196.
Juweid ME, Cheson BD. Role of positron emission
tomography in lymphoma. J Clin Oncol,
2005;21:4577–4580.
43
 Chapter 3: Imaging
Juweid ME, Wiseman GA, Vose JM, et al. Response
assessment of aggressive non-Hodgkin’s lymphoma by
integrated International Workshop Criteria and fluorine-
18-fluorodeoxyglucose positron emission tomography. J
Clin Oncol, 2005;23:4652–4661.
Kazama T, Faria SC, Varavithya V, et al. FDG PET in the
evaluation of treatment for lymphoma: clinical usefulness
and pitfalls. Radiographics, 2005;25:191–207.
Lister TA, Crowther D, Sutcliffe SB, et al. Report of a
committee convened to discuss the evaluation and
staging of patients with Hodgkin’s disease: Cotswolds
meeting. J Clin Oncol, 1989;7:1630–6. Erratum in: J Clin
Oncol 1990;8:1602.
Meyers JL. Pathology of drug-induced lung disease. In
Katzenstein AA, Askin FB, (eds). Katzenstein and Askin’s
Surgical Pathology of Non-neoplastic Lung Disease, 3rd
edn. Philadelphia, PA: Saunders, 1997;4:81–111.
Moon JH, Kim EA, Lee KS, et al. Cytomegalovirus
pneumonia: high-resolution CT findings in ten
non-AIDS immunocompromised patients. Korean J
Radiol, 2000;1:73–78.
Paes FM, Kalkanis DG, Sideras PA, Serafini AN. FDG PET/
CT of extranodal involvement in non-Hodgkin
44
lymphoma and Hodgkin disease. Radiographics,
2010;30:269–291.
Rimmer MJ, Dixon AK, Flower CDR, Sikora K. Bleomycin
lung: computed tomographic observations. Br J Radiol,
1985;58:1041–1045.
Rossi SE, Erasmus JJ, McAdams HP, Sporn TA,
Goodman PC. Pulmonary drug toxicity: radiologic and
pathologic manifestations. Radiographics,
2000;20:1245–1259.
Surbone A, Longo DL, DeVita VT Jr, et al. Residual
abdominal masses in aggressive non-Hodgkin’s
lymphoma after combination chemotherapy:
significance and management. J Clin Oncol,
1988;6:1832–1837.
Terasawa T, Lau, J, Bardet S, et al. Fluorine-18-
fluorodeoxyglucose positron emission tomography
for interim response assessment of advanced stage
Hodgkin’s lymphoma and diffuse large B-cell
lymphoma: a systematic review. J Clin Oncol,
2009;27:1906–1914.
Zissin R, Metser U, Hain D, Even-Sapir E. Mesenteric
panniculitis in oncologic patients: PET-CT findings.
Br J Radiol, 2006;79:37–43.
 Chapter
4
Clinical trials in lymphoma
Thomas M. Habermann and Matthew Maurer
Introduction
Clinical trials are the backbone of the development and
advancement of therapeutic approaches. The pace of
development of new agents, the regulatory overhead,
the costs of data management, inclusion of quality-of-
life assessments, radiologic assessments and central
review, central pathology review, biospecimen acquis-
ition, symptom-control assessments, and the evaluation
of biologic correlates all present significant challenges
for the future development of new therapeutic agents
and regimens. As these therapeutic agents and
approaches become increasingly targeted and the basic
science, clinical risk factors, toxicities, and natural his-
tory of lymphomas broadens, the identification
and assessment of relevant endpoints, be they clinical
(e.g. physical examination, radiologic) or biologic (e.g.
immunologic, genetic, metabolic, etc.), also becomes
increasingly complex. In turn, the more complex the
endpoints and the more targeted the regimens, the
more challenges are presented when designing and
conducting clinical trials and analyzing and interpreting
the results. While this field of clinical trial design is too
broad to give full discussion adequately in this chapter,
the reader is referred to more extensive references on
this area. Given the complexity of issues when design-
ing, monitoring, interpreting, and analyzing data for a
clinical trial, statistician input and collaboration is of
paramount importance. This chapter will focus on the
principles and details of clinical trials from the clinician
perspective. As such, the fundamental considerations of
clinical trial design as well as the types of trials con-
ducted in clinical research will be reviewed.
Background
The practice and science of clinical trials and research
is in its relative infancy. Until 1750, the thinking
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
was that of Galen, who attained an authority that
remained unchallenged. Medicine had a long tradi-
tion of physicians as trusted advisors to patients.
Treatment choices were based not on investigation
but rather on uncontrolled experimentation.
Therapies were primarily evaluated on individual
patients in an ad hoc manner. One of the earliest
and most commonly cited examples of clinical
experiments to evaluate therapeutic interventions
was the treatment of scurvy with oranges and lemons.
The comparison of results from these trials was pri-
marily to those in the literature and other historical
controls. The first strictly controlled clinical trial
through random assignment of patients to treatment
groups was not reported until 1931, in the examina-
tion of tuberculosis patients. A subsequent trial in
tuberculosis patients was also the first to use random
numbers to assign patients to experimental versus
control groups, and it was demonstrated that strep-
tomycin plus bed rest was superior to bed rest
alone. Since then, the practice and development of
clinical trials has expanded greatly. In the United
States, controlled clinical trials were first under the
National Cancer Institute (NCI) under Gordon
Zubrod. Zubrod was one of the key individuals in
the formation of the Eastern Solid Tumor Group
(now the Eastern Cooperative Oncology Group,
ECOG), which published the first randomized trial
in solid malignancies in the USA comparing nitrogen
mustard and thiophosphamide.
In today’s research environment, a primary focus
in the development and conduct of clinical trials is the
safety and protection of the patient involved in such
research. This is of primary importance as we continue
to refine and explore new methods for clinical trial
design and analysis.
45
 Chapter 4: Clinical trials in lymphoma
Informed consent
The unprecedented growth and discovery has resulted
in a dominant role for clinical research, and this in turn
has emphasized the process of informed consent. The
National Cancer Institute (NCI) and the Food and Drug
Administration (FDA) in the USA, and the European
Union Clinical Trials Directive legislation, have all
structured and strengthened the elements of the process.
Safeguards must be in place for patients who enter
clinical trials. Informed consent is one of the key aspects.
The key elements of informed consent are outlined
in Table 4.1. Additional points that need to be brought
to the patient’s attention include unforeseeable risks to
the patient, embryo, or fetus; circumstances in which
the subject’s participation may be terminated by the
investigator without the patient’s consent; the number
of patients; additional costs that may result from study
participation; consequences of withdrawal; and a state-
ment of new findings that may relate to the patient’s
willingness to continue study participation. The con-
sent form must be signed and dated by the patient or
patient’s representative, and a copy must be given to
the patient or patient’s representative.
Table 4.1 Informed consent.
Involves research
Purpose
Duration
Procedures (experimental or not)
Treatment(s)
Description of risks or discomfort
Must include all that are listed in the model consent form
Inconvenience
Description of benefits
Reasonable expected benefits
Benefits to society
Must include all that are listed in the model consent form
Description of the extent of confidentiality of records
If results are published, patient identity remains confidential
Individuals who have access to the medical records: monitors/
auditors, IRB, regulatory authorities, sponsor(s), drug
company/ies
Explanation regarding compensation and/or whether
treatments are available if injury occurs
Voluntary participation
Refusal to participate involves no penalty
May discontinue at any time without loss of benefits
Alternative procedures or treatments
Contacts for
Research-related questions
Information regarding subject’s rights
46 Research-related injury
The investigator must not unduly influence parti-
cipation and/or continuation on protocol and should
fully inform the subject or the subject’s representative
of all pertinent aspects of the trial. The language used
in the oral explanation and in the written consent form
should be non-technical, practical, and understand-
able to the participant. If the patient or patient’s rep-
resentative is unable to read, an impartial witness
should present during the informed consent discus-
sion and should sign and date the consent form along
with the patient, if the patient is capable.
Endpoints
To develop, implement, and analyze a clinical trial
effectively, it is critical to establish the research ques-
tions to be evaluated clearly. A statement of the research
question and reason for conducting the clinical trial
leads to a clear definition of the endpoints to be eval-
uated in the clinical trial. While intuitive, this is impor-
tant to ensure that the endpoints actually address the
goals of the trial. The goals and corresponding end-
points must match the phase and type of trial.
Overall, the choice of the primary endpoint to be
evaluated is critical to answering the primary research
question and motivation for the trial, and typically
drives the sample size and power considerations. The
ultimate goal in treating lymphoma patients is to
extend survival and improve quality of life. Since over-
all survival (OS) is an endpoint that is not always easily
evaluated in the constrained timeframe of a clinical
trial, surrogate endpoints are often used to determine
efficacy. Surrogate endpoints for clinical activity such
as response and progression-free survival (PFS) are
most typically used in evaluating efficacy of experi-
mental regimens. Secondary endpoints to be evaluated
are also defined a priori, but power considerations for
these analyses do not usually impact on the determi-
nation of sample size. However, the sample size of the
trial can be increased moderately to power specific
secondary endpoint analyses of critical interest.
Defining the endpoints used to answer the research
questions is best achieved when considering the treat-
ment interventions themselves that will be evaluated, as
well as the targeted patient population. For example, in
the case of some of the novel therapeutic regimens being
considered, the agents may have cytostatic rather than
cytotoxic effects on the disease. In the past, the focus
has been on cytotoxic agents and high-dose therapies,
but there are new biologic agents that have a range of
targeted activities, including cell surface markers,

modulators of signal transduction pathways, angiogen-
esis inhibitors, growth factor receptor inhibitors, poten-
tiators of apoptosis, modulators of the immune
response, and modulators of gene expression. They
may include cellular therapy, gene therapy, and vaccine
therapy, and the application of clinical trials to these
interventions presents new challenges and opportuni-
ties. With cytostatic agents, the standard endpoint of
clinical response may not be sufficient to determine
efficacy. Instead, the proportion of patients who have
not progressed or relapsed within a specified timeframe
may be used, or other endpoints more appropriate to
capture the effects of a cytostatic agent. In addition,
evaluation of immunotherapeutic regimens may require
similar alternate endpoints to evaluating efficacy. The
timing of the study, initial therapy versus treatment in
relapse, and histology subtype are often the most rele-
vant issues in determining the endpoints.
Response criteria
As histology has been further refined, response criteria
have been standardized. The Modified Cheson criteria
incorporate positron emission tomography (PET) scans
into diffuse large B-cell lymphoma (DLBCL) and
Hodgkin lymphoma (HL). OS is defined as death from
any cause from the time of randomization. Lymphoma-
specific survival is defined from the time of entry to the
study to death as a result of lymphoma or death of
unknown causes attributed to therapy. Progression
base endpoints included the following: PFS is defined
as lymphoma progression or death from any cause.
However, in studies in which failure to respond without
progression is considered an indication for another
therapy, such patients should be censored at that point
for the progression analysis. Event-free survival (EFS) is
defined as time from study entry to any treatment fail-
ure and defined as disease progression or discontinua-
tion of therapy for any reason (progression, toxicity,
refusal, or initiation of new treatment). Time to pro-
gression (TTP) is defined as documented lymphoma
progression or death due to lymphoma. Non-lym-
phoma deaths are censored at the time of death.
Outcome data
After determining the primary and secondary end-
points of interest for the clinical trial, it is important
to identify those variables and resulting outcome data
necessary to answer the relevant research questions at
the end of the trial. Outcome measures will vary
Chapter 4: Clinical trials in lymphoma
depending on the endpoints and phase of the trial.
These measures are on qualitative (categorical), quan-
titative (measurement), or time-to-event data. An
example of a qualitative measure is the classification
of response, where tumor measurements define com-
plete response (CR), partial response (PR), stable dis-
ease (SD), and progressive disease (PD). For those
refractory and/or aggressive patient populations, the
overall response rate may be of interest. For newly
diagnosed and/or more indolent patient populations
where the overall response (CR and PR) is typically
high with standard treatment and thus where a com-
plete response is more interesting, the CR rate may
instead be used as a primary measure of efficacy.
Quantitative endpoints focus on continuous meas-
ures of effect. Examples of this are outcomes such as
changes or percentages of changes in quality-of-life
scores, immunologic parameters, or other biologic
data that can measure clinical effect or adverse reac-
tions. Time-to-event data can also be defined in many
different ways, depending on the needs of the trial, and
it is important to define at the outset, not retrospec-
tively, what constitutes an event as well as the time from
which this is measured (e.g. study entry and diagnosis
need to be defined). Time-to-event measures include
EFS (i.e. time to an event defined as progression, relapse,
and/or death), OS (i.e. time to death from any cause),
and even time to engraftment in stem cell transplanta-
tion. Time to event measures must take into account the
fact that not all patients will have had an event at the time
of analysis, but, having followed the patient for a certain
time, there is the evidence that they are event-free at least
until that time. Statistical methods are used in the esti-
mation and analysis of this type of data to account for
some patients not reaching defined endpoints.
Overall, outcome data can take many different
forms to accommodate the endpoints and goals of
the clinical trial. What is important is to define these
endpoints adequately when designing the trial and to
consider carefully how effectively these outcome data
can be collected on all patients who will enroll onto the
trial. This is a more significant logistical concern in the
context of multi-institution clinical trials, where
standards and care may differ and multiple pieces of
information are being collected.
Evaluating treatment effect
In assessing and determining efficacy in clinical trials,
the classification of the treatment effect on patients is 47
critical for not only identifying the clinical response
 Chapter 4: Clinical trials in lymphoma
rate, but also in defining progression and thereby time
to progression or PFS. The definition of response is
disease- and even subtype-dependent and needs to be
appropriate for the study. These definitions also evolve
over time as new technologies develop to evaluate the
existence of and changes in the disease. The addition of
computerized axial tomography (CT scan) changed
how lymphoma was assessed and measured, and func-
tional imaging – such as PET scans – is now replacing
and/or complementing the CT scan in lymphoma.
PET/CT fusion scans are more readily available; how-
ever, the use of oral and intravenous contrast agents
that are routinely utilized in CT scanning impairs the
quality of the PET scans.
In 1999, an International Working Group (IWG)
developed recommendations for response assessment
for non-Hodgkin lymphoma (NHL) that were adopted
internationally by study groups, industry, regulatory
agencies, and, subsequently, by clinical trial groups for
HL. Despite the publication of standardization of
response criteria for NHL, an examination of current
trial parameters in nine international lymphoma study
groups by the working group for quality management
(WG-QM) of the Competence Network Malignant
Lymphoma found significant differences and missing
definitions. This finding led to an international project
to harmonize trial parameters for malignant lym-
phoma. Revised response criteria for malignant lym-
phomas from the members of the International
Harmonization Project (IHP) of the Competence
Network Malignant Lymphoma were first reported at
the American Society of Hematology in December
2005. These response criteria incorporated functional
imaging via the PET scan. Pre-treatment PET was
recommended but not required in DLBCL, HL, follic-
ular lymphoma (FL), and mantle cell lymphoma
(MCL). Response assessment was recommended for
DLBCL and HL and in FL and MCL only when overall
response rate (ORR)/CR was a primary endpoint of
the study. Routine post-treatment follow-up was not
recommended. A CR was defined as all lesions nega-
tive on the PET scan, a normal lactate dehydrogenase
(LDH), and negative flow cytometry. Complete remis-
sion undefined (CRu) was no longer a response clas-
sification. Relapse should include PET-positive results.
By 2011, multiple groups internationally have been
working to further define when PET scans should be
performed, how to incorporate the maximal standard-
ized uptake value (SUVmax), to which types of lym-
48 phoma PET is most applicable, and how to utilize PET
after the initial pre-treatment PET to assess response
and predict outcome.
Precision and error control
Two key aspects related to the precision of clinical trial
design are interval estimates of the primary endpoint,
referred to as confidence intervals (CI), and the simi-
larity of the study sample to the population to which the
results apply. A CI is a point estimate ± a margin of
error that depends on the degree of confidence one
desires to place on the true value being within the
interval. This is calculated from the data that will
include the true parameter of interest (e.g. response
rate, mean change in quality of life) for the population
with a specified probability. Calculation of the confi-
dence interval is dependent on the primary endpoint
and corresponding outcome measure, but is based on
the statistic (e.g. proportion, mean), a margin of error
that is calculated using a cut-off value for the corre-
sponding level of confidence based on the distribution
of the statistics, as well as the standard error (i.e. stand-
ard deviation of the statistic) calculated from the sample
of patients enrolled on the clinical trial. Typically, 95%
CI are calculated, but other confidence levels are also
useful. These intervals may be obtained from tables
designed for the binomial distribution. There is a
trade-off between level of confidence and the width of
the interval, where the wider the interval, the higher the
confidence level that the true population parameter is in
that interval. Another determinant of CI width (and
thereby precision of the estimate) is sample size; the
larger the sample size, the greater the precision and thus
the narrower the CI. It is important to note that, in the
context of clinical trials, multistage designs can require
different methods of calculating CI to avoid bias.
An essential aspect of precision is the similarity of
the patients eligible for the study versus the target
population of interest. The eligibility criteria deter-
mine the study population, which is typically a subset
of the target population. A trade-off exists based on
restrictive versus broad criteria. The advantages of
restrictive eligibility criteria are that the patients are
more homogeneous, smaller sample sizes may be
required, and there is a reduced likelihood of con-
founding factors. Disadvantages are that the results
may not be able to be generalized to the target pop-
ulation, and fewer patients may be eligible, which may
translate to a longer study duration or non-feasibility
of the enrollment goal. In general, eligibility criteria

should be loosened as the phase of the trial increases
and trials become larger. Phase I trials are usually
more restrictive, because patients should not be
enrolled in a dose-finding trial with a single agent if
there is an alternative treatment that is known to be
effective. Such trials may span multiple histologies if
tolerability is the key determinant of dose level, but the
criteria can be restrictive in terms of who would be
eligible to receive this experimental regimen. Phase II
trials are often conducted in the target population of
interest, but are often more restricted to ensure a base-
line level of health and some homogeneity, to better
evaluate initial evidence of activity. This is an impor-
tant reason why results from phase II trials should not
be used to change clinical practice. Phase III studies
are typically the most liberal in terms of eligibility
criteria as these are usually large trials that are
intended to evaluate regimens against standard of
care, and their results can affect clinical practice.
Because of this, it is critical to design the eligibility
criteria to ensure that patients entered into the clinical
trial are similar to, or representative of, the target
population so that the results can be generalized to
the patient population of interest.
In lymphoma, histology is the most important
eligibility criterion. In 1982, the non-Hodgkin’s
Lymphoma Pathologic Classification Project proposed
a morphologic system that was representative of
at least six classification systems. The Working
Formulation had three major subdivisions: low
grade, intermediate grade, and high grade. A classifi-
cation that was based on immunohistochemistry,
cytogenetics, and molecular genetic characteristics
was developed, the Revised European–American
Lymphoma classification (REAL). The World Health
Organization (WHO) classification, fourth edition, is
now the accepted classification system internationally.
This allows for cross-interpretation of study popula-
tions in different studies.
A second factor is the heterogeneity of the individ-
ual study population. Two phase III USA trials in the
1990s in aggressive lymphoma demonstrated that
phase II trials did not predict superiority of other
agents over CHOP (cyclophosphamide [C], adriamy-
cin [H], vincristine [O], and prednisone [P]). At least
one factor was the heterogeneity in the individual
study populations. Shipp and colleagues developed
the International Prognostic Factor Index (IPI) that
encompassed DLBCL to address the issue of hetero-
geneity in patient study populations. This study
Chapter 4: Clinical trials in lymphoma
defined the lymphoma populations of aggressive lym-
phomas. Subsequent studies have demonstrated the
utility of the IPI in other lymphoma histologies, such
as peripheral T-cell lymphoma, not otherwise speci-
fied. Modifications of the IPI have been demonstrated
to have value in virtually all NHLs. Initially adopted in
the FLs, an adaptation, the FLIPI score and modified
FLIPI score, have replaced the IPI in FL and better
differentiate groups of patients.
Closely related to the issue of precision is that of
error rates and how to control these errors. In clinical
trials there are two types of error rates to control: type I
and type II errors. A type I error is defined as the
probability of concluding that a treatment effect or
difference exists when in truth it does not. The type
II error is the probability of concluding that a treat-
ment effect or difference does not exist when in truth it
does. Both types of errors should be controlled in most
clinical trials. The level of acceptable error rates will
depend on the phase of the trial. The closer the study is
to results that will be generalized to the target popula-
tion, the stricter the acceptable levels are for the rates
of type I and type II errors. Typically, type I and type II
error rates are constrained at 10% each or even 5% type
I and 20% type II in the phase II setting. However, in
the phase III setting the typical constraints are for 5%
type I and 10% type II error rates. These predefined
constraints on error rates affect the resulting sample
size for the clinical trial, and by definition affect the
power of the study. Power is defined as the probability
of concluding that a treatment effect or difference
exists when it truly does exist, and is calculated as 1 –
type II error rate (e.g. type II error rate of 0.10 trans-
lates to 90% power). The constraint on the type I error
rate also affects the decision rule to be used when
determining if the results are significant. This signifi-
cance level is the threshold for the resulting P values
derived using the calculated test statistic for the end-
point(s) of interest. This is intuitive as the P value is
technically the conditional probability of seeing an
effect as large or larger by chance alone given that
the null hypothesis is true. Therefore, the significance
level establishes the cutpoint, at which point it is
believed that the probability of seeing as extreme a
result is due to chance alone under the assumption
that the null hypothesis is true versus our rejection of
the assumption that the null hypothesis is true. In
other words, this translates to the strength of evidence
that it is necessary to see to reject the assumption that
the null hypothesis is true. 49
 Chapter 4: Clinical trials in lymphoma

In addition to the need to constrain type I and Phase I trials

type II errors, it is also important to control potential
The primary aim of a phase I trial is to determine the
inflation of the type I error through multiple interim
dose to be recommended for evaluation in further study
analyses or “looks” at the primary endpoint. Each of
(Figure 4.1). Historically, the assumption has been that
these considerations affects the study design and cor-
“more is better” and thus the recommended dose level is
responding sample size required to evaluate the pri-
defined by the maximum tolerated dose (MTD) of an
mary endpoint of interest. These complexities of
agent. When tolerability is a determinant of the recom-
clinical trial design, and the need to address various
mended dose level, the toxicities to be considered as
aspects of constraining these probabilities of making
excessive (dose-limiting toxicity, DLT) must also be
type I or II errors, further supports close collabora-
defined. In the context of lymphoma trials, it is critical
tion with statisticians in the development of clinical
to define what constitutes a DLT. Leaving it open to any
trials. Additional trial design issues are discussed in
severe (grade 3 or greater) toxicity can mean that events
more detail below. Table 4.2 lists some of the key
such as grade 3 leukopenia may be counted as a DLT
terms used.

Table 4.2 A glossary of terms.

Confidence A range of values calculated from the sample observations that are believed with a particular probability to
interval contain the parameter value. These convey information that the P values do not.
Cox regression The most widely used statistical model in clinical research. This is used when the analysis of a clinical trial is based
model on the time to an event. It models the hazard function as a set of explanatory variables rather than the mean.
Event-free survival Time to event as determined by relapse, progression, or death.
Hazard function The probability that an individual experiences an event (death, impairment, etc.) in a small time interval. It is a
measure of how likely an individual is to experience an event, or a function of the age of the individual.
Intent-to-treat All patients entered in a clinical trial are analyzed together as reflecting that treatment, whether or not they
analysis completed or ever received the treatment. This prevents possible bias from many components.
Interim analysis An analysis, preferably conducted by an independent data-monitoring committee, to assess whether the
ongoing trial can realistically be expected to answer the primary question, taking into account the inclusion rate,
adverse events, previous experience, and the statistical significance as compared to the defined statistical
guidelines in the trial.
Kaplan–Meier A procedure for estimating the survival function for a set of survival times, some of which may be censored
estimator observations. For example, the probability of a patient surviving 2 years after a treatment can be calculated as the
probability of surviving 1 year given that the patient survived the first year.
Overall survival Time to death from any cause.
Power The probability of rejecting the null hypothesis when it is false.
P value An index measuring the strength of the evidence against the null hypothesis. It is not the probability that the
specified hypothesis is true.
Primary endpoint In lymphoma, the primary endpoints are overall survival and progression-free survival.
Randomization The key features of randomization are that the treatment assignment is based on chance alone, the patient
characteristics are balanced or equivalent, and the randomization process provides the foundation for the
statistical tests that are used.
Sample size The number of patients to be included in a study so that the study has a particular power to detect an effect.
Secondary The secondary endpoints in lymphoma include event-free survival, time to progression, etc.
endpoints
Significance level The level of probability at which it is agreed that the null hypothesis will be rejected. This is conveniently set at
0.05. P values need to be much smaller than 0.05 before they can be considered to provide strong or conclusive
evidence against the null hypothesis.

50 Significance test A statistical procedure that results in a P value to a hypothesis.

 Chapter 4: Clinical trials in lymphoma

Preclinical Initial studies on in vitro samples and/or animal studies to obtain information Figure 4.1 Progression of clinical
on dosing and mechanisms of action trial research.

Phase I Trials to identify a recommended dose level to be evaluated in subsequent larger

trials. This is often based on tolerability of the regimen, but can also be based on biologic
markers.

Phase II Trials used to obtain initial evidence of efficacy, and typically incorporate a
decision rule of what evidence is required to declare it promising vs. not. These are not
definitive trials on efficacy of the regimen. Also used to obtain more information on
toxicity profile of the regimen.

Phase III Trials that compare experimental regimen(s) to the standard of care; these trials
are typically large, randomized trials whose outcomes can influence clinical practice.

and thus affect the dose escalation decisions. It is pru-
dent to restrict what constitutes a DLT carefully; often
these are restricted to non-hematologic toxicities, but
they can include time to engraftment (e.g. failure to
engraft by the absolute neutrophil count [ANC] by 28
days could be considered a DLT). The timeframe for
which observed toxicity will be considered in the dose
escalation decisions should also be defined. When
determining the study duration, it is important to con-
sider not only how long it takes to accrue a cohort of
patients, but also for how long each cohort needs to be
evaluated, as defined by the specified toxicity observa-
tion period. Even though a regimen may be given con-
tinuously until a patient progresses, or even when the
regimen is given for 6 months, it is not uncommon for
only the first one to two cycles to be considered when
deciding whether or not a DLT was observed, and thus
whether or not the next cohort can be accrued to the
next dose level.
The most commonly used phase I trial design is the
standard cohort-of-three design. Patients are entered in
a stepwise fashion in cohorts of three, with up to six
patients per dose level, in order to determine the MTD.
If no DLT is observed, then the next cohort of three
patients is treated at the next dose level. If moderate
toxicity (usually one patient with DLT) is observed,
then another cohort of three patients may be treated
at that dose level for a maximum of six patients. If
significant toxicity is observed (usually defined as two
or more patients with DLT), the MTD will have been
exceeded. A subsequent cohort of three patients may be
accrued to the previous dose level if six patients had not
yet been treated to ensure it is the MTD. Accrual must
be stopped after each cohort of three patients is enrolled
at a dose level to assess toxicity before proceeding to the
next dose level. This type of trial design was utilized to
study dose escalation of ProMACE-CytoBOM in
DLBCL, resulting in a phase II study utilizing a 200%
dose level. More recently, this design was utilized in a
trial of R2-CHOP (lenalidomide, rituximab-CHOP)
where the standard dose of CHOP was utilized and
the lenalidomide dose was incrementally increased.
Alternatives to the standard cohort-of-three design
for phase I trials include the continual reassessment
method (CRM) and the accelerated titration method. 51
In brief, the CRM approach utilizes the investigator’s
 Chapter 4: Clinical trials in lymphoma
prior beliefs and the observed data to develop a model
for determining the probability of unacceptable toxicity
at a given dose level. This method uses data from each
treated cohort to choose the dose for each successive
cohort. Advantages of this method include that the
MTD may be determined more quickly and accurately,
but disadvantages are that this approach is more com-
plicated and potentially difficult to implement. Also, the
performance of this design depends on the accuracy of
the model, and an incorrect model can cause problems
with too many patients treated at high dose levels or an
inaccurate estimate of the MTD. In the accelerated
titration method there are different schedules for accel-
erated dose escalation. In brief, one schedule evaluates
one patient at a dose level until a certain (sub-DLT) level
of toxicity is observed, after which it reverts to a stand-
ard cohort-of-three design. Another schedule allows for
intra-patient dose escalation using either a standard or
CRM escalation plan. Advantages of this approach are
that it can require fewer patients and, specifically, fewer
patients at potentially suboptimal levels, and that when
dose escalation is done within patients no patient is
undertreated. Disadvantages are that dose levels are
potentially evaluated with very few patients, and for
the intra-patient dose escalation there are ethical con-
siderations for dosing patients until toxicity.
The designs that have been discussed focus on the
determination of a recommended dose level based on
MTD. In the current research environment, with more
targeted therapies and/or immunotherapies, higher
dose levels may not necessarily be associated with
greater probability of response; i.e. the dose–response
curve may not be consistently increasing, but rather
may be parabolic or some other function where the
curve may reflect decreasing efficacy at higher dose
levels. In such a case, the phase I trial may be designed
to determine a recommended dose level based on
tolerability (as assessed through determination of the
MTD) and also some measure of efficacy or biologic
response. Recent designs have been developed that use
a CRM approach, where the possible outcomes are no
efficacy and no toxicity, efficacy and no toxicity, or
severe toxicity. These types of models are attractive in
that they directly incorporate some measure of efficacy
(usually as measured through some biologic param-
eter) in the dose escalation decisions. However, these
designs also require that the biologic efficacy informa-
tion be obtained relatively quickly so that it can be
used in the dose escalation decision-making process.
52 In practice, many of these endpoints are batched and/
or analyzed at the end of the study. An alternative is to
conduct the study with dose escalation decisions deter-
mined by MTD rules, and utilize the biologic efficacy
data at the end of the study to help determine the
overall recommended dose level. This may require a
specified number of patients at the MTD and sub-
MTD dose levels to ensure sufficient samples for this
type of evaluation by dose level.
Finally, there is the scenario where it is of interest
to determine the recommended dose level only based
upon some biologic or immunologic outcomes. This is
often the case in regimens where previous studies have
determined the MTD to be higher than the dose levels
of interest in the proposed dose-finding trial, and only
changes in these correlative outcomes are of interest in
determining the recommended dose level. This is the
only situation in which multiple dose levels may be
simultaneously evaluated and where patients may be
randomized to dose levels. In all other phase I scenar-
ios, it is completely inappropriate to do so, and each
dose level must be evaluated sequentially. However,
there are other approaches that can be taken that
reflect those of the tolerability-based phase I dose-
finding trials but in which the trial takes into account
a previously defined measure of biologic response.
Overall, there are many issues to consider when
designing phase I clinical trials. Since these are dose-
finding trials, often there is no data yet on efficacy, and
there is no promise that these regimens will be effective.
This affects the types of patients who are ethically
eligible for these types of trials. Also, it is important to
note that phase I trials are for determining recommen-
ded dose levels, not for concluding initial evidence of
efficacy. While phase I trials can provide useful prelimi-
nary data on efficacy, there are still limited numbers of
patients treated at a dose level. Finally, it is important to
note that even though individual agents have been
evaluated in the phase I setting, if their combination
has not, and this is the regimen of interest, this too must
be evaluated in the context of a phase I trial. At a
minimum, this tolerability assessment can be combined
with a phase II trial, where the phase II portion of the
trial would be accrued at the determined MTD.
Phase II trials
The goals of a phase II trial are to assess therapeutic
activity, to further characterize toxicity of the regimen,
and to determine if the treatment warrants further inves-
tigation in the context of a phase III trial. Phase II trials

are typically the first formal evaluation of efficacy of an
experimental therapeutic regimen. With limited resour-
ces and the number of new agents available for evalua-
tion in lymphoma, well-designed phase II trials are even
more important. Therefore, sample sizes are constrained
to reduce the number of patients who are exposed to a
potentially ineffective or even toxic regimen, but a suffi-
cient number of patients are required to effectively eval-
uate evidence of efficacy. Phase II trials are not definitive
evaluations of efficacy, but rather identify those regimens
worthy of further study in the phase III setting.
To evaluate efficacy in the phase II setting, decision
rules are established for the level of evidence required
to indicate that the regimen is promising. Typically,
the outcome measures for documentation of efficacy
are dichotomized as “successes” or “failures,” where
each is defined dependent on characteristics of the
regimen and/or target patient population. In the con-
text of lymphoma trials, this is most often the inci-
dence of a clinical response (partial or complete) to
treatment, although in some settings it may be more
relevant to focus specifically on the incidence of com-
plete responses or other definitions of “success.”
Accounting for constraints on type I and type II
error rates and what true success rate one would expect
if the regimen was not effective (P0) versus what one
would expect if it truly was effective (Pa), a fixed
required sample size and decision criteria are
established.
Multiple possible methods for developing phase II
trials exist. The focus in this review is on the broader
issues to be considered when designing a phase II
study. A primary consideration is whether or not to
include one or more interim analyses, or “stages.” A
one-stage phase II trial accrues patients as rapidly as
possible and then evaluates all endpoints at the end of
the trial. Resulting data are analyzed using the typical
binomial estimators (assuming a “success”/“failure”
variable) and exact binomial confidence intervals.
The primary drawback to the one-stage design is that
there is no formal opportunity for stopping the trial
early. This can be more problematic if results on the
first patients are available and investigators make early
conclusions based on these observations without the
benefit of a-priori criteria based on probabilistic rules.
One-stage designs are best used when the sample sizes
are moderately small and/or when the primary end-
point requires a relatively long follow-up to evaluate it
fully (e.g. response rate by 1 year, if late responses are
typical and should not be missed). However, this
Chapter 4: Clinical trials in lymphoma
design is a concern when a large number of patients
are required, where many patients may be treated with
a regimen that may not have any evidence to date of
potential therapeutic benefit.
Because of these considerations, a multistage trial
is often the most appropriate method for evaluating
the efficacy of these experimental regimens. Most
often, two-stage phase II designs are used, in which
one interim analysis is conducted to permit early stop-
ping for futility and/or toxicity. These designs require
that accrual be temporarily suspended at each stage to
evaluate those patients fully before proceeding to the
next stage. This can cause concern regarding momen-
tum of the trial and timeliness of the study duration,
and modifications to these designs are sometimes used
where the interim analyses are conducted but where
accrual is not halted (note that only the first n1 patients
required for the interim analyses are included to eval-
uate the corresponding decision criteria at that stage).
If accrual is expected to be rapid, there is not much
information on the toxicity profile for this regimen,
and/or the primary endpoint can be evaluated in a
relatively short timeframe, then it is recommended
that accrual be temporarily halted between stages to
allow for the interim analyses. A general rule of thumb
is to estimate how useful an interim analysis would be
if accrual was not halted; if all patients required for the
trial will be accrued by the time results from an interim
analysis are available, then it is prudent simply to make
it a one-stage design (if there are a reasonably small
number of patients [40 or less]) or to impose a halting
of accrual between stages.
Designs such as the Simon optimal or Simon min–
max design are commonly used, and these provide a
rule for stopping in the event of early evidence of lack
of efficacy. The Fleming multistage design also pro-
vides decision criteria to stop accrual when sufficient
evidence indicates lack of activity, but in addition
provides an upper stopping limit, where accrual may
be stopped for positive results. If there is sufficient
early evidence that the regimen is effective, it is often
not of interest to halt accrual; therefore, accrual can be
permitted to continue. This can allow a more complete
evaluation of the toxicity profile in these patients as
well as more precision in the estimated primary end-
point. What this upper stopping limit can provide is a
rule whereby it is possible to report out the early
evidence of the positive results. Without such a rule,
it is inappropriate to report out efficacy results before
the study has completed accrual. 53
 Chapter 4: Clinical trials in lymphoma

Randomized phase II trials type of trial is useful when there is concern regarding
the true success rate in the target population and/or
When more than one experimental therapeutic regimen
when it is unclear what the success rate is for the
is to be evaluated, a randomized phase II study design
standard of care.
provides a venue for simultaneously assessing the effi-
It should be noted that for these phase II screening
cacy of these regimens in the same patient population.
trials, the calculation of sample sizes has special consid-
Since phase II trials are not comparative trials, the
erations depending on the primary efficacy endpoints
purpose of a randomized phase II trial is to simulta-
(e.g. response, survival). The biggest difference in cal-
neously evaluate multiple regimens independently. A
culating sample sizes for standard versus screening
study design and decision criteria are generated for each
phase II trials is that, in the screening trial setting, the
treatment arm, where multiple treatment arms can
type I error is typically not constrained. By definition,
utilize the same study design and criteria as appropriate.
the type I error is the probability of concluding that one
The benefit with these trials is not only evaluating
is more promising than the other(s) when in truth they
multiple experimental treatments simultaneously but
are equivalent. In the context of a screening trial, this
also reducing potential patient selection biases.
type of error is not seen as a problem since it does not
Despite the fact that randomized phase II trials are
necessarily matter which one goes forward to the phase
underpowered for direct comparisons between arms,
III setting if the regimens are truly equivalent.
the success rates between arms can be informally eval-
Just as the hybrid study of phase I and phase II
uated together to identify the most promising regimen
endpoints can be studied in a phase I/II design, a phase
to be carried forward to the phase III setting. This
II/III design can also be utilized in special circum-
screening design approach simultaneously screens
stances. This randomized design utilizes an upfront
two or more regimens to identify the most promising.
phase II element where patients are also randomized to
The criteria for many such designs is based on the
the standard of care arm. Various strategies can be
regimen that results in the highest observed success
used in the first portion of this type of trial to deter-
rate by any amount. This design is appropriate for
mine which regimen(s) is (are) carried forward. Once
prioritizing between two experimental regimens
identified, the study continues accrual to the standard
when there is no a-priori reason to prefer one treat-
of care control arm and any regimens brought forward
ment over the other. Again, since this type of trial does
from the phase II portion of the trial. In the final
not support a direct comparison of efficacy endpoints
analysis, these treatment arms are directly compared
and is underpowered to do so, this is not a definitive
based on the defined efficacy endpoint(s). Like
trial for establishing efficacy of an experimental regi-
randomized phase II studies, statistical use of the
men but rather to identify the most promising regi-
reference arm is unclear and problematic. Different
men. In addition, a flexible screening design has been
strategies in this design have been proposed.
developed that further requires that, for one experi-
mental treatment regimen to be considered more
promising than another, its success rate must be
greater than the other by a prespecified differential. If Phase III trials
the success rates are not different by at least this While efficacy may be formally evaluated in the con-
amount, this design formalizes the ability to use text of phase II trials, these should not be considered
other factors such as quality of life, cost, immunologic definitive trials but simply preludes to the phase III
parameters, or other factors in the decision of which trial. In the phase III setting, a randomized controlled
regimen is most promising and should be carried clinical trial can affect clinical practice in the care of
forward to the phase III setting for a more definitive patients. The major purpose of a randomized con-
evaluation. There is the possibility in the context of trolled trial is to reduce the risks of bias. The phase
this type of design also to include a standard of care III setting is where the standard of care is evaluated,
treatment arm, where there is not direct comparison typically through the comparison of standard treat-
but rather to ensure that the experimental regimen is ment to one or more experimental regimens of inter-
considered more promising. This type of trial is more est. If the standard of care for a target population is no
problematic since there is an inclination and tempta- treatment or a “wait and watch” approach, the control
tion to make these direct comparisons, which are arm can be no treatment or a placebo control. The
54 inappropriate in the phase II setting. However, this decision on whether or not to use a placebo and/or

blinding of the assigned treatment depends on the
potential for bias in the results owing to potential
differences caused by patient and/or treating physician
attitudes. If a placebo is used, it needs to look the same
and be able to be given in the same manner as the
active intervention. The patient and/or treating physi-
cian may be blinded to the true treatment received, as
necessary.
The design of the phase III trial will be determined
first of all by the overall goal, whether the intention is
to assess superiority, equivalence, or non-inferiority.
In the most common type of phase III clinical trial an
experimental regimen is evaluated to determine if it is
superior to another regimen, typically the standard of
care or control arm. In this case, it is of interest to test
the null hypothesis that the treatment arms are equiv-
alent. However, there is increasing interest in evaluat-
ing the equivalence or non-inferiority of one regimen
to another. This is often the case when an active stand-
ard of care exists, where the goal may be to ensure that
an experimental regimen that is possibly less toxic,
easier to administer, and/or is more cost-effective is
at least as efficacious as the standard treatment. In
equivalence trials, the trial tests the null hypothesis
that there is a difference in the efficacy between treat-
ment arms, and in a sense the null and alternate
hypotheses are the reverse of those in the superiority
setting. For non-inferiority trials, it is of primary
interest to ensure that the experimental regimen(s) is
(are) not worse than the standard of care. The exper-
imental regimen(s) may be superior, but this design
focuses on evaluating only if it is at least as effective as
the standard treatment.
Once the overall goals of the phase III trial have been
defined (superiority versus equivalence or non-inferior-
ity), then there are still many other factors to consider in
the design of this type of trial. The most common is the
single-factor trial, where the primary endpoint between
two treatment arms is evaluated and compared. This is
the most straightforward type of design, with patients
randomized to one of two treatment groups, and the
endpoints of interest compared at the end of the trial.
There are other special designs that can be used in the
phase III setting, including multifactor (or multiple
treatment arm), crossover, and factorial designs. The
advantage of a multifactor design is that it addresses
multiple treatment questions in the context of one clin-
ical trial. A specific example is the SWOG 8516 study,
which compared CHOP to three other regimens that
had sufficient evidence of promising efficacy in the
Chapter 4: Clinical trials in lymphoma
phase II setting. These regimens included MACOP-B
(methotrexate [M], adriamycin [A], cyclophosphamide
[C], vincristine [O], prednisolone [P], and bleomycin
[B]); m-BACOD (methotrexate [m], bleomycin [B],
adriamycin [A], cyclophosphamide [C], vincristine
[O], and dexamethasone [D]); and ProMACE-
CytaBOM (prednisone [P], methotrexate [M],
adriamycin [A], cyclophosphamide [C], and etoposide
[E], combined with cytarabine [C], bleomycin [B], vin-
cristine [O], and methotrexate [M]). This trial demon-
strated that all arms were equivalent, and therefore
CHOP remained the standard of care. Subsequently,
phase III trials demonstrated that rituximab-CHOP reg-
imens resulted in superior OS rates compared to CHOP.
The issue of OS as an outcome for randomized clinical
trials with effective subsequent therapies is a complex
issue. A subsequent therapeutic intervention in a patient
who has relapsed that works better in the standard treat-
ment arm than in the experimental arm will lead to a
smaller OS difference.
In the crossover design, every patient gets each
intervention but they are randomized in terms of the
order in which they receive these interventions. These
types of designs are best used when there is a short
duration of effect of the treatments (to avoid carry-
over effects and bias), and the patients are in relatively
stable condition where they can complete both treat-
ments. An advantage of this type of design is that each
patient serves as their own control. This can poten-
tially reduce the required sample size. This design is
most suited to chronic conditions, but for lymphoma
patients the response may depend on the order in
which the different therapies are given. It is possible
to use crossover studies to evaluate interventions to
ameliorate side effects or conditions that affect lym-
phoma patients but generally they are not used when
attempting to improve OS or other measures of dis-
ease control. The disadvantage is that in trials in which
the control arm patients cross over to the experimental
treatment after clinical deterioration, a smaller OS
difference could be observed.
The factorial design can also be used when there
are two or more different comparisons to be made and
the treatments can be combined. For example, one
may wish to determine if the addition of an agent
makes therapy more effective, and also to assess
whether a shorter duration of therapy is as effective
as the standard duration (e.g. 4 versus 6 months). For
this hypothetical example, there are two treatment
questions of interest, and they can be combined as in 55
 Chapter 4: Clinical trials in lymphoma
a two-by-two table, with patients randomized to
receive one of the four combined treatment possibil-
ities. This is the only clinical trial that is designed to
formally evaluate interactions between two treatment
interventions. In this setting, an interaction means that
the effect of one intervention depends on the level of
another intervention. While the ability to assess the
existence of a treatment interaction can be of interest,
this type of trial design also introduces a level of com-
plexity that can make implementation and conduct
logistically difficult. In addition, the existence of a
treatment interaction can require a much larger sam-
ple size for full evaluation.
Another major consideration in the design of
phase III clinical trials is the need for interim
“looks” or analysis of the data. Formal interim anal-
yses of the primary endpoint need to be defined a
priori when designing the trial. Beyond the decision
of the number of interim analyses, as well as the
timing of these analyses, it is important to determine
in the design how to control for the potential inflation
of the type I error rate arising from these interim
looks at the data (i.e. the increased possibility of
rejecting the null hypothesis when it is in fact true,
because of the additional analyses of the hypotheses).
To account for this in the development of the trial
design, we choose what is called an α (or type I error)
spending function; this essentially takes into account
the multiple tests of the primary endpoint. These
spending functions range from simple to complex.
One of the more straightforward spending functions
is that proposed by Pocock, which essentially divides
the targeted type I error rate, α, by the number
of looks, k. This approach is similar to a simple
Bonferroni type of correction seen in multiple com-
parison corrections. This approach uses the same P
value cut-off for all analyses: if the P value is less than
α/k then it is statistically significant. This approach is
easy to use and understand, but the cut-off at the final
analysis can be quite small. Other more complex
approaches are more commonly used, the most com-
mon being the O’Brien–Fleming method. This
approach uses a more complicated statistical algo-
rithm to determine the P value cut-offs to be used
for each analysis, where the cut-offs to determine
statistical significance increase as more patients are
accrued and one gets closer to the final analysis. This
approach, although more complicated in its calcula-
tions, is the more intuitive. Recall that the P value
56 from the test statistic of interest is the probability of
seeing as extreme a result if the null hypothesis was
indeed true. The earlier the interim analysis, the
smaller the P value cut-offs; i.e. the fewer the patients
accrued and data analyzed, the stronger the evidence
has to be to reject the null hypothesis.
As discussed above, the purpose of the phase III trial
is a comparative randomized controlled trial to evaluate
two or more treatment interventions. The choice of
primary endpoint to formally compare between treat-
ment groups will depend on the overall goals of the
study. These primary endpoints can focus on efficacy,
such as PFS, OS, or even response rates. However, it is
possible also to focus on tolerability-based endpoints
such as the incidence of a specific severe toxicity, if that
is what is truly needed to differentiate and determine
the best treatment intervention. Although the choice of
primary endpoint is flexible, the most commonly used
endpoint in the phase III setting for lymphoma studies
is PFS or OS. Given that phase III trials are used to help
define clinical practice for these patients, it is thus
important to focus on the endpoints of ultimate interest
when caring for these patients. One can argue that OS as
well as quality of life is the ultimate goal when treating
patients, but endpoints such as PFS are useful surro-
gates. The choice of endpoint will depend on the target
patient population and the ability to conduct and ana-
lyze the trial in a reasonable timeframe. For example,
OS may best be used when evaluating treatment options
for aggressive lymphoma histologies, but for indolent
histologies this endpoint may take too long to evaluate
adequately with sufficient power. In that setting, it is
likely better to focus instead on PFS. It is of interest that
the immunochemotherapy protocols incorporating rit-
uximab eventually demonstrated an improvement in
OS in the rituximab arms.
The important part of this process is ultimately to
define fully the primary endpoint to be used in compar-
ing the effect of the treatment groups. For example, OS
is typically defined as the time from study entry to death
from any cause. If it is of interest to evaluate survival
from time of diagnosis; this is valid for studies focused
on newly diagnosed patients where lead-time biases and
the potential impact of prior therapies are unlikely to
present a problem. In defining PFS, the event of interest
is often inclusive of both disease progression as well as a
true relapse (disease recurrence after a complete
response). Again, the key to time-to-event analyses is
to define effectively and consistently what constitutes an
event as well as the date from which the time will be
calculated (e.g. study entry or diagnosis). In the analysis

of these time-to-event endpoints between treatment
arms, the differences are typically expressed in terms
of a hazard ratio, which is the ratio of the event rates on
each arm. Many considerations are required when
determining sample size based on a time-to-event end-
point such as the anticipated accrual period, accrual
period, minimum length of follow-up on all patients,
expected loss to follow-up, and, of course, the number
of events expected or median time to the event of
interest. Kaplan–Meier methods and/or Cox regression
models and corresponding log-rank or Wald statistics
are most often used to evaluate and compare time-to-
event endpoints between treatment arms. These meth-
ods formally account for the possibility that not all
patients will have had the event of interest at the time
of the analysis, but the important information that they
have been event-free up until the last evaluation is
incorporated.
Randomization is a key component of the phase
III trial. There are different methods for randomiza-
tion. Other relevant issues include stratification (IPI
in initial randomization, response if a second ran-
domization, etc.) and the different timepoints at
which patients are randomized to various treatments.
Randomization is a controlled assignment of patients
to treatment arms, which removes the potential for
bias in treatment allocation, thus supporting the
validity of statistical tests of significance. Patients
are randomized at a time as close as possible to
the initiation of treatment so as to avoid death, clin-
ical deterioration, complications, ineligibility, or a
change of mind regarding participation in a clinical
trial. Since clinical trials are based upon an intent-to-
treat analysis, this limits further variability.
There are different methods for randomization,
where the method of interest will also depend largely
on whether or not it is necessary to stratify. Stratification
is a process used to insure treatment arms are well
balanced with respect to important baseline prognostic
factors and reduce the potential for confounding varia-
bles when analyzing the data. Stratification prevents
chance imbalances on important factors that can affect
the interpretation of the primary endpoint between
treatment arms. Post hoc methods to adjust for these
imbalances are less credible. Examples of stratification
factors in lymphoma clinical trials are IPI and histology
(if conducted across multiple histologies). In multicen-
ter studies, it is also useful to stratify on center to
account for inherent differences between treating sites.
Chapter 4: Clinical trials in lymphoma
In a phase III setting, rather than randomizing
patients to different initial treatments, all patients
may be given the same initial regimen and are only
randomized to different therapies at the maintenance
phase. This second randomization asks a different
question. Patients who are eligible to be randomized
to maintenance therapy are separated in time and by
response, usually CR or PR, therefore not all of the
patients who entered the study initially will undergo
the second randomization, as they may need to be
responding to be eligible. With the time separation,
long-term questions about the first randomization
may be difficult to answer. Patient selection biases
are introduced because patients in the second random-
ization can be different. Interactions of the different
treatments make the interpretations complex. If
patients are randomized up-front, then patients drop-
ping out can present major issues since patients are
analyzed according to the initial randomization on an
intent-to-treat basis. Comparing only those patients
who received the assigned treatment is not valid since
baseline characteristics are balanced only at the time of
the randomization and the benefits of randomization
are lost.
A modification of this second approach has been
applied. To compare induction treatments without the
confounding effect of maintenance, analyses cannot
simply exclude all patients randomized to a mainte-
nance treatment because the proportion of non-
responding patients relative to the whole population
would be higher and would underestimate the failure-
free survival (FFS) and OS. To achieve an unbiased
estimate, an approach may be applied (weighted Cox
regression) that approximately doubles the information
for patients randomized to observation and uses a
modified variance estimator that is valid for the analy-
sis. As described for weighted Cox regression, the
robust variance estimator provides a proper estimate
of the variance of the relative risk estimate in this setting
and can be implemented using statistical software such
as S-Plus. The concept of the weighted analysis, to
remove the bias that can result from analyzing only a
subset of the patients in two-stage randomized designs,
is consistent with previously proposed methods for the
missing data problem. This approach was recently
applied to the E4494 trial, a DLBCL trial which initially
randomized patients to R-CHOP (rituximab CHOP)
versus CHOP followed by a second randomization
rituximab versus observation.
57
 Chapter 4: Clinical trials in lymphoma
The future
It will be imperative to assure that clinical trials take a
“pragmatic” approach, enrolling populations incorpo-
rating management approaches that reflect the com-
plexity and the diversity of clinical practice to improve
the quality and value of healthcare.
Technology is providing remarkable insights into
the biology of lymphoma. Molecular profiling, single
nucleotide polymorphisms, and other biomarkers are
not only further defining the disease, but are being
incorporated into eligibility criteria or are incorpo-
rated into particular studies.
As further understanding of the pathobiology
improves, collaborative clinical trials from large num-
bers of sites nationally and internationally will be
required.
The costs of drug development, regulatory over-
sight, patient care, and other issues are remarkable and
will influence future trends.
In the 1960s, there was very little literature on
clinical trials methodologies. Many methods were
developed in the 1970s and 1980s by the USA
National Institute of Health, including sample size
estimation, interim data monitoring, and other meth-
ods. Over the past decade, more trials have become
sponsored by industry. The role of the academic health
centers has also been evolving.
Conclusion
The goal of clinical trials is to determine which regi-
mens work, which regimens do not work, and how to
lead to new innovations and approaches. Lymphoma
is a heterogeneous disease. A significant number of
new agents are being developed with different mecha-
nisms of action, molecular targets, and responses.
International standardization of definitions of disease,
response assessment, and statistical interpretation are
essential in the international world in which we live.
Clinical trials are only one application that will lead to
our understanding of therapeutic efficacy in lym-
phoma, but clinical trials determine the standard of
care and future therapeutic directions.
Further reading
General reviews
De Moulin DA. A Short History of Breast Cancer.
58 Dordrecht, Netherland: Kluwer Academie Publishers,
1989.
Friedman LM, Furberg CD, DeMets DL. Fundamentals of
Clinical Trials. New York: Springer-Verlag, 1998.
Green S, Benedetti J, Crowley J. Clinical Trials in Oncology.
Boca Raton: Chapman & Hall/CRC, 2003.
Hill AB. Memories of the British streptomycin trial in
tuberculosis. Control Clin Trials, 1990;11:77–79.
Korn EL, Arbuck SG, Pluda JM, et al. Clinical trial designs
for cytostatic agents: are new approaches needed? J Clin
Oncol, 2001;19(1):265–272.
Piantadosi S. Clinical Trials: A Methodologic Perspective.
New York: John Wiley & Sons, 1997.
Pocock SJ. Clinical Trials: A Practical Approach. Chichester:
John Wiley & Sons, 1983.
Spilker B. Guide to Clinical Trials. New York: Raven Press,
1991.
Silverman WA. Doctoring: from art to engineering. Control
Clin Trials, 1992;13:97–99.
Zubrod CG, Schneiderman S, Frei E, et al. Appraisal of
methods for the study of chemotherapy of cancer in man:
comparative therapeutic trial of nitrogen mustard and
thiophosphoamide. J Chronic Dis, 1960;11:7–33.
Lymphoma-specific
Cheson BD, Pfistner B, Juwied ME, et al. Revised
response criteria for lymphoma. J Clin Oncol,
2007;25:579–586.
Coffier B, Lepage E, Briere J, et al. CHOP chemotherapy plus
rituximab compared with CHOP alone in elderly patients
with diffuse large B-cell lymphoma. N Engl J Med,
2002;346:235–242.
Diem K, Lentner C. (eds) Scientific Tables. Basle,
Switzerland: JR Geigy, 1970.
Federico M, Bellei M, Marcheselli L, et al. Follicular
lymphoma international prognostic index 2: a new
prognostic index for follicular lymphoma developed by
the international follicular lymphoma prognostic factor
project. J Clin Oncol, 2009;27:4555–4562.
Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a
standard regimen (CHOP) with three intensive
chemotherapy regimens for advanced non-Hodgkin’s
lymphoma. N Engl J Med, 1993;328:1002–1006.
Gatter KC, Warnke RA. Diffuse large B-cell lymphoma. In
Jaffe E, Harris NL, Stein H, Vardiman J (eds). World
Health Organization Classification of Tumours: Pathology
and Genetics of Haematopoietic and Lymphoid Tissues.
Lyon: IARC Press, 2001:171–174.
Gordon LI, Harrington D, Andersen J, et al. Comparison of a
second-generation combination chemotherapeutic
regimen (m-BACOD) with a standard regimen (CHOP)
for advanced diffuse non-Hodgkin’s lymphoma. N Engl J
Med, 1992;327:1342–1349.

Habermann TM, Weller EA, Morrison VA, et al.
Rituximab-CHOP versus CHOP alone or with
maintenance rituximab in older patients with
diffuse large B-cell lymphoma. J Clin Oncol,
2006;24:3121–3127.
Juweid ME, Stroobants S, Hoekstra OS, et al. Use of positron
emission tomography for response assessment of
lymphoma: consensus of the imaging subcommittee of
international harmonization project in lymphoma. J Clin
Oncol, 2007;25:571–578.
Nowakowski GS, Laplant B, Habermann TM, et al.
Lenalidomide can be safely combined with R-CHOP
(R2CHOP) in the initial chemotherapy for aggressive
B-cell lymphomas: phase I study. Leukemia, 2011, Jul 01.
[Epub ahead of print]
Pfister B, Diehl V, Greb A, et al. International harmonization
of trial parameters in malignant lymphoma. Eur J
Haematol, 2005;75:53–54.
Solal-Céligny P, Roy P, Colombat P, et al. Follicular
lymphoma international index. Blood,
2004;104:1258–1265.
Swerdlow SH, Campo E, Harris HL, et al. (eds). WHO
Classification of Tumours of Haematopoietic and
Lymphoid Tissues. IARC: Lyon 2008.
The International Non-Hodgkin’s Lymphoma Prognostic
Factors Project. A predictive model for aggressive
non-Hodgkin’s lymphoma. New Engl J Med,
1993;329:987–994.
The Non-Hodgkin’s Lymphoma Pathologic
Classification Project. National Cancer Institute
sponsored study of classifications of non-Hodgkin’s
lymphomas: summary and description of a working
formulation for clinical usage. Cancer,
1982;49:2112–2135.
Phase I, II, III trials
Bryant J, Day R. Incorporating toxicity considerations into
the design of two-stage phase II clinical trials. Biometrics,
1995;51:1372–1383.
Conaway MR, Petroni GR. Bivariate sequential designs for
phase II trials. Biometrics, 1995;51:656–664.
Fan SK, Chaloner K. Optimal designs and limiting optimal
designs for a trinomial response. J Stat Plan Inference,
2004;126:347–360.
Garrett-Mayer E. The continual reassessment method
for dose-finding studies: a tutorial. Clin Trials,
2006;3(1):57–71.
Gooley T, Martin P, Fisher L, et al. Simulation and design
tool for phase I/II clinical trials: an example from bone
marrow transplantation. Control Clin Trials,
1994;15:45–462.
Chapter 4: Clinical trials in lymphoma
Green S, Dahlberg S. Planned versus attained design in phase
II clinical trials. Stat Med, 1992;11:853–862.
Herson J, Carter SK. Calibrated phase II clinical trials in
oncology. Stat Med, 1986;5:441–447.
Hunsberger S, Rubinstein LV, Dancey J, Korn EL. Dose
escalation trial designs based on a molecularly targeted
endpoint. Stat Med, 2005;24:2171–2181.
Korn EL. Nontoxicity endpoints in phase I trial designs for
targeted, non-cytotoxic agents. J Natl Cancer Inst,
2004;96(13):977–978.
Liu P-Y, Dahlberg S, Crowley J. Selection of designs for pilot
studies based on survival endpoints. Biometrics,
1993;49:391–308.
O’Quiqley J, Pepe M, Fisher L. Continual reassessment
method: a practical design for phase 1 clinical trials in
cancer. Biometrics, 2002;58(1):245–249.
Sargent DJ, Goldberg RM. A flexible design for multiple
armed screening trials. Stat Med, 2001;20(7):1051–1060.
Simon R, Wittes RE, Ellengerg SS. Randomized phase II
clinical trials. Cancer Treatment Rep, 1985;69:1375–1381.
Simon R, Freidlin B, Rubinstein L, et al. Accelerated titration
designs for phase I clinical trials in oncology. J Natl
Cancer Inst, 1997;89(15):1138–1147.
Storer B. Design and analysis of phase I clinical trials.
Biometrics, 1989;45:925–938.
Thall PF, Russel KE. A strategy for dose-finding and safety
monitoring based on efficacy and adverse outcomes in
phase I/II clinical trials. Biometrics, 1998;54:251–264.
Zhang W, Sargent DJ, Mandrekar S. An adaptive dose-
finding design incorporating both toxicity and efficacy.
Stat Med, Published online 11 October 2005.
Interpretation
Cox DR. Regression models and life-tables. J R Stat Soc Series
B, 1972;34:187–200.
Fleming TR. Clinical trials: discerning hype from substance.
Ann Int Med, 2010;153:400–406.
Kaplan E, Meier P. Nonparametric estimation from
incomplete observations. J Am Stat Assoc,
1958;53:457–481.
Lunceford JK, Davidian M, Tsiatis AA, et al. Estimation of
survival distributions of treatment policies in two-stage
randomization designs in clinical trials. Biometrics,
2002;58:48–57.
O’Brien PC, Fleming TR. A multiple testing procedure for
clinical trials. Biometrics, 1979;35:549–556.
Therneau TM, Grambsch PM. Modeling Survival Data:
Extending the Cox Model. New York, Springer-Verlag,
2000;161–163.
59
 Chapter 4: Clinical trials in lymphoma
Wahed AS, Tsiatis AA. Optimal estimator for the survival
distribution and related quantities for treatment policies
in two-stage randomization designs in clinical trials.
Biometrics, 2004;60:124–133.
Future issues
Cerhan SM, Ansell SM, Fredericksen ZS, et al. Genetic
variation in 1253 immune and inflammation genes
and risk of non-Hodgkin lymphoma. Blood,
2007;110:4455–4463.
DeMets DL, Califf RM. A historical perspective on clinical
trials innovation and leadership. Where have academics
gone? JAMA, 2011;305:713–714.
Korn EL, Freidlin B, Abrams JS. Overall survival as
the outcome for randomized clinical trials with
60
effective subsequent therapies. J Clin Oncol,
2011;29:2439–2442.
Lenz G, Wright G, Dave SS, et al. Stromal gene signatures
in large-B-cell lymphomas. N Engl J Med,
2008;359:2313–2323.
Meyer RM. Generalizing the results of cancer clinical trials.
J Clin Oncol, 2009;27:187–189.
Moss AJ, Francis CW, Ryan D. Collaborative clinical trials.
N Engl J Med, 2011;364:789–791.
Rosenwald A, Wright G, Chan WC, et al. The use of
molecular profiling to predict survival after
chemotherapy for diffuse large-B-cell lymphoma.
N Engl J Med, 2002;346:1937–1947.
Ware JH, Hamel MB. Pragmatic trials-guides to better
health care? N Engl J Med, 2011;364:1685–1687.
 Chapter
5
Hodgkin lymphoma
Stephanie Sasse and Andreas Engert
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Introduction
Hodgkin lymphoma (HL) is a malignant lymphoma
characterized by the presence of mononucleated
Hodgkin and multinucleated Reed–Sternberg (HRS)
cells in the classical variant (cHL), while lymphocytic
and histiocytic (L&H) cells are pathognomonic for the
nodular lymphocyte-predominant HL (NLP-HL). In
the majority of cases, HRS cells are most likely derived
from preapoptotic germinal-center B-cells, but have lost
most of the B-cell-typical genes, while L&H cells are
reported to originate from antigen-selected germinal
center B-cells and are characterized by the expression
of multiple B-cell markers. Many questions concerning
the pathogenesis of HL are still unanswered. Multiple
signaling pathways and transcription factors show
deregulated activity in HRS cells, including the
NF-κB-, Jak-Stat- and PI3k-pathway, and signaling
molecules such as Notch-1, which are physiologically
not activated in B-cells. Additionally, the interaction of
HRS cells with the surrounding inflammatory cells, the
microenvironment, seems to play an important role in
HRS cell growth and survival.
The prognosis of affected patients with HL depends
on stage of disease and other clinical risk factors. The
development of stage- and risk-adapted treatment regi-
mens based on modern polychemotherapy and radio-
therapy has improved the outcome dramatically over
the past few decades. Current strategies aim at reducing
therapy-associated complications while maintaining
high cure rates. The increasing knowledge on the biol-
ogy of HL offers the opportunity to develop targeted
treatment approaches.
Epidemiology
With an annual incidence of 2–3 per 100 000 in
Europe and the USA, HL accounts for approximately
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
30% of all malignant lymphoma. Slightly more men
than women are affected. In industrialized countries
the onset of disease shows a bimodal distribution
with a first peak in the third decade and a second,
much smaller peak, after the age of 50.
Clinical presentation of Hodgkin
lymphoma
Usually, indolent swellings localized in the cervical or
supraclavicular region (60–70%) are the first symptoms
of disease; axillary lymph nodes are observed in about
30% of newly diagnosed patients. Almost two-thirds of
patients with newly diagnosed classical HL have radio-
graphic evidence of intrathoracic involvement. Bone
marrow involvement occurs in less than 10% of newly
diagnosed patients. Details on organ involvement at
diagnosis are given in Table 5.1.
About 40% of patients, especially those with initial
abdominal involvement or advanced stage disease,
demonstrate B-symptoms. Other symptoms comprise
pain at the site of nodal involvement shortly after
drinking alcohol, pruritus, or fatigue.
Compared with cHL, NLP-HL usually begins as a
localized, slow-growing, and rather benign entity with
involvement of only one peripheral nodal region.
Pathology
Nodular lymphocyte-predominant
Hodgkin lymphoma (nLPHL)
The lymph node shows a lymphoid proliferation with
a nodular growth pattern. In most cases this is prom-
inent but in some cases the nodularity may be difficult
to discern. A proportion of cases will have areas of
preserved normal nodal architecture with reactive
61
 Chapter 5: Hodgkin lymphoma

Table 5.1 Anatomic sites of disease involved in untreated

patients with Hodgkin lymphoma. (Modified from Gupta et al.,
1999.)

Anatomic site Involvement (%)

Waldeyer`s ring 1–2
Cervical nodes 60–70
Axillary nodes 30–35
Mediastinum 50–60
Hilar nodes 15–35
Para-aortic nodes 30–40
Iliac nodes 15–20
Mesenteric nodes 1–4
Inguinal nodes 8–15 Figure 5.1 Nodular lymphocyte-predominant Hodgkin’s
Spleen 30–35 lymphoma (lymph node). The lymphoid proliferation has a nodular
growth pattern. The nodules consist mainly of small lymphocytes,
Liver 2–6 with scattered large cells and histiocytes.
Bone marrow 1–4
Total extranodal 10–15

Figure 5.3 Nodular lymphocyte-predominant Hodgkin’s

Figure 5.2 Nodular lymphocyte-predominant Hodgkin’s
lymphoma (lymph node). The large cells have lobulated nuclei and
intermediate-sized nucleoli (“popcorn cells”).
follicles, some of which may show features of progres-
sive transformation of germinal centers (Figure 5.1).
The abnormal nodules show a proliferation of small
mantle-type cells with round or slightly irregular nuclei
and scanty cytoplasm. Histiocytes are usually present
either singly or in small clusters. Within this back-
ground is a scattered population of large cells with
abundant cytoplasm and large lobulated nuclei with
open chromatin and prominent eosinophilic nucleoli.
62 These neoplastic cells are variously termed either L&H
cells after a previous designation of this type of HL
lymphoma (lymph node). The “popcorn cells” stain for CD20. They are
surrounded by CD20-negative T-cells. The remainder of the small
lymphocytes in the nodule are small B-cells.
(lymphocytic and histiocytic) in the Lukes–Butler clas-
sification or as popcorn cells in reference to the com-
plexity of nuclear convolutions (Figure 5.2). A few cells
may have large inclusion-like nucleoli and resemble
classical HRS cells, but these should not be numerous.
If significant numbers of HRS-like cells are present, the
diagnosis of cHL, nodular lymphocyte-rich type should
be excluded. Variable numbers of popcorn cells can be
found in the interfollicular zone. Occasionally these
interfollicular cells can be strikingly numerous.
Immunophenotypically the popcorn cells express
CD45 and the B-cell antigens CD20 (Figure 5.3),
CD79a, and PAX-5. They are positive for the B-cell

Figure 5.4 Nodular lymphocyte-predominant Hodgkin’s
lymphoma (lymph node). The nodules are delineated by underlying
follicular dendritic cell meshworks stained for CD21, which wrap
around the large cells and an attached rim of small lymphocytes.
transcription factors OCT2 and BOB1. Staining for
CD20 is usually strong (while CD79a may give a
weak pattern) and is accentuated by the presence of a
surrounding rim of CD20-negative T-cells. The pop-
corn cells are positive for BCL-6 but do not express
CD10. They contain immunoglobulin heavy and light
chains (almost always kappa light chain-restricted)
with the presence of J-chain. The cells are negative
for BCL-2 protein and P53 is not overexpressed. The
cells may be positive for epithelial membrane antigen
(EMA). The cells are negative for CD30 and CD15 in
paraffin sections. The cells show no evidence of
Epstein–Barr virus (EBV) infection.
The nodules of nLPHL have an underlying mesh-
work of follicular dendritic cells that can be demon-
strated by staining for CD21 or CD23 (Figure 5.4). The
small B-cells are predominantly IgD-positive with
scattered T-cells. The cells around the popcorn cells
are follicular T-helper cells with expression of CD4,
CD57, and BCL-6. The L&H cells with their T-cell
rings are frequently encircled by follicular dendritic
cell (FDC) processes. With time the background cel-
lular population can change with loss of small B-cells
and increased numbers of T-cells.
Residual germinal center cells may be present
either singly or in small clusters. CD30-positive
B-immunoblastic cells are seen in the interfollicular
areas but these are not part of the neoplastic
population.
In addition to the classical nodular form five other
variants have been described with different patterns.
The serpiginous/interconnected nodular form is
Chapter 5: Hodgkin lymphoma
characterized by nodules with irregular outlines with
some fusion of nodules. In the nodular and extranod-
ular variant the L&H cells are present in the periphery
of the nodules and in the area between the nodules.
Many cases with this pattern are associated with IgD
expression by the L&H cells. The nodular and T-cell-
rich pattern shows depletion of small B-cells within the
nodules associated with an increase in small T-cells.
There are two variants where the growth is more
diffuse – the T-cell-rich B-cell lymphoma-like variant
has a pattern where the L&H cells are present in a
background rich in small T-cells and histiocytes. In
the diffuse “moth-eaten” pattern the background
resembles the classical nodular form with a small
B-cell-rich background, but the nodular growth pat-
tern is lacking. Frequently more than one pattern is
present in an individual case. The clinical significance
of these patterns remains uncertain, but a T-cell-rich
B-cell lymphoma-like pattern has been associated with
increased rates of relapse.
nLPHL needs to be differentiated from reactive
changes in the follicles, particularly progressive trans-
formation of the germinal centers. In the latter con-
dition the classical finding of strongly CD20-positive
cells surrounded by rosettes of T-cells is not present.
Differentiation from cHL can be problematic but is
achieved by careful assessment of the immunopheno-
type of the neoplastic cells. Distinction between
nLPHL and diffuse large B-cell lymphoma of T-cell
histiocyte-rich type also may be problematic, but
nLPHL usually shows at least some residual nodularity
or follicular dendritic cell meshwork and there is usu-
ally at least a sprinkling of IgD-positive small B-cells
around the popcorn cells. Residual CD57-positive
T-cells also favour nLPHL. Distinction from other B-
cell lymphomas with a nodular growth pattern such as
follicular and mantle cell lymphoma should not be
difficult as these lack the characteristic L&H cells.
Classical Hodgkin lymphoma (cHL)
The hallmark of cHL is the presence of Reed–
Sternberg (R–S) cells or their variants. These are set
in a background that consists of a polymorphous mix
of inflammatory cells. The background cellular popu-
lation together with other changes determines the
morphological subclassification of cHL (Figure 5.5).
R–S cells are large with abundant pale eosinophilic
cytoplasm. The nuclei of the classical form of R–S cells
are bilobed and contain large round eosinophilic nucle- 63
oli that are surrounded by a thin clear rim. Variants
 Chapter 5: Hodgkin lymphoma
Figure 5.5 Classical Hodgkin’s lymphoma. Reed–Sternberg cells are
large with abundant cytoplasm. Classical variants have a bilobed
nucleus with prominent large inclusion-type nucleoli.
from the classical R–S cell include mononuclear forms
(sometimes termed Hodgkin cells) and cells with more
complex multilobated nuclei. Some R–S cells have more
deeply staining eosinophilic cytoplasm with pyknotic
nuclei (often referred to as mummified cells).
Immunophenotypically the R–S cells and variants
express CD30. Expression of CD15 is variable and is
highly dependent on the sensitivity of the detection
method employed, but is probably positive in 75–85%.
Positive cases frequently show staining in only a pro-
portion of the neoplastic cell population. R–S cells are
now considered to be derived from mature B-cells and
they express IRF4/Mum1 and the B-cell-specific tran-
scription factor PAX-5, but they are negative for CD45
and only about 40% express CD20, with a smaller
proportion expressing CD79a. Expression of CD20 is
frequently heterogeneous in intensity, and in most
cases the staining is not seen in all neoplastic cells.
Staining for the transcription factor OCT2 and its
coactivator BOB1 is usually absent, although a small
proportion of cases may be positive for one or other of
the two. Expression of both together is not seen. The
cells are negative for J-chain. R–S cells stain with anti-
bodies against fascin but most, although not all, ana-
plastic large cell lymphomas (T and null types) also
stain with antibodies to this antigen.
The R–S cells are usually surrounded by a corona
of small reactive T-cells, making assessment of T-cell
antigen expression difficult. In general R–S cells do not
express T-cell-related antigens, although aberrant
expression of these may be seen in a small number of
64 cases. There is no staining for ALK kinase protein, and
R–S cells are generally negative for EMA.
Figure 5.6 Nodular lymphocyte-rich classical Hodgkin’s lymphoma
(lymph node). The lymphoid population has a modular/follicular
growth pattern. The nodules contain a background of small
lymphocytes with scattered large cells.
Expression of EBV latent membrane protein-1
(LMP-1) is variable and is dependent on morphological
subtype. R–S can overexpress p53 protein. Expression of
bcl-2 protein is variable and both P53 and BCL-2 pro-
tein expression may be prognostically important,
although this remains controversial. Proliferation in
the R–S cells is high and there is also significant prolif-
erative activity within the background reactive lym-
phoid population.
Lymphocyte-rich classical Hodgkin
lymphoma (LR-cHL)
In this subtype the background population is predom-
inantly composed of small lymphocytes (Figures 5.6
and 5.7). The growth pattern is usually nodular. The
nodules are lymphoid follicles with expanded mantle
zones. They may contain residual germinal centers.
The R–S cells are scattered within the expanded mantle
zones. This subtype may therefore mimic the morpho-
logical appearance of NLP-HL. Eosinophils are scanty
or absent. A rare diffuse form also exists.
Immunophenotypically the R–S cells show the typ-
ical immunophenotype with expression of CD30 and
CD15 (in distinction from the cells of NLP-HL)
(Figures 5.8 and 5.9). The R–S cells are rimmed by
T-cells but the majority of the associated lymphocyte-
rich background consists of mantle-type small B-cells.
The nodules also contain disrupted follicular dendritic
cell meshworks (highlighted by staining for CD21 or
CD23).

Figure 5.7 Nodular lymphocyte-rich classical Hodgkin’s lymphoma
(lymph node). The large cells have the morphology of Reed–
Sternberg cells rather than popcorn cells, helping to distinguish this
from nodular lymphocyte-predominant Hodgkin’s lymphoma.
Figure 5.9 Nodular lymphocyte-rich classical Hodgkin’s lymphoma
(lymph node). Further distinction from lymphocyte-predominant
Hodgkin’s lymphoma is the strong positive staining of the Reed–
Sternberg cells for CD30 (popcorn cells are CD30-negative).
Mixed cellularity Hodgkin lymphoma
(MC-HL)
In this subtype the nodal architecture is effaced by a
polymorphous mixture of cells within which classical
R–S cells are found. This background contains lym-
phocytes, histiocytes, plasma cells, eosinophils, and
neutrophils in proportions that vary from case to
case. The nodal capsule is not thickened and there
are no well-formed sclerotic fibrous bands formed of
mature collagen, although some interstitial fibrosis
may be seen. In some cases the nodal architecture is
not completely effaced and the infiltrate occupies the
interfollicular areas (Figure 5.10).
Chapter 5: Hodgkin lymphoma
Figure 5.8 Nodular lymphocyte-rich classical Hodgkin’s lymphoma
(lymph node). The Reed–Sternberg cells are negative for CD20.
Figure 5.10 Mixed-cellularity Hodgkin’s lymphoma. Reed–
Sternberg cells are present within a mixed background rich in small
lymphocytes, histiocytes, eosinophils, and plasma cells. There is no
nodule formation and no mature, banded sclerosis.
Immunophenotypically the R–S cells show the typ-
ical antigen profile but expression of EBV LMP-1 is
frequent.
Nodular sclerosis Hodgkin lymphoma
(NS-HL)
The lymph node usually shows thickening of the cap-
sule. There are fibrosclerotic bands that extend from the
capsule into the node pulp with formation of cellular
nodules. The degree of nodularity varies between cases
and the nodules may contain residual lymphoid fol-
licles. Within the cellular nodules R–S cells are present 65
in variable numbers. There may be confluent sheets of
 Chapter 5: Hodgkin lymphoma
Figure 5.11 Nodular sclerosing Hodgkin’s lymphoma (lymph
node). The node architecture is effaced. There are cellular nodules
that are separated from each other by sclerotic bands. The nodules
contain Reed–Sternberg cells in a background of small lymphocytes,
histiocytes, eosinophils, and plasma cells.
R–S cells and there may be areas of necrosis. In
formalin-fixed tissue a proportion of the R–S cells
may show artifactual separation from the surrounding
rim of lymphocytes (lacunar cells) (Figure 5.11).
A grading system has been formulated by the
British National Lymphoma Investigation (BNLI) for
assessment based on the number of R–S cells and the
background cellular population. In grade 1, 75% of the
nodules contain scattered R–S cells in a lymphocyte-
rich polymorphous infiltrate. In grade 2, at least 25%
of the nodules contain sheets of R–S cells with a
relatively lymphocyte-depleted background. There is
frequently associated necrosis in the nodules. Grade 2
lesions have, in the past, been called nodular sclerosis
with lymphocyte depletion.
Immunophenotypically the R–S cells show a typi-
cal antigen profile, but EBV LMP-1 is less frequently
expressed (10–40%).
Lymphocyte-depleted Hodgkin
lymphoma (LD-HL)
The lymph node is effaced by a population that contains
a relative predominance of R–S cells over background
cells. In some cases the R–S cells show marked pleo-
morphism with a sarcomatous appearance. In a pro-
portion of cases there is diffuse fibrosis (without nodule
formation) with scattered R–S cells (Figure 5.12).
66 Immunophenotypically the R–S cells show the
characteristic antigen profile.
Figure 5.12 Lymphocyte-depleted Hodgkin’s lymphoma. A
proliferation of pleomorphic cells including occasional Reed–
Sternberg cells in a background that shows a paucity of lymphocytes.
Rare patterns of nodal infiltration
In some cases the pattern of infiltration does not fall into
one of the classical patterns described above. In some
cases the infiltrate is subtle, with the HRS cells present in
the interfollicular area. This pattern may indicate early
infiltration of the node and can be seen when small
nodes from the margin of an involved area are sampled.
In a few cases the background can contain a prominent
histiocytic proliferation with the HRS cells surrounded
by granulomatous-like inflammation that can be mis-
taken for sarcoid or toxoplasmosis. Occasionally HRS
cells are seen within areas of monocytoid B-cell hyper-
plasia. In very rare cases the HRS cells can be seen within
sinusoids in a pattern that can mimic anaplastic large
cell lymphoma.
B-cell lymphoma, unclassifiable, with
features intermediate between diffuse
large B-cell lymphoma and classical
Hodgkin lymphoma
The current 2008 fourth edition of the WHO classifica-
tion of tumors of the hemopoietic and lymphoid tissues
recognized the presence of a group of lymphomas in
which there are overlapping features between diffuse
large B-cell lymphoma (DLBCL) and cHL. These lym-
phomas, previously designated “gray-zone lymphomas,”
are most frequently encountered in the mediastinum,
consistent with the probable overlap in cell of origin
between primary mediastinal large B-cell lymphoma
and cHL. These cases show either morphological

and/or immunophenotyic features that make differen-
tiation between DLBCL and cHL impossible.
Morphologically the cells are frequently seen in
sheets and may be associated with sclerosis. The cells
are pleomorphic and there are often areas that have a
distinctly cHL appearance adjacent to other areas that
are more typical of DLBCL. Immunophenotypically, the
cells demonstrate a DLBCL/cHL overlap, with expres-
sion of CD45 and preservation of the B-cell repertoire
(CD20 and CD79a) associated with expression of CD30
and, in most cases, CD15. The B-cell transcription fac-
tors OCT2 and BOB1 are frequently expressed together.
Molecular pathology
and cytogenetics
The molecular and genetic analysis of the malignant
tumor cells in this lymphoma, the HRS cells, has long
been hampered by the paucity of the neoplastic cells
(1–2%) in the lymphoma specimen. However, advan-
ces in tissue microdissection techniques have enabled a
more detailed molecular genetic analysis of HRS cells.
HRS cells in cHL have a pronounced lineage infi-
delity concerning the expression of antigen markers
and frequently express myeloid (CD15), B-cell (Pax-
5), plasma cell (MUM-1 and syndecan), and activation
markers (CD30). The analysis of IgVH genes, however,
demonstrated that the vast majority of HRS cells har-
bor clonal Ig rearrangements that represent a molec-
ular marker of B-cell derivation. Interestingly, a
proportion of these cases (25%) harbors destructive
Ig rearrangements with crippling mutations. Since
such crippling mutations would result in apoptotic
cell death in reactive germinal center (GC) B-cells, it
has been postulated that HRS cells might be derived
from preapoptotic GC B-cells. Additional evidence
came from the analysis of composite lymphomas con-
sisting of simultaneous HL and non-HL tumors that
demonstrated a clonal relationship between the two
lymphomas and a mutational pattern characteristic of
GC B-cells. Expression of T-cell markers also occurs in
a subset of cHL and a proportion of these cases harbor
clonal T-cell receptor rearrangements and germline Ig
configuration, pointing to the existence of rare T-cell-
derived HL. In contrast to cHL, in NLP-HL, L&H cells
show an immunophenotype consistent with B-cell
derivation and harbor clonal and somatically mutated
IgVH genes with ongoing mutations.
Although derived from B-cells, HRS cells usually
lack B-cell lineage marker expression. This fact was
Chapter 5: Hodgkin lymphoma
confirmed in gene expression profiling studies that
revealed global downregulation of the B-cell lineage
gene repertoire, including surface molecules, kinase
signaling, and B-cell-specific transcription factors
(OCT2, BOB1, and PU1). Recently, it was demonstra-
ted that the loss of B-cell identity in cHL results from
the aberrant expression of the transcription factors
ABF-1 and Id2, leading to downregulation of B-cell
genes, but also to expression of genes not usually asso-
ciated with B-cell lineage. A characteristic feature of
HRS cells is constitutive NFκB activity, which is medi-
ated by inactivating mutations of the NFκB inhibitor
IκBα in 30% of the cases, or, to a lesser extent, of IκBε.
Additionally, genomic amplification of a component of
NFκB (c-Rel) is frequently demonstrated in cHL.
HL cases have been demonstrated to harbor com-
plex clonal chromosomal abnormalities with fre-
quently hyperdiploid karyotypes. Some recurrent
chromosomal gains in cHL involve JAK-2 (9p23–24),
REL/BCL11a (2p13–16) and MDM2 (12q14), provid-
ing possible explanations for the constitutive NF-κB
and STAT member family expression in HL. In a
thorough investigation of NLP-HL involving micro-
dissection, degenerate oligonucleotide-primed poly-
merase chain reaction (DOP-PCR), and comparative
genome hybridization (CGH) analysis chromosomal
regions with a gain included 1p, 1q, 2q, 3p, 3q, 4q, 5q,
6p, 6q, 8q, 11q, 12q and X, and losses were demon-
strated at 17/17p. Chromosomal translocations occur
in a fraction of HL; in NLP-HL, BCL-6 rearrangements
were identified in up to 48% of cases.
To identify potential transforming events able to
rescue HRS cell precursors from apoptosis, candidate
genes like TP53, CD95 (Fas), and N-RAS were inves-
tigated, but no mutations were found. HRS cells,
nevertheless, are resistant to Fas-mediated apoptosis,
but no alterations of downstream effectors of the Fas
cascade were found. Very recently, inactivating muta-
tions of A20, an NFkB inhibitory protein, were
described in a significant subset of classical HL.
Approximately 40% of cHL cases in the western
world are associated with infection of EBV, whereas in
some geographic regions like Latin America such
involvement can reach up to 90% of cases depending
on age and histological subtype. The transforming
capabilities of the viral latent membrane protein 1
(LMP-1) activating NFκB can be considered a poten-
tial transforming event after infection of HRS cell
precursors by rescuing them from apoptosis. It is note-
worthy that the viral latent membrane protein 2 67
 Chapter 5: Hodgkin lymphoma
(LMP-2) is also expressed in infected HRS cells and is
able to mimic B-cell receptor signaling normally defec-
tive in HRS cells.
HRS cells in cHL commonly express caspase-3, a
component of the extrinsic and intrinsic apoptosis
pathways. Functional studies, however, demonstrated
that caspase-3 is non-functional in cultured HRS cells.
c-FLIP and X-linked inhibitor of apoptosis molecules,
both of which represent NFκB target genes, inhibit
caspase-3 activation, therefore hampering effective
apoptosis. In line with these findings, small interfering
RNAs directed to c-FLIP were able to reconstitute the
apoptosis pathway.
The microenvironment of cHL sets the stage for a
crosstalk between clonal and non-clonal elements. A
micromilieu is built up in which cytokine and che-
mokine production promotes survival and escape
from immune surveillance. For example, HRS cells
produce Th2-related cytokines (IL-4 and IL-13) and
Th1 and CD8 inhibitory cytokines (IL-10 and TGFβ)
and, thus, create a favorable milieu for HRS to resist
cell-mediated apoptosis. Adding to this scenario,
the production of several chemokines, including
TARC, MDC, IP-10, and eotaxin, by infiltrating
cells supports the favorable microenvironment for
the tumor clone. Of note, there is an overall higher
expression of chemokines in EBV-related cHL cases,
possibly reflecting additional mechanisms of ham-
pered immunosurveillance in this cHL subgroup.
Diagnosis, staging, and risk
stratification
Diagnosis and staging
An excisional biopsy of a suspicious lymph node is
indispensable for the diagnosis of HL. HL patients are
treated according to stage and risk factors. The histo-
logical subtype – except NLP-HL – does not influence
the treatment decision. The stage of disease is assessed
with the Cotswolds classification, a modified version
of the Ann Arbor classification, which was published
by Lister (Table 5.2).
The clinical staging includes chest X-ray, abdomi-
nal sonography, CT scans of the neck, thorax, abdo-
men and pelvis, bone marrow biopsy, and bone
marrow or skeletal radionuclide imaging. In some
cases, additional tests such as MRI, PET, or a liver
biopsy might be indicated. So far, the role of 18FDG-
68 PET in the initial diagnostic procedure has not been
clearly defined.
Risk stratification
The prognostic factors used by the German Hodgkin
Lymphoma Study Group (GHSG) for risk stratifica-
tion of HL patients include stage, B-symptoms, bulky
disease of mediastinal lymphoma manifestation
defined as greater than one-third mediastinal widen-
ing, extranodal manifestation, at least three involved
nodal areas, and an elevated erythrocyte sedimentation
rate. In North America, most centers treat patients
according to the traditional classifications of early
(CS I–IIA or B) and advanced stages (CS III–IVA or
B, CS I–II A or B with bulky disease defined as >10 cm
maximum dimension of nodal mass). European study
groups such as the EORTC (European Organisation
for Research and Treatment of Cancer), the GELA
Table 5.2 The Cotswold staging classification for Hodgkin
lymphoma (extracted and adapted from Lister et al., 1989).
Cotswold staging classification
Stage I Involvement of a single lymph node region or
lymphoid structure (e.g. spleen, thymus,
Waldeyer‘s ring) or involvement of a single
extralymphatic site (IE)
Stage II Involvement of two or more lymph node regions
on the same side of the diaphragm; localized
contiguous involvement of only one extranodal
organ or site and lymph node region(s) on the
same side of the diaphragm (IIE)
The number of anatomic regions involved
should be indicated by a subscript (e.g. II3)
Stage III Involvement of lymph node regions on both
sides of the diaphragm (III), which may also be
accompanied by involvement of the spleen
(IIIS) or by localized contiguous involvement
of only one extranodal organ site (IIIE) or both
(IIISE)
III1 With or without involvement of splenic, hilar,
celiac, or portal nodes
III2 With involvement of para-aortic, iliac, and
mesenteric nodes
Stage IV Diffuse or disseminated involvement of one or
more extranodal organs or tissues, with or
without associated lymph node involvement
Designations applicable to any disease stage.
A No symptoms
B Fever (temperature, >38°C), drenching night
sweats, unexplained loss of 10% of body weight
within the preceding 6 months
X Bulky disease (a widening of the mediastinum
by more than one-third of the presence of a
nodal mass with a maximal dimension greater
than 10 cm)
E Involvement of a single extranodal site that is
contiguous or proximal to the known nodal site
 Chapter 5: Hodgkin lymphoma

Table 5.3 Definition of treatment groups according to the EORTC/GELA and GHSG.

Treatment groups EORTC/GELA GHSG

Early stage favorable CS I–II without risk factors (supradiaphragmatic) CS I–II without risk factors
Early stage unfavorable (intermediate) CS I–II with ≥1 risk factors (supradiaphragmatic) CS I, CSIIA ≥1 risk factors;
CS IIB with C/D but without A/B
Advanced stage CS III–IV CS IIB with A/B;
CS III–IV
Risk factors (RF) A large mediastinal mass A large mediastinal mass
B age ≥50 years B extranodal disease
C elevated ESR* C elevated ESR*
D ≥4 involved regions D ≥3 involved areas
Abbreviations: GHSG, German Hodgkin Lymphoma Study Group; EORTC, European Organisation for Research and Treatment of Cancer;
GELA, Groupe d’Etude des Lymphomes de l’Adulte.
* Erythrocyte sedimentation rate (≥50 mm/h without or ≥30 mm/h with B-symptoms).

(Groupe d’Etudes des Lymphomes de l’Adulte), and
the GHSG classify patients into early favorable,
early unfavorable, and advanced stages depending
on the clinical factors listed in Table 5.3. A recently
performed retrospective analysis of the GHSG con-
firmed the need to differentiate an early favorable
and unfavorable risk group, and identified the risk
scores of the GHSG, the EORTC and of the
National Comprehensive Cancer Network (NCCN)
as equally suitable to define the early unfavorable
risk group.
To define the risk to patients with advanced disease
more precisely, the “International Prognostic Score”
(IPS) was developed. The IPS consists of seven factors,
which were significantly related to an unfavorable prog-
nosis when present at initial diagnosis of HL: serum
albumin <4 g/dL, hemoglobin <10.5 g/dL, male sex, age
≥45 years, stage IV disease, leukocytosis >15 000/mm³,
lymphocytopenia <600/mm³ and/or <8% of white cells.
Response assessment
In addition to CT restaging, metabolic imaging with
2-[18F]fluoro-2-deoxyglucose–positron emission
tomography (18FFDG–PET) has gained an increasing
relevance in response assessment of HL patients. A
significant prognostic value of negative 18FDG–PET
scans after completion of the chemotherapy courses
could be verified in the HD15 trial of the GHSG. HL
patients included in the HD15 trial were randomly
assigned to receive six or eight courses of a bleomycin,
etoposide, adriamycin, cyclophosphamide, vincristine,
procarbazine, prednisone (BEACOPP)-escalated-
based chemotherapy. Residual disease in the CT scan
after completion of chemotherapy was further
assessed by 18FDG–PET scan. Only those patients
with 18FDG–PET-positive residual lesions received
consolidation radiotherapy. For those patients with a
negative 18FDG–PET scan, a progression-free survival
(PFS) of 92.1% could be documented after a follow-up
of 36 months; this was about the same as the PFS of
those patients with complete remission in the CT
scan. The negative predictive value of 18FDG–PET
scan determined in this analysis was 94.6% (95% CI,
92.7–96.6%). Thus, this preliminary analysis indicates
that consolidation radiotherapy can be omitted in
advanced stage HL patients with 18FDG–PET-negative
residual disease after treatment with BEACOPP-
escalated. The negative predictive value of 18FDG–
PET scan after completion of chemotherapy in early
favorable and unfavorable HL is still being evaluated;
so far there are data supporting a PET-guided radio-
therapy approach in early favorable and unfavorable
HL out of a clinical trial.
Response-adapted chemotherapy treatment of
HL might also be developed on the basis of interim
18
FDG–PET scan results. Several phase II trials reported
a negative predictive value of over 90% for interim
18
FDG–PET scans; additionally, a strong correlation
between result of the early interim 18FDG–PET scan
and PFS could be identified. However, the negative
predictive value might vary dependent on the chemo-
therapy regimen. The ongoing HD18 trial is evaluating
a response-adapted, BEACOPP-escalated-based chemo-
therapy approach in advanced HL patients dependent 69
on the interim 18FDG–PET scan results.
 Chapter 5: Hodgkin lymphoma
Treatment
First line treatment
Early-stage favorable Hodgkin lymphoma
Until the 1990s, early-stage favorable HL was treated
with extended-field radiation (EF-RT). Because of the
high incidence of relapse (25–30%) after EF-RT alone
and fatal long-term effects after radiotherapy, new
treatment strategies were developed combining
short-duration chemotherapy with radiotherapy.
The Southwest Oncology Group (SWOG) reported
in 2001 that patients treated with combined modality
treatment, consisting of three cycles of doxorubicin and
vinblastine followed by subtotal nodal radiotherapy
(STNI) had a significantly better outcome in terms of
freedom from treatment failure (FFTF) than those
patients receiving subtotal lymphoid irradiation alone.
The EORTC/GELA H7F and H8F trials demonstrated a
significantly higher event-free survival (EFS) for a com-
bined modality treatment consisting of six courses of
EBVP (epirubicin, bleomycin, vinblastine, prednisone)
or three courses of mechlorethamine, vincristine, pro-
carbazine, prednisone, doxorubicin, bleomycin, and
vinblastine (MOPP/ABVD) followed by involved-field
radiation (IF-RT) than for subtotal nodal radiotherapy
alone. Additionally, the H8F trial showed an overall
survival benefit for those patients treated with the com-
bined modality arm. The combined modality treat-
ment, including ABVD (doxorubicin, bleomycin,
vinblastine, dacarbazine), which is less toxic than
MOPP/ABVD, as a chemotherapy regimen was estab-
lished in the HD7 trial by the GHSG. In this trial, two
cycles of ABVD plus EF-RT were shown to be superior
to EF-RT alone in terms of freedom of treatment failure
(FFTF) (Table 5.4). The results of the H7F trial addi-
tionally indicated that the integration of ABVD/ABVD-
like chemotherapy allows the radiation field and dose to
be reduced without compromising efficacy.
To reduce treatment-associated toxicity, the subse-
quent trials evaluated strategies to reduce treatment
intensity. In the HD10 trial of the GHSG, possible
reduction of chemotherapy from four to two cycles of
ABVD in combination with IF-RT and a dose reduction
of IF-RT from 30 Gy to 20 Gy were examined. Neither
the two chemotherapy arms nor the different radiation
doses differed significantly with respect to FFTF or
overall survival (OS). Based on these results, two
70 courses of ABVD followed by 20 Gy were defined as
standard treatment for early-stage HL.
The aim of the HD13 trial was to examine if the
presumably less effective chemotherapy substances, bleo-
mycin and dacarbazine, can be omitted from the ABVD
regime. Because of the significantly increased number of
relapses and progressions, the treatment arms without
dacarbazine had to be closed prematurely after an interim
analysis in June 2006. The assessment of the comparison
of AVD with ABVD is currently being performed.
The question as to whether a consolidating radio-
therapy is indispensable in early-stage favorable
Hodgkin lymphoma is still a matter of debate. The avail-
able clinical trials including the H.6 trial, published by
Meyer et al., 2011, and the retrospective analysis of Hay
et al., 2012, have reported a significantly reduced tumor
control in those treatment arms without radiotherapy.
Therefore combined modality treatment can still be
regarded as standard. Whether radiotherapy can be omit-
ted in those patients with PET-negativity after two
courses of ABVD has been addressed in the following
prospective clinical trials. In the ongoing HD16 trial of
the GHSG, the experimental arm of the HD16 trial
evaluates if a consolidating radiotherapy can be omitted
in the case of a negative 18FDG–PET scan after two
courses of ABVD. In the HF10 trial conducted by
EORTC/GELA, the standard treatment arm consists of
three courses of ABVD followed by involved node (IN)-
radiotherapy. In the experimental arm, patients with
negative 18FDG–PET scan after two courses of ABVD
received two additional courses of ABVD without con-
solidating radiotherapy. In the case of positive interim
18
FDG–PET, two courses of BEACOPP-escalated are
administered followed by IN-RT. Based on a recently
reported first interim analysis, which showed more relap-
ses and progessions in the non-RT PET-negative arm, it
was concluded that it is unlikely that the study will show
that chemotherapy alone is non-inferior to combined
modality treatment; therefore, it was decided to treat all
18
FDG–PET-negative patients with standard combined
modality treatment, regardless of randomization.
Similarly to the H10 trial, the results of the “RAPID”
trial of the UK “National Cancer Research Institute”
(NCRI) indicate that also those patients with a negative
18
FDG-PET after chemotherapy might benefit from
consolidating radiotherapy (Radford et al., 2012). The
question of a PET-guided radiotherapy approach will
hopefully be answered by the HD16 trial (Andrè M.,
et al., 2012).
A combined modality approach consisting of two
courses of ABVD followed by 20 Gy IF-RT is regarded
as the standard of care for patients with early-stage
 Chapter 5: Hodgkin lymphoma

Table 5.4 Selected trials for early-stage favorable Hodgkin lymphoma.

Trial Therapy regimen Number of Outcome References

patients
SWOG #9133 A. 3 (dox.+ vinbl.) + STLI 165 94% FFTF; 98% OS Press et al.,
(36–40 Gy) 2001
B. STLI (36–40 Gy) 161 81% FFTF; 96% OS; [3 years]
Milan A. 4 ABVD + STLI 65 97% FFP; 93% OS Bonadonna
1990–97 B. 4 ABVD + IF-RT 68 97% FFP; 93% OS; [5 years] et al., 2004
EORTC/ A. 6 EBVP + IF-RT (36 Gy) 168 88% EFS; 92% OS Noordijk
GELA H7F B. STNI 165 78% EFS; 92% OS; [10 years] et al., 2006
EORTC/ A. 3 MOPP/ABV + IF-RT 542 93% EFS; 97% OS Fermé et al.,
GELA H8F (36 Gy) 2007
B. STNI 68% EFS; 92% OS; [10 years]
EORTC/ A. 6 EBVP + IF-RT (36 Gy) 783 87% EFS; 98% OS Noordijk
GELA H9F B. 6 EBVP + IF-RT (20 Gy) 84% EFS; 98% OS; [4 years] et al., 2005
C. 6 EBVP C closed because of high relapse rate
GHSG HD7 A. EF-RT 30 Gy (40 Gy IF) 305 67% FFTF; 92% OS Engert et al.,
B. 2 ABVD + EF-RT 30 Gy 312 88% FFTF; 94% OS; [7 years] 2007
(40 Gy IF)
GHSG HD10 A. 4 ABVD + IF-RT (30 Gy) 1190 87.2% FFTF; 94.4% OS Engert et al.,
B. 4 ABVD + IF-RT (20 Gy) 89.9% FFTF; 94.7% OS 2010
C. 2 ABVD + IF-RT (30 Gy) 85.5% FFTF; 93.6% OS
D. 2 ABVD + IF-RT (20 Gy) 85.9% FFTF; 95.5% OS [8 years]
GHSG HD13 A. 2 ABVD + IF-RT (30 Gy) Treatment arms B and D prematurely closed;
B. 2 ABV + IF-RT (30 Gy) Publication in progress
C. 2 AVD + IF-RT (30 Gy)
D. 2 AV + IF-RT (30 Gy)
GHSG HD16 A. 2 ABVD + IF-RT (20 Gy) Ongoing
(PET+/–)
B. 2 ABVD +/– IF-RT
(PET +/–)
EORTC/GELA A. 3 ABVD + IN-RT (PET +/–) Ongoing: after first interim analysis, all patients
H10 F B. 2 ABVD + 2 ABVD in the experimental arm receive consolidating
(PET–) RT
2 ABVD + 2 BEACOPP esc.
+ IF-RT (PET+)
Abbreviations: SWOG, Southwest Oncology Group; EORTC, European Organisation for Research and Treatment of Cancer; GELA, Groupe
d‘Etude des Lymphomes de l’Adulte; GHSG, German Hodgkin Lymphoma Study Group; EF/IF-RT, extended/involved-field radiotherapy;
STLI, subtotal lymphoid irradiation; STNI, subtotal nodal irradiation; IN-RT, involved node radiotherapy; FFTF, freedom of treatment failure;
RFS, relapse-free survival; RT, radiotherapy; FFP, freedom from progression; EFS, event-free survival; OS, overall survival.

favorable HL. Treatment-associated toxicity might be
further reduced by implementing a targeted drug such
as the anti-CD30 immunoconjugate bentuximab vedo-
tin and by further reducing radiotherapy intensity.
Early-stage unfavorable disease
Early-stage unfavorable HL has been shown to be
associated with a significantly worse outcome than
early-stage favorable HL and has therefore to be treated
more intensively. So far, combined modality treatment
has to be regarded as standard in early-unfavorable
HL. However, best chemotherapy, optimal number of
cycles, and the radiotherapy regime have not been
clearly defined and there is an ongoing desire to opti-
mize therapy in this risk group.
Several trials have shown that the reduction of field
size to IF-RT does not compromise the efficacy of treat-
ment. The Milan trial, which compared subtotal nodal
irradiation (STNI) with IF-RT after four courses of
ABVD in patients with early-stage favorable and unfav-
orable stages (see Table 5.4), reported a similar treat-
ment outcome in both arms. The H8U, conducted by
EORTC/GELA, randomized patients to six cycles of
MOPP/ABV + 36 Gy IF-RT, four cycles of MOPP/
ABV + 36 Gy IF-RT, and four cycles of MOPP/ABV 71
+ STNI. There was no difference between the arms in
 Chapter 5: Hodgkin lymphoma

Table 5.5 Selected trials for early-stage unfavorable Hodgkin lymphoma.

Trial Therapy regimen Number Outcome References

of patients
EORTC/ A. 6 EBVP + IF-RT (36 Gy) 194 68% EFS; 79% OS Noordijk
GELA B. 6 MOPP/ABV + IF-RT (36 Gy) 195 88% EFS; 87% OS [10 years] et al., 2006
H7U
EORTC/ A. 6 MOPP/ABV + IF-RT (36 Gy) 335 82% EFS; 88% OS Fermé et al.,
GELA B. 4 MOPP/ABV + IF-RT (36 Gy) 333 80% EFS; 85% OS 2007
H8U C. 4 MOPP/ABV + STNI 327 80% EFS; 84% OS; [10 years]
GHSG A. 2 COPP + ABVD + EF-RT 532 80% FFTF; 80% SV Sasse et al.,
HD8 (30 Gy) + Bulk (10 Gy) 2012
B. 2 COPP + ABVD + IF-RT (30 Gy) + 532 86% FFTF; 87% SV; [10 years]
Bulk (10 Gy)
EORTC/ A. 6 ABVD + IF-RT 808 94 % EFS; 96% SV Noordijk
GELA B. 4 ABVD + IF-RT 89 % EFS; 95% SV et al., 2005
H9U C. 4 BEACOPP bas.+ IF-RT 91 % EFS; 93% SV; [4 years]
GHSG A. 4 ABVD + IF-RT (30 Gy) 343 88% FFTF; 95% SV Eich et al.,
HD11 B. 4 ABVD + IF-RT (20 Gy) 339 83% FFTF; 95% OS 2010
C. 4 BEACOPP bas. + IF-RT (30 Gy) 332 89% FFTF; 96% OS
D. 4 BEACOPP bas. + IF-RT (20 Gy) 337 89% FFTF; 97% OS; [5 years]
GHSG A. 4 ABVD + IF-RT (30 Gy) 818 89% FFTF; Engert et al.,
HD14 B. 2 BEACOPP esc. + 2 ABVD + IF-RT 805 95% FFTF; [4 years] 2010
(30 Gy)
EORTC/ A. 4 ABVD + IN-RT 30 Gy (PET +/–) Ongoing, after first interim analysis: all
GELA B. 2 ABVD + 4 ABVD (PET–) patients in the experimental arm receive
H10U C. 2 ABVD + 2 BEACOPP esc. + IN-RT consolidating RT
30 Gy (PET+)
GHSG 2 BEACOPP esc. + 2 ABVD ⇒PET scan Ongoing
HD17 A. PET +: 20 Gy IF-RT/20 Gy IN-RT
B. PET–: 20 Gy IF-RT/no RT
Abbreviations: SWOG, Southwest Oncology Group; EBVP, epirubicin, bleomycin, vinblastine, prednisone; EORTC, European Organisation for
Research and Treatment of Cancer; GELA, Groupe d’Etude des Lymphomes de l’Adulte; GHSG, German Hodgkin Lymphoma Study Group;
ECOG, Eastern Cooperative Oncology Group; EF-/IF-RT, extended-/involved-field radiotherapy; IN-RT, involved node radiotherapy; STNI, subtotal
nodal irradiation; FFTF, freedom from treatment failure; RFS, relapse-free survival; RT, radiotherapy; EFS, event-free survival; OS, overall survival.

terms of response rates, EFS, or OS. The HD8 trial of
the GHSG, the largest trial investigating radiotherapy
reduction, confirmed these results: patients with early-
stage unfavorable disease were randomized to two alter-
nating cycles of COPP (cyclophosphamide, vincristine,
procarbazine, prednisone)/ABVD followed by 30 Gy
radiotherapy in either EF (arm A) or IF (arm B) tech-
nique. Final results at 5 years and the follow-up analysis
at 10 years did not disclose significant differences
between the two arms in terms of FFTF and OS; how-
ever, more toxicity was reported in the patients who
were treated with EF-RT (Table 5.5).
Despite the excellent initial remission rates obtained
with ABVD and radiotherapy, approximately 15% of
patients in the early unfavorable stage relapse within
5 years and about another 5% have primary progressive
72 disease. Efforts have been made to improve the efficacy
of chemotherapy by altering drugs and schedules as well
as the number of cycles. Alternation or hybridization of a
MOPP-like regimen with ABVD did not produce better
outcomes when compared with data reported for ABVD
alone nor did the increase of chemotherapy courses.
Because of these disappointing outcome rates more
intensive chemotherapy regimes, which were originally
developed for the treatment of advanced stages, have
been evaluated for the treatment of early unfavorable
stage (Table 5.5). The HD11 trial performed by GHSG
and the H9U trial conducted by EORTC/GELA com-
pared four cycles of ABVD and four cycles of
BEACOPP-baseline. The H9U trial additionally ana-
lysed whether six cycles of ABVD are more effective
than four cycles of ABVD. At a median follow-up of 4
years, the treatment arms of the H9U trial did not show
any difference with respect to EFS and OS. The analysis
of the GHSG HD11 trial did not show any significant
difference between ABVD and BEACOPP-baseline

either. Further treatment intensification was evaluated
in the HD14 trial, which randomized patients to two
cycles of BEACOPP-escalated plus two cycles of ABVD
or four cycles of ABVD followed by 30 Gy IF-RT. After a
median follow-up of 4 years a significant difference of
PFS was detected, implicating an improved tumor
control by two courses of BEACOPP-escalated followed
by two courses of ABVD (Table 5.5). Despite an
increased hematotoxicity, no differences with regard
to treatment-related deaths and the incidence of
secondary neoplasia were documented between the
treatment arms. Therefore, “2+2” is regarded as the
standard chemotherapy regimen in patients with
early-unfavorable HL patients and with an age of up
to 60 years.
The question of whether a PET-guided radiotherapy
approach is feasible after “2+2” is being addressed in the
current HD 17 trial of the GHSG. In the HD17 trial, in
those patients with positive 18FDG–PET scan, consoli-
dative 30 Gy IF-RT is compared with 30 Gy IN-RT.
18
FDG–PET-negative patients are randomized in 30 Gy
IF-RT and no further treatment. In the H10U trial of the
EORTC/GELA, standard treatment for patients with
early-unfavorable HL consists of four courses of
ABVD followed by IN-RT. In the experimental treat-
ment arms, patients with negative 18FDG–PET scan
after two courses ABVD received four additional courses
of ABVD without consolidating radiotherapy and
18
FDG–PET scan-positive patients are treated with two
courses of BEACOPP-escalated followed by IN-RT.
After the first interim analysis of the H10U trial, it was
decided to treat 18FDG–PET-negative patients with the
standard combined modality treatment, because more
relapses and progressions were documented in the non-
RT PET-negative arm; therefore it was concluded that
this trial will not show a non-inferiority of the exper-
imental 18FDG–PET-negative treatment arm.
Based on the results of the HD14 trial, a com-
bined modality treatment consisting of two courses of
BEACOPP-escalated followed by two courses ABVD
and 30 Gy IF-RT is regarded as the standard treatment
for patients with early-stage unfavorable HL within the
GHSG up to the age of 60 years. A less toxic, but equally
effective treatment approach might be developed by
implementing a targeted drug such as brentuximab
vedotin. Four cycles of ABVD followed by 30Gy of
IF-RT is a suitable but less effective alternative to
“2+2”+ 30 Gy IF-RT in those patients who do not
qualify for “2+2”.
Chapter 5: Hodgkin lymphoma
Advanced Hodgkin lymphoma
Before the introduction of polychemotherapy, only very
few patients with advanced HL could be cured. The use
of combinations such as MOPP or similar regimens led
to long-term remission of approximately 50%. The
introduction of ABVD by Bonadonna and colleagues
in 1975 resulted in a significant increase of FFTF and
OS to 63% and 82% after 5 years. The hybrid regimen
MOPP/ABV and the alternating MOPP/ABVD regi-
men were demonstrated to be equally effective as com-
pared with ABVD, but more toxic than ABVD alone in
terms of acute toxicity and the incidence of secondary
acute leukemia or myelodysplastic syndrome (MDS).
Therefore ABVD was accepted as standard chemother-
apy used in a combined modality strategy.
To improve the results in this group of patients
further, new regimens were developed by integrating
additional drugs such as etoposide and by increasing
dose intensity with the support of colony-stimulating
factors and with modern antibiotics. These new
approaches include regimens such as Stanford V,
MEC, BEACOPP-baseline and BEACOPP-escalated
(Table 5.6). A randomized comparison of Stanford V
with MEC and ABVD published by Chisesi et al. in
2002 showed a clear inferiority of the Stanford V pro-
tocol, but no significant difference between ABVD and
MEC could be detected. The conflicting results con-
cerning the efficacy of Stanford V might be partially
explained by the use of less radiotherapy in the random-
ized setting. In a randomized phase III trial of the
Eastern Cooperative Oncology Group (ECOG) the effi-
cacy and toxicity of Stanford V was compared with
ABVD; at a median follow-up of about 5 years, PFS
and OS of the Stanford V and the ABVD treatment arm
showed no significant difference.
In 1992, the GHSG designed the BEACOPP regimen
in a baseline and an escalated version and compared
these variants with COPP/ABVD in the HD9 trial. IF-
RT was used for bulky disease at diagnosis and for
residual disease after eight cycles of chemotherapy. At
a follow-up of 10 years, FFTF and OS rates were 82%
and 86% in the BEACOPP-escalated group, 70% and
80% in the BEACOPP-baseline group, and 64% and
75% in the COPP/ABVD. Because of the considerable
superiority of the BEACOPP-escalated regime over
BEACOPP-baseline and COPP/ABVD in the HD9
trial, the GHSG defined eight courses of BEACOPP-
escalated as the new treatment standard for advanced
HL. However, in the HD9 trial the BEACOPP-escalated
73
 Chapter 5: Hodgkin lymphoma

Table 5.6 First line polychemotherapy regimens.

Seldom/ Drug combination References

former
used
regimen
MOPP Mechlorethamine, Oncovin Hagenbeek et al.,
(vincristine), Procarbazine, 2000; Zittoun
Prednisone et al., 1985;
Ferme et al., 2000
COPP Cyclophosphamide, Oncovin (vin- Engert et al.,
cristine), Procarbazine, Prednisone 2003
EBVP Epirubicin, Bleomycin, Vinblastine, Carde et al., 1997;
Prednisone Nordijk et al.,
2005
MEC Mechlorethamine, CCNU Chisesi et al.,
(Lomustine), Vindesine, Alkeran, 2002
Prednisone, Epidoxorubicin,
Vincristine, Procarbazine,
Vinblastine, Bleomycin
Commonly Drug combinations Dose (mg/m²) Route Schedule (days) References
used
regimen
ABVD (cycle Adriamycin (doxorubicin) 25 IV 1+15 Bonadonna et al.,
length 28 Bleomycin 10 IV 1+15 2004; Sieber et al.,
days) Vinblastine 6 IV 1+15 2002; Diehl et al.,
Dacarbazine 375 IV 1+15 2005; Klimm et al.,
2005
BEACOPP baseline/escalated
baseline/ Bleomycin 10 IV 8 Nordijk et al., 2005;
escalated Etoposide 100/200 IV 1–3 Klimm et al., 2005;
(cycle Adriamycin 25/35 IV 1 Diehl et al., 2003;
length 21 Cyclophosphamide 650/1250 IV 1 Diehl et al., 2004
days) Oncovin (vincristine) 1.4 (max. 2 mg) IV 8
Procarbazine 100 PO 1–7
Prednisone 40 PO 1–14
G-CSF (for escalated regimen) from day 8
BEACOPP-14 Like BEACOPP baseline with a Sieber et al., 2003
cycle length of only 14 days
[Prednisone 80 mg/m² day 1–7,
G-CSF from day 8]
Stanford V Mechlorethamine 6 IV Wk 1, 5, 9 Horning et al.,
(12 weeks) Adriamycin 25 IV Wk 1, 3, 5, 7, 9, 11 1999; Horning
Vinblastine 6 IV Wk 1, 3, 5, 7, 9, 11 et al., 2002;
Vincristine 1.4 (max. 2 mg) IV Wk 2, 4, 6, 8, 10, 12 Chisesi et al., 2002
Bleomycin 5 IV Wk 2, 4, 6, 8, 10, 12
Etoposide 60 × 2 IV Wk 3, 7, 11
Prednisone 40 PO Wk 1–10,
G-CSF every 2 days after
dose reduction
or delay

regimen was associated with more acute and long-term
toxicity than ABVD such as a higher incidence of sec-
ondary leukemia and infertility.
74 The subsequent GHSG trials, HD12 and HD15,
aimed to reduce therapy-associated toxicity by modifying
the eight courses of BEACOPP-escalated regimen and by
evaluating the role of consolidative radiotherapy.
The HD12 trial of the GHSG compared eight
courses of BEACOPP-escalated with four courses of
BEACOPP-escalated followed by four courses of

BEACOPP-baseline, with or without consolidative
radiation to initial bulky and residual disease. In the
final analysis of the HD12 trial, published by
Borchmann et al., 2012, no significant FFTF-difference
could be detected between the two chemotherapy
arms. However, in the experimental chemotherapy
arm more patients progressed, but there was no differ-
ence regarding the treatment-related deaths. Those
patients who did not receive a consolidating radiother-
apy had an inferior FFTF. In the HD15 trial, patients
were randomized between eight courses of BEACOPP-
escalated, six courses of BEACOPP-escalated, and eight
courses of BEACOPP-14. The BEACOPP-14 regimen
is a time-intensified BEACOPP-baseline regimen given
in 14-day intervals with the support of G-CSF. To
develop a response-adapted therapy and to evaluate
the negative predictive value of 18FFDG–PET, only
those patients with PET-positive residual tumors
received 30 Gy IF-RT. With a median follow-up of 38
months, 3-year PFS was 92.1% for PET(–) patients and
86.1% for PET(+) patients. The negative predictive
value for 18FFDG–PET was 94.6%. The 5-year analysis
of the different chemotherapy arms revealed a signifi-
cant FFTF-benefit of the chemotherapy arm containing
6 courses of BEACOPP-escalated (5-year FFTF with
6 courses BEACOPP-escalated 89.3% vs. 84.4% with
8 courses BEACOPP-escalated vs. 85.4% with 8 courses
BEACOPP-14). Furthermore, compared to the previous
standard, 6 courses of BEACOPP-escalated resulted in a
significantly improved OS (5-year-OS 95.3% vs. 91.9%)
and a reduced treatment-associated mortality. Therefore,
6 courses of BEACOPP-escalated were defined as the new
GHSG standard chemotherapy regimen in advanced
stage HL patients (publication in progress).
The comparison of eight courses of ABVD with four
courses of BEACOPP-escalated followed by four courses
of BEACOPP-baseline (“4+4”) in the EORTC 2012
resulted in a significantly higher 4-year PFS in the “4
+4”-arm (Carde et al., 2012). Similarly, the Italian
HD2000 trial, which compared 8 courses of ABVD
with 4 courses of BEACOPP-escalated and two courses
of BEACOPP-baseline, showed a superior PFS in the
BEACOPP-escalated arm, but no significant OS benefit
(Federico et al., 2009). In order to receive more data
regarding the efficacy of ABVD in comparison with
BEACOPP-escalated, a network meta-analysis that
included 10,111 patients with advanced-stage HL was
performed; this analysis showed an OS benefit of 10%
at 5 years for BEACOPP-escalated over ABVD (Skoetz
et al., 2013).
Chapter 5: Hodgkin lymphoma
A response-adapted 18FFDG–PET-guided treat-
ment approach is being further evaluated by the
GHSG in the ongoing HD18 trial: those patients
with a negative 18FFDG–PET scan after two courses
of BEACOPP-escalated are randomized between
four and only two further courses of BEACOPP-
escalated. 18FFDG–PET(+) patients receive four fur-
ther courses of BEACOPP-escalated and are
randomized between additional versus no addi-
tional rituximab application. Only those patients
with 18FFDG–PET(+) residual tumor receive a con-
solidative radiotherapy.
Several study groups still regard the less toxic
ABVD regimen as standard of care for advanced
stage HL and are evaluating an alternative treatment
approach of escalating treatment in those patients
with positive 18FFDG–PET after two courses of
ABVD; so far, only preliminary results of this treat-
ment approach are available.
Treatment of advanced HL is still a matter of
debate. Based on the results of the HD9 trial and the
HD15 trial, the GHSG regards six courses of
BEACOPP-escalated as standard of care for advanced
HL. A less toxic, but equally effective approach might
be a modified BEACOPP-escalated regimen including
brentuximab vedotin followed by a PET-guided radio-
therapy strategy.
Treatment of primary progressive
and relapsed Hodgkin lymphoma
Patients with refractory or relapsed disease after first
line treatment still have the chance of being cured with
an adequate salvage therapy.
Relapse after initial radiotherapy
Conventional anthracycline-containing chemother-
apy is the treatment of choice for patients who
relapse after initial radiotherapy for early-stage dis-
ease. According to a retrospective analysis of the
GHSG published by Rueffer et al., 2005, the survival
of these patients is comparable to that of patients
with advanced-stage disease initially treated with
chemotherapy.
Relapse after initial chemotherapy
High-dose chemotherapy followed by autologous stem 75
cell transplantation (HDCT/ASCT) has become the
 Chapter 5: Hodgkin lymphoma

treatment of choice for HL patients who relapse or are ASCT, but their reported outcome was significantly
primary refractory to the first line chemotherapy. Two worse than the outcome of patients with relapsed HL.
prospective, randomized trials have shown that A further prognostic factor and a significant pre-
HDCT/ASCT results in an improved disease control dictive marker for the outcome after HDCT/ASCT is
compared to conventional chemotherapy. In the BNLI the chemosensitivity to second line induction therapy.
trial, patients with relapsed or refractory HL were Therefore the optimizing of the reinduction chemo-
treated with conventional-dose mini-BEAM (carmus- therapy is an important approach to improve the out-
tine, etoposide, cytosine arabinoside, melphalan) or come of relapsed HL.
high-dose BEAM and ASCT. The 3-year EFS was sig- There are several salvage chemotherapy regimens
nificantly better after high-dose chemotherapy (53% that have been reported to be effective in single-arm trials
versus 10%). The HDR1 multicenter trial conducted (Tables 5.7 and 5.8). Since the Dexa-BEAM regimen has
by the GHSG and the EBMT (European Group for
Blood and Bone Marrow Transplantation) compared
four cycles of Dexa-BEAM with two cycles of Dexa- Table 5.7 Second line polychemotherapy regimens.
BEAM followed by BEAM and ASCT. The freedom
from second failure (FF2F) at 3 years was significantly Salvage Drug combination References
better in the transplanted group than in the group regimen
receiving conventional salvage chemotherapy (55% ESHAP Etoposide, methylpredniso- Aparicio
versus 34%). The OS did not differ significantly lone, high-dose cytarabine, et al., 1999
cisplatin
between the two treatment arms in both trials.
The benefit from HDCT/ASCT is significantly ASHAP Doxorubicin, Solu-Medrol, Rodriguez
high-dose cytarabine, et al., 1999
influenced by prognostic factors, such as the duration cisplatin
of the first remission. As a result of a retrospective
ICE Ifosfamide, carboplatin, Moskowitz
analysis performed by the GHSG a prognostic score etoposide et al., 2001
for relapsed HL was developed; relapse within 12 DHAP Dexamethasone, high-dose Josting et al.,
months after initial therapy, stage III or IV disease, cytarabine, cisplatin 2005
and a low hemoglobin level (less than 12 g/dL for men GVD Gemcitabine, vinorelbine, Bertlett et al.,
or less than 10.5 g/dL for women) at relapse were pegylated liposomal 2007
identified as unfavorable prognostic factors. doxorubicin
Prospective phase II trials indicate that patients IGEV Ifosfamide, gemcitabine, Santoro
with primary refractory HL also benefit from HDCT/ vinorelbine et al., 2007

Table 5.8 Results of selected phase II trials with salvage chemotherapy regimens.

Salvage Number Response Complete Successful PBSC References

regimen of patients rate (%) remissions (%) collection (%)
ESHAP 22 73 9/22 NR Aparicio
et al., 1999
ASHAP 56 70 19/56 (34%) NR Rodriguez
et al., 1999
ICE 65 85 17/65 (26%) 86 Moskowitz
et al., 2001
DHAP 102 89 21/102 (21%) 96 Josting et al.,
2002
GVD 91 70 17/91 (19%) NR Bertlett et al.,
2007
IGEV 91 81 49/91 (54%) 99 Santoro
et al., 2007
NR, not reported.
76

been shown to be too toxic and since the successful stem
cell collection rate with modified BEAM-regimens is
reported to be rather low, new salvage chemotherapy
regimens have been developed. Platin-based regimens
include ESHAP (etoposide, methylprednisolone, cytara-
bine, cisplatin), ASHAP (doxorubicin, methylpredniso-
lone, cytarabine, cisplatin), DHAP (dexamethasone,
high-dose cytarabine, cisplatin), and ICE (ifosfamide,
carboplatin, etoposide). The response rates of these che-
motherapy regimens range from 70% to 89%, including
complete remission (CR) rates from about 21% to 34%;
no severe acute treatment-related complications have
been described, with myelosuppression being the most
relevant toxicity. Gemcitabine-based regimens, such as
IGEV (ifosfamide, gemcitabine, vinorelbine) and GVD
(gemcitabine, vinorelbine, pegylated liposomal doxoru-
bicin), have been shown to induce similar overall
response rates and a high rate of successful stem cell
mobilization. In the phase II trial evaluating the efficacy
and safety of IGEV, a CR rate of 54% could be docu-
mented; impressively, 60% of the primary refractory
patients responded.
So far, there are no randomized trials comparing the
efficacy of these different second line regimens. The
recently reported prospective, randomized HRD2 trial
performed by the GHSG, the EORTC, and other
European groups confirmed the efficacy of DHAP,
and indicates that two courses of DHAP directly fol-
lowed by HDCT/ASCT are equally effective when com-
pared to a more aggressive sequential induction
treatment of two courses of DHAP followed by high-
dose cyclophosphamide, methotrexate plus vincristine,
and etoposide. The documented 3-year-FFTF rates
were 71% in the standard arm consisting of two courses
of DHAP followed by HDCT/ASCT versus 65% in the
experimental arm; the evaluated 3-year OS rates were
87% versus 80%. Thus, two courses of DHAP followed
by HDCT/ASCT are considered current standard of
care for patients with relapsed HL.
A further increase in efficacy of the salvage induc-
tion treatment might be achieved by combining the
induction chemotherapy with a targeted therapy. In
the ongoing HDR3i trial the GHSG evaluates the
combination of DHAP with the mTor inhibitor evero-
limus (RAD001) compared to DHAP. The outcome of
relapsed HL might be additionally improved by treat-
ing those patients with an increased risk to relapse
after HDCT with a maintenance treatment. In a
placebo-controlled trial the GHSG is currently exam-
ining if the relapse rate after ASCT can be significantly
Chapter 5: Hodgkin lymphoma
reduced with a maintenance treatment consisting of
the deacetylase inhibitor panobinostat.
For patients with primary refractory HL a tandem
ASCT might be an appropriate approach. In a prospec-
tive multicenter phase II trial of the French lymphoma
study group (“Groupe d’Etude des Lymphomes de
l’Adulte” (GELA)/SFGM), a tandem ASCT resulted in
a 5-year OS of 46% in primary refractory and poor-risk
relapsed HL.
Salvage radiotherapy
Salvage radiotherapy alone offers a treatment option for
a selected subset of patients with localized relapse in
previously non-irradiated areas. In a retrospective anal-
ysis of the GHSG published by Josting et al. (2005), the 5-
year freedom from second failure (FF2F) and OS rates
were 28% and 51% after salvage radiotherapy alone. The
retrospective analysis of Campbell et al. (2005) reported
10-year FF2F and OS rates of 33% and 46% for those
patients with salvage radiotherapy-treated recurrent HL
in limited stage. So far, there are no data of controlled
trials analyzing the role of salvage radiotherapy in a
selected group of patients with relapsed HL.
Summing up, HDCT, consisting of two courses of
reinduction therapy such as DHAP, followed by ASCT
is the treatment of choice for patients with relapsed HL
after first line chemotherapy. For patients with primary
refractory HL, a tandem ASCT might be a more appro-
priate treatment approach. A selected group of patients
with localized relapse in previously non-irradiated areas
can be effectively treated with salvage radiotherapy alone.
Those patients with relapsed disease after initial radio-
therapy should receive an anthracycline-containing che-
motherapy regime.
Relapse after high-dose
chemotherapy and autologous stem
cell transplantation
Allogeneic stem cell transplantation
Allogeneic stem cell transplantation is an experimental
treatment option resulting in long-term remissions in
a subset of young HL patients with relapse after ASCT.
However, the role of allogeneic stem cell transplanta-
tion is controversial.
The existence of a relevant “graft versus HL” (GVL)
effect has been one point of this discussion. There are no
prospective randomized trials comparing the efficacy of 77
 Chapter 5: Hodgkin lymphoma
allogeneic stem cell transplantation to a second ASCT or
to salvage radiochemotherapy. Small retrospective, non-
randomized analyses reporting a lower relapse rate for
allogeneic stem cell transplantation than for ASCT sug-
gested that allogeneic stem cell transplantation may be
associated with a clinically significant GVL activity.
More convincing evidence for the existence of GVL
activity comes from the use of donor lymphocytes
(DLIs) in HL patients who had relapsed after allogeneic
transplantation. Single center retrospective analyses
reported response rates to DLIs of up to 50%.
Unfortunately, the majority of these HL patients did not
achieve long-term benefits so that further studies are
necessary to optimize the application of DLIs in HL.
The retrospective analyses reported by Sureda et al.
(2008) and by Sarina et al. (2010) supported the evidence
of a GVL effect; those patients developing chronic “graft
versus host disease” (GvHD) had a significantly lower risk
of relapse compared to patients with no chronic GvHD.
There was no impact on treatment-related mortality.
OS and EFS data reported for myeloablative
allogeneic stem cell transplantation were rather disap-
pointing because of unacceptably high treatment-related
mortality. To reduce treatment-related toxicity, different
reduced-intensity conditioning (RIC) regimens have
been developed. A retrospective analysis performed by
the European Group for Blood and Bone Marrow
Transplantation (EBMT) and reported by Sureda et al.
(2008) showed a significantly decreased non-relapse-
related mortality (NRM) in relapsed HL patients receiv-
ing RIC compared to patients treated with myeloablative
conditioning (1-year incidences of NRM 23% versus
46%; P = 0.001). Despite a significantly higher relapse
rate (57.3% versus 30.4% after a median time of 6
months), the 5-year OS rate was significantly higher in
the RIC group (28% versus 22%; P = 0.003). In patients
receiving a reduced intensity conditioning, relapse rates
seem to be higher with less-intensive non-myeloablative
regimens than with more intensive regimens like the
combination of fludarabine and melphalan. A promis-
ing new approach reported by Burroughs et al. (2007) is
the use of haploidentical donors in combination with
cyclophosphamide administration before and after
allogeneic transplantion.
Multivariate analyses of several retrospective studies
have identified the pre-transplant disease status as the
most relevant prognostic factor for outcome after alloge-
neic transplantation. The largest retrospective analysis
reported by Robinson et al. (2009) identified chemore-
78 fractory disease and poor performance status as risk
factors for poor OS and PFS. It is still a matter of dis-
cussion whether achievement of CR should be considered
as a precondition for allogeneic transplantation.
Allogeneic transplantation might be an appropriate
strategy in those patients with relapsed HL after ASCT
who are in a good performance status and have chemo-
sensitive disease. Retrospective analyses reported long-
term PFS rates of about 20% after reduced intensity
conditioning allogeneic transplantation. However, pro-
spective clinical studies are required to define clear
indications for this treatment option.
Targeted therapy of HL
Based on the increasing knowledge of the pathogenesis
of HL, new treatment approaches are currently being
developed which target cell surface molecules, modulate
the function of certain intracellular proteins, or modu-
late the immune response of the microenvironment. An
effective targeted therapy, given alone or in combina-
tion with chemotherapy, might improve the long-term
outcome of relapsed and primary refractory patients. As
a component of first line treatment a biologically based
drug might eliminate residual lymphoma cells after
chemotherapy. Additionally, combination of chemo-
therapy with a targeted therapy might help to reduce
the toxicity of the currently used regimens.
In this chapter a selection of new drugs, which were
either proved to be effective or which showed promis-
ing results in clinical trials are discussed.
Anti-CD30 monoclonal antibodies/immunotoxins
Because of its dense and highly restricted expression
on HRS cells, the surface antigen CD30 is an ideal
target for targeted therapy.
The clinical results of the first-generation anti-
CD30 monoclonal antibodies SGN-30 and MDX-060
were rather disappointing. By coupling the anti-CD30
antibody cAC10 to the antitubulin agent monomethyl
auristatin E (SGN-35: brentuximab vedotin), a potent
anti-CD30 targeted therapy was developed. In the first
phase I trial published by Younes et al. (2010), the
maximum tolerated dose (MTD) of SGN-35 (brentux-
imab vedotin) administered every 3 weeks was 1.8 mg/
kg; of 12 patients who received the MTD, six patients
had an objective response. In a subsequent phase II
trial 102 patients suffering from HL relapse after autol-
ogous stem cell transplantation were included. The
response rate was 73% with a median duration of
response of 6.7 months (Younes et al., 2012). Based

on these results brentuximab vedotin received appro-
val from the FDA and the EMA for treatment of HL
patients who relapsed after autologous stem cell trans-
plantation or received at least two prior systemic
therapies and are not eligible for autologous stem
cell transplant. Subsequently, further clinical trials
have been initiated or are being planned in order to
implement brentuximab in the first line or second
line treatment of HL.
Protein-specific “small molecules”
Histone deacetylase inhibitors (HDAC inhibitors)
Histone acetyltransferases (HATs) and histone deace-
tylases mediate acetylation and deacetylation of spe-
cific lysine residues on histone and non-histone
proteins and thereby regulate the expression and the
functional status of a variety of proteins that are
involved in cell proliferation, differentiation, survival,
angiogenesis, and immunity. Overexpression of differ-
ent HDACs has been reported in a number of human
cancers. HDAC inhibitors have been shown to induce
growth arrest and apoptosis of transformed cells, while
not-transformed cells are resistant to HDAC inhibi-
tors in vitro. Several HDAC inhibitors have antiproli-
ferative activity in HL-derived cell lines. In addition to
the antiproliferative and proapoptotic effect, HDAC
inhibitors might have additional modes of action such
as an immunomodulatory effect.
Promising clinical results in HL could be docu-
mented with the oral pan-HDAC inhibitor panobino-
stat. In a phase I trial published by Prince et al. (2007),
eight of 20 HL patients included in the trial achieved a
PR. The final analysis of a recently finished phase II trial
showed a response rate of 27% (Younes et al., 2012).
A double-blind, placebo-controlled phase III trial eval-
uating the safety and efficacy of panobinostat as main-
tenance treatment after ASCT is currently ongoing.
mTOR inhibitors
The mammalian target of rapamycin kinase mTORC1
regulates protein synthesis by phosphorylation of two
translation initiation factors (4E-binding protein 1 and
p70S6K), which lead to an increased synthesis of pro-
teins involved in cycle progression, proliferation, angio-
genesis, and survival pathways. An increased or
constitutive activation of mTORC1, for example
through the phosphatidylinositol 3-kinase (PI3K)/AKT
pathway, can be documented in different cancers. In HL
cell lines, the inhibition of mTORC1 induces cell cycle
arrest and apoptosis.
Chapter 5: Hodgkin lymphoma
The potential therapeutic value of mTOR inhibi-
tion in HL could be documented with the oral mTOR
inhibitor everolimus. In a phase II trial published by
Johnston (2010), treatment of relapsed HL with daily
doses of 10 mg everolimus resulted in a clinical
response in nine of 19 enrolled patients.
Targeting the microenvironment
Rituximab
While the expression of the CD20 antigen on HRS cells
is rather rare, CD20 is highly expressed on reactive
B-cells of the microenvironment. The interaction of
these reactive B-cells with HRS cells via CD40 or CD80
is assumed to play an important role for survival of HRS
cells. Thus, depletion of reactive B-cells with rituximab
might improve treatment efficacy in classical HL. In a
pilot study performed by the M.D. Anderson Cancer
Center, an overall response rate of 23% (5/22 enrolled
patients) in patients with relapsed classical HL could be
achieved with rituximab monotherapy. The following
phase II trial reported by Wedgwood et al. (2007) indi-
cated that first line treatment results might be improved
by combining ABVD with rituximab. The ongoing
HD18 trial of the GHSG is evaluating if those patients
with positive interim 18FFDG–PET scan benefit from
additional treatment with rituximab.
Lenalidomide
Lenalidomide, a thalidomide derivative, represents an
attractive option for the treatment of HL as it targets
several signal pathways that regulate survival of HRS
cells. In addition to direct antiproliferative and pro-
apoptotic effects, an immunomodulatory and antian-
giogenetic activity of lenalidomide has been described.
Two phase II trials and several named patient programs
have shown promising results with single-agent lenali-
domide treatment in relapsed HL, so that a phase I/II
study of lenalidomide combined with conventional che-
motherapy (adriamycin, vinblastine, dacarbazine) for
elderly HL patients has been initiated by the GHSG.
In addition to the anti-CD30 immunoconjugate
brentuximab vedotin, which has substantially
changed the treatment options in HL patients,
additional promising targeted therapy approaches
include, CD20-specific monoclonal antibodies, his-
tone deacetylase inhibitors such as panobinostat,
mTOR inhibitors such as everolimus, and the
79
thalidomide-derivative lenalidomide.
 Chapter 5: Hodgkin lymphoma
Nodular lymphocyte-predominant
Hodgkin lymphoma (NLP-HL)
Lymphocyte-predominant HL is a rare disease with
an estimated incidence of 1.5 per million. The
neoplastic cell in NLP-HL is the LP cell, which is
immunophenotypically distinguished from RS cells
by the strong expression of CD20 and negativity for
CD15 and CD30. About 70–80% of patients with
NLP-HL are initially diagnosed with limited stage
disease. Despite an excellent response to combined
modality treatment, patients with NLP-HL tend to
relapse continuously over decades. Current treat-
ment strategies should take into account the favor-
able prognosis of these patients and late toxicity
effects after a combined modality treatment. For
limited stage NLP-HL an excellent outcome could
be documented in retrospective analyses, without
any significant difference for treatment with IF-RT,
EF-RT, or a combined modality treatment
approach, so that the EORTC and the GHSG cur-
rently recommend IF-RT at a dose of 30 Gy as
standard of care for stage IA NLP-HL.
In selected cases, especially in children with stage
IA disease, a watch-and-wait strategy after initial
lymph node resection might be an appropriate strat-
egy. In retrospective analyses, patients with early-
unfavorable and advanced stage NLP-HL had a similar
response rate and tumor control as classical HL
patients in early-unfavorable and advanced stage; the
GHSG therefore recommends treatment of NLP-HL in
early-unfavorable and advanced stage in an analogous
manner to classic HL.
NLP-HL patients with suspected relapse should
undergo a renewed biopsy to exclude a transforma-
tion into a more aggressive NHL. A localized relapse
of NLP-HL might be treated with radiotherapy.
In two phase II trials reported by Ekstrand et al.
(2003) and Schulz et al. (2008), the anti-CD20 anti-
body rituximab resulted in impressive response rates
so that rituximab might be considered as an alter-
native treatment option in limited stage relapse of
NLP-HL. NLP-HL patients with advanced stage
relapse show a poorer response to single-agent rit-
uximab and should be treated with a more intensive
salvage therapy. An additional promising treatment
option for relapsed NLP-HL, the anti-CD20 anti-
body Ofatumumab, is currently being evaluated in
a phase II trial of the GHSG.
80
Treatment of Hodgkin lymphoma
in children and adolescents
Biologic features unique to childhood or adolescent
HL have not been described; only the relative inci-
dence of histological subtypes varies.
Similarly to treatment of adult HL, children suffering
from HL are allocated to three risk groups according to
their stage of disease. With risk-adapted, combined
modality treatment an OS of over 90% in children suffer-
ing from HL can be achieved. However, because of the
high rate of treatment-induced developmental disturban-
ces and the high risk of developing long-term complica-
tions, radiotherapy dose and field have been reduced and
new chemotherapy regimens such as OEPA (vincristine,
etoposide, prednisone, adriamycin), OPPA (vincristine,
procarbazine, prednisone, doxorubicin), COPDAC
(cyclophosphamide, vincristine, prednisone, dacarba-
zine), or ABVE-PC (adriamycin, bleomycin, vincristine,
etoposide, prednisone, cyclophosphamide) have been
designed. By replacing alkylating agents such as procarba-
zine and by increasing the number of agents and thereby
reducing the cumulative dose of individual agents, long-
term toxicity such as infertility and secondary leukemia
might be less frequent.
In a recently published trial performed by the
German Society of Paediatric Oncology and
Hematology–Hodgkin’s Disease (GPOH-HD), girls ini-
tially received two courses of OPPA, followed by two or
four courses of COPP in the intermediate or advanced
stage, while boys were treated with potentially less
gonadotoxic regimens, including two courses of
OEPA, followed by two or four courses of COPDAC.
All patients received consolidative IF-RT, except those
with early-stage disease who were in CR after chemo-
therapy. The 5-year EFS was 92% in early stage, 90.2%
in intermediate stage, and 84.7% in advanced stage HL,
without any significant difference between boys and
girls. These data suggest that OPPA-COPP and
OEPA-COPDAC are equally effective in intermediate-
and advanced-stage HL. So far there is no international
consensus concerning the optimal chemotherapy regi-
men and the number of chemotherapy courses
required; furthermore, so far the role of consolidative
radiotherapy has not been clearly defined.
Several trials of the British Paediatric Oncology
Group (POG) and the American Children’s Oncology
Group (COG) indicated that documentation of early
response to chemotherapy by FDG–PET scan is a

suitable parameter to guide treatment intensity.
Treatment outcome and toxicity of response-adapted
therapy is further evaluated in the EuroNet-PHL-C1
trial initiated by the European Network for Paediatric
Hodgkin Lymphoma (EuroNet-PHL).
Published data focusing on the treatment of adoles-
cent HL patients (15–19 years old) are scarce, and it is
still a matter of discussion whether adolescent HL
patients should be treated according to adult or pediatric
protocols. There are some retrospective analyses indicat-
ing a better outcome of adolescent than of adult HL
patients. Most of the prospective trials including adoles-
cents have not performed age-specific survival analyses.
Subgroup analysis of the DAL-HD-90 trial performed
by the German Paediatric Hodgkin Study Group
(GPOH-HD) and reported by Schellong et al. (1999)
indicated that treatment of adolescent HL patients with
a risk-adapted, combined modality approach including
chemotherapy regimens such as OEPA or OPPA results
in EFS rates which do not differ significantly from those
achieved for children. Therefore it seems to be reason-
able to develop a risk- and response-adapted treatment
approach for adolescent HL patients. Randomized trials
are indispensable to define the optimal treatment regi-
men for adolescent HL patients.
In summary, standard of care for childhood HL is a
combination chemotherapy combined with low-dose,
limited-field radiation. A promising approach to realize
individualized, risk-adapted treatment is early response-
adapted therapy. The optimal treatment regimen for
adolescents has to be determined in randomized trials.
Treatment-associated toxicity
As a consequence of the impressive long-term remis-
sion rates in HL, the reduction of treatment-related
complications has become increasingly important.
Short-term toxicity
Acute treatment-associated adverse effects include nau-
sea and vomiting, hematotoxicity, infections, mucositis,
and neurological complications such as polyneuro-
pathy. Less frequent acute complications are pulmonary
dysfunction or acute cardiac failure. The severity of
these therapy-associated complications depends on
the chemotherapy regimen and the additional applica-
tion of radiotherapy. Chemotherapy with BEACOPP-
escalated results in a significantly higher incidence of
grade 3 and 4 leukopenia, thrombocytopenia, and ane-
mia than treatment with BEACOPP-baseline, with
Chapter 5: Hodgkin lymphoma
COPP-AVBD or with ABVD alone; the HD9 trial
additionally showed a higher infection rate for the
BEACOPP-escalated regimen. To reduce the infection
rate, an obligatory antibiotic prophylaxis and G-CSF
application were incorporated into the HD18 protocol.
Bleomycin-based chemotherapy regimens, such as
ABVD or BEACOPP, as well as mediastinal irradia-
tion, may induce acute pneumonitis followed by lung
fibrosis.
Acute cardiac toxicity as a result of mediastinal
radiotherapy or an anthracycline-containing chemo-
therapy regimen is rather rare. It most frequently
presents with coronary heart disease or acute cardiac
failure owing to therapy-induced cardiomyopathy.
Long-term toxicity
Similar to short-term treatment-related toxicity, long-
term complications vary in frequency and severity
dependent on the chemotherapy regimen and the addi-
tional application of radiotherapy. Chemotherapy-
associated long-term toxicity particularly depends on
the cumulative dose of certain chemotherapeutic agents.
The most serious long-term treatment-associated
complications are the development of secondary neo-
plasia and cardiovascular disease. Other late effects
contributing to morbidity and/or mortality of long-
term survivors of HL are pulmonary dysfunction,
infectious complications, hypothyroidism, infertility,
and fatigue.
Secondary malignancies
The most frequent secondary malignancies are solid
tumors, particularly breast cancer and lung cancer.
Leukemia and NHL account for 20–25% of the
observed secondary cancers. While the risk of devel-
oping secondary leukemia is mainly confined to the
initial 5 years of follow-up, the incidence of secondary
solid tumors continuously increases up to a cumula-
tive risk of 23% at a follow-up of 25 years and exceeds
the cumulative risk of leukemia and NHL. However,
the relative risk of solid cancer varies significantly
dependent on the age at diagnosis and the attained age.
Treatment with alkylating chemotherapy has been
identified as the most important risk factor for the
development of secondary leukemia, with the cumu-
lative alkylator dose being the strongest determinant of
risk. Additionally, treatment with topoisomerase
inhibitors has been associated with an increased risk 81
of secondary acute myeloid leukemia (AML).
 Chapter 5: Hodgkin lymphoma
Radiotherapy, chemotherapy with alkylating
agents, and smoking are the most relevant risk factors
for the pathogenesis of secondary lung cancer.
Smoking can be considered as an amplifying risk fac-
tor for treatment-associated lung cancer.
According to the case-control study reported by
van Leeuwen et al. (2003), increased risk of breast
cancer after HL is largely attributable to radiotherapy
of the chest, with a significant correlation of the
relative risk to the radiotherapy dose. A risk reduc-
tion of breast cancer can be documented after chemo-
therapy with gonadotoxic substances such as
alkylating agents.
A strategy to reduce the incidence of secondary
malignancies is the reduction of treatment intensity,
for example by implementing targeted drugs or by
reducing radiation field size and radiation dose.
Further prevention strategies include smoking cessation
programs and counselling on sun-safety practice. Since
early detection through screening tests can reduce mor-
tality in several cancer types, several screening tests have
been evaluated in patients surviving HL. As a result of
one of these screening trials, annual MRI or screening
mammography is recommended by a UK screening
programme beginning 8 years after mediastinal radio-
therapy and an age of at least 25 years. Screening pro-
tocols for other secondary cancers have to be developed
and established in treatment guidelines.
Cardiac toxicity
Treatment-related long-term cardiac toxicity com-
prises a wide spectrum of cardiac diseases such as
coronary artery disease, cardiomyopathy, pericardial
disease, valvular disease, arrhythmia, and autonomic
dysfunction. Among these cardiac abnormalities, cor-
onary artery disease accounts for 66% of all cases of
fatal cardiac events in HL survivors. Mediastinal irra-
diation was reported as the most important risk factor
for developing cardiopulmonary complications.
Additionally, anthracycline-containing chemotherapy
regimens are associated with an increased risk for
congestive heart failure and myocardial infarction
and may further increase radiation-related cardiac
toxicity. Strategies to reduce treatment-associated car-
diac disease focus on improving radiation therapy
techniques and further reduction of dose and field
size. To detect subclinical cardiac abnormalities, non-
invasive cardiac screening tests such as resting and
82 stress echocardiogram have been evaluated; a stand-
ardized screening procedure regarding the required
screening tests, the timing in relationship to initial
HL treatment, and the frequency needs to be defined.
Fertility
Sterility is a long-term treatment-related complication
with serious impact on the quality of life of young HL
patients. A high incidence of primary ovarian insuffi-
ciency or oligospermia/azoospermia has been docu-
mented after treatment with alkylating agents such as
cyclophosphamide or procarbazine or after radiother-
apy to the pelvis or to the testes. Further risk factors for
female infertility are advanced-stage HL and age over 30.
Pre-treatment semen analysis showed that HL itself
is associated with a significant impairment of sperm
quality. Post-treatment dysspermia can be observed in
up to 90% of HL patients depending on the intensity of
the chemotherapy treatment. Therefore cryopreserva-
tion prior to therapy is strongly recommended for
male HL patients.
So far, no standardized procedures for assessment
and for preservation of female fertility after HL
treatment have been established. Available data of
prospective trials suggest that female HL patients
receiving six courses of BEACOPP-escalated do not
benefit from administration of oral contraceptives or
gonadotropin-releasing hormone agonists. Cryopre-
servation prior to therapy is another approach to
preserve fertility, but it has still to be regarded as an
experimental procedure.
In summary, the most relevant acute treatment-
related adverse effects are mucositis, bone marrow depres-
sion, infections, polyneuropathy, and, less frequently,
acute pneumonitis and acute cardiac toxicity. Delayed
treatment-associated complications include the develop-
ment of secondary malignancy or cardiac disease, pulmo-
nary dysfunction, infections, infertility, hypothyroidism,
and fatigue. In particular, second malignancies and car-
diovascular disease represent the leading causes of excess
mortality of long-term HL survivors. To reduce
treatment-related toxicity, ongoing trials evaluate the
development of response-adapted, less intensive treatment
protocols. Regular cancer and cardiac screening tests
might also contribute to reduce mortality of HL survivors.
Additionally, clinical research should focus on the preser-
vation of quality of life of long-term HL survivors.
Summary
Classical HL is a B-cell lymphoma characterized by the
presence of HRS cells which derive from germinal center
 Chapter 5: Hodgkin lymphoma

CR/PR
ABVD X 2
observe
+IFRT
Favorable Consider CR/PR
clinical trial
relapse
Early stage
2 + 2 + IF-RT (up to NR/PD
CS−IIA the age of 60 years)
w/o bulk ABVD × 4 + RT in the
Unfavorable case of High dose
contraindications. therapy + ASCT
NR/PD
Clinical staging +IFRT
Consider
clinical trial
CS IIA with
bulk, CSII−IV

NR/PD
Advanced relapse
stage

Polychemotherapy
6 courses of BEACOPP
escalated + RT for
residual disease (up to
the age of 60 years).
observe
Consider CR/PR
clinical trial

Figure 5.13 Treatment algorithm for classic Hodgkin Lymphoma. CR, Complete remission; PR, partial remission; NR, no response; PD,
progressive disease.

B-cells, but have lost most of the B-cell typical genes.
Pathognomonic for the less frequent NLP-HL are L&H
cells. Both variants share common features, i.e. the neo-
plastic cells are surrounded by a large number of reactive
non-clonal hematopoietic cells. In addition to a dereg-
ulation of multiple signaling pathways and transcription
factors, the interaction of HRS cells with these inflam-
matory cells seems to play an important role in the
pathogenesis of classical HL.
Treatment strategies for patients with HL are based
on an individual risk factor assessment (Figure 5.13).
The major prognostic factors used for risk stratifica-
tion include stage of disease, bulky disease, and
B-symptoms. According to the risk stratification
applied by most European groups, including the
GHSG and the EORTC, patients are assigned to
early-stage favorable, early-stage unfavorable or
advanced-stage risk groups.
The recommended treatment for patients with an
early-stage favorable disease is a combined modality
treatment consisting of two courses of ABVD followed
by 20 Gy IF-RT based on the results of the HD10 trial.
The current standard of care for HL patients with
early-stage unfavorable disease within GHSG consists
of two courses of BEACOPP-escalated followed by two
courses of ABVD and 30 Gy IF-RT. In the case of
contraindications, four cycles of ABVD followed by
30 Gy IF-RT can be given.
Patients with advanced-stage disease are treated
more aggressively with six to eight courses of che-
motherapy. In many countries ABVD is still
considered the standard regime for treatment of
advanced HL; based on the data of the HD9 trial
and the HD15 trial, the GHSG regards six courses
of BEACOPP-escalated as standard.
To date, over 80% of all patients achieve long-term
disease-free survival after first line therapy. The poor
prognosis of relapsed HL was markedly improved by
the introduction of effective high-dose chemotherapy 83
followed by ASCT.
 Chapter 5: Hodgkin lymphoma
Ongoing trials aim at reducing treatment-
associated toxicity while maintaining or even improv-
ing outcome. In addition to risk factor assessment at
diagnosis, response-adapted, 18FFDG–PET-guided
treatment approaches might be an important step
towards an individualized HL treatment and thereby
might help to reduce treatment intensity in those HL
patients with good response to chemotherapy.
The combination of chemotherapy/radiotherapy
with an effective targeted therapy such as brentuximab
vedotin might also contribute to decreased therapy-
associated toxicity. Furthermore, other targeted therapy
approaches, including the HDAC inhibitor panobino-
stat, and mTOR inhibitors such as everolimus or lena-
lidomide might be effective tools to circumvent
chemotherapy resistance in relapsed or refractory HL.
Further reading
Andrè M, Reman U, Federico M, et al., Interim analysis of
the randomized EORTC/LYSA/FIL intergroup H10 trial
on early PET-scan driven adaptation in stage I/II
Hodgkin Lymphoma. Ash, 2012, 623:549.
Bonadonna G, Zucali R, Monfardini S, et al. Combination
chemotherapy of Hodgkin’s disease with adriamycin,
bleomycin, vinblastine, and imidazole carboxamide
versus MOPP. Cancer, 1975;36(1):252–259.
Bonadonna G, Bonfante V, Viviani S, et al. ABVD plus
subtotal nodal versus involved-field radiotherapy in
early-stage Hodgkin’s disease: long-term results. J Clin
Oncol, 2004;22(14):2835–2841.
Borchmann P, Haverkamp H, Diehi V, et al. Eight cycles of
escalated-dose BEACOPP compared with four cycles of
escalated-dose BEACOPP followed by four cycles of
basline-dose BEACOPP with or without radiotheraphy in
patients with advanced-stage Hodgkin’s lymphoma: final
analysis of the HD12 trial of the German Hodgkin Study
Group. J Clin Oncol, 2011;29(32):4234–4242.
Brauninger A, Wacker HH, Rajewsky K, et al. Typing the
histogenetic origin of the tumour cells of lymphocyte-
rich classical Hodgkin’s lymphoma in relation to tumour
cells of classical and lymphocyte-predominance
Hodgkin’s lymphoma. Cancer Res, 2003;63:1644–1651.
Brink AA, Oudejans JJ, van den Brule AJ et al. Low p53 and
high bcl-2 expression in Reed-Sternberg cells predicts
poor clinical outcome for Hodgkin’s disease: involvement
of apoptosis resistance? Mod Pathol, 1998;11:376–383.
Buettner M, Greiner A, Avramidou A, et al. Evidence of
abortive plasma cell differentiation in Hodgkin and Reed-
Sternberg cells of classical Hodgkin lymphoma. Hematol
Oncol, 2005;23:127–132.
84 Carde PP, Karrasch M. Fortpied C, et al. ABVD (8 cycles)
versus BEACOPP (4 escalated cycles ≥ baseline) in stage
III–IV high-risk Hodgkin lymphoma: First results of
EROTC 20012 intergroup randomized phase III clinical
trial. J Clin Oncol, 2012 ASCO Annual Meeting. Abstract
8002.
DeVita VT, Jr., Hubbard SM. Hodgkin’s disease. N Engl J
Med, 1993;328(8):560–565.
Diehl V, Stein H, Hummel M, et al. Hodgkin’s lymphoma:
biology and treatment strategies for primary, refractory,
and relapsed disease. Hematology Am Soc Hematol Educ
Program, 2003:225–47.
Eich HT, Diehl V, Gorgen H, et al. Intensified chemotherapy
and dose-reduced involved-field radiotherapy in patients
with early unfavorable Hodgkin’s lymphoma: final
analysis of the German Hodgkin Study Group HD11 trial.
J Clin Oncol, 2010;28(27):4199–4206.
Engert A, Schiller P, Josting A, et al. Involved-field
radiotherapy is equally effective and less toxic compared
with extended-field radiotherapy after four cycles of
chemotherapy in patients with early-stage unfavorable
Hodgkin’s lymphoma: results of the HD8 trial of the
German Hodgkin’s Lymphoma Study Group. J Clin
Oncol, 2003;21(19):3601–3608.
Engert A, Franklin J, Eich HT, et al. Two cycles of
doxorubicin, bleomycin, vinblastine, and dacarbazine
plus extended-field radiotherapy is superior to
radiotherapy alone in early favorable Hodgkin’s
lymphoma: final results of the GHSG HD7 trial. J Clin
Oncol, 2007;25(23):3495–3502.
Engert A, Diehl V, Franklin J, et al. Escalated-dose
BEACOPP in the treatment of patients with
advanced-stage Hodgkin’s lymphoma: 10 years of
follow-up of the GHSG HD9 study. J Clin Oncol,
2009;27(27):4548–4554.
Engert A, Borchmann P, Pluetschow A, et al. Dose-
escalation with BEACOPP escalated is superior to ABVD
in the combined-modality treatment of early unfavorable
Hodgkin lymphoma: final analysis of the German
Hodgkin Study Group (GHSG) HD14 trial. Blood,
2010;116(21):336.
Engert A, Kobe C, Markova J, et al. Assessment of residual
bulky tumor using FDG-PET in patients with advanced-
stage Hodgkin lymphoma after completion of
chemotherapy: final report of the GHSG HD15 trial.
Blood, 2010;116(21):336.
Engert A, Plutschow A, Eich HT, et al. Reduced treatment
intensity in patients with early-stage Hodgkin’s
lymphoma. N Engl J Med, 2010;363(7):640–652.
Fan G, Kotylo P, Neiman RS, Braziel RM. Comparison of
fascin expression in anaplastic large cell lymphoma and
Hodgkin disease. Am J Clin Pathol, 2003;119:199–204.
Fan Z, Natkunam Y, Bair E, et al. Characterisation of variant
patterns of nodular lymphocyte predominant Hodgkin
lymphoma with immunohistologic and clinical
correlation. Am J Surg Pathol, 2003;27:1346–1356.

Federico M, Luminari S, Iannitto E, et al. ABVD compared
with BEACOPP compared with CEC for the initial
treatment of patients with advanced Hodgkin’s
lymphoma: results from the HD2000 Gruppo Italiano per
lo studio dei linfomi Trial. J Clin Oncol, 2009;10;27
(5):805–811.
Fermé C, Eghbali H, Meerwaldt JH, et al. Chemotherapy plus
involved-field radiation in early-stage Hodgkin’s disease.
N Engl J Med, 2007;357(19):1916–1927.
Garcia JF, Camacho FI, Garcia JF et al. Hodgkin and Reed-
Sternberg cells harbour alterations in the major tumor
suppressor pathways and cell-cycle checkpoints: analysis
using tissue microarrays. Blood, 2003;101:681–689.
Gordon LI, Hong F, Fisher RI, et al. A randomized phase III
trial of ABVD vs. Stanford V +/- radiation therapy in
locally extensive and advanced stage Hodgkin’s
lymphoma: an intergroup study coordinated by the
Eastern Cooperative Oncology Group (E2496). Blood,
2010;116(21):185.
Hay A, Klimm B, Chen BE, et al., Treatment of Stage I–II A
Non-Bulky Hodgkin’s Lymphoma (HL): An Individual
Patient-Data Comparison of German Hodgkin Study
Group (GHSG) HD10 and HD11 Combined-Modality
Therapy (CMT) and NCIC Clinical Trials Group (NCIC
CTG) HD.6 ABVD Alone. Blood (ASH Annual Meeting
Abstracts), 2012;623:548.
Jaffe ES, Stein H, Swerdlow SH, et al. B cell lymphoma,
unclassifiable, with features intermediate between diffuse
large B cell lymphoma and classical Hodgkin lymphoma.
In Swerdlow SH, Campo E, Harris NL, et al. (eds). WHO
Classification of Tumours of Haemopoietic and Lymphoid
Tissues. Lyon: IARC, 2008:267–268.
Josting A, Rudolph C, Mapara M, et al. Cologne high-dose
sequential chemotherapy in relapsed and refractory
Hodgkin lymphoma: results of a large multicenter study
of the German Hodgkin Lymphoma Study Group
(GHSG). Ann Oncol, 2005;16(1):116–123.
Josting A, Muller H, Borchmann P, et al. Dose intensity of
chemotherapy in patients with relapsed Hodgkin’s
lymphoma. J Clin Oncol, 2010;28(34):5074–5080.
Klimm B, Goergen H, Fuchs M, et al. Risk factors in early-
stage Hodgkin Lymphoma effectively define an
unfavorable subgroup with inferior outcome: an analysis
of international staging definitions (publication in
progress).
Linch DC, Winfield D, Goldstone AH, et al. Dose
intensification with autologous bone-marrow
transplantation in relapsed and resistant Hodgkin’s
disease: results of a BNLI randomised trial. Lancet,
1993;341(8852):1051–1054.
Lorigan P, Radford J, Howell A, et al. Lung cancer after
treatment for Hodgkin’s lymphoma: a systematic review.
Lancet Oncol, 2005;6(10):773–779.
Chapter 5: Hodgkin lymphoma
MacLennan KA, Bennett MH, Vaughan HB, Vaughan HG.
Diagnosis and grading of nodular sclerosing Hodgkin’s
disease: a study of 2190 patients. Int Rev Exp Pathol,
1992;33:27–51.
Mauz-Korholz C, Hasenclever D, Dorffel W, et al.
Procarbazine-free OEPA-COPDAC chemotherapy in
boys and standard OPPA-COPP in girls have comparable
effectiveness in pediatric Hodgkin’s lymphoma: the
GPOH-HD-2002 study. J Clin Oncol, 2010;28
(23):3680–3686.
Montalban C, Garcia JF, Abraira V et al. Influence of
biologic markers on the outcome of Hodgkin’s
lymphoma: a study by the Spanish Hodgkin’s Lymphoma
Study Group. J Clin Oncol, 2004;22:1664–1673.
Morschhauser F, Brice P, Ferme C, et al. Risk-adapted
salvage treatment with single or tandem autologous stem-
cell transplantation for first relapse/refractory Hodgkin’s
lymphoma: results of the prospective multicenter H96
trial by the GELA/SFGM study group. J Clin Oncol,
2008;26(36):5980–5987.
Ng AK, Mauch PM. Late complications of therapy of
Hodgkin’s disease: prevention and management. Curr
Hematol Rep, 2004;3(1):27–33.
Nogova L, Reineke T, Brillant C, et al. Lymphocyte-
predominant and classical Hodgkin’s lymphoma: a
comprehensive analysis from the German Hodgkin Study
Group. J Clin Oncol, 2008;26(3):434–439.
Noordijk EM, Carde P, Dupouy N, et al. Combined-
modality therapy for clinical stage I or II Hodgkin’s
lymphoma: long-term results of the European
Organisation for Research and Treatment of Cancer H7
randomized controlled trials. J Clin Oncol,
2006;24(19):3128–3135.
Press OW, LeBlanc M, Lichter AS, et al. Phase III
randomized intergroup trial of subtotal lymphoid
irradiation versus doxorubicin, vinblastine, and subtotal
lymphoid irradiation for stage IA to IIA Hodgkin’s
disease. J Clin Oncol, 2001;19(22):4238–4244.
Santoro A, Magagnoli M, Spina M, et al. Ifosfamide,
gemcitabine, and vinorelbine: a new induction regimen
for refractory and relapsed Hodgkin’s lymphoma.
Haematologica, 2007;92(1):35–41.
Sasse S, Klimm B, Goergen H, et al. Comparing long-term
toxicity and efficacy of combined modality treatment ng
extended- or involved-field radiotherapy in early-stage
Hodgkin’s oma. Ann Oncol. 2012;23(11):2953–2959.
Schmid C, Pan L, Diss T, Isaacson PG. Expression of B-cell
antigens by Hodgkin’s and Reed-Sternberg cells. Am J
Pathol, 1991;139:701–707.
Schmitz N, Pfistner B, Sextro M, et al. Aggressive
conventional chemotherapy compared with high-dose
chemotherapy with autologous haemopoietic stem-cell
transplantation for relapsed chemosensitive Hodgkin’s 85
 Chapter 5: Hodgkin lymphoma

disease: a randomised trial. Lancet, Sureda A, Robinson S, Canals C, et al. Reduced-intensity

2002;359(9323):2065–2071.
Schmitz R, Hansmann ML, Bohle V, et al. TNFAIP3 (A20) is
a tumor suppressor gene in Hodgkin lymphoma and
primary mediastinal B cell lymphoma. J Exp Med,
2009;206:981–989.
Sieniawski M, Reineke T, Josting A, et al. Assessment of male
fertility in patients with Hodgkin’s lymphoma treated in
the German Hodgkin Study Group (GHSG) clinical trials.
Ann Oncol, 2008;19(10):1795–1801.
Skoetz N, Trelle S, Rancea M, et al. Effect of initial treatment
strategy on survival of patients with advanced-stage
Hodgkin’s lymphoma: a systematic review and network
meta-analysis. Lancet Oncol. 2013;14(10):943–952.
Smolewski P, Robak T, Krykowski E et al. Prognostic factors
in Hodgkin’s disease: multivariate analysis of 327 patients
from a single institution. Clin Cancer Res,
2000;6:1150–1160.
Spector N, Milito CB, Biasoli I, et al. The prognostic value of
the expression of bcl-2, p53 and LMP-1 in patients with
Hodgkin’s lymphoma. Leuk Lymphoma,
2005;46:1301–1306.
conditioning compared with conventional allogeneic
stem cell transplantation in relapsed and refractory
Hodgkin’s lymphoma: an analysis from the Lymphoma
Working Party of the European Group for Blood
and Bone Marrow Transplantation. J Clin Oncol,
2008;26(3):455–462.
van Leeuwen FE, Chorus AM, van den Belt-
Dusebout AW, et al. Leukemia risk following
Hodgkin’s disease: relation to cumulative dose of
alkylating agents, treatment with teniposide
combinations, number of episodes of chemotherapy,
and bone marrow damage. J Clin Oncol,
1994;12(5):1063–1073.
Younes A, Gopal AK, Smith SE, et al. Results of a pivotal
phase II study of brentuximab vedotin for patients with
relapsed or refractory Hodgkin’s lymphoma. J Clin Oncol.
2012;30(18):2183–2189.
Younes A, Sureda A, Ben-Yehuda D, et al. Panobinostat in
patients with relapsed/refractory Hodgkin’s lymphoma
after autologous stem-cell transplantation: results of a
phase II study. J Clin Oncol. 2012;30(18):2197–2203

86
 Chapter
6
Follicular lymphoma
Kristian Bowles, Daniel Hodson, and Robert Marcus
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Introduction
Follicular lymphoma (FL) is the second most common
lymphoma after diffuse large B-cell lymphoma
(DLBCL). It has an annual incidence of 4 per 100 000
(USA and Europe) and accounts for 22% of all cases of
non-Hodgkin’s lymphoma (NHL). As discussed in
Chapter X the incidence changes with geography and
may be up to 10-fold lower in Asia. The disease is
generally characterized by the insidious onset of lym-
phadenopathy, usually without extranodal disease or
B-symptoms. Although long-term disease-free sur-
vival (DFS) is seen in some patients treated for early-
stage disease, the majority of patients are incurable
with conventional therapy. Although usually readily
responsive to treatment, remissions are temporary and
the disease follows a relapsing and remitting course.
Successive remissions become harder to achieve and of
shorter duration. Most patients receive several lines of
treatment before finally succumbing to refractory dis-
ease or high-grade transformation. In addition, a small
proportion of patients die from complications of
therapy.
The behavior of FL shows considerable variability.
In some cases the disease follows an aggressive chemo-
refractory course while in others the disease can be
controlled for 15 years or more. Studies comparing the
median survival of FL patients over the past few dec-
ades suggest that patient survival is improving. A
number of new therapeutic modalities have been
introduced in recent years and may explain some of
this improvement. In particular, the development of
the anti-CD20 antibody rituximab appears to be
changing the natural history of FL. Its use is associated
with survival improvements both at first presentation
and in relapsed disease.
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
Clinical presentation
The majority of patients with FL present with painless
lymphadenopathy. The median age at presentation is
approximately 60 years. Only 15–20% present with
stage I or II disease. Over 40% of patients have stage
IV disease at diagnosis. Apart from lymphadenopathy
many patients are asymptomatic at presentation. In
contrast to high-grade lymphoma, constitutional (“B”)
symptoms and extranodal involvement are rare.
Histology
Classical FL recapitulates the cellular composition of the
normal lymphoid follicle germinal center, being com-
posed of cells with centroblast and centrocyte morpho-
logy in variable ratios. The majority of cases show a
pronounced follicular growth pattern in which the fol-
licles are usually closely packed throughout the node,
frequently extending into the hilum and through the
lymph node capsule into perinodal fat. The follicular
structures are uniform in size and shape, have poorly
formed mantles, fail to show the normal distinction
between light and dark zones, and have few tangible
body macrophages. Neoplastic cells are found between
the follicular structures to a variable degree and may
only be discernible by immunocytochemical staining
but, when present, can be useful in distinguishing neo-
plastic from reactive proliferations (Figures 6.1 and 6.2).
FLs are graded according to the number of centro-
blasts present within a standard (0.159 mm2) high-
power field (hpf)[Table 6.1; Figures 6.3 and 6.4].
In a proportion of cases the follicular growth pat-
tern is lost and the proliferation adopts a diffuse
growth pattern. In almost all cases some residual fol-
licularity will be seen, although this might only be
87
 Chapter 6: Follicular lymphoma

Table 6.1 Grading of follicular lymphoma (WHO 4th edition).

Grade 1–2 0–15 centroblasts per hpf

Grade 1 0–5 centroblasts per hpf
Grade 2 6–15 centroblasts per hpf
Grade 3 >15 centroblasts per hpf
Grade 3a Centrocytes present
Grade 3b Centroblasts with sheet-like growth pattern

Figure 6.1 Follicular lymphoma (lymph node). The node

architecture is effaced by a lymphoid proliferation with a follicular
growth pattern. The follicles have a monotonous shape and size, and
the distinction between the germinal center and the surrounding
cells is lost. There is no differentiation into dark and light zones (loss of
zoned pattern) and macrophages are scanty.

Figure 6.2 Follicular lymphoma (lymph node). A component of

neoplastic centrocytes and centroblasts can be seen in the
interfollicular compartment.

Figure 6.3 Follicular lymphoma, grade 2 (lymph node). The

neoplastic follicle contains a mixture of cells, with centrocytes
predominating but with scattered centroblasts.

follicular growth is in the range 75–25% and focally

Figure 6.4 Follicular lymphoma, grade 3b (lymph node). The
neoplastic follicles contain an essentially pure population of
centroblasts but the follicular growth pattern is retained.
discernible by staining for follicular dendritic cells
88 (FDCs). Cases are reported as follicular if there is
>75% follicular growth, follicular and diffuse when
follicular when follicular growth is <25%. Cases of FL
where the areas of diffuse growth contain increased
numbers of cells with centroblast morphology (>15
per hpf) are considered to represent DLBCL with FL.
Other histological patterns have been described that
take into account an unusual growth pattern (floral
variant), morphological characteristics (monocytoid
differentiation, plasma cell differentiation, signet ring
cell morphology, immunoglobulin inclusions), or pres-
ence of stromal changes (sclerosis, PAS [periodic acid–
Schiff]-positive extracellular material).

Immunophenotypically the cells of FL express the
pan-B-cell markers CD19, CD20, CD22, CD79a, and
PAX5 and have surface immunoglobulin (sIg). In the
majority of cases they have a germinal center cell
phenotype with expression of CD10 and BCL-6 pro-
tein. A proportion of cases show loss of CD10. In
many cases the neoplastic cells in the interfollicular
compartment show downregulation of expression of
CD10 and BCL-6. Recent gene expression analysis has
highlighted expression of germinal-center B-cell-
expressed transcript (GCET) as a characteristic of fol-
licle center cells. There is usually no staining for
MUM1, although rare cases may express this antigen,
usually in cases with higher numbers of centroblasts.
In contrast to reactive germinal centers, the cells of FL
express BCL-2 protein in the majority (about 85%) of
cases. This is more frequently the case in grade 1–2
lymphoma (85–90%) but is found in only 50% of grade
3 FLs. A proportion of cases express CD23 but the cells
are negative for CD5 and cyclin D1. The neoplastic
follicles contain an FDC meshwork that is highlighted
by staining for CD21 or CD23 and contains a variable
component of reactive T-cells, including those of fol-
licular T-helper type and histiocytes (Figure 6.5).
Proliferation is variable but is generally low in grade
1 and higher in cases that are grade 3. The proliferation
does not show the organized zoned pattern of reactive
follicles.
The 4th edition of the WHO classification recog-
nizes clinicopathological variants of classical FL deter-
mined by age or site at presentation. Pediatric FL
characteristically involves lymph nodes (particularly
A B
Figure 6.5 Follicular lymphoma stained for BCL-2 protein (lymph node). The cells of follicular lymphoma show strong staining for BCL-2
protein (a) in distinction to the cells of reactive germinal centers, which are (with the exception of the intrafollicular T-cells) BCL-2-negative (b).
Chapter 6: Follicular lymphoma
cervical), Waldeyer’s ring, testis, or other extranodal
sites, as is usually at an early stage. These lymphomas
often have large follicles, are more frequently grade 3
and lack BCL-2 expression. There is an appreciation
that lymphomas that arise in the intestine, predomi-
nantly duodenum, behave differently from nodal FL.
They present frequently as polyps and may involve
multiple sites within the intestine but infrequently dis-
seminate (stage IE or IIE) and are associated with an
excellent prognosis. Other extranodal FL may also rep-
resent a variant with predominantly localized disease.
Primary cutaneous follicle center lymphoma is a
distinct and separate entity from nodal and non-cuta-
neous extranodal FL.
It has recently been recognized that some nodes
that have preserved architecture can harbor neoplastic
follicle center cells occupying some or all of the ger-
minal center in a variable proportion of the follicles.
While this pattern may be seen in nodes from patients
with concurrent or prior classical FL, it is also encoun-
tered in a minority of patients with no history of
lymphoma. In these cases the term intrafollicular neo-
plasia or in-situ FL has been used. The atypical cell
population is composed of centrocytes for the most
part and is highlighted by the presence of strong stain-
ing for BCL-2 protein. The significance of the finding
of in-situ FL is unknown.
Molecular and cytogenetics
The cytogenetic hallmark of FL, the t(14;18)(q32;q21)
chromosome translocation, is present in roughly 85%
89
 Chapter 6: Follicular lymphoma
of grade 1 and 2 tumors. By virtue of this transloca-
tion, which is easily detectable by fluorescence in-situ
hybridization (FISH) techniques, the BCL-2 proto-
oncogene is juxtaposed to enhancer sequences of the
IgH gene promoter region leading to BCL-2 deregula-
tion and overexpression in the neoplastic follicles. This
rearrangement occurs during the recombination of
variable (V), diversity (D), and joining (J) regions in
precursor B-cells. In reactive follicles, the BCL-2 pro-
tein is not expressed, thus enabling the removal of
B-cells with non-functional B-cell receptors via apop-
tosis. The constitutive activation of the antiapoptotic
BCL-2 via the t(14;18) bypasses this physiological
mechanism and, hence, FL may be viewed as a disease
of programmed cell death deregulation. A relatively
new, but highly controversial, finding is the presence
of the translocation in FL samples not only in the
neoplastic lymphoid cells, but also in microvascular
endothelial cells within the tumor. More recently,
t(14;18) has also been demonstrated in histiocytic or
dendritic cell neoplasms from patients simultaneously
suffering from t(14;18)-positive FL, thus providing
evidence for the transdifferentiation capacity of the
neoplastic FL clone. However, according to data
obtained from mouse models the introduction of a
t(14;18) in a lymphoid cell may not be sufficient to
cause a lymphoma on its own. Instead, this event may
rather constitute a prerequisite for the acquisition of
secondary chromosomal aberrations by prolonging
the lifespan of a germinal center B-cell. In this respect,
it is highly interesting that circulating t(14;18)-positive
B-cells can be found in the peripheral blood of
up to 70% of healthy individuals, and that cases of
“in-situ” FLs have been described in which clonal
t(14;18)-positive “FL-like” cells partly colonize reac-
tive follicles in lymph nodes. The significance of both
findings towards the development of overt FL is not
yet clear; however, in both conditions the clonal
FL-like B-cells may carry the t(14;18) as the sole aber-
ration, not yet having acquired additional genetic
lesions.
Secondary chromosome aberrations are well
defined in FL. In particular, (partial) trisomies of chro-
mosomes 1q, 7, 12, 18, and X as well as deletions in the
short arm of chromosome 1 (1p) and the long arm of
chromosome 6 (6q) have been reported. The acquisi-
tion of these secondary chromosomal alterations
appears to be non-random, and certain aberration cas-
cades may be determined by genetic alterations in the
90 early steps of progression. In a large cohort of FL cases,
the concept of varied cytogenetic pathways central to
pathogenesis has emerged from the analysis of secon-
dary chromosomal alterations defined by the acquisi-
tion of aberrations like 6q–, +7, or der(18)t(14;18). The
presence of a deletion in the long arm of chromosome 6
(6q–) has been associated with inferior outcome in FL.
FL frequently progresses to DLBCL, and this event
seems to be associated with certain recurring chromo-
some aberrations. Transformed FLs frequently acquire
deletions in the short arms of chromosomes 9 and 17,
the sites of the p16INK4A and TP53 tumor suppressor
genes. The additional introduction of a t(8;14)(q24;
q32)/MYC rearrangement in FL leads to a highly
aggressive “Burkitt-like” neoplasia, sometimes with a
precursor B-cell phenotype.
FL grade 3A, in accordance with its morphological
and immunophenotypical similarities to FL grades 1
and 2, is also frequently t(14;18)-positive, although to
a slightly lesser extent than its grade 1 and 2 counter-
parts. In contrast, FL3B is infrequently rearranged for
BCL-2 and BCL-6, and 50% of DLBCL with an addi-
tional FL3B component harbor BCL-6 rearrange-
ments. FLs occurring in the pediatric population, as a
rule, are t(14;18)-negative and generally present as
localized disease and with higher histological grade.
Recently, it was shown that t(14;18)-negative FLs of
grades 1 and 2 do display a gene expression profile
that, although maintaining the typical global signature
of FL, differ from their t(14;18) positive counterparts.
Gene expression profiling emerged as a powerful
tool to identify robust prognostic subgroups in FL. The
analysis of a large cohort of FL resulted in the definition
of two gene expression signatures with highly predictive
potential. One of these signatures, termed “immune
response 1 signature,” was associated with a favorable
clinical course, while the “immune response 2 signature”
conferred inferior survival times. As suggested by their
names, both signatures do not appear to be derived from
the malignant B-cells, but instead from reactive tumor-
infiltrating cells. In particular, the immune response 1
signature may primarily be derived from certain T-cell
subsets and the immune response 2 signature from
macrophages and dendritic cells. A mathematical
model combining these two signatures predicts that the
relative levels of subsets of these cells in the lymphoma
specimen are crucial for the survival times in FL, arguing
for a prominent role of immunologic cross-talk between
the malignant B-cells and infiltrating bystander cells.
These findings were recently confirmed by a RT-PCR-
based gene expression profiling attempt.

Staging investigations
Treatment decisions critically depend upon the dis-
tinction between early-stage disease and advanced dis-
ease. Staging investigations should therefore include
whole body CT scanning, full blood count, biochem-
ical assessment, including LDH, and bone marrow
biopsy. Marrow involvement by FL is typically patchy
and may not be identified by a bone marrow aspirate
alone. It is more readily detected as a paratrabecular
infiltrate on the trephine biopsy. The incidence of CNS
involvement by FL is rare, and routine examination of
the cerebrospinal fluid unnecessary.
FDG–PET is able to detect FL of all grades with
high sensitivity and specificity, and PET–CT per-
formed after immunochemotherapy is predictive of
progression-free survival (PFS). Although staging by
PET may be more accurate than CT it is not currently
known whether PET-based staging or restaging leads
to any improvement in treatment outcome. The role of
PET in FL is currently under investigation.
Prognostic factors
This topic is covered in greater detail in Chapter 2.
The International Prognostic Index (IPI), devel-
oped for use in high-grade lymphoma treated with
anthracycline-based chemotherapy, has been applied
to FL. However, its use is limited by the fact that in FL
extranodal disease is rare and elevation of the LDH is
uncommon. As such, only a small minority of patients
are classified as high risk and the index does not
effectively identify those cases of FL at high risk. To
improve upon the IPI, the Follicular Lymphoma
International Prognostic Index (FLIPI) was developed
from multivariate analysis of 4167 cases of FL (Table
6.2). The index includes age, stage, hemoglobin level,
Table 6.2 Overall survival according to risk group defined by the Follicular Lymphoma International Prognostic Index
(FLIPI).
FLIPI score Risk Proportion
group of patients (%)
0–1 Low 36
2 Intermediate 37
3–5 High 27
Overall survival of 1795 patients as described by Solal-Celingy et al. by FLIPI score. Criteria for FLIPI score: (1) age greater than 60
years; (2) Ann Arbor stage III–IV; (3) number of nodal sites greater than four; (4) serum LDH elevated; (5) hemoglobin less than
12 g/dL.
Chapter 6: Follicular lymphoma
LDH, and involvement of four or more nodal sites. It
distinguishes three roughly equal sized risk groups –
low, intermediate, and high – based on scores of 0–1, 2,
and ≥3, respectively. The 5- and 10-year percentage
overall survivals (OS) for low-, intermediate-, and
high-risk groups, respectively, are 91, 78, 52 and 71,
51, 36, respectively. Notably, there is no plateau in the
survival curve of any risk group. Although the FLIPI
was generated from patient cohorts prior to the intro-
duction of rituximab therapy, its predictive value in
patients receiving rituximab-chemo as first line ther-
apy has now been confirmed. An alternative prognos-
tic index termed “FLIPI2” or F2 has since been
developed specifically to identify factors predicting
prognosis in rituximab-treated patients. Using a pro-
spective patient cohort treated with up-front rituxi-
mab-based chemoimmunotherapy, the five most
predictive variables were found to be age >60 years,
β2-microglobulin (B2M) above normal, hemoglobin
<12 g/dL, bone marrow involvement and largest nodal
diameter >6 cm. These factors define low-, intermedi-
ate-, and high-risk groups with 5-year PFS of 80%,
51%, and 18%, and 5-year OS of 98%, 88%, and 77%,
respectively, in patients treated with rituximab chemo-
therapy. As yet there are no data proving its superi-
ority to FLIPI.
Treatment of early-stage disease
FL is exquisitely sensitive to radiation and in the
minority (15–20%) of patients who present with
stage I or II and small volume disease radiotherapy
alone has the potential to be curative. Numerous stud-
ies confirm long-term DFS in a proportion of patients
treated with involved-field radiotherapy (IF-RT)
alone. Relapses, when they occur, do so at sites outside
5-year overall 10-year overall
survival (%) survival (%)
90.6 70.7
77.6 50.9
52.5 35.5
91
 Chapter 6: Follicular lymphoma
the original field of radiation and usually do so within
the first 15 years. Beyond this time relapses are
extremely rare.
In a series of 460 patients with stage I/II FL treated
with IF-RT, DFS rates were 56% at 5 years and 41% at
10 years. Beyond 15 years only 2% of patients relapsed.
As expected, the majority of relapses occurred outside
the original field of radiation. Similar results were seen
in a series of 80 patients from M.D. Anderson treated
with IF-RT alone. PFS at 15 years was 66% for stage I
patients and 26% for stage II. Beyond 17 years no
relapses were seen, suggesting that this approach may
be curative in some patients.
In an attempt to improve long-term PFS the role of
adjuvant chemotherapy has been examined in single
arm studies. Seymour reported a 10-year time to treat-
ment failure (TTF) of 72% in stage I–III FL patients
treated with 10 cycles of CVP-bleo (cyclophospha-
mide, vincristine, prednisolone, and bleomycin) and
IF-RT. Whilst these results are superior to historical
controls treated with IF-RT there was a significant
increase in secondary malignancy. No randomized
trial has compared IF-RT with combined modality
treatment. The Trans Tasman Radiation Oncology
Group is currently conducting such a study to examine
IF-RT ± six cycles of CVP plus Rituximab. Clearly any
improvement in disease control must be balanced
against immediate and late toxicity of this approach
and in particular the risk of secondary malignancy.
The efficacy of rituximab in advanced-stage FL, com-
bined with its low toxicity, suggests that immunother-
apy may ultimately prove beneficial in early-stage
disease, either alone or in combination with radio-
therapy. In retrospective analysis the addition of
Rituximab to IFRT has been reported to improve
PFS. However, this approach is yet to be validated by
a prospective or randomized trial.
Because of the indolent nature of the disease
and concerns over late treatment-related toxicity
other researchers have advocated the approach of “no
initial therapy” in early-stage disease. No prospective,
randomized trial compares the use of upfront IF-RT
with watchful waiting. A retrospective analysis of the
Surveillance, Epidemiology and End Results Database
included 6568 patients with early-stage FL diagnosed
between 1973 and 2004. The authors compared the
long-term outcomes of patients who received initial
radiotherapy with those who did not. Multivariate
analysis showed a clear and independent association
92 between upfront radiation and improvement in both
disease-specific survival (hazard ratio [HR] 0.65) and
OS (HR 0.73). In the absence of data from a prospec-
tive randomized study these data suggest that upfront
radiotherapy should remain the standard of care
for early stage FL. Historically a radiation dose of
30–45 Gy has been used. However, a recent rando-
mised phase III study by Lowry et al. compared 24Gy
with 40–45Gy in patients requiring local control of
their disease. This revealed no difference in response
rate, in-field progression, PFS or OS making 24 Gy the
current recommended dose. Based upon reports of
high response rates in patients with relapsed FL treated
with just 4 Gy IF-RT a randomized UK trial compared
24 Gy with 4 Gy IF-RT. This trial has been closed early
due to inferior local PFS in the 4Gy arm. Although
lower doses may be useful in the context of palliation
24 Gy remains the standard of care for IF-RT delivered
with curative intent.
Patients with fully resected stage I disease have a
good prognosis, with up to 80% achieving long-term
remission. It is not known whether radiotherapy con-
fers any additional advantage or whether PET–CT
may identify those requiring further treatment.
Summary
At present 24Gy IF-RT is considered standard of care
for small volume stage I and II FL. In around 40% of
such patients, radiotherapy will be curative. In com-
pletely excised stage I disease the toxicity of treatment
may outweigh the potential reduction in risk of
recurrence.
Treatment of advanced disease
in asymptomatic patients
The majority of patients with FL present with advanced
disease and many of these are asymptomatic at diagno-
sis. In contrast to early-stage FL conventional treatment
is not curative and the aim of treatment is disease
control. Three randomized trials, conducted by the
NCI, GELA, and BNLI, have confirmed that early
treatment with conventional chemotherapy offers no
survival advantage in asymptomatic patients with FL.
Young and colleagues randomized 99 patients with
indolent NHL to watchful waiting versus aggressive
combined modality treatment with ProMACE–MOPP
(prednisone, methotrexate, doxorubicin, cyclophos-
phamide, etoposide plus mechlorethamine, vincristine,
procarbazine, and prednisone) followed by total nodal

irradiation. No survival difference was detected
between the two arms. In a GELA study, Brice and
colleagues randomized 193 patients with low tumor
burden FL to receive watchful waiting, prednimustine,
or interferon. “Low tumor burden” was defined as: (1)
largest nodal or extranodal mass less than 7 cm; (2)
fewer than three nodal sites with a diameter greater
than 3 cm; (3) absence of systemic symptoms; (4) no
substantial splenic enlargement; (5) no serous effusion;
(6) no risk of organ compression; (7) no leukemia or
blood cytopenia. After 5 years of follow-up no differ-
ence in survival was detected between the three arms.
The largest trial, run by the BNLI, randomized 309
patients with asymptomatic FL to “watchful waiting” or
immediate chlorambucil. Eligibility criteria for this trial
were defined by the absence of the following: (1) pruri-
tus or B-symptoms; (2) rapid generalized disease pro-
gression in the preceding 3 months; (3) life-endangering
organ involvement; (4) bone marrow infiltration result-
ing in bone marrow depression (defined as hemoglobin
<100 g/L, white cell count <3·0×109/L, or platelet count
<100×109/L); (5) localized bone lesions detected on
radiography or isotope scan; (6) renal infiltration; and
(7) radiological evidence of liver involvement. Bulk dis-
ease alone was not an exclusion criterion. Although the
chance of complete remission was increased by imme-
diate treatment (63% versus 27%), no difference in OS
or lymphoma-related death was observed between the
two arms. The mean time to treatment in the observa-
tion arm was 2.6 years. Furthermore, in the observation
arm 19% of patients had required no treatment by 10
years. Elderly patients (over 70 years) had an even
greater chance of requiring no therapy, with 40%
untreated at 10 years after diagnosis (Figure 6.6).
Whilst there is no role for conventional chemother-
apy in asymptomatic FL it is possible that newer, less
toxic agents such as rituximab may prolong the time to
disease progression and requirement for chemotherapy.
The preliminary analysis, presented in abstract by
Ardeshna and colleagues, of a randomized study in 462
patients comparing short course or prolonged single
agent rituximab with a “watch and wait” policy in stage
2–4 asymptomatic FL showed that rituximab both short
and long course significantly delayed the need to initiate
new therapy (median 33 months versus not reached at 4
years). Whether such an approach will alter the long-
term course of the disease remains to be determined.
Watchful waiting remains the appropriate manage-
ment for asymptomatic patients with advanced FL, in
Chapter 6: Follicular lymphoma
100 Observation (n = 151)
Chlorambucil (n = 158)
Cumulative survival (%)
80
60
40
20
0
0 4 8 12 16 20 24
Time (years)
Figure 6.6 Chlorambucil versus observation alone in asymptomatic
patients with advanced stage follicular lymphoma. Figure taken from
Ardeshna et al. 2003, with permission.
whom up to 25% may regress spontaneously. Data from
trials of immunotherapy, immunochemotherapy, and
radioimmunotherapy may alter this approach in the
future.
First line treatment of advanced
disease in symptomatic patients
The indications to initiate treatment in FL are gener-
ally considered to be overt progression of bulky dis-
ease, compromise of a vital organ, bone marrow
failure, and B-symptoms. Patients are treated with
the expectation that the disease will pursue a relapsing
and remitting course and that patients may require
several lines of treatment during the course of their
disease. Until recently the standard strategy was to
treat to remission or best response and then manage
with regular surveillance until progression. However,
postinduction maintenance therapy with rituximab is
now advocated. At disease progression, restaging and
repeat biopsy (to exclude high-grade transformation)
should be performed to guide subsequent therapy. For
many years standard first line treatment has been
alkylator-based therapy, frequently in combinations
including vinca alkaloids, corticosteroids, and anthra-
cyclines. No single alkylator-based regimen shows
superiority in terms of OS over any other, although
more intensive regimens such as CHOP have higher
response rates. Recently, the addition of rituximab to
alkylator-based regimens has been shown to improve
all measurable outcomes, including PFS and OS in
several randomized trials. 93
 Chapter 6: Follicular lymphoma
Conventional chemotherapy
Single agent chlorambucil or cyclophosphamide induces
response rates up to 80% although few complete
responses are seen. The addition of corticosteroid, vincris-
tine, or anthracycline may increase the rates of complete
remission but does not affect OS. In 1994 Kimby and
colleagues randomized 259 patients with untreated, symp-
tomatic low-grade NHL between ChP (chlorambucil +
prednisolone) and CHOP (cyclophosphamide, doxorubi-
cin, vincristine, and prednisolone). Although the response
rate was greater in the CHOP-treated patients (60% versus
36%), 5-year OS was equivalent (54% versus 59%). Most
recently, Peterson randomized 228 patients with previ-
ously untreated FL to receive either daily oral cyclophos-
phamide or CHOP–bleomycin. Responses were seen in
91% of patients; however, no significant differences in
either complete remission (CR) (66% versus 60%) or OS
(44% versus 46%) were seen. The purine analog fludar-
abine has been shown to have considerable activity against
FL as first line treatment when used both as a single agent
and in combination with either cyclophosphamide or
mitoxantrone. Although rates of response, including com-
plete remissions and molecular remissions, are seen, no
trial has identified a survival advantage compared to
alkylator-based therapy. Chemotherapy alone has now
been superseded, where available, by combined immuno-
chemotherapy regimens.
Immunotherapy
The introduction of monoclonal antibody therapy, in
particular rituximab, has led to major improvements in
the treatment of FL. Although optimally used in combi-
nation with chemotherapy, single agent rituximab has
shown significant activity and minimal toxicity in trials
of previously untreated patients. Hainsworth treated 62
patients with newly diagnosed indolent NHL (61% FL)
with four weekly doses of rituximab followed by 6-
monthly maintenance up to 2 years. At 6 weeks response
rates of 47% were seen, rising to 73% (37% CR) with
continued maintenance therapy. The median PFS was 34
months. Similar results were reported by the SAKK
group. This study randomized patients with FL (64
naïve and 138 previously treated) to single agent ritux-
imab with or without rituximab maintenance. Response
rates of 67% (9% CR) were reported for chemo-naïve
patients, with durable responses in nearly half of
responding patients. Promising new approaches to
94 improve the efficacy of immunotherapy include the
addition of the immunomodulatory agent lenalidomide.
A recent phase II study (presented in abstract by Fowler
and colleagues) of previously untreated stage III/IV
indolent lymphoma treated for 6 months with lenalido-
mide/rituximab reported response rates of 86%, with 16
of 17 FL patients achieving a complete remission. Future
randomized studies will be needed to establish the true
impact of this regimen.
Single agent rituximab is a potential treatment
option in patients with newly diagnosed FL who are
unable to tolerate immunochemotherapy.
Combined immunochemotherapy
The standard of care for first line treatment of FL is
now considered to be combined immunochemother-
apy. Four phase III trials and a Cochrane meta-analy-
sis have confirmed clear survival benefit from the
addition of rituximab to a variety of alkylator-contain-
ing regimens both with and without the inclusion of
anthracycline (Table 6.3).
Marcus et al. randomized 321 patients with previ-
ously untreated advanced FL to eight cycles of CVP
(cyclophosphamide, vincristine, and prednisolone)
chemotherapy with or without rituximab. At a median
follow-up of 53 months, patients treated with rituxi-
mab had significantly superior outcomes; TTF (27
months versus 7 months), time to progression (TTP)
(32 versus 15 months), overall response rate (ORR;
81% versus 57%) and complete response (CR; 41%
versus 10%). Notably, the improvement in TTP was
seen across all prognostic groups. At 4 years OS was
increased (83% versus 77%).
The German Lymphoma Study Group (GLSG)
randomized 428 patients with untreated advanced-
stage FL to receive 6–8 cycles of CHOP chemotherapy
with or without rituximab. A significantly improved
ORR (96% versus 90%) and PFS were seen in the ritux-
imab arm. In addition, rituximab led to an improvement
in survival, with six deaths versus 17 deaths in the first 3
years. Although the numbers were small this achieved
statistical significance. Survival advantage for immuno-
chemotherapy over chemotherapy alone was confirmed
in another phase III study conducted by the OSHO
group. They randomized 358 patients with previously
untreated advanced indolent NHL to receive eight cycles
of MCP (mitoxantrone, cyclophosphamide, and predni-
solone) with or without rituximab. Although the trial
included mantle cell lymphoma and lymphoplasmacytic
lymphoma, 201 patients had FL. The FL subgroup
showed an ORR 92.4% versus 75% and CR 50% versus
25% (P < 0.0001). Median EFS was 26 months in the
 Chapter 6: Follicular lymphoma

Table 6.3 Phase III trials of immunochemotherapy in first line treatment.

Study Treatment n ORR (%) CR (%) Response duration OS (%)

Marcus (2005, 2008) CVP 159 57 10 15 months 77
R-CVP 162 81 41 32 months 83
(TTP) (4 years)
Hiddeman (2005) CHOP 205 90 17 31 months 92
R-CHOP 223 96 20 Not reached 97
(TTF) (3 years)
Herold (2007) MCP 96 75 25 26 74
R-MCP 105 92 50 Not reached 87
(EFS) (4 years)
Salles (2008) CHVP/IFN 175 85 34 37% Survival benefit seen
R-CHVP/IFN 184 94 63 53% in FLIPI high risk
(EFS 5 years)
ORR, Overall response rate; CR, complete response; TTP, time to progression; TTF, time to treatment failure; EFS, event-free survival; OS,
overall survival.

control arm but was not reached at 47 months in the
rituximab arm. Most notably the addition of rituximab
to MCP in patients with FL was associated with a sig-
nificantly improved OS at 4 years (87% versus 74%;
P = 0.01). Finally, the FL-200 study conducted by
GELA randomized 359 patients with previously
untreated FL and high tumor burden to 12 cycles of
CHVP/interferon (cyclophosphamide, doxorubicin, eto-
poside, prednisolone, and alpha-interferon) or six cycles
of CHVP/interferon with six doses of rituximab.
Patients receiving rituximab had a significantly higher
response rate at 6 months (ORR 94% versus 85%; com-
plete remission/complete remission – unconfirmed (CR/
Cru; 76% versus 49%) and prolonged EFS at 5 years
(53% versus 37%). A prolongation of OS at 5 years was
seen in patients with FLIPI risk score 3–5 but did not
achieve statistical significance in low-risk patients.
The survival benefit of adding rituximab to chemo-
therapy for first line treatment of FL has been exam-
ined in a recent Cochrane meta-analysis. Of note, this
combined survival data from the above randomized
trials was composed prior to the updated survival data
from the GELA and Marcus trials. Even without this
data the analysis favored immunochemotherapy with
a hazard ratio of 0.63 (95% CI: 0.51–0.79).
Whilst the benefit of adding rituximab to chemo-
therapy is established, the optimal chemotherapy regi-
men to use remains unclear. A randomised trial
comparing R-CVP versus R-CHOP versus R-FM for
the initial treatment of patients with advanced-stage
follicular lymphoma appeared to favour R-CHOP.
After a median follow-up of 34 months, 3-year TTFs
were 46%, 62%, and 59% for the respective treatment
groups with significantly higher toxicity in those receiv-
ing R-FM. Overall survival at three years was 95% for all
patients. A meta-analysis conducted by Itchaki and col-
leagues concluded that the addition of anthracycline to
either chemotherapy or immunochemotherapy was
associated with no improvement in OS. Rummel and
colleagues have compared R-CHOP and R-bendamus-
tine in a study of 549 patients with grade 1/2 FL and
mantle cell lymphoma. Patients were randomized to
receive six cycles of R-CHOP or R-bendamustine (90
mg/m2 day 1+2 every 28 days). Improvement in median
PFS (69.5 months versus 31.2 months) was seen in the R-
bendamustine arm. In addition to its superior PFS, R-
bendamustine was better tolerated and showed reduced
toxicity compared to R-CHOP. Response by FLIPI score
was not presented and it is not known if the benefit
extends to all prognostic subgroups. OS was not different
at the time of analysis and longer follow-up is awaited. It
is not clear from current evidence if bendamustine is
optimally used upfront or better reserved for later relap-
ses where it is also associated with high response rates.
All patients who require treatment for stage III and
IV FL should receive remission induction treatment of
rituximab combined with chemotherapy (e.g. CVP,
CHOP, bendamustine). In those unable to tolerate che-
motherapy, single agent rituximab should be considered.
Rituximab maintenance
The benefit of rituximab as maintenance therapy
was originally shown following treatment for
relapsed FL. It has also been shown following first 95
 Chapter 6: Follicular lymphoma
line treatment in patients who received single agent
rituximab and in patients who received chemother-
apy without rituximab. Until recently it was
unclear whether patients receiving first line immu-
nochemotherapy would benefit from rituximab
maintenance. This question was addressed by the
GELA sponsored PRIMA study. The study
randomized 1217 patients who had achieved a min-
imum of partial remission (PR) following treat-
ment with immunochemotherapy (either R-CVP,
R-CHOP, or R-FCM [rituximab, fludarabine,
cyclophosphamide, and mitoxantrone]) to mainte-
nance therapy with rituximab (375 mg/m2 every 8
weeks for 2 years) or no further therapy. At 3 years
maintenance rituximab was associated with pro-
longed PFS (74.9% versus 57.6%). No difference
was observed in OS and quality of life assessment
was similar in the maintenance and control arms.
Little additional toxicity was seen in the mainte-
nance group. Longer follow-up data, particularly
that addressing OS and the response to salvage
therapy, is currently awaited. A similar effect was
seen in a retrospective population-based analysis of
patients treated before and after the introduction of
maintenance rituximab as standard in British
Columbia; at 3 years PFS was 62% and 83%,
respectively, with no identifiable change in OS.
Autologous stem cell transplant
(ASCT) in first remission
Four prospective randomized trials have examined the
benefit of ASCT in first remission. The GELA study
compared ASCT with IFN following conventional
induction chemotherapy and found no difference in
either PFS or OS. The GLSG compared ASCT with
IFN maintenance following induction therapy with
CHOP or MCP. Autografting was associated with
prolonged PFS at 5 years (64.7% versus 33.3%)
although data for OS was not presented. The updated
GOELAMS study also showed improved PFS in
patients randomized to ASCT (64% versus 39% at 9
years). However, partly because of increased toxicity in
the ASCT arm, this failed to translate into a survival
advantage (80% versus 81% at 9 years). The GITMO
study was the only one to examine the use of ASCT
following the use of upfront rituximab. This study
confirmed that the improvement in PFS associated
with ASCT is still seen after rituximab (68% versus
96 31% at 4 years). However, as with previous studies the
combination of increased toxicity in the ASCT and the
favorable response to salvage treatment in the control
arm meant that no difference in OS was observed.
Importantly, all these studies predate the introduction
of rituximab maintenance.
Radioimmunotherapy
Radioimmunotherapy (RIT) has shown impressive results
in the first line treatment of FL. Kaminski treated 76
previously untreated patients with stage III/IV FL with
131
I-tositumomab and reported a 95% response rate (75%
CR) and PFS of 59% and 40% at 5 and 10 years, respec-
tively. For those patients achieving a CR the median PFS
was 10.9 years. One case of secondary myelodysplastic
syndrome (MDS) was reported. Although these results
seem impressive it is important to note that the median
age of the patients included in this study was only 49
years. No randomized trial has compared RIT with the
current standard of care – immunochemotherapy.
RIT has also been examined in the context of post-
induction consolidation. In a SWOG study 90 patients
were treated with six courses of CHOP chemotherapy
followed by 131I-tositumomab consolidation and
reported a response rate of 91% (69% CR) and 5-year
PFS of 67%. This compared favorably to their historical
controls treated with CHOP alone. A single phase III
multicenter randomized controlled trial – the FIT
study – compared consolidation with 90Y-ibritumomab
tiuxetan to no further treatment following first line
chemotherapy in 414 patients with advanced-stage FL.
RIT consolidation resulted in high (77%) PR-to-CR
conversion, leading to a final CR rate of 87%. In those
randomized to receive RIT, median PFS was prolonged
by 2 years compared to the observation group. After 5
years of follow-up, PFS (RIT versus observation) was
57% versus 43% in patients achieving a CR and 38%
versus 14% in patients achieving a PR following induc-
tion therapy. As only 15% of patients received rituximab
as a component of their induction therapy it remains to
be determined whether RIT consolidation improves PFS
in patients treated with immunochemotherapy and
whether consolidation with RIT is superior to mainte-
nance therapy with rituximab. No significant difference
in OS was observed and at 5 years a trend towards
increased risk of secondary MDS was seen (six cases in
the RIT arm versus one case in the observation arm).
Summary: first line treatment
of advanced lymphoma
Observation alone remains an appropriate strategy
for asymptomatic patients with advanced FL. Those

patients who do require treatment should receive an
alkylator-based regimen in combination with rituxi-
mab. The inclusion of an anthracycline improves
response rate but does not improve OS. There is some
evidence that the efficacy of R-bendamustine may be at
least as good as, if not better than, R-CHOP, and the
toxicity lower. The PRIMA study suggests that main-
tenance rituximab prolongs the duration of remission in
patients induced with R-chemotherapy. ASCT is not
indicated in first remission. The use of RIT is associated
with prolonged remissions but its role in the era of
immunotherapy and rituximab maintenance has yet
to be defined.
Management of relapsed
follicular lymphoma
The management of patients with relapsed disease is
influenced by previous therapy, duration of remission,
and performance status. Similar to the situation in
patients with newly diagnosed FL, observation alone
may be appropriate for relapsed patients who are asymp-
tomatic. The risk of transformation to high-grade lym-
phoma is approximately 20% at 5 years. Repeat biopsy is
therefore important to exclude transformation at each
recurrence. Restaging of the disease is also recommen-
ded at each relapse to establish the extent of the lym-
phoma and to provide a baseline from which to define
response to therapy. There are a number of treatment
options available to clinicians in the management of
symptomatic relapsed FL, but the heterogeneity of the
patient population and specifically the nature of pre-
vious treatments and the duration of previous remis-
sions, combined with a lack of good quality randomized
controlled trial data, means that treatment at relapse is
far from standardized.
Immunochemotherapy
Several recent randomized trials confirm that the addi-
tion of rituximab to chemotherapy in relapsed FL is
associated with improvement in both response rate
and survival. Van Oers randomized 465 patients with
relapsed FL to R-CHOP or CHOP and saw improved
PFS (33.1 months versus 20.2 months) and OS (85%
versus 77%) at 3 years. Forstpointner randomized 65
patients with relapsed FL to R-FCM or FCM alone and
also observed improved PFS and OS in the rituximab
arm. Most of this data comes from patients not pre-
viously exposed to rituximab. However, where
patients relapse following a long (>2 years) remission
Chapter 6: Follicular lymphoma
induced by R-chemo, it seems reasonable to retreat
with rituximab.
Bendamustine appears to be an increasingly prom-
ising drug in relapsed FL. In a phase III randomized
trial from Rummel and colleagues of relapsed indolent
lymphoma therapy (46% of patients had FL), benda-
mustine–rituximab (BR) was compared with fludara-
bine–rituximab (FR). The BR combination was
associated with significantly improved ORR (83.5%
versus 52.5%), CR rate (38.5% versus 16.2%), and
median PFS (30 versus 11 months). Serious adverse
events were similar for the BR and FR groups (17.4%
and 22.2%), and OS did not differ with a median
follow-up period of 33 months. Furthermore, in an
unplanned subanalysis, rituximab maintenance ther-
apy significantly prolonged OS and PFS in a group of
40 patients who received this treatment.
Fludarabine is an effective agent in patients who
relapse following first line alkylator-based therapy. It
showed improved PFS in comparison with CVP but did
not significantly improve response rate or OS. In patients
with refractory and relapsed FL response rates appear to
be improved by combining fludarabine with an anthracy-
cline, an alkylating agent, or both. However, no direct
comparison was made to alkylator therapy alone, and
superiority in terms of OS has not been demonstrated.
Fludarabine treatment is also associated with significantly
greater toxicity than alkylator- or anthracycline-based
regimens and its benefits as second line treatment must
be weighed against the risks of infection, stem cell toxicity
(which may compromise future attempts at stem cell
harvesting), and treatment-related MDS.
In relapsed patients unable to tolerate conventional
chemotherapy, single agent rituximab induces a response
in approximately 50% of patients (who had not previously
received antibody therapy), with a median TTF of 13
months. Single agent rituximab is also active in patients
who have previously responded to rituximab therapy,
with a response rate in this group of 40% and a median
time to progression estimated at 18 months.
Although the benefit of rituximab seems clear the
choice of chemotherapy to use in combination will be
influenced by, among other things, prior treatment (in
particular, anthracycline exposure), duration of first
remission, need for future stem cell harvest, and
patient fitness. Therapeutic options include alkylators,
anthracyclines, bendamustine, and fludarabine, either
alone or in combination. Of these, bendamustine
appears to combine efficacy with a favorable side effect 97
profile.
 Chapter 6: Follicular lymphoma
Rituximab maintenance
The benefit of rituximab extends beyond the reinduc-
tion stage. There is clear evidence that extended use of
rituximab as maintenance therapy after reinduction
prolongs both PFS and OS. The van Oers study
involved a double randomization initially comparing
six cycles of CHOP versus R-CHOP. Those patients
achieving CR or PR were subjected to a second ran-
domization between observation only or rituximab
maintenance every 3 months for 2 years. The use of
rituximab maintenance was associated with a pro-
longed PFS (3.7 versus 1.3 years). Importantly, this
benefit in PFS was seen both in patients reinduced
with CHOP (3.1 versus 1.0 years) and in patients
induced with R-CHOP (4.1 versus 1.9 years). Thus,
in both rituximab-exposed and rituximab-naïve
patients 2 years of extended rituximab maintenance
is associated with roughly 2 years of additional PFS. At
5 years a trend towards improved OS was also seen
(74.3 versus 64.7; P = 0.07). This benefit was evident
despite the fact that the majority of patients in the
observation arm were treated with rituximab at sub-
sequent relapse. Extended rituximab therapy was asso-
ciated with a significant increase in grade 3 or 4
infections (9.7 versus 2.4%).
The ability of maintenance rituximab to prolong
PFS has also been shown by the GLSG. Patients with
relapsed FL who responded to either FCM or R-FCM
were randomized to either rituximab maintenance
(given as two courses of 4 weekly doses at 3 and 9
months) or to observation alone. This showed a clear
benefit in duration of response (17 months versus not
reached at 26 months median follow-up). Recent data
from the SAKK 35/98 trial demonstrated that as few as
four doses of rituximab maintenance given at 2-month
intervals significantly improved EFS when compared
to no maintenance. Although the benefit of rituximab
maintenance in relapsed FL is clear there is no con-
sensus on the optimum dose or duration of therapy.
Summary of immunochemotherapy
as second line treatment
Alkylating agents, anthracyclines, bendamustine, and
fludarabine are all effective in a proportion of patients
with relapsed FL. Response rates appear to be higher
when these drugs are used in combination, although at
the cost of greater short- and long-term toxicity.
98 Rituximab should be used in patients who have previ-
ously responded well to rituximab (>2 year remission)
and in those who are rituximab-naïve. Rituximab
should also be considered as maintenance in patients
who respond to second line therapy. It remains unclear
which are the optimal drugs to use in combination with
rituximab. Patients who have achieved durable remis-
sions from previous therapies may be considered for a
regimen containing the same agents at relapse.
Chemotherapy or antibodies that have previously failed
or induced only short remissions should be avoided.
Patients with clinically aggressive or advanced disease
at relapse should be considered for more intensive com-
bination regimens.
Novel agents
The activity of the proteasome inhibitor bortezomib in
relapsed FL has led investigators to study the combi-
nation of rituximab, bendamustine, and bortezomib.
Whilst this appears to be an effective regimen, no
randomized trial compares such combination therapy
with single or dual agent therapy. Single arm studies
suggest that the immunomodulatory agent lenalido-
mide has activity in relapsed FL when used as a single
agent or in combination with rituximab. The success
of rituximab has led to the development of several new
anti-CD20 antibodies. Ofatumumab is a fully human-
ized anti-CD20 monoclonal. Compared to rituximab
it is able to induce potent complement-dependent
cytotoxicity in vitro. However, a recent study in
patients with rituximab-resistant FL has shown disap-
pointing response rates. Its role in combination with
CHOP chemotherapy as first line treatment for FL is
currently being examined. Another anti-CD20 mono-
clonal GA101 has been engineered to induce potent
antibody-dependent cell-mediated cytotoxicity. Early
clinical data shows GA101 is active as a single agent in
up to 50% of patients treated with relapsed FL, includ-
ing cases resistant to rituximab-based regimens.
Inotuzumab ozogamicin (CMC-544) is a humanized
anti-CD22 antibody conjugated to calicheamicin, a
potent cytotoxic antitumor antibiotic that binds
DNA (producing double-stranded DNA breaks). In a
cohort of heavily pre-treated patients with FL who
were rituximab-refractory or had relapsed within 6
months of rituximab regimens, an ORR of 66% has
recently been reported. There are a number of other
monoclonal antibodies directed against B-cell antigens
in development. Presently randomized trials are under
way to further define the role of these newer anti-B-cell
monoclonal antibodies as single agents or in combi-
nation with chemotherapy in the management of FL.

Signaling through the B cell receptor is a requirement
for the survival of most normal B cells and many B cell
lymphomas. A number of small molecule kinase
inhibitors that target pathways downstream of the B
cell receptor are currently in early stage clinical trials
for FL. Of particular interest are Idelalisib (formerly
CAL-101) a selective inhibitor of the delta isoform of
PI3-Kinase and Ibrutinib, an irreversible inhibitor of
Bruton's tyrosine Kinase.
High-dose therapy and autologous
stem cell transplantation
The European CUP (Chemotherapy Unpurged
Purged) trial is the only randomized control trial that
has compared chemotherapy and ASCT in relapsed
FL. Between 1993 and 1997, 89 patients with relapsed
or refractory FL who had achieved a CR or PR follow-
ing induction therapy (the majority of patients treated
with CHOP), had a WHO performance status of 0–2,
and had limited bone marrow infiltration (defined as
<20% B-cells in the marrow aspirate) were random-
ized to receive three further chemotherapy courses, an
unpurged ASCT or a purged ASCT. As the numbers in
the three groups were small the two transplant groups
(unpurged and purged) were combined and compared
with chemotherapy, and an OS benefit for ASCT was
reported (Hazard ratio 0.40 [range 0.18–0.89],
P = 0.026), with an OS at 2 years of 71% in the ASCT
group and 46% in the chemotherapy group.
Importantly, this trial was conducted prior to the
introduction of rituximab to first line therapy.
A retrospective analysis of 100 consecutive patients
with relapsed or refractory FL treated with HDCT/
ASCT from 1993 to 2008 in Calgary showed an EFS
and OS at 5 years of 57% and 71%; at 10 years these are
projected to be 55% and 63%, respectively. With a
median follow-up of 63 months, severe toxicity
included two treatment-related deaths and four deaths
from secondary acute myeloid leukemia (AML)/MDS.
The Calgary group observed improvement in EFS for
patients treated since 2000 in their cohort, which they
hypothesized may be related to the more frequent use of
rituximab within 6 months of ASCT since 2000, either
with stem cell mobilization or preceding reinduction
therapy, reflecting an “in vivo” purging of autografts.
In addition, a number of phase II studies suggest an
apparent survival benefit for relapsed FL treated with
ASCT when compared to historical controls. In some
studies with prolonged follow-up there is also a
Chapter 6: Follicular lymphoma
suggestion of a plateau in survival curves. A combined
series from St Bartholomews and Dana Faber reported a
plateau in OS at 48% at 12 years. Although most of these
studies recruited patients prior to the introduction of
upfront rituximab, a recent case series has suggested
that ASCT remains an effective treatment for relapsed
patients who have received prior rituximab.
The role of rituximab at the time of autograft for
relapsed FL has been investigated in the Lym1 study.
This study shows that rituximab in vivo purging followed
by 2 years maintenance postautograft gives a superior PFS
compared to controls, with a suggestion that both the
purging and the maintenance contributed to the overall
improvement. No benefit in OS has yet been reported.
Perhaps the most contentious issue is the optimum
timing of ASCT for relapsed FL. The Rohatimer series
showed a clear association of improved survival with
patients receiving ASCT in second (as opposed to
later) remissions, suggesting that early ASCT is indi-
cated. However, this series (in line with others) showed
a high rate of late treatment-related toxicity, with a
12% mortality due to secondary MDS/AML. When
taken in the context of the recent improvements in
outcome for relapsed FL patients treated with ritux-
imab and other newly introduced treatment modal-
ities, this high toxicity means that ASCT may be best
reserved for patients with early relapse, aggressive dis-
ease, and those who have failed antibody treatment.
The exact role for ASCT remains to be defined in
randomized clinical trials.
Allogeneic stem cell transplantation
in FL (allo-SCT)
Allogeneic stem cell transplant (allo-SCT) is a poten-
tially curative therapy for relapsed FL and appears an
attractive proposition, particularly in a selected group
of younger patients. However, in contrast to other hem-
atologic malignancies where relapsed disease is often
untreatable, relapsed FL may still be amenable to treat-
ment with many other less toxic lines of therapy, includ-
ing novel agents. This fact must be weighed against the
significant morbidity and mortality attached to allo-
SCT, and the timing of transplant is a key clinical
decision. While historically much of the decreased dis-
ease recurrence seen after allo-SCT appears to have
been offset by the toxicity of the treatment, the risk–
benefit ratio for selected patients may be becoming
more favorable as outcomes of allo-SCT for FL appear
to be showing improvement over the past decade. 99
 Chapter 6: Follicular lymphoma
Unfortunately the absence of good trial evidence
makes it difficult to make specific recommendations in
the area of allo-SCT. There is limited data available to
guide patient selection and the optimum timing of
transplantation in the natural history of disease. Case
series and registry data of allo-SCT suggest that a
reduced intensity conditioning approach presently
appears to be preferable to myeloablative conditioning
regimens. The data also suggest that, while a sibling
donor for a reduced intensity conditioning (RIC)-allo-
SCT is preferable, a suitable unrelated HLA-compat-
ible donor can result in similar outcomes.
Chemorefractory disease at time of transplant consis-
tently appears as a poor prognostic factor in a number
of publications. An overview of the increasing num-
bers of studies reporting results from a number of
different conditioning regimens, in heterogeneous
groups of what appear to be generally poor risk
patients under the age of 65, suggests that non-relapse
mortality rates vary between about 15% and 35%
depending on the study.
Summary of allogeneic transplantation
Allogeneic stem cell transplant should be discussed with
younger patients with aggressive disease who have
relapsed or progressed despite conventional treatments
(chemotherapy, antibody therapies, and ASCT), and
who are considered otherwise fit enough for the proce-
dure. The significant toxicities mean that allogeneic
transplantation will be an option for only a minority
of patients. The optimum conditioning regimen is
unclear but reduced intensity strategies appear to be
associated with improved outcomes compared with
myeloablative conditioning protocols, at least in part
because of lower non-relapse mortality. A suitable unre-
lated HLA-matched donor appears to be a reasonable
alternative if an HLA-matched sibling donor is not
available. Allogeneic transplantation remains an exper-
imental therapy and patients should be treated where
possible within the context of a clinical trial.
Radioimmunotherapy
In patients with relapsed and refractory FL RIT shows
significant clinical activity, with remission rates
reported to be similar to those seen with R-CHOP. In
relapsed disease durable (>18-month) remissions have
been achieved in about 50% of patients with iodine-131
rituximab radioimmunotherapy. Furthermore, in
100 patients with rituximab-refractory FL, treatment with
Y90-ibritumomab has been shown to induce a response
in about 50% of patients with a median duration of 6
months. As with RIT following first line FL treatment,
in relapsed disease RIT remission consolidation has not
been directly compared with rituximab maintenance
strategies.
The principal toxicities of RIT are neutropenia and
thrombocytopenia, and it is necessary that patients
have <25% marrow tumor infiltration and good tri-
lineage hematopoiesis with well-preserved platelet
counts to have RIT. Treatment on the whole, however,
is well tolerated and may be most appropriate in
patients considered unfit for standard immunochemo-
therapy regimens. Careful consideration should be
given before treating patients who have received mul-
tiple prior therapies, particularly fludarabine. There
also remain concerns about the long-term risks of
MDS/AML in patients treated with RIT, particularly
those who have had previous purine analog therapy.
In relapsed FL, RIT appears effective and generally
well tolerated. In selected patients, such as those not
considered fit for immunochemotherapy or those who
are refractory or relapse early following rituximab-con-
taining regimens, RIT may be highly efficacious.
High-grade transformation of FL
Transformation of FL to high-grade lymphoma
appears to be an early event in the course of the disease
and is mainly observed in patients with known adverse
prognostic factors or those who do not achieve CR
after initial treatment. Furthermore, transformed
FL carries a poor prognosis despite treatment with
aggressive chemotherapy. In a retrospective analysis
of 220 patients over 15 years, Bastion and colleagues
reported a probability of transformation of 22%
at 5 years and 31% at 10 years, and observed a
plateau after 16 years. Predictive factors for transfor-
mation were non-achievement of CR after initial ther-
apy, low serum albumin level (<35 g/L), and beta
2-microglobulin level greater than 3 mg/L at diagnosis.
Transformation accounted for 44% of deaths and was
associated with a poor outcome, with a median
survival time of 7 months.
Conflicting data exists as to whether the timing or
nature of first line therapy influences the risk of high-
grade transformation. In contrast to other published
reports, a case series from British Columbia observed a
significant correlation between initial treatment regi-
men and risk of high-grade transformation. This series
included patients treated on two serial phase II studies;
the first cohort received multiagent chemotherapy

(bleomycin, cisplatin, etoposide, doxorubicin, cyclo-
phosphamide, vincristine, and prednisolone) followed
by 25 Gy IF-RT, while the second cohort received the
combination of an alkylator and a purine analog. The
10-year risk of transformation in these two cohorts
was 18% and 30%, respectively (P = 0.001). No differ-
ence in OS was seen. The authors hypothesized that
patients with advanced stage FL may, at diagnosis,
already harbor areas of occult transformation that
may be eradicated by a more aggressive initial treat-
ment regimen.
Data is limited as to the most appropriate course of
treatment in transformed FL. Anthracycline-based
combinations (e.g. CHOP or FMD) should be used
to induce remission induction. Intuitively, it seems
reasonable to add rituximab to the remission induc-
tion regimen, although data is lacking. In the event of
an inadequate response, previous anthracycline expo-
sure or progressive disease therapeutic options include
salvage chemotherapy (e.g. etoposide, methylpredni-
sone, cytarabine, cisplatin [ESHAP]; ifosfamide, car-
boplatin, etoposide [ICE], IVE ± rituximab) to
establish disease response and subsequent ASCT in
those with responsive disease. In patients treated this
way, 5-year survival is approximately 30%. RIT and
allo-SCT are more experimental strategies, but may
also be effective in selected individuals. Patients pre-
senting in transformation are managed differently to
those who transform following previous therapies.
Patients whose first presentation is with high-grade
transformation should generally be managed as high-
grade lymphoma, and the management of high-grade
lymphoma is described in more detail in Chapter 12.
Summary of relapsed follicular lymphoma
At present relapsed FL remains an incurable disease for
the majority of patients. Despite the increasing number
of treatment options, the optimal timing of the various
therapies has yet to be defined. Intervention is generally
recommended only in the context of symptomatic or
progressive disease. Rebiopsy and restaging of the lym-
phoma is recommended at each relapse. The choice of
therapy will be principally dictated by the tempo of
disease, response to previous therapy, and the patient’s
performance status. Chemotherapy or antibody combi-
nations that have provided a previous remission of 2
years or more should be considered again. Rituximab-
naïve patients should receive rituximab-based immu-
nochemotherapy. Rituximab should also be given as
maintenance in patients with chemosensitive disease.
Chapter 6: Follicular lymphoma
Patients with aggressive disease and those who relapse
early after immunochemotherapy should be considered
for RIT or ASCT. Younger patients who relapse despite
the above should be considered for an allogeneic trans-
plant. Where possible, patients should be managed in
the context of a clinical trial.
Further reading
Al-Tourah AJ et al. Population-based analysis of incidence
and outcome of transformed non-Hodgkin’s lymphoma.
J Clin Oncol, 2008;26:5165–5169.
Ardeshna KM et al. Long-term effect of a watch and wait
policy versus immediate systemic treatment for
asymptomatic advanced-stage non-Hodgkin lymphoma:
a randomised controlled trial. Lancet, 2003;362:516–522.
Bastion Y et al. Incidence, predictive factors, and outcome of
lymphoma transformation in follicular lymphoma
patients. J Clin Oncol, 1997;15:1587–1594.
Buske C et al. The Follicular Lymphoma International
Prognostic Index (FLIPI) separates high-risk from
intermediate- or low-risk patients with advanced-stage
follicular lymphoma treated front-line with rituximab
and the combination of cyclophosphamide, doxorubicin,
vincristine, and prednisone (R-CHOP) with respect to
treatment outcome. Blood, 2006;108:1504–1508.
Federico M et al. Follicular lymphoma international
prognostic index 2: a new prognostic index for follicular
lymphoma developed by the international follicular
lymphoma prognostic factor project. J Clin Oncol,
2009;27:4555–4562.
Federico M et al. R-CVP versus R-CHOP versus R-FM for
the initial treatment of patients with advanced-stage
follicular lymphoma: results of the FOLL05 trial
conducted by the Fondazione ltaliana Linfomi. J Clin
Oncol 2013;31:1506–1513.
Fisher RI et al. New treatment options have changed the
survival of patients with follicular lymphoma. J Clin
Oncol, 2005;23:8447–8452.
Forstpointner R et al. The addition of rituximab to a
combination of fludarabine, cyclophosphamide,
mitoxantrone (FCM) significantly increases the response
rate and prolongs survival as compared with FCM alone
in patients with relapsed and refractory follicular and
mantle cell lymphomas: results of a prospective
randomized study of the German Low-Grade Lymphoma
Study Group. Blood, 2004;104:3064–3071.
Gyan E et al. High-dose therapy followed by autologous
purged stem cell transplantation and doxorubicin-based
chemotherapy in patients with advanced follicular
lymphoma: a randomized multicenter study by the
GOELAMS with final results after a median follow-up of
9 years. Blood, 2009;113:995–1001. 101
 Chapter 6: Follicular lymphoma
Haas RL et al. High response rates and lasting remissions
after low-dose involved field radiotherapy in indolent
lymphomas. J Clin Oncol, 2003;21:2474–2480.
Harris NL, et al. Follicular lymphoma. In Swerdlow SH,
Campo E, Harris NL, et al. (eds) WHO Classification of
Tumours of Haemopoietic and Lymphoid Tissues. Lyon:
IARC, 2008:220–226.
Herold M et al. Rituximab added to first-line mitoxantrone,
chlorambucil, and prednisolone chemotherapy followed
by interferon maintenance prolongs survival in patients
with advanced follicular lymphoma: an East German
Study Group Hematology and Oncology Study. J Clin
Oncol, 2007;25:1986–1992.
Hiddemann W et al. Frontline therapy with rituximab added
to the combination of cyclophosphamide, doxorubicin,
vincristine, and prednisone (CHOP) significantly
improves the outcome for patients with advanced-stage
follicular lymphoma compared with therapy with CHOP
alone: results of a prospective randomized study of the
German Low-Grade Lymphoma Study Group. Blood,
2005;106:3725–3732.
Jegalian AG, et al. Follicular lymphoma in situ: clinical
implications and comparisons with partial
involvement by follicular lymphoma. Blood,
2011;118:2976–2984.
Kaminski MS et al. 131I-tositumomab therapy as initial
treatment for follicular lymphoma. N Engl J Med,
2005;352:441–449.
Ladetto M et al. Prospective, multicenter randomized
GITMO/IIL trial comparing intensive (R-HDS) versus
conventional (CHOP-R) chemoimmunotherapy in
high-risk follicular lymphoma at diagnosis: the
superior disease control of R-HDS does not translate into
an overall survival advantage. Blood,
2008;111:4004–4013.
Leahy MF, Turner JH. Radioimmunotherapy of relapsed
indolent non-Hodgkin lymphoma with 131I-rituximab
in routine clinical practice: 10-year single-institution
experience of 142 consecutive patients. Blood,
2011;117:45–52.
Lenz G et al. Myeloablative radiochemotherapy followed by
autologous stem cell transplantation in first remission
prolongs progression-free survival in follicular
lymphoma: results of a prospective, randomized trial of
the German Low-Grade Lymphoma Study Group. Blood,
2004;104:2667–2674.
Lowry L et al. Reduced dose radiotherapy for local control in
non-Hodgkin lymphoma: a randomised phase III trial.
Radiother Oncol 2011;100:86–92.
Marcus R et al. CVP chemotherapy plus rituximab
compared with CVP as first-line treatment for advanced
follicular lymphoma. Blood, 2005;105:1417–1423.
102
Marcus R et al. Phase III study of R-CVP compared with
cyclophosphamide, vincristine, and prednisone alone in
patients with previously untreated advanced follicular
lymphoma. J Clin Oncol, 2008;26:4579–4586.
Mann RB, Berard CW. Criteria for the cytologic
subclassification of follicular lymphomas: a proposed
alternative method. Hematol Oncol, 1983;1:187–192.
Morris E et al. Outcomes after alemtuzumab-containing
reduced-intensity allogeneic transplantation regimen for
relapsed and refractory non-Hodgkin lymphoma. Blood,
2004;104:3865–3871.
Morschhauser F et al. Phase III trial of consolidation therapy
with yttrium-90-ibritumomab tiuxetan compared with
no additional therapy after first remission in advanced
follicular lymphoma. J Clin Oncol, 2008;26:5156–5164.
Pettengell R et al. Rituximab purging and/or maintenance
in patients undergoing autologous transplantation for
relapsed follicular lymphoma: a prospective randomized
trial from the lymphoma working party of the European
group for blood and marrow transplantation. J Clin
Oncol 2013;31:1624–1630.
Pugh TJ, Ballonoff A, Newman F, Rabinovitch R. Improved
survival in patients with early stage low-grade follicular
lymphoma treated with radiation: a Surveillance,
Epidemiology, and End Results database analysis. Cancer,
2010;116:3843–3851.
Rohatiner AZ et al. Myeloablative therapy with autologous
bone marrow transplantation for follicular lymphoma at
the time of second or subsequent remission: long-term
follow-up. J Clin Oncol, 2007;25:2554–2559.
Rummel MJ et al. Bendamustine plus rituximab versus
CHOP plus rituximab as first-line treatment for
patients with indolent and mantle-cell lymphomas:
an open-label, multicentre, randomised, phase 3
non-inferiority trial. Lancet 2013;381:1203–1210.
Salles G et al. Rituximab combined with chemotherapy and
interferon in follicular lymphoma patients: results of the
GELA-GOELAMS FL2000 study. Blood,
2008;112:4824–4831.
Salles G et al. Rituximab maintenance for 2 years in patients
with high tumour burden follicular lymphoma
responding to rituximab plus chemotherapy (PRIMA):
a phase 3, randomised controlled trial. Lancet,
2011;377:42–51.
Schouten HC et al. High-dose therapy improves
progression-free survival and survival in relapsed
follicular non-Hodgkin’s lymphoma: results from the
randomized European CUP trial. J Clin Oncol,
2003;21:3918–3927.
Schulz H et al. Chemotherapy plus rituximab versus
chemotherapy alone for B-cell non-Hodgkin’s

lymphoma. Cochrane Database Syst Rev, 2007;
CD003805.
Sebban C et al. Standard chemotherapy with
interferon compared with CHOP followed by
high-dose therapy with autologous stem cell
transplantation in untreated patients with
advanced follicular lymphoma: the GELF-94
randomized study from the Groupe d’Etude des
Lymphomes de l’Adulte (GELA). Blood,
2006;108:2540–2544.
Solal-Celigny P et al. Follicular lymphoma international
prognostic index. Blood, 2004;104:1258–1265.
Thomson KJ et al. T-cell-depleted reduced-intensity
transplantation followed by donor leukocyte infusions to
promote graft-versus-lymphoma activity results in
excellent long-term survival in patients with multiply
relapsed follicular lymphoma. J Clin Oncol,
2010;28:3695–3700.
Trotman J et al. Positron emission tomography-computed
tomography (PET-CT) after induction therapy is
highly predictive of patient outcome in follicular
Chapter 6: Follicular lymphoma
lymphoma:analysis of PET-CT in a subset of
PRIMA trial participants. J Clin Oncol
2011;29: 3194–3200.
van Oers MH et al. Rituximab maintenance improves
clinical outcome of relapsed/resistant follicular non-
Hodgkin lymphoma in patients both with and without
rituximab during induction: results of a prospective
randomized phase 3 intergroup trial. Blood,
2006;108:3295–3301.
van Oers MH et al. Rituximab maintenance treatment of
relapsed/resistant follicular non-Hodgkin’s lymphoma:
long-term outcome of the EORTC 20981 phase III
randomized intergroup study. J Clin Oncol,
2010;28:2853–2858.
Witzig TE et al. Randomized controlled trial of
yttrium-90-labeled ibritumomab tiuxetan
radioimmunotherapy versus rituximab
immunotherapy for patients with relapsed or
refractory low-grade, follicular, or transformed B-cell
non-Hodgkin’s lymphoma. J Clin Oncol,
2002;20:2453–2463.
103
 Chapter
7
MALT and other marginal zone lymphomas
Emanuele Zucca and Francesco Bertoni
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Extranodal marginal zone B-cell
lymphoma (MALT lymphoma)
Clinical features
Lymphoma of the mucosa-associated lymphoid tissue
(MALT lymphoma) comprises about 7–8% of all non-
Hodgkin lymphomas (NHLs). It is a neoplasm of adults
with a median age at presentation of about 60 years and
with a slightly higher proportion of females than males.
The presenting symptoms are essentially related to the
primary location. Few patients present with elevated
lactate dehydrogenase (LDH) or β2 microglobulin lev-
els. Constitutional B-symptoms are extremely uncom-
mon. MALT lymphoma usually remains localized for a
prolonged period within the tissue of origin, but dis-
semination to multiple sites is not uncommon and has
been reported in up to one-quarter of cases, with either
synchronous or metachronous involvement of multiple
mucosal sites or non-mucosal sites such as the bone
marrow, the spleen, or the liver. Regional lymph nodes
can also be involved. Bone marrow involvement is
reported in up to 20% of cases.
The stomach is the commonest localization, repre-
senting about one-third of the cases. Other typical
presentation sites include the salivary glands, the
orbit, the thyroid, and the lung; the frequency of
occurrence in different organs is shown in Table 7.1.
Within the stomach, MALT lymphoma is often
multifocal, possibly explaining the reports of relapses
in the gastric stump after surgical excision. Gastric
MALT lymphoma can often disseminate to the small
intestine and to the splenic marginal zone. Concomitant
gastrointestinal (GI) and non-GI involvement can be
detected in approximately 10% of cases. Disseminated
disease appears to be more common in non-GI MALT
104 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
lymphomas, in which it has been reported in up to one-
quarter of the patients. It has been postulated that dis-
semination of MALT lymphoma may be a result of
specific expression of special homing receptors or adhe-
sion molecules on the surface of most B-cells of MALT,
whether normal or transformed.
MALT lymphoma usually has a favorable outcome,
with overall survival (OS) at 5 years higher than 85% in
most series. Patients at high risk according to the
International Prognostic Index (IPI) were found to
carry a worse prognosis in retrospective series and in
one prospective study. Patients with lymph node or
bone marrow involvement at presentation, but not
those with involvement of multiple mucosal sites, may
have a worse prognosis. If initially localized, the disease
is generally slow to disseminate. Recurrences may
involve either extranodal or nodal sites. The median
time to progression has been reported to be better for
the GI than for the non-GI lymphomas, but with no
significant differences in OS between the groups.
Localization may have prognostic relevance because of
organ-specific clinical problems that determine partic-
ular management strategies, but also because of specific
local pathogenic mechanisms, as suggested by the
reports of different frequency of different chromosomal
translocations at distinct anatomic locations.
In a radiotherapy study from Toronto, gastric and
thyroid MALT lymphomas had the best outcome,
whereas distant failures were more common for
other sites. In a multicenter series from the
International Extranodal Lymphoma Study Group
(IELSG), patients with the disease initially presenting
in the upper airway appeared to have a slightly poorer
outcome (Table 9.1), but their small number pre-
vented any definitive conclusion. In general, despite
frequent relapses, MALT lymphomas most often
 Chapter 7: MALT and other MZLs

Table 7.1 Main clinical features and survival of MALT lymphomas at different presentation sites.

Extranodal site Frequency Stage I Elevated Nodal Bone marrow Overall

LDH involvement involvement survival
at 5 years
GI localization
Stomach 33% 88% 2% 4% 8% 82–88%
Intestine 3–9% 56% 11% 44 0% 59–100%
Non-GI localization
Skin 8–10% 82% 9% 0 9% 84–100%
Ocular adnexa* 10–12% 84% 26% 10% 13% 90–94%
Salivary glands 16% 83% 17% 11% 9% 96–100%
Lung 6–10% 60% 27% 27% 7% 84–100%
Upper airways 5–10% 50% 42% 33% 33% 46–80%
Breast 2–3% 100% 33% 0% 0% 100%
Thyroid 4% 60% 10% 40% 0% 100%
Multiple mucosal 10–30% Not applicable 26% 45% 33% 77% (43–93)
sites**
Derived from Pinotti et al., 1997, Thieblemont et al., 2000, Zucca et al., 2003, de Boer et al., 2008, Mazloom et al., 2010, Zucca et al., 2010.
* Recalculated from the published split data on orbital and conjunctival cases (Zucca et al., 2003).
** With or without nodal and/or bone marrow involvement.

maintain an indolent course. Histologic transforma-

tion to large-cell lymphoma is reported in about 10%
of cases, usually late during the course of the disease
and independently from dissemination. It is unknown
whether the incidence of transformation is different at
different primary anatomic sites.

MALT lymphoma pathology

In early cases the neoplastic cells infiltrate around the
periphery of lymphoid follicles outside the mantle
zone in the region that would normally be occupied
by marginal zone B-cells (such as are found in the
Peyer’s patches of the terminal ileum). As the disease
progresses the lymphoma may overrun the lymphoid
follicles and residual germinal centers may be difficult
to identify, although the lymphoma usually retains at
least a focal nodular growth pattern. Even in those
cases with preserved lymphoid follicles, colonization
of the reactive germinal centers is frequently encoun-
tered (Figure 7.1).
The morphology of the cells is variable and this
cellular variation can be seen within a single lesion.
Classically the cells are described as having moderate
amounts of cytoplasm with small irregular nuclei, fine
chromatin, and indistinct nucleoli that resemble the
Figure 7.1 Extranodal marginal B-cell (MALT) lymphoma (stomach).
There is a perifollicular infiltrate of small lymphoid cells with irregular
nuclei. Glands are infiltrated and destroyed to form lymphoepithelial
lesions.
centrocyte of the germinal center. However, the cells
may have less abundant cytoplasm and round nuclei
resembling small lymphocytes or have more abundant
cytoplasm with crisp cytoplasmic borders and round
nuclei giving the cell a monocytoid appearance.
Plasma cell differentiation is frequent and may be so 105
pronounced as to raise the possibility of a plasma cell
 Chapter 7: MALT and other MZLs
Figure 7.2 Extranodal marginal B-cell (MALT) lymphoma (stomach).
Some cases show prominent plasma cell differentiation. In this case
the plasma cells are distended by accumulated immunoglobulin.
neoplasm (Figure 7.2). Scattered large cells with abun-
dant cytoplasm, vesicular nuclei, and prominent
nucleoli are present within the infiltrate.
In cells associated with epithelial structures the
infiltrate lies between the follicle and the epithelium,
and neoplastic cells infiltrate and destroy the epithe-
lial structures to form lymphoepithelial lesions
(Figure 7.3). The disruption of the epithelial struc-
tures is associated with morphological changes in the
epithelial cells that impart an eosinophilic appear-
ance to their cytoplasm.
Involved lymph nodes show either a perifollicular
infiltrate or, in more advanced cases, a nodular pro-
liferation of similar cells. These cases may mimic nodal
marginal zone lymphoma (MZL).
Immunophenotypically MALT lymphomas express
surface immunoglobulin (usually IgM, less often IgA or
IgG). The lymphocytes express CD19, CD20, CD22,
CD79a, and PAX5. Approximately half of cases express
CD43. They are usually negative for CD5, CD23, CD10,
and cyclin D1. There is expression of BCL-2 protein. A
small percentage of cases show expression of CD5 and
these may be associated with a more aggressive presen-
tation including a leukemic phase.
MALT lymphomas need to be distinguished from
other low-grade lymphomas, including follicular lym-
phoma (FL), mantle cell lymphoma, and small lym-
phocytic lymphoma/B-cell chronic lymphocytic
leukemia (B-CLL). This can usually be done by mor-
phology and immunophenotypic studies. Distinction
106 from underlying reactive conditions that are known to
be precursor lesions to the development of MALT
Figure 7.3 Extranodal marginal B-cell (MALT) lymphoma (salivary
gland). MALT lymphomas at all sites share a common morphological
appearance. The cells infiltrate associated epithelial structures to form
lymphoepithelial lesions.
lymphoma (e.g. Hashimoto’s thyroiditis in the thyroid
and Sjögren’s syndrome in the salivary glands) is
essential.
The molecular biology of MALT
lymphomas
The molecular genetic and cytogenetic features of
extranodal lymphomas are in keeping with their
marked heterogeneity with respect to morphology,
immunophenotype, clinical presentation, and clinical
outcome.
Extranodal marginal zone B-cell lymphomas of
MALT-type represent mature B-cell tumors with pro-
ductively rearranged immunoglobulin heavy-chain
(IgH) genes. Interestingly, the VH4, VH3, and VH1
family genes known to be involved in autoantibody
formation are preferentially used, suggesting a possi-
ble derivation from autoreactive B-cell clones. The
rearranged IgVH genes are usually mutated and often
show a mutation pattern indicative of selective antigen
pressure. Recent studies provide evidence that MALT-
type lymphomas from special localizations (gastric
and salivary gland), in contrast to any other lym-
phoma subtype, frequently express B-cell receptors
with strong homology to the rheumatoid factor (RF).
This finding underscores the strong link between
chronic antigenic stimulation, autoimmunity, and
development of MALT-type lymphomas.
Recurrent genetic aberrations in MALT-type lym-
phomas include trisomies of chromosomes 3 and 18,
DNA gains at 6p and 8q, and losses at 6q23 (TNFAIP3),

and translocations t(1;14)(p22;q32), t(11;18)(q21;q21),
t(14;18)(q32;q21), affecting MALT1, t(3;14)(q27;q32)
and t(3;14)(p13;q32). In addition, extranodal MALT-
type lymphomas lack translocations common in other
low-grade B-cell lymphomas, e.g. t(11;14)(q13;q32), or
t(14;18)(q32;q21) affecting BCL2. In some MALT-type
lymphomas, disease progression and high-grade trans-
formation is associated with MYC rearrangements and
inactivation/deletion of the TP53 and CDKN2A genes.
Three translocations, t(1;14)(p22;q32), t(11;18)
(q21;q21), and t(14;18)(q32;q21) are of particular
interest, since they appear to be specific for this subtype
of B-cell lymphoma. In addition, they target a common
oncogenic pathway resulting in nuclear factor kappa B
(NFκB) activation. The translocation t(11;18)(q21;q21)
generates a chimeric fusion between the BIRC3/API2
and MALT1 genes, whereas translocations t(1;14)(p22;
q32) and t(14;18)(q32;q21) result in deregulated
expression of BCL10 and MALT1, respectively, by jux-
taposing them to regulatory sequences of the IgH
gene cluster. In fact, gene expression analyses recently
demonstrated the preferential upregulation of NFκB
target genes in translocation-positive MALT lympho-
mas in contrast to translocation-negative tumors. The
translocation t(11;18)(q21;q21) is specifically associ-
ated with MALT-type lymphoma and has not been
detected in any other lymphoma subtype or in uncom-
plicated Helicobacter pylori-associated gastritis.
Cytogenetically, the t(11;18) is almost always the sole
chromosomal aberration, in contrast to t(11;18)-
negative tumors that often show other alterations
such as trisomies 3, 12, and 18 and allelic imbalances.
Importantly, gastric MALT lymphomas with t(11;18)
do not respond to H. pylori eradication therapy and
also rarely undergo high-grade transformation. These
data suggest that lymphomas carrying the t(11;18)(q21;
q21) may constitute a distinct subgroup of MALT-type
lymphoma that may require specific treatment strat-
egies. On the molecular level, the BIRC3–MALT1
fusion product consists of three BIR (Baculovirus
Inhibitor of apoptosis protein Repeat) domains present
in the N-terminus of the BIRC3 protein and a variable
part of the MALT1 protein, which always contains the
caspase-like p20 domain. The chimeric protein
BIRC3–MALT1 effectively activates the NFκB survival
pathway in vitro and it has been proposed that con-
stituents of the BIR domain may mediate the oligome-
rization of the BIRC3–MALT1 fusion protein, thereby
conferring its ability to activate NFκB.
Chapter 7: MALT and other MZLs
The translocations t(1;14)(p22;q32) and t(1;2)
(p22;p12) occur in less than 4% of MALT-type lym-
phomas. BCL10, the target gene of these transloca-
tions, specifically links antigen receptor signaling to
the NFκB pathway. Wild-type BCL10 protein is
expressed in the cytoplasm in normal germinal center
and marginal zone B-cells. MALT-type lymphomas
with translocations t(1;14)(p22;q32) and t(1;2)(p22;
p12) show strong nuclear BCL10 labeling, as is also
the case in t(11;18)-positive tumors. This suggests
that nuclear BCL10 expression may confer oncogenic
activity.
In addition to the t(11;18)(q21;q21), MALT1 is
also targeted by the translocation t(14;18)(q32;q21)
and by genomic amplifications. MALT1 alone is
not able to activate NFκB, but acts synergistically
with BCL10. Upon antigen receptor stimulation, a
member of the membrane-associated guanylate kin-
ase (MAGUK) family of scaffolding proteins such
as CARD11/CARMA1 is recruited to the receptor
complex. CARD11 subsequently recruits BCL10
and induces its oligomerization. This is followed by
BCL10-mediated MALT1 oligomerization, which
activates the IκB kinase complex and, through a
cascade of events, results in NFκB activation.
Interestingly, t(14;18)(q32;q21)-positive MALT-type
lymphomas coexpress BCL10 and MALT1 protein
in the cytoplasm, with perinuclear accumulation of
BCL10. An additional recurrent translocation, t(3;14)
(p13;q32), was reported in MALT-type lymphomas
from different localizations, resulting in the deregu-
lated expression of the Forkhead Box P1 (FOXP1)
gene by juxtaposing it to the IgH regulatory elements.
In subsequent studies, FOXP1 translocations were
also identified in cases of diffuse large B-cell lympho-
mas (DLBCL), suggesting that this genetic event may
not be confined to MALT lymphomas alone.
Interestingly, the frequencies of the translocations
t(11;18), t(14;18), t(1;14), and t(3;14), and also
the number of cases with numerical aberrations,
clearly differ between tumors arising at different
localizations, e.g. MALT lymphomas of the lung
and MALT lymphomas of the parotid gland or the
ocular adnexae, pointing to the possible importance
of local factors influencing the transformation
pathways.
Another recurrent lesion contributing to NFκB
activation is the inactivation of TNFAIP3 gene, map-
ped at 6q23, that codes for a negative regulator.
107
 Chapter 7: MALT and other MZLs

Malt lymphoma staging disease seem to have the same long-term outcome as
those with localized disease.
Since patients presenting with lymphoma dissemi-
Outside clinical trials, aggressive staging proce-
nated to multiple mucosal sites may have a favorable
dures should be tailored to the individual patient
outcome similar to the patients with localized disease,
according to the clinical conditions (presenting local-
the traditional Ann Arbor staging system, which is
ization, age, intended treatment, performance status,
mainly based on the extension of nodal areas, is not
and symptoms). Staging procedures should always
optimal. Alternative staging systems (Table 7.2) for
comprise a complete clinical history and physical
extranodal lymphomas were proposed in the 1970s,
examination with a careful evaluation of all lymph
but a general consensus has not yet been achieved.
node regions, inspection of the upper airway and ton-
The main problem in staging patients with MALT
sils, thyroid examination, and clinical evaluation of the
lymphoma is the number of extranodal sites to be
size of liver and spleen. Standard chest radiographs
explored at diagnosis. The finding of asymptomatic
and a CT scan of thorax, abdomen, and pelvis should
dissemination in patients with apparently localized
be performed. Bone marrow biopsy must be per-
disease, and the relatively high proportion of patients
formed at diagnosis. Laboratory tests should include
with early dissemination, suggests the need for exten-
complete blood counts with cytological examination,
sive staging procedures in all patients with MALT
LDH, and β2 microglobulin levels, evaluation of renal
lymphoma. However, patients with disseminated

Table 7.2 Staging systems for GI tract lymphoma.

Ann Arbor Musshoff Lugano (Rohatiner, TNM modifications in the Lymphoma extension
(Carbone, (Musshoff, 1994) Paris system
1971) 1977) (Ruskone-Fourmestraux,
2003)
IE IE1 I = confined to GI tract T1m N0 M0 Mucosa
(single primary or
multiple, non-
contiguous)
IE IE2 T1sm N0 M0 Submucosa
IE T2 N0 M0 Muscularis propria
T3 N0 M0 Serosa
IIE IIE1 II = extending into Tl–3 Nl M0 Perigastric lymph nodes
abdomen
IIE IIE2 II1 = local nodal Tl–3 N2 M0 More distant regional nodes
involvement
II2 = distant nodal Tl–3 N3 M0 Extra-abdominal lymph nodes
involvement
IIE IIE IIE = penetration of T4 N0 M0 Invasion of adjacent structures
serosa to involve
adjacent organs or
tissues
IIIE IIIE IV = disseminated T1–4 N3 M0 Lymph nodes on both sides of the
extranodal diaphragm, and/or additional
involvement or extranodal sites with non-
concomitant continuous involvement of
supradiaphragmatic separate GI site
nodal involvement
IVE IVE T1–4 N0–3 M1 Or non-continuous involvement
T1–4 N0–3 M2 of non-GI sites
Tl–4 N0–3 M0–2 BX Bone marrow not assessed
Tl–4 N0–3 M0–2 B0 Bone marrow not involved
108 Tl–4 N0–3 M2 B1 Bone marrow involvement
 Chapter 7: MALT and other MZLs

Table 7.3 Site-specific investigations that may be useful for the staging of extranodal
lymphomas of MALT type.

Stomach Gastroduodenal endoscopy with multiple gastric biopsies from all the visible
lesions and the non-involved areas with a complete mapping
Histological and histochemical examination for H. pylori (Genta stain or
Warthin–Starry stain of antral biopsy specimen) and serology studies if
histology results are negative
FISH for the t(ll;18) translocation
Endoscopic ultrasound
Small intestine Endoscopy
(IPSID) Small bowel double-contrast X-ray
Campylobacter jejuni search in the tumor biopsy by PCR,
immunohistochemistry or in-situ hybridization
Large intestine Colonoscopy
Endoscopic ultrasound
Ocular adnexa MRI (or CT scan) and ophthalmologic examination
Chlamydophila psittaci in the tumor biopsy and blood
Mononuclear cells by PCR
Lung Bronchoscopy with bronchoalveolar lavage
Salivary gland, ENT examination and ultrasound
tonsils, parotid
Thyroid Ultrasound ±CT scan of the neck and thyroid function tests
Breast Mammography and MRI
Skin Borrelia burgdorferi in the tumor biopsy by PCR
CT, Computed tomography; ENT, ear, nose, and throat; FISH, fluorescent in situ hybridization; IPSID,
immunoproliferative small intestinal disease; MRI, magnetic resonance imaging; PCR, polymerase chain
reaction.

and liver function, and hepatitis C virus (HCV) and
HIV serology. Utility of positron emission tomogra-
phy (PET) scanning remains unclear, with conflicting
reports on 18FDG avidity of extranodal MZLs, and,
thus, the exam is not currently recommended. Then,
depending upon the particular clinical presentation,
staging investigations should focus on the specific
organs suspected of being involved (Table 7.3).
Diagnosis and staging of gastric MALT lymphoma
The stomach is the most common and best-known
MALT lymphoma site. The most frequent presenting
symptoms of gastric MALT are non-specific upper
GI complaints (dyspepsia, epigastric pain, nausea,
and chronic manifestations of GI bleeding such as
anemia), often leading to an endoscopy that usually
reveals non-specific gastritis or peptic ulcer, with
mass lesions being unusual. Diagnosis is based on
the histopathologic evaluation of the gastric
biopsies.
The best staging system is still controversial. The
Ann Arbor staging system, routinely used for NHL,
was developed for Hodgkin’s disease and is not
adequate for the specific problems posed by GI tract
lymphomas. A variety of alternative systems have
therefore been proposed, as summarized in
Table 7.2. We have largely used the modification of
the Blackledge system known as the “Lugano” staging
system. However, this system was proposed before
the wide use of endoscopic sonography and does not
accurately describe the depth of infiltration in the
gastric wall, a parameter that is highly predictive for
the MALT lymphoma response to anti-Helicobacter
therapy.
There is a general consensus that initial staging
procedures should include a gastroduodenal endos-
copy, with multiple biopsies from each region of the
stomach, duodenum, and gastroesophageal junction,
as well as from any abnormal-appearing site. Fresh
biopsy and washings material should be available for
cytogenetic studies in addition to routine histology
and immunohistochemistry. Fluorescent in situ
hybridization (FISH) analysis or a molecular assay
for the detection of t(11;18) is recommended for 109
 Chapter 7: MALT and other MZLs
Table 7.4 Complete remission and relapse rates after antibiotic therapy for localized (stage I–II) gastric MALT lymphomas in recent series
with at least 3 years of median follow-up.
Author Number of
patients
Fischbach et al. Gut, 2004 90
Montalban et al. Ann Oncol, 24
2005
Wundisch et al. J Clin Oncol, 120
2005
Kim et al. Br J Cancer, 2007 111
Terai et al. Tohoku J Exp Med, 74
2008
Andriani et al. Dig Liv Dis, 60
2009
Stathis et al. Ann Oncol, 2009 105
CR, complete lymphoma remission after H. pylori; relapse rate, histological relapse rate among patients with previous CR.
identifying disease that is unlikely to respond to anti-
biotic therapy. The presence of active infection must
be determined by histochemistry and Helicobacter
pylori breath test; serologic studies are recommended
when the results of histology are negative. Monoclonal
stool antigen test has been proposed as a reliable and
accurate alternative method for H. pylori detection.
Endoscopic ultrasound is recommended in the initial
follow-up for evaluation of depth of infiltration and
presence of perigastric lymph nodes. Deep infiltration
of the gastric wall is associated with a higher risk of
lymph node involvement and a lower response rate
with antibiotic therapy.
Staging should include complete blood counts,
basic biochemical studies (see above), CT of the
chest, abdomen, and pelvis, and a bone marrow
biopsy. Although the disease usually remains localized
in the stomach, systemic dissemination and bone mar-
row involvement should be excluded at presentation,
since prognosis is worse with advanced-stage disease
or with unfavorable IPI score.
Staging procedures in non-gastric MALT lymphoma
Presentation with multiple MALT localizations is
more frequent in patients with non-GI lymphoma,
and a careful initial staging is therefore important.
Staging should include basic laboratory studies (see
above), total-body CT scans, and a bone marrow
biopsy. In addition, the different anatomic sites may
require specific procedures, as summarized in
110
Table 7.3.
Stage I CR rate Median follow-up Relapse rate
(%) (years) (%)
90 62 4 7
24 91 4 4
120 80 6 17
111 85 3 6
74 89 4 5
50 79 5 23
100 65 6 19
Management and follow-up
A treatment algorithm for early- and advanced-stage
MALT lymphoma is shown in Figure 7.4.
Helicobacter pylori eradication in gastric MALT
lymphoma
A major change in the management of gastric lym-
phoma occurred following the identification of H.
pylori as the etiologic agent for gastric MALT lym-
phoma. Surgical resection, often followed by postop-
erative radiotherapy or chemotherapy, was standard
up to the early 1990s. After the initial report by
Wotherspoon and colleagues in 1993, eradication of
H. pylori with antibiotics as the sole initial treatment of
localized (i.e. confined to the gastric wall) H. pylori-
positive MALT lymphoma has now become the stand-
ard therapeutic approach.
Durable lymphoma remissions have been docu-
mented in multiple studies for 60–100% of patients
with localized H. pylori-positive gastric MALT lym-
phoma treated with antibiotics (Table 7.4 summarizes
the results of recent studies with long follow-up).
Differences in the response criteria adopted in the
individual studies to evaluate the lymphoma eradica-
tion after antibiotic therapy may explain the wide
range of reported remission rates. The histologic
remission can usually be documented within 6 months
from the H. pylori eradication, but sometimes the
period required is more prolonged and the therapeutic
response may be delayed up to more than 1 year.
 Chapter 7: MALT and other MZLs

Figure 7.4 A treatment algorithm for

Gastric MALT lymphoma, stage IE−IIE early- and advanced-stage MALT
lymphoma. PPI, proton pump inhibitor;
H. pylori assessment EGD, esophagogastroduodenoscopy;
IF-RT, involved field radiotherapy.
H. pylori-positive
H. pylori-negative or
t(11; 18)-negative or
H. pylori-positive with t(11; 18)
undetermined

Consider trial of antibiotics

Antibiotics and PPI and PPI

H. pylori test at 2 months

(if positive for H. pylori, treat with 2nd line Antibiotic-resistant
antibiotics regimen)
EGD and biopsies at 3−6 months

After H. pylori eradication IF-RT

Negative for Positive for

lymphoma lymphoma,
asymptomatic Positive for lymphoma,
symptomatic or with other
treatment indications [overt
Follow-up with EGD progression, deep invasion, nodal
and biopsy Monitor by EGD involvement, t(11; 18)]
q6 months × 2 years, and biopsy
then yearly q3–6 months
Chlorambucil or
other alkylating agent
and/or rituximab
if radiotherapy
not feasible

Several effective anti-H. pylori programs are avail-
able. The choice should be based on the epidemiology
of the infection in the different countries, taking into
account the locally expected antibiotic resistance. The
most commonly used regimen is triple therapy with a
proton pump inhibitor (e.g. omeprazole, lansoprazole,
pantoprazole, or esomeprazole) in association with
amoxicillin and clarithromycin. Metronidazole can
substitute for amoxicillin in penicillin-allergic indivi-
duals. Other regimens that include bismuth or H2-
receptor antagonists with antibiotics are also effective.
The role of additional chemotherapy after H. pylori
eradication was investigated in a randomized study
(whose power, however, has been limited from not
having reached the planned accrual). In this study,
chlorambucil conferred no benefit, with progression-
free survival (PFS) and OS rates similar for observed-
only and chlorambucil-treated patients.
Anti-Helicobacter therapy in gastric diffuse
large B-cell lymphoma
Anti-Helicobacter therapy may also be of benefit in
some cases of DLBCL of the stomach since, in the
subset of cases that may have been derived from a
MALT lymphoma, antibiotics may eliminate a resid-
ual or relapsed low-grade component that can be
responsible for tumor recurrence following antigen
stimulation. Cases of regression of high-grade lesions
after anti-H. pylori therapy have been reported,
suggesting that high-grade transformation is not ne-
cessarily associated with the loss of H. pylori depend-
ence. However, relying solely on antibiotic therapy
for gastric large-cell lymphomas cannot be advised
outside clinical trials until large-scale prospective
studies have validated its use as first line therapy,
and at present we recommend treating them as local- 111
ized DLBCL.
 Chapter 7: MALT and other MZLs

Figure 7.4 (cont.)

Gastric MALT lymphoma, stage III–IV

H. pylori eradication if
infection present

Asymptomatic Symptomatic or with other

treatment indications (impending
organ damage, overt progression,
bulky disease, patient preference)
Wait and see:
EGD and biopsies
every 6 months
Chemotherapy
and/or rituximab,
or consider
including in
clinical trials

Non-gastric MALT lymphoma

Consider anti-infectious therapies against
Borrelia, Chlamydophila, Campylobacter or
HCV−associated lymphomas

Stage IE-IIE Stage III–IV

Resection or IF-RT Symptomatic or with other

Asymptomatic
treatment indications

Wait and see

Chemotherapy
and/or rituximab,
or consider
including in
clinical trials

Post-treatment histologic evaluation become a useful tool if its reproducibility can be

The interpretation of residual lymphoid infiltrate in
post-treatment gastric biopsies can be very difficult,
and there are no uniform criteria in the literature for
the definition of histologic remission.
The Wotherspoon score shown in Table 7.5 can
be very useful to express the degree of confidence in
the MALT lymphoma diagnosis on small gastric
biopsies, but it is difficult to apply in the evaluation
of the response to therapy, and for this reason other
criteria have been proposed. The lack of standard
reproducible criteria can affect the comparison of
the results of the different clinical trials. Copie–
Bergmann and colleagues have proposed a new
(GELA) histological grading system with the aim of
112 providing relevant information to the clinician. This
system, which is also summarized in Table 7.5, may
confirmed on larger series.
Factors predicting the lymphoma response
to H. pylori eradication
Endoscopic ultrasound can be useful to predict the
lymphoma response to H. pylori eradication. There is
a significant difference between the response rates of
lymphomas restricted to the gastric mucosa and those
with less superficial lesions. The response rate is high-
est for the mucosa-confined lymphomas (approxi-
mately 70–90%) and then decreases markedly and
progressively for the tumors infiltrating the submu-
cosa, the muscularis propria, and the serosa. In cases
with documented nodal involvement, response is
highly unlikely. Nearly all gastric lymphomas with
the t(11;18) translocation fail to respond to H. pylori
 Chapter 7: MALT and other MZLs

Table 7.5 The Wotherspoon and GELA systems for the histological evaluation of gastric MALT lymphoma endoscopic biopsies.

Wotherspoon’s score (for diagnosis and post-treatment evaluation)

0 Normal Scattered plasma cells in LP
1 Chronic active gastritis Small clusters of lymphocytes in LP, no lymphoid follicles, no LEL
2 Chronic active gastritis with Prominent lymphoid follicles with surrounding mantle zone and plasma cells, no LEL
lymphoid follicles
3 Suspicious lymphoid infiltrate, Lymphoid follicles surrounded by small lymphocytes that infiltrate diffusely in LP and
probably reactive occasionally into epithelium
4 Suspicious lymphoid infiltrate, Lymphoid follicles surrounded by centrocyte-like cells that infiltrate diffusely in LP and
probably lymphoma into epithelium in small groups
5 Low-grade MALT lymphoma Dense diffuse infiltrate of centrocyte-like cells in LP with prominent LEL
GELA grading score (for post-treatment evaluation)
CR Complete histological remission Normal or empty LP and/or fibrosis with absent or scattered plasma cells and small
lymphoid cells in the LP, no LEL
pMRD Probable minimal residual Empty LP and/or fibrosis with aggregates of lymphoid cells or lymphoid nodules in the
disease LP/MM and/or SM, no LEL
rRD Responding residual disease Focal empty LP and/or fibrosis with dense, diffuse, or nodular lymphoid infiltrate,
extending around glands in the LP, focal or absent LEL
NC No change Dense, diffuse, or nodular lymphoid infiltrate, LEL usually present
LEL, Lymphoepithelial lesions; LP, lamina propria; MM, muscularis mucosa; SM, submucosa.

eradication antibiotic therapy, although not necessar-
ily to chemotherapy or immunotherapy. Moreover,
the cases carrying this translocation rarely undergo
high-grade transformation and have been reported to
have a significantly longer relapse-free survival irre-
spective of treatment modality. There are no large-
scale studies to suggest the utility of any specific treat-
ment strategy in this subgroup. H. pylori eradication
may have some benefit on symptoms, but is usually
unable to induce a lymphoma regression. The treat-
ment of antibiotic-resistant gastric MALT lymphoma,
as discussed later, is a controversial issue, with no
evidence to suggest that one therapeutic approach is
better than another.
Clinical and molecular follow-up
A number of molecular follow-up studies have shown
that post-antibiotic histological and endoscopic remis-
sion does not necessarily mean a cure. The long-term
persistence of monoclonal B-cells after histologic
regression of the lymphoma has been reported in
about half of the cases, suggesting that H. pylori erad-
ication suppresses but does not eradicate the lymphoma
clones. Transient histological and molecular relapses
have been reported during long-term follow-up of
antibiotic-treated patients, but without the stimulus
from H. pylori this usually remains a self-limited event
and does not imply a frank clinical progression. The
clinical relevance of the detection of monoclonal B-cells
by molecular methods remains unclear. In the long-
term follow-up of some cases with minimal residual
disease neither lymphoma clinical growth nor histolog-
ical transformation was documented despite a persistent
clonal population, suggesting that a watch-and-wait
policy could be feasible and safe and that these patients
do not necessarily require additional treatment. On the
other hand, cases of lymphoma recurrence following
H. pylori reinfection have been reported, suggesting
that residual dormant tumor cells can be present despite
histological remission.
More recent long-term follow-up studies suggest
that even minimal residual histological disease detected
after macroscopic regression may remain clinically dor-
mant, and that a “wait-and-see” policy seems safe and
could be taken, at least for patients agreeable to frequent
endoscopies. Relapses have also been documented in
the absence of H. pylori reinfection, indicating the pres-
ence of lymphoma clones that may have escaped the
antigenic drive. The frequency of histological transfor-
mation into DLBCL is unclear, but its long-term risk
seems lower than in other indolent lymphomas at least
for the primary gastric presentations. Several cases of
synchronous or metachronous gastric adenocarcino- 113
mas in patients with gastric MALT lymphomas have
 Chapter 7: MALT and other MZLs
been documented. Indeed, a Dutch nationwide tumor
registry study has shown a sixfold increased risk of
gastric adenocarcinoma in patients with gastric MALT
lymphomas in comparison with the general population.
A strict follow-up is therefore strongly advisable, and
histological evaluation of repeated biopsies continues to
be the fundamental follow-up procedure. We perform a
breath test 2 months after treatment to document H.
pylori eradication and repeat post-treatment endoscop-
ies with multiple biopsies every 6 months for 2 years,
then yearly to monitor the histological regression of the
lymphoma.
Management of H. pylori-negative or
antibiotic-resistant cases
No definitive guidelines exist for the management of
the subset of H. pylori-negative cases and for the
patients who fail antibiotic therapy. In two retrospec-
tive series of patients with gastric low-grade MALT
lymphoma, no statistically significant difference was
apparent in survival between patients who received
different initial treatments (including chemotherapy
alone, surgery alone, surgery with additional chemo-
therapy or radiation therapy, or antibiotics against H.
pylori). Surgery has been widely and successfully used
in the past. It leads to excellent long-term local control,
but its role should be redefined in view of the promis-
ing results of the conservative approach. Whether
stage I patients treated with radical gastrectomy need
further treatment is unclear, but a wait-and-see policy
is most often appropriate and for the majority of
patients other treatment approaches are preferred. A
modest dose of involved-field radiotherapy (IF-RT;
30 Gy given in 4 weeks to the stomach and perigastric
nodes) gives excellent disease control and it might be
the treatment of choice for patients with stage I–II
MALT lymphoma of the stomach without evidence
of H. pylori infection, or with persistent lymphoma
after antibiotics.
Patients with disseminated disease should be con-
sidered for systemic treatment (i.e. chemotherapy and/
or immunotherapy with an anti-CD20 monoclonal
antibody). In the presence of advanced-stage disease,
chemotherapy is an obvious choice, but only a few
compounds and regimens have been tested specifically
in MALT lymphomas. Oral alkylating agents (either
cyclophosphamide or chlorambucil, with median
treatment duration of 1 year), can result in a high
rate of disease control. Phase II studies have demon-
114
strated some antitumor activity of the purine analogs
fludarabine and cladribine, which might, however, be
associated with an increased risk of secondary myelo-
dysplastic syndrome (MDS), and of a combination
regimen of chlorambucil/mitoxantrone/prednisone.
Aggressive anthracycline-containing chemotherapy
should be reserved for patients with high tumor bur-
den (bulky masses, high IPI score).
The activity of the anti-CD20 monoclonal anti-
body rituximab has been demonstrated in a phase II
study, with a response rate of about 70%, and this may
represent an additional option for the treatment of
systemic disease. The efficacy of the combination of
rituximab with chlorambucil has been explored in a
randomized study of the International Extranodal
Lymphoma Study Group (IELSG) in gastric (failing
antibiotics) or non-gastric MALT lymphomas
(IELSG19, NCT00210353). In comparison with chlor-
ambucil alone, chlorambucil plus rituximab results in
increased complete remission and event-free survival
(EFS) rates but 5-year OS was identical in both groups.
Similar approaches can be considered for gastric
lymphoma patients experiencing a frank relapse after
an initial lymphoma response to antibiotics. Local
relapses may benefit from IF-RT, while systemic relap-
ses may require systemic treatment.
Management of non-gastric localizations
MALT lymphomas have been described in various non-
GI sites, such as salivary glands, thyroid, skin, conjunc-
tiva, orbit, larynx, lung, breast, kidney, liver, prostate,
and even the intracranial dura. In general, MALT lym-
phoma arises in mucosal sites where lymphocytes are
not normally present and where the MALT is acquired
in response either to autoimmune processes such as
Hashimoto’s thyroiditis and Sjögren’s syndrome or to
chronic infectious conditions. H. pylori gastritis is the
best studied condition, but other infectious agents have
more recently been implicated in the pathogenesis of
MALT lymphomas arising in the skin (Borrelia burg-
dorferi), in the ocular adnexa (Chlamydophila psittaci)
and in the small intestine (Campylobacter jejuni).
Among viruses, the involvement of HCV infection in
the development of some MZLs (especially the splenic
type) has recently been proposed.
Non-gastric MALT lymphomas have been diffi-
cult to characterize because these tumors are distrib-
uted so widely throughout the body that it is difficult
to assemble adequate series for any given site. One-
quarter of non-GI MALT lymphomas have been
reported to present with involvement of multiple

mucosal sites or non-mucosal sites such as bone
marrow. Non-GI MALT lymphomas, despite pre-
senting with stage IV disease more often than the
gastric variant, frequently have a quite indolent
course regardless of treatment type, but they are
significantly more prone to relapse than the primary
gastric cases (most often at other mucosal sites). A
multicentric retrospective survey of 180 non-gastric
cases showed no evidence of a clear advantage for any
type of therapy and, despite the high proportion of
cases with disseminated disease, which should
require a systemic approach, no clear advantage was
associated with any chemotherapy program.
In general, the treatment of non-gastric MALT
lymphoma can be approached in the same way as
that of the H. pylori-negative cases described above.
The sole use of moderate-dose radiotherapy for
patients with localized disease results in a high rate of
local control and often cure. Effective radiotherapy is
frequently possible for both common and rare presen-
tations of the disease. Side effects of RT are mild and
reversible. For patients with localized non-gastric
MALT lymphoma, given the unique biologic behavior
with a tendency to relapse in MALT tissues and an
indolent course, RT is often the first line treatment of
choice. The emerging literature on localized MALT
lymphomas confirms a high rate of local control in
MALT lymphoma, with a high proportion of patients
likely to be cured of the disease. The moderate doses of
radiation required for cure (25–35 Gy) are generally
associated with a minimal risk of long-term toxicity,
although special considerations are needed for partic-
ular localizations such as the eye or the lung.
For some patients with stage III or IV disease,
radiotherapy can also be an effective therapy in pro-
viding local disease control, but the optimal manage-
ment of disseminated MALT lymphomas is less clearly
defined. Because no curative treatment exists, expect-
ant observation can be an adequate initial policy in
most patients. In general, the treatment should be
“patient-tailored,” taking into account the site, the
stage, and the clinical characteristics of the individual
patient. When systemic treatment is needed, enroll-
ment in controlled clinical trials is advisable. As men-
tioned above, only one randomized study has thus far
been conducted (IELSG19) that showed a clinical ben-
efit from combining chlorambucil and rituximab, but
with no impact on OS.
Systemic antibiotic treatment should be tried first
in patients with B. burgdorferi-associated cutaneous
Chapter 7: MALT and other MZLs
lymphomas. Similarly, the finding that C. psittaci is
associated with MALT lymphoma of the ocular adnexa
may provide the rationale for the antibiotic treatment
with doxycycline of localized lesions, and preliminary
encouraging results have been reported, but this
approach remains investigational and will need to be
confirmed by larger clinical studies.
Immunoproliferative small
intestinal disease (IPSID)
This condition, known in the past as alpha heavy chain
disease or Mediterranean lymphoma, is nowadays
considered a special subtype of MALT lymphoma.
The distinguishing feature of IPSID is the synthesis
of alpha heavy chain that is secreted and detectable in
the serum, urine, saliva, and duodenal fluid in approx-
imately two-thirds of cases; in the remainder, the
heavy chain protein is demonstrable by tumor cell
immunohistochemistry but not secreted. Most of the
cases have been described in the Middle East, espe-
cially in the Mediterranean area, where the disease is
endemic and affects young adults of both sexes, but
predominantly males. A few cases have been reported
from industrialized western countries, usually among
immigrants from the endemic area.
The pathologic features of IPSID recognize a specific
type of MALT lymphoma that is characterized by thick-
ening of the wall of the proximal small bowel. The
morphological appearances are similar to typical
MALT lymphoma but there is profound plasma cell
differentiation. There are scant lymphoid cells that
may only become obvious following staining for
CD20. These are predominantly around the crypts, and
lymphoepithelial lesions are seen. Immunophenotypical
characteristics are the same as for other types of MALT
lymphoma but these cells show synthesis of IgA without
either light chain.
The natural course of IPSID is usually prolonged,
often over many years, including a potentially rever-
sible early phase, with the disease usually confined to
the abdomen. If untreated it degenerates, with high-
grade transformation, into large B-cell lymphoma.
Since the histology of mucosal lesions and mesen-
teric lymph nodes can be discordant, with a higher
grade in the latter, staging laparotomy can be useful
in the evaluation of the mesenteric nodal involve-
ment, but surgery has no therapeutic role because
of a generally diffuse intestinal involvement. It
has been known since the 1970s that early IPSID 115
 Chapter 7: MALT and other MZLs

phases could achieve durable remissions with sus- Pathology of splenic MZL
tained antibiotic treatment (such as tetracycline or
The splenic pathology shows that the disease is pre-
metronidazole and ampicillin for at least 6 months).
dominantly within the white pulp, with a less pro-
Recently, the presence of C. jejuni has been demon-
nounced nodular and interstitial infiltration of
strated in IPSID tumor sections. At an early stage,
the red pulp. The white pulp is partially or com-
antibiotic treatment directed against C. jejuni may
pletely effaced by the neoplastic cells. These areas
lead to lymphoma regression, but there is no clear
show an inner zone of small lymphocytes that
evidence that antibiotics alone are of benefit in
resemble mantle cells that either surround residual
the advanced phases. Although the early studies
germinal centers or, more commonly, replace them.
reported that aggressive chemotherapy is not well
Around this small cell layer there is a further zone
tolerated by patients with advanced disease and
composed of slightly larger cells that have more
severe malabsorption, anthracycline-containing reg-
abundant cytoplasm. Large transformed cells with
imens, combined with nutritional support plus anti-
more abundant cytoplasm and vesicular nuclei that
biotics to control diarrhea and malabsorption, may
contain nucleoli are scattered within the outer layer.
offer the best chance of cure, with 5-year survival
Plasma cell differentiation may occur. The red pulp
rates up to 70%.
nodules are frequently composed of cells resem-
bling those of the inner zone of the white pulp
(Figure 7.5).
Primary splenic marginal zone Spread to the lymph nodes of the spleen hilum is
lymphoma frequent. These nodes may show a nodular pattern
based on pre-existing follicles or may show a zoned
Clinical features appearance similar to that seen in the splenic white
Splenic MZL is a very rare disorder, comprising less pulp. The bone marrow shows interstitial infiltration
than 1% of all lymphomas. Up to two-thirds of by small lymphocytes usually with a nodular compo-
patients with splenic MZL have circulating villous nent. In the interstitium cells can frequently be seen
lymphocytes with characteristic fine short cytoplasm within sinusoids, although this pattern is not specific
polar projections. When these are more than 20% of to splenic MZL.
the lymphocyte count, the term “splenic lymphoma Immunophenotypically the cells show surface
with villous lymphocytes” is commonly used. Despite immunoglobulin, usually IgM and IgD. They express
relevant geographical variations, HCV seems to be
involved in lymphomagenesis of a portion of cases.
An association with malaria and with Epstein–Barr
virus (EBV) infection has been shown in tropical
Africa, and the tropical cases are characterized by a
high percentage of circulating villous lymphocytes.
Most patients with primary splenic MZL are over 50,
with a similar incidence in males and females. The
disease usually presents with massive splenomegaly,
which produces abdominal discomfort and pain.
Lymphadenopathy is usually minimal or absent. The
diagnosis may be made at splenectomy, performed to
establish the cause of unexplained spleen enlargement.
B-symptoms are not uncommon: anemia, thrombocy-
topenia, or leukocytosis are reported in one-quarter of
cases. Autoimmune hemolytic anemia can be found in
Figure 7.5 Splenic marginal zone B-cell lymphoma (spleen). The
up to 15% of patients. Involvement of the splenic hilar white pulp shows infiltration by a population of lymphoid cells that
lymph nodes and/or the liver is reported in 25–30% of surround residual reactive germinal centers with a biphasic pattern.
cases. Frequently, a small clonal B-cell population may There is an inner rim of cells with scanty cytoplasm and an outer rim
of cells with larger nuclei and more abundant cytoplasm. Scattered
116 be detected by peripheral blood flow cytometry even in larger cells are seen in the outer layer. The red pulp contains small
the absence of overt lymphocytosis. nodular aggregates of lymphoma cells.

CD20 and CD79a and are usually negative for CD5,
CD23, CD10, and cyclin D1. There is expression of
BCL-2 protein. Proliferation shows a characteristic
pattern with infrequent cells staining for KI67/MiB1
in the inner, small cell zone and scattered cells positive
in the outer zone.
Splenic MZLs have to be distinguished from other
small B-cell lymphomas, almost all of which can adopt
an organizational pattern that mimics that of splenic
MZL. This can usually be accomplished by the immu-
nocytochemical studies noted here.
Molecular pathogenesis of splenic MZL
As this is a relatively newly recognized lymphoma
entity, its precise molecular pathogenesis is largely
unknown. Recent studies indicate that these tumors
may harbor unmutated or mutated IgVH genes, thus
being derived from naïve or antigen-experienced
B-cells. The pattern of IgVH gene usage (biased
usage of VH1–2 gene family) and the finding of
somatic hypermutations in some cases suggest an
antigen-driven lymphomagenesis. In line with these
findings, a possible relationship between HCV infec-
tion and splenic MZL has recently been established.
Cytogenetic and molecular genetic studies have
demonstrated that approximately 45% of splenic
MZL harbor allelic losses in the 7q32–q33 chromo-
some region and these cases are possibly more
aggressive clinically. Other recurrent cytogenetic
alterations include trisomies 3 and 18, +12q and
translocations t(6;14)(p12;q32), and t(2;7)(p12;q22)
with deregulation of Cyclin D3 and CDK6.
Gene expression profiling studies confirmed the
homogeneity of this disease entity on the molecular
level and point to different pathways that may be
involved in lymphomagenesis. The molecular signa-
ture of splenic MZL includes upregulation of several
genes involved in NFκB activation, B-cell receptor
and tumor necrosis factor (TNF) signaling, and of
genes associated with the splenic microenvironment
(SELL, LPXN). Genes located in the 7q31–7q32 dele-
tion region have been reported to be downregulated.
Two large studies – one cytogenetic study and a
comprehensive genome-wide analysis of DNA copy
number alterations – confirmed that DNA copy
number imbalances in splenic MZLs are indeed spe-
cific for the entity and differ from other lymphomas,
including MALT-type MZL. More recent studies
have identified recurrent somatic mutations affect-
ing the NFκB and NOTCH pathways.
Chapter 7: MALT and other MZLs
Clinical course and management
The clinical diagnosis is based on a combination of
features, including lymphocyte morphology, immu-
nophenotype, cytogenetic abnormalities, bone mar-
row histology, and, when available, spleen histology.
A set of minimum diagnostic criteria for splenic MZL
have been proposed and should be followed. Diagnosis
can be based either on spleen histology with immuno-
phenotypic features not diagnostic for chronic lym-
phocytic leukemia or on the presence of typical blood
and bone marrow morphology, immunophenotype
not consistent with the diagnosis of chronic lympho-
cytic leukemia and intrasinusoidal infiltration by
CD20-positive cells (if spleen histology is unavailable).
Nowadays, an increasing number of cases is diagnosed
based on the examination of bone marrow and periph-
eral blood specimens only.
According to the WHO classification, peripheral
lymph node involvement is typically absent. However,
patients with disseminated MZLs can be observed and a
precise diagnosis can be very difficult in cases presenting
with concomitant splenic, extranodal, and nodal
involvement. In a retrospective French series of 124
patients with non-MALT-type MZL, four clinical sub-
types were observed: splenic (48% of cases), nodal (30%),
disseminated (splenic and nodal, 16%), and leukemic
(neither splenic nor nodal, 6%). Even when the disease
is restricted to the cases presenting with splenomegaly,
nearly all patients have bone marrow involvement, often
accompanied by involvement of peripheral blood
(defined as the presence of absolute lymphocytosis of
more than 5%). Serum paraproteinemia is observed in
about 10–25% of cases and is most frequently of IgM
type, posing the problem of differential diagnosis with
lymphoplasmacytic lymphoma/Waldenström’s macro-
globulinemia. The two diseases often present with sim-
ilar clinical features (splenomegaly, bone marrow
lymphoplasmacytic infiltration, anemia), but marked
hyperviscosity and hypergammaglobulinemia are
uncommon in splenic MZL.
The clinical course is most usually indolent, with
5-year OS ranging from 65% to 80%. Histological trans-
formation is rare, often associated with B-symptoms,
disease dissemination, and poorer outcome. A prognos-
tic model for clinical use was recently proposed, based
on the results of a large cooperative retrospective study
of 309 patients from Italy. Using three readily available
parameters, hemoglobin less than 120 g/L, LDH higher
than normal, and serum albumin less than 35 g/L, three 117
prognostic groups can be identified: low-risk patients
 Chapter 7: MALT and other MZLs
(5-year cause-specific survival 88%) with no adverse
factors, intermediate-risk patients (5-year cause-specific
survival 73%) with one adverse factor, and high-risk
patients (5-year cause-specific survival 50%) with two
or three adverse factors.
The reported largest series shows that most patients
can initially be managed with a wait-and-see policy, and
they do not seem to have a worse outcome. When
treatment is needed this is usually because of large
symptomatic splenomegaly or cytopenia. Splenectomy
has been considered the treatment of choice: it allows a
reduction/disappearance of circulating tumor lympho-
cytes and recovery of the lymphoma-associated cytope-
nia. Responses to splenectomy occur in approximately
90% of patients and the benefit often persists for several
years; time to next treatment can be longer than 5 years.
Adjuvant chemotherapy after splenectomy may result
in a higher rate of complete response, but there is no
evidence of a survival benefit.
Chemotherapy alone may be considered for
patients who require treatment but have contraindica-
tions to splenectomy, and for the patient with clinical
progression after spleen removal. Alkylating agents
(chlorambucil or cyclophosphamide) have been
reported to be active and can be used as single agents
or in combination (as in the CVP and CHOP regi-
mens). Among the purine analogs, fludarabine has
been shown to be effective but, curiously, cladribine
seems not to be active. Rituximab, alone or in combi-
nation with chemotherapy, induces good responses in
cases refractory to standard chemotherapy. In retro-
spective series of rituximab monotherapy, sustained
responses are reported in up to 90% of patients, with
rapid regression of splenomegaly and improvement of
cytopenias. OS seems comparable to that reported
following splenectomy. Since splenic MZL patients
are often elderly and at high risk surgically, frontline
treatment with rituximab alone or in combination
with chemotherapy has become more widely used.
Very interesting is the observation from one study
that some patients with splenic lymphoma with villous
lymphocytes and HCV infection obtained a complete
response after treatment with interferon alpha, alone
or in combination with ribavirine. Interferon treat-
ment had no antitumor effect on HCV-negative
splenic lymphomas. Analogous to H. pylori infection
in gastric MZL, it appears that HCV may be respon-
sible for an antigen-driven stimulation of the lym-
phoma clone. This report suggests that all patients
118 should be tested for HCV infection, and antiviral
therapy should be considered in the positive cases
before any decision is reached about more aggressive
therapeutic approaches.
Nodal marginal zone lymphoma
Clinical features
In contrast with mucosa-based extranodal MALT lym-
phoma, nodal MZL (previously known as monocytoid
B-cell lymphoma) is typically lymph node-based. This
type of lymphoma is exceedingly rare, accounting for
less than 10% of all MZLs and less than 1% of all NHLs.
It has also been associated with HCV infection in some
epidemiologic studies. The clinical data are sparse and
have been largely drawn from pathologic series rather
than clinical ones. Nodal MZL is a disease of older
people, with the median age at presentation in the
sixth decade, and affects both sexes, with a slight
female predominance. Whether this lymphoma car-
ries a worse prognosis in comparison with other MZLs
remains controversial.
The commonest presenting feature is localized –
most often in the neck – or sometimes a generalized
adenopathy. Bone marrow is involved at presentation
in less than half of the cases. Transformation to high-
grade lymphoma has been described in some cases.
Pathology and genetics
The histologic pattern of lymph node involvement by
primary nodal MZL is often indistinguishable from
that of extranodal MZL of MALT type, and this lym-
phoma has previously been thought to be a nodal
variety of MALT lymphoma. The tumor cell morphol-
ogy is heterogeneous and is similar to the lymph node
involvement of extranodal MZL and splenic MZL.
Sometimes the tumor cells resemble monocytes;
indeed, the term monocytoid lymphoma was used in
the past to indicate this type of lymphoma.
Analysis of the IgH genes suggests a prevalence of
cases with mutated IgH genes, but, similarly to splenic
MZL, unmutated cases do exist. No specific genomic
aberration is known to occur in nodal MZL. The most
common alterations, such as gain of 3q, are also
present in extranodal MZL and splenic MZL.
Management
Because no curative treatment exists, most patients
can initially be managed with expectant observation.

There is at present no consensus about the best treat-
ment, individual cases being managed differently
according to site and stage. Treatment options may
include single-agent chlorambucil or fludarabine or
combination chemotherapy regimens (such as CVP
or CHOP). Rituximab has single-agent efficacy and
can be combined with chemotherapy. Anti-HCV
treatment may induce lymphoma regression in
some HCV-infected patients. Autologous stem cell
transplantation has been used in younger patients
with adverse prognostic factors or with large-cell
transformation.
Further reading
Arcaini L, Lazzarino M, Colombo N, et al. Splenic marginal
zone lymphoma: a prognostic model for clinical use.
Blood, 2006;107:4643–4649.
Berger F, Felman P, Thieblemont C, et al. Non-MALT
marginal zone B-cell lymphomas: a description of clinical
presentation and outcome in 124 patients. Blood,
2000;95:1950–1956.
Bennett M, Schechter GP. Treatment of splenic marginal
zone lymphoma: splenectomy versus rituximab. Semin
Hematol, 2010;47:143–147.
Bertoni F, Zucca E. State-of-the-art therapeutics: marginal
zone lymphoma. J Clin Oncol, 2005;23:6415–6420.
Bertoni F, Conconi A, Capella C, et al. Molecular follow-up
in gastric mucosa-associated lymphoid tissue
lymphomas: early analysis of the LY03 cooperative trial.
Blood, 2002;99:2541–2544.
Chen LT, Lin JT, Tai JJ, et al. Long-term results of anti-
Helicobacter pylori therapy in early-stage gastric high-
grade transformed MALT lymphoma. J Natl Cancer Inst,
2005;97:1345–1353.
Copie-Bergman C, Gaulard P, Lavergne-Slove A, et al.
Proposal for a new histological grading system for post-
treatment evaluation of gastric MALT lymphoma. Gut,
2003;52:1656.
Copie-Bergman C, Wotherspoon AC, Capella C, et al. Gela
histological scoring system for post-treatment biopsies of
patients with gastric MALT lymphoma is feasible and
reliable in rountine practice. Br J Haematol,
2013;160:47–52.
de Boer JP, Hiddink RF, Raderer M, et al. Dissemination
patterns in non-gastric MALT lymphoma.
Haematologica, 2008;93:201–206.
Dreyling M, Thieblemont C, Gallamini A, et al. ESMO
Consensus conferences: guideliness on malignant
lymphoma. part 2: marginal zone lymphoma, mantle
cell lymphoma, T-cell lymphoma. Ann Oncol,
2013;24:857–877.
Chapter 7: MALT and other MZLs
Farinha P, Gascoyne RD. Molecular pathogenesis of
mucosa-associated lymphoid tissue lymphoma. J Clin
Oncol, 2005;23:6370–6378
Ferreri AJ, Dolcetti R, Du MQ, et al. Ocular adnexal MALT
lymphoma: an intriguing model for antigen-driven
lymphomagenesis and microbial-targeted therapy. Ann
Oncol, 2008;19:835–846.
Hancock BW, Qian W, Linch D, et al. Chlorambucil versus
observation after anti-Helicobacter therapy in gastric
MALT lymphomas: results of the international
randomised LY03 trial. Br J Haematol, 2009;144:367–375.
Iannitto E, Tripodo C. How I diagnose and treat splenic
lymphomas. Blood, 2011;117:2585–2595.
Kiel MJ, Velusamy T, Betz BL, et al. Whole-genome
sequencing identifies recurrent somatic NOTCH2
mutations in splenic marginal zone lymphoma. J Exp
Med 2012;209:1553–65.
Lecuit M, Abachin E, Martin A. et al. Immunoproliferative
small intestinal disease associated with Campylobacter
jejuni. N Engl J Med 2004;350:239–248.
Martinelli G, Laszlo D, Ferreri AJ, et al. Clinical activity of
rituximab in gastric marginal zone non-Hodgkin’s
lymphoma resistant to or not eligible for anti-
Helicobacter pylori therapy. J Clin Oncol
2005;23:1979–1983.
Matutes E, Oscier D, Montalban C, et al. Splenic marginal
zone lymphoma proposals for a revision of diagnostic,
staging and therapeutic criteria. Leukemia,
2008;22:487–495.
Mazloom A, Medeiros LJ, Mclaughlin PW, et al. Marginal
zone lymphomas: factors that affect the final outcome.
Cancer, 2010;116:4291–4298.
Mollejo M, Camacho FI, Algara P, et al. Nodal and splenic
marginal zone B cell lymphomas. Hematol Oncol,
2005;23:108–118.
Montalban C, Abraira V, Arcaini L, et al. Simplification of risk
stratification for splenic marginal zone lymphoma: a point-
based score for practical use. Leuk Lymphoma, 2013.
Rinaldi A, Mian M, Chigrinova E, et al. Genome wide DNA-
profiling of marginal zone lymphomas identifies subtype-
specific lesions with an impact on the clinical outcome.
Blood, 2011;117:1595–1604.
Rossi D, Trifonov V, Fangazio M, et al. The coding genome
of splenic marginal zone lymphoma: activation of
NOTCH2 and other pathways regulating marginal zone
development. J Exp Med 2012;209:1537–1551.
Ruskone-Fourmestraux A, Dragosics B, Morgner A,
Wotherspoon A, De Jong D. Paris staging system for
primary gastrointestinal lymphomas. Gut,
2003;52:912–913.
Salido M, Baro C, Oscier D, et al. Cytogenetic aberrations
and their prognostic value in a series of 330 splenic 119
 Chapter 7: MALT and other MZLs
marginal zone B-cell lymphomas: a multicenter study of
the Splenic B-Cell Lymphoma Group. Blood,
2010;116:1479–1488.
Stathis A, Chini C, Bertoni F, et al. Long-term outcome
following Helicobacter pylori eradication in a
retrospective study of 105 patients with localized gastric
marginal zone B-cell lymphoma of MALT type. Ann
Oncol 2009;20:1086–1093.
Stathis A, Bertoni F, Zucca E. Treatment of gastric marginal
zone lymphoma of MALT type. Expert Opin
Pharmacother, 2010;11:2141–2152.
Thieblemont C, Berger F, Dumontet C, et al. Mucosa-
associated lymphoid tissue lymphoma is a disseminated
disease in one third of 158 patients analyzed. Blood,
2000;95:802–806.
Thieblemont C, Felman P, Callet-Bauchu E, et al. Splenic
marginal-zone lymphoma: a distinct clinical and
pathological entity. Lancet Oncol, 2003;4:95–103.
Tsang RW, Gospodarowicz MK. Radiation therapy for
localized low-grade non-Hodgkin’s lymphomas. Hematol
Oncol, 2005;23:10–17.
Wenzel C, Dieckmann K, Fiebiger W, et al. CD5 expression
in a lymphoma of the mucosa-associated lymphoid tissue
(MALT)-type as a marker for early dissemination and
120
aggressive clinical behaviour. Leuk Lymphoma,
2001;42:823–829.
Williams ME, Hong F, Kahl B, et al. Results of E4402
(RESORT): a randomized phase III study comparing two
different rituximab dosing strategies for low tumor
burden indolent B-cell lymphoma. Proc ASCO, 2012;
abstract 8007.
Zucca E, Dreyling M. Gastric marginal zone lymphoma of
MALT type: ESMO Clinical Practice Guidelines for
diagnosis, treatment and follow-up. Ann Oncol, 2010;21
Suppl 5:v175–176.
Zucca E, Bertoni F, Roggero E, Cavalli F. The gastric
marginal zone B-cell lymphoma of MALT type. Blood,
2000;96:410–419.
Zucca E, Conconi A, Pedrinis E, et al.
Nongastric marginal zone B-cell lymphoma of
mucosa-associated lymphoid tissue. Blood,
2003;101:2489–2495.
Zucca E, Conconi A, Laszlo D, et al. Addition of
rituximab to chlorambucil produces superior
event-free survival in the treatment of patients with
extranodal marginal-zone B-cell lymphoma: 5-year
analysis of the IELSG-19 Randomized Study. J Clin
Oncol 2013;31:565–572.
 Chapter
8
Small lymphocytic lymphoma/chronic
lymphocytic leukemia
Peter Hillmen
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Introduction
Small lymphocytic lymphoma (SLL) and chronic lym-
phocytic leukemia (CLL) are in the midst of a period of
huge change resulting from advances on several fronts.
One important development is the appreciation that
SLL and CLL are two manifestations of the same dis-
order. Throughout this review, therefore, it should be
assumed that SLL is managed in a similar manner to
CLL, although the studies drawn upon and recom-
mendations made will be based mainly on publications
on the diagnosis and therapy for CLL.
There have been major advances in our understand-
ing both of the pathophysiology of CLL/SLL and of the
mechanism by which the disease becomes resistant to
conventional therapies. This has coincided with the
application of novel approaches to define remissions,
including the use of modern imaging techniques, which
have never previously been applied in CLL, and the
development of techniques to detect minimal residual
disease, particularly by multiparameter flow cytometry.
To some extent a major driver of these changes has been
the development of novel therapeutic approaches which
yield higher proportions of complete remissions. We
have now moved from standard therapies that achieve
complete responses in less than 10% of patients to novel
approaches that result in greater than 70% complete
responses. There is now the prospect of achieving
response rates which for other hematopoietic malignan-
cies are associated with a prolongation in survival – and
even cures! Treatment paradigms are evolving rapidly
towards risk stratification by molecular prognostic fac-
tors and tailoring therapy to an individual patient’s
disease. These changes have largely happened over the
past 5 years, and this rapidity of movement results in
two principal problems: firstly, randomized clinical tri-
als cannot be performed rapidly enough to address the
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
wide variety of relevant questions, and, secondly, the
application of these modern tests is not straightforward,
neither in their availability in most health economies
nor in their application to clinical practice. It is not
unreasonable to believe, however, that considering a
curative intent for more than just a small minority of
patients is now a realistic prospect. The current and
future management of patients with CLL will require
treatment stratification by individual patient risk, the
application of techniques to detect minimal residual
disease, and the intelligent use of the variety of novel
therapies that are currently being developed. We now
stand on the threshold of a paradigm shift in our treat-
ment of CLL. This results in part from a huge amount of
work reported over the past decade regarding the patho-
physiology of CLL. This is revolutionizing the treatment
of all aspects of CLL. There are now a number of novel
therapeutic targets that are at last yielding real progress
in the treatment of CLL. These include therapies that
target the downstream signaling from the B-cell recep-
tor, a key proliferative signal in CLL, and those that
induce apoptosis, again a process that is dysregulated
in CLL. The challenge now is how best to utilize these
new agents alone and in logical combinations to induce
prolonged remissions with minimal toxicity. This chap-
ter will summarize these changes in our approach to
CLL.
B-cell small lymphocytic lymphoma/
chronic lymphocytic leukemia
The lymph node architecture is effaced by a diffuse
proliferation of small lymphocytes that have scanty
cytoplasm and round nuclei with minimal nuclear irreg-
ularity (Figures 8.1 and 8.2). The chromatin is clumped
and nucleoli are not seen. Within the small-cell
121
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
Figure 8.1 B-cell small lymphocytic lymphoma/chronic
lymphocytic leukemia (lymph node). The nodal architecture is
effaced by a diffuse proliferation of small cells with scanty cytoplasm
and round nuclei. Occasional larger cells are present singly and in
small groups (proliferation centers/pseudofollicles).
population are scattered larger cells, which are fre-
quently clustered into discrete areas that appear pale
when the node is examined at low power. These are
termed proliferation centers or pseudofollicles and con-
tain a mixture of intermediate-sized cells with round
nuclei and central eosinophilic nucleoli (prolympho-
cytes) and large cells with a moderate amount of pale
cytoplasm, oval nuclei, and large central eosinophilic
nucleoli (paraimmunoblasts). In some cases residual
normal germinal centers may be seen entrapped within
the diffuse lymphomatous proliferation. A proportion
of cases may show plasmacytic differentiation.
Immunophenotypically the cells show surface
immunoglobulin (sIg) of IgM type, usually with coex-
pression of IgD. The cells are positive for CD20
(although this can be weak) and CD79a. The majority
of cases show expression of CD5, CD43, and CD23 but
without staining for cyclin D1 or CD10. A proportion
of cases may lack CD23 or, rarely, CD5. Staining for
BCL-2 protein is invariably positive. Proliferation is
low and, where present, is usually most prominent in
the proliferation centers.
B-cell prolymphocytic leukemia (B-PLL)
Lymph nodes are not usually enlarged in this lympho-
proliferation, and the histopathology usually relates to
the appearances within the spleen and bone marrow. In
the bone marrow there is infiltration of the interstitium
122 with a non-paratrabecular distribution. In the spleen
the red and white pulp is effaced by a diffuse population
Figure 8.2 B-cell small lymphocytic lymphoma/chronic
lymphocytic leukemia (bone marrow). There is an infiltrate of small
lymphocytes in the bone marrow with a nodular and interstitial
growth pattern.
of cells that are larger than normal lymphocytes. There
is moderately abundant cytoplasm, and the nuclei con-
tain single central eosinophilic nucleoli.
Immunophenotypically the cells are positive for
CD20 and CD79a. There is variable expression of
CD5, and CD23 is usually negative. The cells do not
express CD10. Distinction between true B-PLL and a
nucleolated version of mantle cell lymphoma that
presents in leukemic phase is important. True B-PLL
is negative for cyclin D1, while this protein is expressed
in the nuclei of nucleolated mantle cell lymphoma.
Molecular pathology and cytogenetics
Chronic lymphocytic leukemia of B-cell lineage
(B-CLL) does not seem to constitute a homogeneous
genetic entity. In particular, recent molecular studies
investigating the occurrence of somatic hypermuta-
tions in the variable regions of the immunoglobulin
heavy chain (IgVH) genes provide evidence that B-CLL
may consist of at least two important subgroups. In
roughly 50% of cases, the tumor cells do not display
somatic mutations and, hence, are derived from
B-cells that have not undergone a germinal center
reaction. In the remaining cases somatic mutations
are present, indicating a germinal center passage of
the tumor cells. The IgVH-mutated form of B-CLL is
clearly associated with a better prognosis. In contrast
to many other B-cell non-Hodgkin’s lymphomas
(NHLs), reciprocal chromosome translocations are
not a frequent feature in B-CLL, although the t(1;6)
(p35;p25) has recently been identified as a recurrent
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia

translocation in rare cases. While translocations stratifying patients to specific therapies. In contrast,
appear to be uncommon in B-CLL, chromosomal the more recently described prognostic factors can
imbalances at several loci characterize this leukemia more accurately predict the outcome for individual
on the genetic level. In particular, frequent aberrations patients, and these have now been validated by large
include deletions in chromosomal band 13q14 that prospective studies. These newer biological variables,
lead to the loss of microRNA miR-15a/miR-16–1 clus- which are now being used to stratify treatment in
ter, detected in up to 40% of cases, trisomy 12, and current clinical studies, fall into four principal types.
deletions in 17p13 and 11q22–23. Deletions in 17p
1. Those that are inherent to the individual patient’s
frequently target the TP53 gene, and 11q deletions
disease and which will not alter during the course
commonly result in the loss of one copy of the ATM
of their illness – such as somatic mutation of the
(ataxia telangiectasia-mutated) gene as well as the miR-
immunoglobulin gene.
34b/miR-34c cluster. The del(11q) has also been asso-
2. Genetic abnormalities that develop during the
ciated with mutation of the spliceosome gene SF3B1.
disease course and which are indicative of genetic
Both aberrations have been shown to be of prognostic
evolution, and in some cases therapeutic
significance in defining patient subgroups with infe-
resistance, of disease – such as deletion of the short
rior prognosis and are commonly found in the IgVH-
arm of chromosome 17, studied by fluorescent in
unmutated B-CLL variant. Interaction of the miR and
situ hybridization (FISH).
TP53 aberrations has been associated both with CLL
pathogenesis and patient prognosis.
Despite the presence of IgVH-mutated and IgVH-
unmutated B-CLL subsets, there is evidence from Table 8.1 Rai staging system for CLL.
DNA microarray studies that the two B-CLL sub- Stage Risk Clinical Median
groups share a homogeneous gene expression profile. features survival
B-CLL, therefore, can be viewed as a single entity with
0 Low Lymphocytes >5 >10 years
a distinct transcriptional profile, distinguishing it from × 109/L and
other low-grade lymphomas. >40% lympho-
In contrast to both B-CLL and mantle cell lym- cytes in the
marrow
phoma, IgVH genes in B-PLL are generally mutated.
According to older reports, the most prevalent cyto- I Intermediate Stage 0 plus 7 years
enlarged lymph
genetic marker is a rearranged chromosome 14 with nodes
14q32 breaks. Other genetic aberrations in B-PLL II Stages 0 or I with
include trisomy 12 and structural rearrangements of an enlarged liver
1p32 and 6q21. Previously, a t(11;14)(q13;q23) trans- or spleen
location was considered to be a recurring aberration in III High Stages 0, I, or II 1.5–4
this tumor, but a critical review of these cases points to with hemoglo- years
bin <100 g/L
a significant overlap with a more refined classification
of these tumors as mantle cell lymphoma. IV Stages 0, I, II, or III
with platelets
<100 × 109/L

Therapy
Table 8.2 Binet staging system for CLL.
Treatment of CLL by risk stratification
Conventional prognostic variables, such as clinical Stage Clinical features Median
stage (either Rai or Binet staging; Tables 8.1 and 8.2) survival
or lymphocyte doubling time, have been used to pre- A <3 areas of lymphadenopathy and 12 years
dict the proportion of patients who will progress to no anemia or thrombocytopenia
treatment or who will respond well to therapy. These B 3 or more areas of lymphadeno- 7 years
variables are useful for predicting the outcomes of pathy and no anemia or
thrombocytopenia
groups of patients (for example, when assessing the
merits of different trials), but are not useful for pre- C Hemoglobin <100 g/L and/or 2–4 years
platelets <100 × 109/L 123
dicting the outcome of individual patients or for
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
3. Prognostic factors that may vary during a patient’s
disease or even in response to infections or similar
acute events – such as CD38.
4. Those that are related to tumor bulk – such as
LDH, β2-microglobulin or thymidine kinase.
The factors that are most appropriately used for strat-
ification into different treatment intentions are those
that are inherent to a patient’s CLL (group 1), as they
predict whether or not a patient will eventually
develop treatment resistance. In contrast, genetic
abnormalities (group 2) that explain chemotherapy
resistance, for example disruption of the p53 pathway,
are probably most appropriate to tailor an individual
patient’s treatment, for example by using therapies
that are independent of the p53 pathway.
Therefore the two most extensively studied and
robust prognostic markers are the presence or absence
of somatic mutations in the immunoglobulin gene of
the CLL cell and chromosomal abnormalities detected
by FISH. Mutational status and FISH can be used to
predict the course of a patient’s disease, and are now
being evaluated to determine whether patients with
poor-risk disease should have therapy initiated before
disease progression, as is currently recommended, or
whether different therapeutic approaches should be
applied to different risk groups. Patients with CLL
cells that have no or few (>98% homology with germ-
line sequence) somatic mutations in the immunoglo-
bulin disease (VH-unmutated) have similar response
rates to therapy compared with patients with mutated
CLL but remain in remission for a shorter period, with
a poorer overall survival. In addition, it appears that
when patients with unmutated CLL relapse after initial
therapy there are frequently abnormalities of the p53
pathway (deletion of either 17p or 11q), leading to a
poorer response to conventional therapy and subse-
quently a poorer survival (Figure 8.3). It is not clear
whether the treatment itself creates abnormalities in the
p53 pathway or simply selects for a pre-existing sub-
clone that was present at a low level prior to starting the
initial therapy. In contrast, when mutated CLL pro-
gresses after therapy, these poor-risk chromosomal
abnormalities appear to occur only rarely. When
patients either present with or develop a dysfunctional
p53 pathway (either 17p or 11q deleted), they are inher-
ently resistant to therapies that damage DNA or inter-
fere with its repair, such as alkylating agents or purine
analogs. These patients have a very low response rate to
conventional therapy, and this may explain the mecha-
124 nism of resistance in most, if not all, patients who fail to
100
Patients surviving (%)
75
VH homology
< 98% (n = 123)
50
VH homology
11p-
≥ 98% (n = 109)
25 (n = 48)
17p-
(n = 20)
0
0 24 48 72 96 120 144 168 192
Months
Figure 8.3 Survival of patients with CLL according to cytogenetics
(from Doehner et al. 2000).
respond to fludarabine-based combination therapies. In
these cases it is logical to use treatments that do not
depend upon an intact p53 pathway for their activity,
such as high-dose steroids or monoclonal antibodies.
It is logical to treat unmutated CLL intensively,
with the aim of achieving more profound remissions,
and then to consolidate remissions with therapies that
are not dependent upon an intact p53 pathway for
their function. In contrast, for good-risk mutated
CLL it is logical to de-escalate therapy to reduce tox-
icity, because when patients relapse they are more
likely to respond to second line therapy. For example,
the LRF CLL4 trial in the UK demonstrated that CLL
can be divided into three different risk groups by the
molecular characteristics of the leukemic cells.
1. “Good risk” CLL can be defined by the presence of
somatic mutations in the immunoglobulin gene of
the CLL cell, excluding those cases utilizing the VH
segment VH3–21, and comprises approximately
30% of patients requiring therapy.
2. Standard-risk patients can be defined as those cases
with unmutated immunoglobulin genes or
utilizing VH3–21, or 11q deletion, comprising
approximately 65% of patients.
3. Poor-risk patients are those with greater than 20%
CLL cells which have loss of the short arm of
chromosome 17 (17p) on FISH, comprising about
5% of patients.
Minimal residual disease
The NCI Working Group (NCI-WG) 2008 response
criteria in CLL defined a complete response on
purely clinical examination, blood counts, and
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia

Table 8.3 Criteria for NCI Working Group complete response. Table 8.4 Comparison of methods of residual disease
monitoring in CLL.
B-symptoms Absent
MRD flow Allele-specific
Lymph nodes Not palpablea
cytometry oligonucleotide
Liver/spleen Not palpablea PCR
Peripheral blood lymphocytes ::: 4 × 109/L Applicable >95% 85–95%
Peripheral blood neutrophils ::: 1.5 × 109/L patients

Hemoglobin >110 g/L Sensitivity limit 0.01% 0.001%

Bone marrow aspirate <30% lymphocytesb Quantitative 0.1–0.01% 0.01%

range
c
Bone marrow trephine No nodules
Cost and Moderate Initially high, follow-
a
No requirement for imaging; complexity up low
b
no requirement for immunophenotyping;
c
this can equate to up to 2% CLL cells. Pre-treatment Preferable Essential
material
required
Turn-round Hours Weeks
morphological examination of the bone marrow. A time
complete response was defined as a patient in whom
the clinical examination was normal, with an almost
normal blood count and a morphologically normal
100
bone marrow (Table 8.3). These criteria have proved All MRD-negative patients (n = 24)
to be extremely useful to allow comparison between Median survival not reached
80
% patients surviving

the results of trials from the various collaborative

groups. However, there can be as many as 2% CLL
cells in the marrow of a patient who is in an NCI
complete remission.
The advent of therapies that result in profound
reductions in levels of CLL has necessitated the devel-
opment of techniques that can detect extremely low
levels of CLL. The most important approaches for
assessment of minimal residual disease (MRD) are
molecular techniques, such as allele-specific oligonu-
cleotide polymerase chain reaction (ASO-PCR) direc-
ted against the immunoglobulin gene of the CLL
clone, or multiparameter, four-color flow cytometry
(MRD flow). Both of these techniques will detect as
low as a single CLL cell in 10 000 leukocytes or more.
ASO-PCR is slightly more sensitive than MRD flow
but has several disadvantages (Table 8.4), which means
that MRD flow is likely to become the standard
approach used in routine practice if and when CLL
patients are routinely tested for MRD. In all reports of
MRD in CLL, patients who achieve an MRD-negative
response have a better progression-free survival (PFS)
and overall survival (OS) than patients who remain
MRD-positive, regardless of the therapy used to
achieve such a profound response, including combi-
nation chemotherapy, immunochemotherapy, mono-
clonal antibody-based therapy or stem cell
transplantation (Figure 8.4).
60 P = 0.002
40 MRD-positive patients (n = 67)
Median survival 19 months
20
0
0 12 24 36 48 60 72 84 96
Months
Figure 8.4 Impact of minimal residual disease on overall survival in
CLL. Ninety-one patients with relapsed or refractory CLL were treated
with alemtuzumab in an attempt to eradicate detectable CLL using a
highly sensitive MRD flow cytometric assay. Twenty-four patients
achieved MRD negativity: 18 with alemtuzumab monotherapy, 4
following autologous stem-cell transplantation after alemtuzumab
and the remaining 2 patients following combined alemtuzumab and
fludarabine. The overall survival of patients achieving MRD negativity
was significantly better than those remaining MRD-positive.
However, in all of these series the aim of therapy
was to eradicate MRD and therefore they do not
prove beyond doubt that MRD eradication is critical
(the patients achieving MRD negativity may have
been biologically better risk and therefore could
have had a better survival regardless of therapy com-
pared to their more resistant counterparts). The next
series of clinical trials will therefore address in a
randomized fashion whether attempting to eradicate
MRD is an important endpoint of therapy. 125
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia

Table 8.5 Indications for therapy.

(1) Any one of the following symptoms:

(a) weight loss >10% body weight over 6 months
(b) extreme fatigue (WHO performance status 2 or worse)
(c) fevers of over 39.8°C for over 2 weeks without evidence of infection
(d) night sweats without evidence of infection.

(2) Evidence of progressive bone marrow failure with either progressive anemia (unless there is another cause) or thrombocytopenia
(platelets <100 × 109/L). If the cytopenia is stable, then treatment may not be immediately required.
(3) Autoimmune anemia and/or thrombocytopenia poorly responsive to corticosteroids or other standard therapy.
(4) Massive (i.e. >6 cm below the costal margin) or progressive or symptomatic splenomegaly.
(5) Massive lymphadenopathy (i.e. >10 cm in longest diameter) or progressive or symptomatic lymphadenopathy.
(6) Progressive lymphocytosis with an increase of >50% over a 2-month period, or anticipated doubling time of <6 months. (Doubling
time is not used as a single treatment parameter if initial lymphocyte count <30 000/mcl.)
(7) Marked hypogammaglobulinemia or the development of a monoclonal pattern, in the absence of any of the above criteria for active
disease, is not sufficient for protocol therapy.

Chemotherapeutic approach after their initial diagnosis. Therefore many patients

are well over 70 when they first require treatment and
The time to initiate therapy in CLL is defined by the
will probably have one or more significant comor-
NCI-WG criteria of 2008 (Table 8.5). Effectively,
bidities. Therefore relatively intensive chemoimmu-
patients require therapy if they are symptomatic
notherapeutic approaches, such as fludarabine,
from their CLL, or if they have cytopenias caused by
cyclophosphamide, and rituximab (FCR), may be con-
marrow replacement or, rarely, a rapidly evolving dis-
sidered too toxic. The key question is which patients
ease because of refractory immune cytopenias. The
should be treated with less intensive treatment. There
absolute level of lymphocytosis in the absence of any
is now a series of phase II and phase III trials in
other features indicating active disease should not be
progress aimed at optimizing therapy for those unfit
used to justify initiation of therapy.
for FCR. The criteria for entry into these trials vary.
There is no proven benefit for the initiation of
For example, the German CLL Study Group have
therapy with asymptomatic Binet stage A CLL,
applied the Cumulative Illness Rating Scale (CIRS) to
although the randomized trials only tested relatively
define those who are fit for FCR-like therapy (CIRS
ineffective therapies, namely alkylating agent-based
≤6) and those unfit for FCR (CIRS score >6). In con-
therapy, against no therapy in unselected early-stage
trast, the NCRI CLL subgroup in the UK have defined
patients. The advent of better prognostic markers and
criteria that are being used to define patients consid-
more effective therapy raises the question whether a
ered unfit for FCR-like therapy. The current trial for
subset of poor-risk patients may benefit from early
such patients is the RIAltO study, which compares
intervention. This question is currently being
chlorambucil plus ofatumumab with bendamustine
addressed by collaborative CLL trials groups.
plus ofatumumab, and defines patients as being unfit
Factors defining the choice of therapy for FCR if they have any one of the following:
The next question after a patient has fulfilled the
a. Age 75 or greater
criteria for treatment is what is the aim of treatment
b. WHO performance status 2 or 3
for the patient as this will have a major impact on the
c. Cardiac impairment (NYHA class II)
choice of treatment. This depends on both the charac-
teristics of the disease as well as patient-specific d. Respiratory impairment (bronchiectasis or
factors. moderate COPD)
e. Renal impairment (estimated GFR <30 mL/min)
Patient-specific factors f. Any other significant comorbidity that makes
The median age at diagnosis in CLL is 71 years, and R-FC (rituximab, fludarabine, cyclophosphamide)
126
many patients do not require therapy for some time inappropriate.
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
This broadly captures the patient population that
should be considered for alternatives to FCR, although
the view of the patient and their social circumstances
will also influence their choices.
It is also important to consider the biological
characteristics of the patient’s CLL when selecting
the most appropriate therapy. Currently, chromoso-
mal analysis by FISH influences the choice of treat-
ment. FISH is used to identify patients with deletion
of the short arm of chromosome 17 (17p deletion),
which is the location of the TP53 gene. This is the
only abnormality at present that is routinely being
used to modify therapeutic choices. It is clear that an
intact TP53 gene is necessary for the cell to react
appropriately to genotoxic therapies, such as chemo-
therapy. Therefore it is not surprising that patients
with 17p deletion respond poorly to chemotherapy.
Hallek et al. demonstrated in the German CLL8 Trial
that 90% of previously untreated patients without
17p deletion will respond to FCR, with almost 50%
achieving a complete remission (CR), whereas only
65% of patients with a 17p deletion will respond, with
only 5% achieving a CR. Therefore, patients with 17p
deletion should be considered for alternative thera-
pies that do not depend on an intact p53 pathway for
their efficacy. At present monoclonal antibodies,
such as alemtuzumab, and corticosteroids are the
routinely available therapies that work in a p53-inde-
pendent manner. FISH is widely available but,
increasingly, p53 mutational analysis is also being
used in clinical trials and is likely to enter routine
practice in the near future. A proportion of patients
have mutated p53 with normal FISH. It appears that
these patients have a similarly poor outcome to those
with 17p-deleted disease and in future are likely to be
included in the 17p-deleted/p53 dysfunctional cate-
gory of CLL.
At present neither the other poor biological
markers of CLL, such as deletion of the long arm of
chromosome 11 (11q del) or unmutated immunoglo-
bulin genes, nor the good biological markers, such as
isolated deletion of the long arm of chromosome 13
(13q del) or mutated immunoglobulin genes, influ-
ence the choice of treatment. However, there are
ongoing trials specifically using these criteria to influ-
ence therapy.
The aim of therapy will depend on the fitness of the
patient as well as the prognosis in each individual’s case.
At present the more effective therapies in terms of depth
and duration of remission are more intensive, such as
the combination of fludarabine, cyclophosphamide,
and rituximab (FCR). Even in patients considered fit
enough for this therapy, only three-quarters complete
the full six cycles. The median age at diagnosis of CLL is
71 years and, therefore, with current therapies, the aim
of treatment in most is disease control rather than
eradication of detectable disease and prolonged remis-
sion. However, studies assessing the quality of life in
patients following therapy have demonstrated that
better remissions are associated with a better quality of
life for longer. Therefore if a patient is fit enough for
FCR-like therapy then they should be offered this.
Alkylating agents in CLL
Randomized controlled trials in the 1990s demonstra-
ted that alkylating agents used as monotherapy were
equally effective when compared to combinations of
alkylating agents with steroids and vinca alkaloids
(CVP) or with the inclusion of an anthracycline
(CHOP). Randomized trials comparing purine analogs
with alkylating agents have all demonstrated a signifi-
cantly higher response rate (in the region of 15% com-
plete response for fludarabine monotherapy, compared
to <5% for chlorambucil) and improved PFS for purine
analogs (mainly fludarabine), but no improvement in
OS (median PFS for fludarabine monotherapy in the
range of 18–24 months). It appears that this lack of
improvement in OS is largely due to the crossover
from chlorambucil to fludarabine. Therefore, until
recently chlorambucil remained the therapy most com-
monly used as the initial therapy in CLL, and it remains
an option for elderly and medically unfit patients.
Choice of first line therapy in CLL
Previously untreated patients without 17p deletion
who are fit
Improved proportions of patients achieving complete
remissions following the combination of FCR were first
reported approximately 10 years ago by Keating and
colleagues. This has subsequently been shown to result
in significantly better PFS and OS when compared to
the previous gold standard fludarabine plus cyclophos-
phamide in the German CLL Study Group CLL8 Trial.
The latest updates from the German CLLSG CLL8 Trial
indicate that patients treated with FCR as their initial
therapy have a median time to progression in excess of 6
years. Therefore FCR is now the gold standard for
patients who have no significant comorbidities. 127
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
Previously untreated patients without 17p deletion
with significant comorbidity
The definition of who should and should not receive
FCR is not well defined. This has led to individual
differences in practice between hematologists. Most
base their decision on a combination of the comorbid-
ities suffered by the patient, their age, and, after a
discussion with the patient, the patient’s aims of ther-
apy. The gold standard for patients who are considered
unfit for FCR is less clear and varies between countries
and even within countries. The only randomized con-
trolled trial for elderly patients was the German CLL5
trial in which single agent chlorambucil was compared
with single agent fludarabine in patients over 65 years
of age. A major problem with this trial is that even
in younger fit patients fludarabine has not been shown
to be better than chlorambucil and is therefore
unlikely to be better in a population with comorbid-
ities. Eichhorst et al. demonstrated in the German
CLL5 Trial that there was no significant difference
in survival between the two arms but, if anything,
the trend was in favor of chlorambucil. It appears
that chlorambucil monotherapy remains the most fre-
quently used therapy for this group of patients and at
present there is no good data to refute this approach.
Two phase II trials tested the addition of rituximab to
chlorambucil for this group of patients and appeared
to show that the overall response rate (ORR) increased
from about 65% in previous trials to 80%, but there
was only an apparently modest increase in complete
remissions and no patients achieved eradication of
detectable MRD. Larger randomized phase III trials
have now completed recruitment (both the
Complement-1 Trial: chlorambucil versus chlorambu-
cil + ofatumumab and the German CLL11 Trial: chlor-
ambucil versus chlorambucil + rituximab versus
chlorambucil + GA-101 [a novel anti-CD20 anti-
body]). The results of these studies are awaited to see
if the addition of anti-CD20 antibodies to alkylating
agents leads to an improvement in PFS.
The other question that is being addressed in eld-
erly patients is whether there is a better chemotherapy
backbone than chlorambucil. Bendamustine is an
alkylating agent that may have some purine analog
activity. Knauf et al. reported a randomized trial of
bendamustine versus chlorambucil as initial therapy,
demonstrating a much higher response rate (68% ver-
sus 31%) and a longer median PFS for bendamustine
(21.6 months compared to 8.3 months for chlorambu-
128
cil). However, in this trial the outcome of the patients
receiving chlorambucil was very poor compared to
those in other trials, raising some doubts about the
results of the trial conclusions. Fischer et al. reported a
more attractive approach in the combination of bend-
amustine plus rituximab that appears to have high
response rates in phase II trials and is quickly becom-
ing widely used for patients in whom there are con-
cerns over their ability to tolerate FCR. However, the
bendamustine data is virtually all in patients who are
younger and fitter so we have relatively little data on
the tolerability of bendamustine in the elderly and
those with comorbidities. In addition, bendamustine
is an intravenous rather than an oral therapy, making
it less attractive for the elderly. The German group
have just completed recruiting a randomized phase III
trial comparing bendamustine, rituximab (BR) with
FCR but, again, this is in patients considered fit for
FCR. The NCRI group in the UK have just initiated a
trial comparing bendamustine + ofatumumab with
chlorambucil + ofatumumab in patients considered
unfit for FCR, but the results of this trial will not be
available for several years. Therefore there is uncer-
tainty over the “standard” therapy for patients who are
considered unfit for FCR but globally, at present,
probably single agent chlorambucil remains the most
widely used.
Previously untreated patients with 17p deletion
and/or p53 mutation
The analysis of genomic aberrations by FISH reveals
that approximately 7% of previously untreated
patients have a deletion of the short arm of chromo-
some 17 (17p deletion), which corresponds to the
locus of the TP53 gene. Most patients with 17p dele-
tion will have a mutation of the TP53 gene on the other
allele and therefore no functional p53 protein. In fact,
a smaller proportion of patients will have no obvious
17p deletion but will have two TP53 mutations, one on
each allele, and therefore have inactivated p53 in their
CLL cells. This is critical as p53 function is required for
the normal cellular response to the DNA damage
resulting from chemotherapy, i.e. DNA repair, cell
cycle arrest, or apoptosis. Therefore patients with 17p
deletion and/or TP53 mutation do not respond nor-
mally to chemotherapy, resulting in much poorer out-
comes compared to patients with an intact p53. For
example, Oscier et al. reported the LRF CLL4 trial,
in which patients with previously untreated CLL
were treated with either chlorambucil or fludarabine-
containing chemotherapy: the complete remission rate
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia

Table 8.6 Randomized controlled trials comparing fludarabine monotherapy with fludarabine plus cyclophosphamide.

Group Fludarabine Fludarabine plus P value

monotherapy cyclophosphamide
Eichhorst et al. (German CLL No. evaluable 151 148
Study Group) ORR 84.1% 95.3% 0.002
CR 8.6% 20.3% 0.004
Median PFS 21.0 46.7 0.003
(months)
Catovsky et al. (LRF CLL4) No. evaluable 176 176
ORR 81% 94%
CR 15% 39%
% PFS at 3 years 31% 59% <0.0005
Flinn et al. (Intergroup Trial No. evaluable 121 125
E2997) ORR 49.6% 70.4% 0.001
CR 5.8% 22.4% 0.0002
Median PFS 17.7 41.0% <0.001
(months)
ORR, Overall response rate; CR, complete response; PFS, progression-free survival.

for 17p deleted patients was less than 5% and all these
patients had died within 4 years of their initial therapy.
This has led to the search for therapies that do not
depend on an intact p53 to be effective. The current
therapies that are p53-independent are either mono-
clonal antibodies or high doses of corticosteroids as
well as allogeneic stem cell transplantation (SCT).
The current trials of therapy for 17p-deleted
CLL have combined alemtuzumab (Campath) with
high-dose steroids (i.e. CamPred in the UK NCRI
CLL206 trial [Pettitt et al., 2012] and CamDex in
the German CLLSG CLL2O Trial [Stilgenbauer
et al., 2011]). Both of these studies have resulted
in higher response rates and complete remission
rates when compared to historical experience with
conventional chemotherapy, even including FCR.
With CamPred 17 patients were treated in the
frontline setting and all 17 patients responded,
with 65% achieving a complete response. With
CamDex in the frontline setting 97% of patients
responded, whereas only 20% have thus far been
reported to have experienced a CR although the
follow-up in this trial is less mature. It is also
clear that the depth of remission and tolerability
of such therapies is better if patients have them as
their first therapy rather than in second or subse-
quent line. It is therefore important to test for 17p
deletion prior to any line of therapy so that the
patient is not exposed to ineffective but potentially
toxic therapy. It is also clear that, despite higher
response rates, patients are continuing to relapse
early and at present allogeneic stem cell transplant
in the first remission is advised for the minority
of patients where this is an option. We are also
looking at improving the effectiveness of therapy
by adding other potentially non-p53-dependent
therapies, such as lenalidomide (the UK
NCRI CLL210 trial has just opened which looks
at alemtuzumab [Campath], dexamethasone, and
lenalidomide [CamDexRev]), with either the
patients going onto allogeneic SCT if applicable
or being randomized to lenalidomide maintenance
or observation.
Of considerable interest at present is the activity of
the novel agents in development for CLL in 17p-
deleted CLL. It seems that the B-cell receptor pathway
antagonists, either the Bruton’s tyrosine kinase inhib-
itor (PCI-32765 or ibrutinib) or the PI3 kinase d
(CAL-101/GS-1101 or idelalisib) may well be active
in 17p-deleted CLL and could alter the treatment of
these poor-risk patients.
Monoclonal antibody therapy in CLL
Until recently the only monoclonal antibody that was
approved for use in CLL as a monotherapy was alem-
tuzumab (Campath or MabCampath), which was
licensed for fludarabine refractory CLL. The response
rates to single agent alemtuzumab in relapsed, refrac-
tory CLL range between 33% and 50%, with up to 25%
129
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
of patients achieving a complete response. The most
important predictor of response to alemtuzumab is the
presence or absence of significant lymphadenopathy.
Patients with massive lymphadenopathy (single lymph
nodes greater than 5 cm in diameter) have a very low
response rate. In these patients a more effective strat-
egy is to control the lymphadenopathy with an alter-
native therapy, such as high-dose methylprednisolone,
prior to alemtuzumab therapy. Two recently reported
phase II trials of subcutaneous alemtuzumab in flu-
darabine-refractory CLL suggest a similar efficacy
compared to when the drug is given intravenously
but with a much improved toxicity profile. In August
2012, Genzyme surrendered the license for alemtuzu-
mab, pending regulatory approval to re-introduce it as
a treatment for multiple sclerosis. Alemtuzumab is still
available for CLL through an Access Programme but it
is likely that it’s use will fall despite a long history of
effectiveness.
Ofatumumab (Arzerra), a second generation fully
human anti-CD20 antibody, was approved by FDA in
2009 and the European Medicines Agency in 2010 for
CLL that is refractory to fludarabine and alemtuzu-
mab. Ofatumumab is well tolerated with infusion
reactions being the major complication with approx-
imately 50% of patients responding with a median
progression free survival of approximately 6 months.
Ofatumumab provides a useful option for patients
who have failed conventional chemotherapies partic-
ularly those with cytopenias preventing further mye-
lotoxic therapy.
Rituximab (Rituxan or MabThera) has also been
used in large numbers of patients with CLL, both alone
and in combination. Rituximab as a single agent used
at the conventional dose of 375 mg/m2 weekly for 4
weeks has little efficacy in relapsed or refractory CLL,
with only partial remissions observed in a minority of
patients, and these remissions only persist for a few
months. The partial response rate increases with
higher doses of rituximab but complete responses
were not achieved and the doses used were extremely
high (up to 2250 mg/m2). The use of “conventional”
doses of rituximab has also been reported in untreated
CLL, with higher response rates (up to 50% of
patients), but still very few complete responses, and
these are not durable. Rituximab therefore appears to
have no role as monotherapy in CLL, but has an
important place in combination with chemotherapy
as the initial therapy for CLL, at least in a proportion of
130 patients (see above).
Choice of therapy in relapsed CLL
Although in some patients the choice of the initial
therapy is difficult, the choice of therapy in relapsed
and refractory disease is often even more challenging.
It will depend on a variety of factors, including the
following:
1. Previous treatment received: it is often difficult to
deliver intensive therapies, such as FCR, to patients
who have previously received six cycles of FCR,
especially if this was relatively recently. In addition,
there are increasing concerns over the potential
long-term bone marrow consequences of multiple
episodes of purine analogs. Unless the remission
after such therapy is prolonged (at least 2 years and
perhaps even 3 years), it is probably better to
consider alternative therapies if available.
2. Quality of the latest remission: the depth and
duration of the previous remission will define
whether a similar type of therapy as was previously
used should be re-used. For example, patients who
have a remission lasting less than 2 years after FCR
are considered to have relapsed early and should
have a change of therapy where possible. In this
situation alemtuzumab-based therapy may be an
option, either alone or in combination with
corticosteroids. In patients who are considered
unsuitable for such therapy, by way of either their
age and/or fitness, the aim of therapy will probably
be different and, if it is palliative, steroids alone or
therapies such as single agent ofatumumab might
be considered.
3. Physical condition of the patient: not only age
and frailty but also the history of infections
(patients with recurrent lower chest infections are
a particular concern) and the degree of cytopenias
will influence the choice of therapy. Whether a
patient can be treated with intensive therapies such
as FCR, might be considered for an allogeneic stem
cell transplant, or even less intensive therapies such
as bendamustine-based treatments is very much an
individual clinical decision.
4. Degree of marrow recovery: after FCR-like
therapies the degree of bone marrow recovery is
variable. Many patients might have a mild to
moderate degree of thrombocytopenia or,
occasionally, neutropenia and anemia, after
completion of their prior therapy. A quarter of
patients will not have completed the six cycles of
FCR previously, often because of persistent
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
cytopenias. In these patients an alternative to
fludarabine-based therapies is preferred, and
treatments such as bendamustine with rituximab
might be considered.
5. Philosophy of the patient: the central role of the
patient and his or her family in deciding which
therapy to choose cannot be underestimated. For
example, some patients will be searching for a
prolonged durable remission or even cure almost
regardless of the effort and risk involved, and such
patients are likely to want very intensive therapies,
such as allogeneic stem cell transplant, or might be
prepared to travel distances to enter clinical trials
of novel therapies (of which there are currently
many [see below]). On the other hand, some
patients may prefer a less intensive approach with
less disruption, may only want oral therapies
rather than parenteral treatment, or even a purely
palliative approach.
6. p53 pathway abnormalities: it is important to
repeat the cytogenetic examination by FISH and
TP53 mutation analysis (if available) prior to any
change or reinitiation of therapy. The frequency of
p53 pathway abnormalities increases with
increasing numbers of prior therapies and the
degree of resistance to treatment, and when they
occur in relapsed/refractory disease their
implications are similar to the frontline.
7. The availability of licensed and trial treatments:
there are now a number of approved therapies for
both frontline and relapsed CLL, including
chlorambucil, steroids (available but not approved
for CLL), fludarabine-based therapy, cladribine,
deoxycoformycin, bendamustine, rituximab in
combination, ofatumumab, alemtuzumab, and a
number of novel therapies in early- and late-stage
clinical trials as well as the possibility of allogeneic
SCT. The choice of therapy will obviously depend
on whether this treatment is approved locally and
whether there is funding.
Role of allogeneic stem cell transplantation in CLL
CLL is currently incurable and most patients, even
those with a good response to treatment, will eventu-
ally relapse and die as a result of their CLL. The current
exception is allogeneic stem cell transplant, which
offers the opportunity of cure to the patient.
However, this comes with a relatively high price both
in terms of the risk of mortality and of significant
toxicity. The success of allogeneic SCT in CLL depends
on a number of factors, including the patient’s age and
comorbidities as well as the depth of remission at the
time of the transplant. It is now possible to define a
group of patients who are unlikely to respond with
durable remissions to conventional therapy and whose
expected survival is poor. Since allogeneic stem cell
transplant is potentially curative it should be consid-
ered at an early stage as soon as a patient is known to
have a poor prognosis. The rationale for considering
transplantation earlier rather than later is that the
chances of a successful outcome are related to a large
degree to the quality of remission going into the trans-
plant and to the physical condition of the patient at
that time. There are clear European Guidelines for
who should be selected for allogeneic SCT as follows:
– the presence of deletion of chromosome 17p
– relapse within 2 years of FCR-like therapy
– relapse within 12 months of fludarabine
monotherapy (or similar).
Novel therapeutic agents
There has been an explosion in the number of novel
therapeutic targets in CLL and an even greater num-
ber of potential agents. This has been driven to a large
extent by our changing and increasing understanding
of the pathophysiology of CLL. It is clear now that
CLL is not, as previously often thought, simply the
accumulation of a population of mature lympho-
cytes. It is now apparent that CLL is actually a very
proliferative disorder, demonstrated by the finding
that in some patients up to 2% of the CLL population
is being “born” every day. This proliferation is
driven, at least in part, by the interaction between
the CLL cell and its microenvironment, and is medi-
ated by stimulation, presumably due to antigen,
through the B-cell receptor (the surface immunoglo-
bulin on the CLL cell). This proliferation is counter-
balanced by a high rate of apoptosis within the clone.
The rate of growth (simply the lymphocyte doubling
time) is dependent on the balance between prolifer-
ation and apoptosis. This gives us several potential
novel therapeutic targets that are proving to be
extremely promising and may well redefine the way
we treat CLL. The novel therapeutic approaches are
as follows.
1. Interfering with the microenvironment and its
interaction with the CLL cell
Drugs such as lenalidomide have activity in CLL and it 131
appears likely that this is mediated through its influence
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
over the interaction between CLL cells and their micro-
environment. In addition, other therapies, such as
targeting CXCR4 with plerixafor, are being tested.
2. Targeting downstream signaling from the B-cell receptor
It is becoming clear that signaling through the B-cell
receptor is of central importance in the pathophysiol-
ogy of CLL, and it appears that engagement of the
B-cell receptor leads to a proliferative signal being
passed through a series of tyrosine kinases. This pro-
vides a series of novel targets in CLL in that interfering
with the transmission of this stimulus offers potential
novel therapeutic targets. There are several different
potential targets in this signaling pathway, including
Src kinase, Bruton’s tyrosine kinase (Btk) and
phosphatidyl-inositol-3-kinase δ (PI3kd). The most
advanced agents in this class are two orally active
drugs, namely ibrutinib (formerly known as PCI-
32765), which inhibits Btk, and idelalisib (GS-1101,
formerly known as CAL-101), which inhibits PI3kd.
Both of these agents have shown a similar pattern of
response when used as a single agent in relapsed and
refractory CLL.
Pattern of response. Patients experience an almost
immediate improvement in their constitutional
symptoms associated with CLL and within days or a
couple of weeks notice an improvement in, or in
some cases, resolution of, their lymphadenopathy.
At the same time there is an increase in the lympho-
cytosis to sometimes two or three times the starting
value, and this lymphocytosis peaks at approximately
2 months after which it then begins to fall. In some
cases the lymphocytosis falls to below the starting
point and may normalize, whereas in others it
plateaus. Therefore patients experience a “nodal
response” with lymphocytosis and, over time, this
usually converts to a partial remission or even com-
plete remission. The pattern of response is intriguing
and is the subject of further studies. It appears that
these agents initially redistribute the CLL cells from
the tissue compartment into the peripheral blood and
then perhaps, because of the inhibition of prolifera-
tion, there is a natural decay of the remaining lym-
phocytes. Effectively they appear to tip the balance of
proliferation:apoptosis in favor of cell death.
Side-effect profile. These drugs are generally asso-
ciated with relatively mild side effects that are some-
what variable. For example, inhibition of Btk appears
132 to result in a self-limiting and usually relatively mild
diarrhea. GS-1101 (inhibition of PI3kd) can lead to
usually mild and reversible abnormalities in liver func-
tion tests. Grade 3 or 4 serious adverse events are rare.
Combination with other agents. The combination
of these agents with either chemotherapy, such as
bendamustine with or without rituximab, or with
anti-CD20 antibodies (rituximab or ofatumumab)
seeks to ameliorate or resolve the lymphocytosis and
is currently being studied in a number of randomized
phase III trials.
3. Apoptosis
The other side of the CLL pathophysiological see-saw
is cell death or apoptosis. CLL cells invariably have a
high expression of BCL-2 that in part leads to the
protection of the cell against apoptosis. It appears
that this overexpression of BCL-2 is caused by the
loss of two microRNAs, miR15a and miR16–1, that
negatively regulate the transcription of BCL-2.
Therefore BCL-2 is a potential target in CLL, but
unfortunately previous attempts to target BCL-2
with either antisense technology, namely oblimersen,
or small molecules, for example with obatoclax, have
failed to show sustained responses and the subse-
quent trials failed to demonstrate significant enough
activity for approval. These attempts to inhibit the
activity of BCL-2 were proof of principle that this
approach might be an effective therapeutic strategy if
a selective and high-affinity inhibitor could be
identified.
The recent report of the BH3 mimetic, navitoclax
or ABT-263, is an alternative strategy to inhibiting
the BCL-2 family of proteins. Navitoclax binds to
BCL-2, BCL-xL and BCL-w, releasing the proapop-
totic molecules, BAX and BAK, which then oligo-
merize on the mitochondrial outer membrane,
triggering apoptosis. Roberts et al. reported a phase
I trial of navitoclax in relapsed and refractory CLL,
demonstrating clinical activity, with 19 of 21 patients
having at least a 50% reduction in their lymphocyto-
sis and with nine (35%) of 26 patients achieving a
partial remission. Also, the activity of navitoclax
appears to be seen in patients with abnormalities of
the p53 pathway (i.e. those with deletion of chromo-
some 17p). However, the main dose-limiting side
effect of navitoclax is thrombocytopenia, almost cer-
tainly because of its activity against BCL-XL, which is
expressed by platelets. The next generation of BH3
mimetics, ABT-199 or GDC-0199, is specific for
BCL-2 and has relatively little activity against BCL-
XL. These agents promise to be effective without the
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
complication of thrombocytopenia and are currently
being studied in subsequent phase I trials. These
agents may be the most logical to combine with the
B-cell receptor antagonists.
4. Cyclin-dependent kinase inhibition
As a major component of the pathophysiology of CLL
is cell proliferation, then cyclin-dependent kinases
(CDK) are a potential target for therapy in CLL. The
first CDK inhibitor to enter clinical trials was flavopir-
idol, which inhibits a wide range of CDKs, including
CDK9, which is implicated in CLL. Flavopiridol
demonstrated considerable activity in vitro but little
activity in vivo, probably because of very high protein
binding. However, when the dosing strategy was
modified to give a bolus over 30 minutes followed by
a 4-hour infusion rather than continuous infusion
over 7 days, the protein binding was to some extent
overcome, and there was clinical activity in a phase II
trial. In over 100 patients the ORR to flavopiridol was
47%, with a median PFS of 12 months. However, a
major issue for flavopiridol is tumor lysis syndrome
(TLS), with acute hyperkalemia not infrequently
requiring immediate hemodialysis. TLS occurred in
19% of patients in a subsequent phase II trial that
recruited 113 patients. The development of flavopir-
idol is currently not being pursued, but an alternative
CDK inhibitor, dinaciclib (SCH727965), is now enter-
ing clinical studies and, although dinaciclib is better
tolerated, there is still a concern regarding TLS.
5. Immune-mediated approaches
Allogeneic SCT is potentially curative in CLL and is
used for patients with poor-risk disease, defined either
with a p53 pathway abnormality or by early relapse or
refractoriness to conventional therapy. However, the
vast majority of patients are not eligible for allogeneic
SCT owing either to age and/or comorbidities. Thus
there has been considerable interest in trying to harness
the immune system to control CLL without the toxicity
of allogeneic SCT. One of the most promising
approaches is the development of chimeric antigen
receptor specifically targeting CLL cells. A single case
report last year led to an almost frenzied excitement in
this approach as the patient, despite experiencing major
toxicity, had an extremely good response. However,
there are major logistic and safety issues to overcome
before this approach can be considered in any but a few
centers and this appears to be some way off the clinic.
6. Novel monoclonal antibodies
The addition of an anti-CD20 antibody, namely rit-
uximab, to chemotherapy in CLL has been a major
step forward, with the combination of FCR becoming
the standard approach for younger fitter patients.
FCR has been shown to lead to not only an improve-
ment in PFS but also, for the first time in a random-
ized trial in CLL, to an improvement in OS. However,
the expression of CD20 on CLL cells is classically low,
raising concerns that rituximab may not be the opti-
mal antibody in CLL. In addition, there is some
evidence that the use of rituximab in CLL leads to a
loss of CD20 expression on CLL cells, with the sug-
gestion that this is due to the specific removal of
CD20 from the cell surface by the reticuloendothelial
system or so-called “shaving.” This led to attempts to
develop more effective anti-CD20 antibodies and two
of these are either approved, in the case of ofatumu-
mab, or in phase III trials (GA-101). Ofatumumab is
a fully human antibody that targets a different epi-
tope on CD20 than rituximab, with closer binding to
the cell membrane. Ofatumumab shows single agent
activity in relapsed and refractory CLL, with approx-
imately half of patients responding, even heavily pre-
treated and cytopenic patients. However, there are no
studies directly comparing rituximab with ofatumu-
mab and the doses of ofatumumab used as a single
agent in the relapsed/refractory setting are extremely
high (22.3 g over a 6-month period), so it is not
entirely clear whether ofatumumab is simply a
more effective antibody than rituximab or whether
this is a dose phenomenon. There are ongoing
randomized phase III trials involving ofatumumab
in CLL at a lower dose in combination with chlor-
ambucil and, although the results are awaited, again
there is no comparison between ofatumumab and
rituximab.
GA-101 is a novel glyco-engineered anti-CD20
antibody that has properties of a type II antibody. All
of the other anti-CD20 antibodies, including rituxi-
mab and ofatumumab, are type I antibodies, which
means that they cross-link CD20 antigens and lead to
colocalization in the lipid rafts and probably to the
shaving phenomenon (the removal of CD20 antigens
from the cell surface after opsonization with anti-
body). GA-101 is a type II antibody which has high-
affinity binding to CD20. GA-101 has low comple-
ment-dependent cytotoxicity (CDC) related to the rec-
ognition of the CD20 epitope and the lack of CD20 133
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia

Initial therapy

Localized, stage I−II (rare) Advanced, stage III–IV

Involved-field radiotherapy
Watchful waiting, with
treatment at progression

Symptomatic Asymptomatic
No comorbidity
Watchful
Fludarabine/cyclophosphamide (FC)
waiting, with
With comorbidity
treatment at
Chlorambucil monotherapy
progression
Under investigation
Rituximab + fludarabine ± cyclophosphamide
Alemtuzumab consolidation

Relapsed or progressive disease

Chemoimmunotherapy Investigational
As above Stem cell transplantation
Pentostatin/cyclophosphamide Various monoclonal antibodies
±Rituximab (PCR)
Alemtuzumab Lenalidomide
Bendamustine ± Rituximab BCL-2 antisense therapy
Pulse corticosteroids B-cell receptor pathway inhibitors

Figure 8.5 Therapeutic approaches in small lymphocytic lymphoma/CLL. As standard therapies for front-line and for relapsed or refractory
disease are not yet established, treatment of these patients on a clinical trial is encouraged.

localization into lipid rafts after binding of the mono-
clonal antibody to CD20, but increased antibody-
dependent cellular cytotoxicity (ADCC) related to an
improved binding of GA-101 to the different allotypes
of FcgRIIIa expressed by natural killer (NK) cells and
monocytes. Compared with rituximab, the increased
direct cell death induction related to an elbow hinge
amino exchange of the Fab region and type II binding
of the CD20 epitope, GA-101 has shown promising
activity in small phase I trials and is now being studied
in a randomized phase III trial in combination with
chlorambucil, compared to both chlorambucil alone
134 and chlorambucil plus rituximab (the German CLL11
Trial).
Recommended therapeutic
approaches in CLL
As a general rule, patients should be offered entry into
well-designed clinical trials if available and if the
patient is eligible. A treatment algorithm is shown in
Figure 8.5.
Conclusion
Whilst the therapy of CLL has improved over the past
decade or so the current treatments have many prob-
lems. The most effective treatments are associated with
significant short- and long-term toxicities, which means
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
that they are only an option for a proportion of younger
fitter patients. In these patients they are not curative and
potentially lead to long-term toxicities that may com-
promise subsequent therapeutic interventions. The
treatment of more elderly patients or those with comor-
bidities, who form the majority of patients with CLL,
has hardly advanced over the past few decades. The
increased understanding of the pathophysiology of
CLL and the rapid development of a variety of novel
targeted therapies has the potential to change the treat-
ment landscape in CLL completely. The challenge that
we now face is how best to utilize these new agents to
confer the most rapid and effective change in the treat-
ment paradigm of CLL to optimize outcomes for all
patients with the disease.
Further reading
Andritsos LA, Byrd JC, Hewes B, et al. Preliminary results
from a phase I/II dose escalation study to determine the
maximum tolerated dose of plerixafor in combination
with rituximab in patients with relapsed chronic
lymphocytic leukemia. Haematologica, 2010;95
(Suppl.2): Abstract 0772.
Binet JL, Auquier A, Dighiero G, et al. A new prognostic
classification of chronic lymphocytic leukaemia derived
from a multivariate survival analysis. Cancer,
1981;48:198–206.
Bologna L, Gotti E, Manganini M, et al. Mechanism of
action of type II, glycoengineered, anti-CD20
monoclonal antibody GA101 in B-chronic lymphocytic
leukemia: whole blood assays in comparison with
rituximab and alemtuzumab. J Immunol,
2011;186(6):3762–3769.
Brentjens RJ, Rivière I, Park JH, et al. Safety and persistence
of adoptively transferred autologous CD19-targeted T
cells in patients with relapsed or chemotherapy refractory
B-cell leukemias. Blood, 2011;118:4817–4828.
Burger JA, Ghia P, Rosenwald A, Caligaris-Cappio F. The
microenvironment in mature B-cell malignancies: a
target for new treatment strategies. Blood,
2009;114(16):3367–3375.
Byrd JC, Lin TS, Dalton JT, et al. Flavopiridol
administered using a pharmacologically derived
schedule is associated with marked clinical efficacy in
refractory, genetically high-risk chronic lymphocytic
leukemia. Blood, 2007;109:399–404.
Byrd JC, Shinn C, Waselenko JK, et al. Flavopiridol induces
apoptosis in chronic lymphocytic leukemia cells via
activation of caspase-3 without evidence of bcl-2
modulation or dependence on functional p53. Blood,
1998;92:3804–3816.
Caligaris-Cappio F. Role of the microenvironment in
chronic lymphocytic leukaemia. Br J Haematol,
2003;123(3):380–388.
CLL Trialists’ Collaborative Group. Chemotherapeutic
options in chronic lymphocytic leukemia: a meta-analysis
of the randomized trials. JNCI J Natl Cancer Inst,
1999;91:861–868.
Davids MS, Lannutti BJ, Brown JR, and Letai AG. BH3
profiling demonstrates that restoration of apoptotic
priming contributes to increased sensitivity to PI3K
inhibition in stroma-exposed chronic lymphocytic
leukemia cells. Blood (ASH Annual Meeting Abstracts),
2011;118:974.
Döhner H, Stilgenbauer S, Benner A, et al. Genomic
aberrations and survival in chronic lymphocytic
leukemia. N Engl J Med, 2000;343(26):1910–1916.
Dreger P, Corradini P, Kimby E, et al. Chronic Leukemia
Working Party of the EBMT. Indications for allogeneic
stem cell transplantation in chronic lymphocytic
leukemia: the EBMT transplant consensus. Leukemia,
2007;21(1):12–17.
Dreger P, Dohner H, Ritgen M, et al. Allogeneic stem cell
transplantation provides durable disease control in poor-
risk chronic lymphocytic leukemia: long-term clinical
and MRD results of the German CLL Study Group
CLL3X trial. Blood, 2010;116:2438–2447.
Eichhorst BF, Busch R, Hopfinger G, et al. The German CLL
Study Group. Fludarabine plus cyclophosphamide versus
fludarabine alone in first-line therapy of younger patients
with chronic lymphocytic leukemia. Blood,
2006;107:885–891.
Eichhorst BF, Busch R, Stilgenbauer S, et al. German CLL
Study Group (GCLLSG). First-line therapy with
fludarabine compared with chlorambucil does not result
in a major benefit for elderly patients with advanced
chronic lymphocytic leukemia. Blood,
2009;114(16):3382–3391.
Fabbri M, Bottoni A, Shimizu M, et al. Association of a
microRNA/TP53 feedback circuitry with pathogenesis
and outcome of B-cell chronic lymphocytic leukemia.
JAMA, 2011;305:59–67.
Fischer K, Cramer P, Busch R, et al. Bendamustine
combined with rituximab in patients with relapsed and/
or refractory chronic lymphocytic leukemia: a
multicenter phase II trial of the German Chronic
Lymphocytic Leukemia Study Group. J Clin Oncol,
2011;29(26):3559–3566.
Flynn JM, Jones JA, Andritsos L, et al. Update on the phase I
study of the cyclin dependent kinase inhibitor dinaciclib
(SCH 727965) in patients with relapsed or refractory
chronic lymphocytic leukemia (CLL): confirmation of
clinical activity and feasibility of long-term
administration. Blood, 2010;116:1396. 135
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
Furman RR, Byrd JC, Brown JR, et al. CAL-101, an isoform-
selective inhibitor of phosphatidylinositol 3-kinase P110,
demonstrates clinical activity and pharmacodynamic
effects in patients with relapsed or refractory chronic
lymphocytic leukemia. Blood (ASH Annual Meeting
Abstracts), 2010;116:55.
Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the
diagnosis and treatment of chronic lymphocytic
leukemia: a report from the International Workshop on
Chronic Lymphocytic Leukemia updating the National
Cancer Institute-Working Group 1996 guidelines. Blood,
2008;111:5446–5456.
Hallek M, Fischer K, Fingerle-Rowson G, et al. Addition of
rituximab to fludarabine and cyclophosphamide in
patients with chronic lymphocytic leukaemia: a
randomised, open-label, phase 3 trial. Lancet,
2010;376(9747):1164–1174.
Herman SE, Gordon AL, Wagner AJ, et al.
Phosphatidylinositol 3-kinase-δ inhibitor CAL-101
shows promising preclinical activity in chronic
lymphocytic leukemia by antagonizing intrinsic and
extrinsic cellular survival signals. Blood,
2010;116(12):2078–2088.
Herman SE, Gordon AL, Hertlein E, et al. Bruton tyrosine
kinase represents a promising therapeutic target for
treatment of chronic lymphocytic leukemia and is
effectively targeted by PCI-32765. Blood,
2011;117(23):6287–6296.
Jaglowski SM, Flynn JM, Jones JA, et al. Flavopiridol is an
effective therapy to bridge patients with chronic
lymphocytic leukemia (CLL) to reduced-intensity
conditioning allogeneic stem cell transplant. Blood,
2010;116:2383.
Keating MJ, O’Brien S, Albitar M, et al. Early results of a
chemoimmunotherapy regimen of fludarabine,
cyclophosphamide, and rituximab as initial therapy for
chronic lymphocytic leukemia. J Clin Oncol,
2005;23(18):4079–4088.
Knauf WU, Lissichkov T, Aldaoud A, et al. Phase III
randomized study of bendamustine compared with
chlorambucil in previously untreated patients with
chronic lymphocytic leukemia. J Clin Oncol,
2009;27(26):4378–4384.
Lanasa MC, Andritsos L, Brown JR, et al. Interim analysis of
EFC6663, a multicenter phase 2 study of Alvocidib
(flavopiridol), demonstrates clinical responses among
patients with fludarabine refractory CLL. Blood,
2010;116:58.
Lin TS, Ruppert AS, Johnson AJ, et al. Phase II study of
flavopiridol in relapsed chronic lymphocytic leukemia
demonstrating high response rates in genetically high-
risk disease. J Clin Oncol, 2009;27:6012–6018.
Messmer BT, Messmer D, Allen SL, et al. In vivo
136 measurements document the dynamic cellular kinetics of
chronic lymphocytic leukemia B cells. J Clin Invest,
2005;115(3):755–764.
O’Brien S, Burger JA, Blum KA, et al. The Bruton’s tyrosine
kinase (BTK) inhibitor PCI-32765 induces durable
responses in relapsed or refractory (R/R) chronic
lymphocytic leukemia/small lymphocytic lymphoma
(CLL/SLL): follow-up of a phase Ib/II study. Blood, (ASH
Annual Meeting Abstracts), 2011;118:983.
O’Brien SM, Cunningham CC, Golenkov AK, et al. Phase I
to II multicenter study of oblimersen sodium, a Bcl-2
antisense oligonucleotide, in patients with advanced
chronic lymphocytic leukemia. J Clin Oncol,
2005;23:7697–7702.
O’Brien SM, Claxton DF, Crump M, et al. Phase I study of
obatoclax mesylate (GX15–070), a small molecule pan-
Bcl-2 family antagonist, in patients with advanced
chronic lymphocytic leukemia. Blood, 2009;113:299–305.
Oltersdorf T, Elmore SW, Shoemaker AR, et al. An inhibitor
of Bcl-2 family proteins induces regression of solid
tumours. Nature, 2005;435:677–681.
Oscier D, Wade R, Davis Z, et al. Prognostic factors
identified three risk groups in the LRF CLL4 trial,
independent of treatment allocation. Haematologica,
2010;95(10):1705–1712.
Pettitt AR, Jackson R, Carruthers S, et al. Alemtuzumab in
combination with methylprednisolone is a highly
effective induction regimen for patients with chronic
lymphocytic leukemia and deletion of TP53: final results
of the National Cancer Research Institute CLL206 trial. J
Clin Oncol, 2012;30(14):1647–1655.
Porter DL, Levine BL, Kalos M, Bagg A, June CH. Chimeric
antigen receptor-modified T cells in chronic lymphoid
leukemia. N Engl J Med, 2011;365(8):725–733.
Rai KR, Sawitsky A, Cronkite EP, et al. Clinical staging of
chronic lymphocytic leukemia. Blood, 1975;46:219–234.
Rai KR, Peterson BL, Appelbaum FR, et al. Fludarabine
compared with chlorambucil as primary therapy for
chronic lymphocytic leukemia. N Engl J Med,
2000;343:1750–1757.
Roberts AW, Seymour JF, Brown JR, et al. Substantial
susceptibility of chronic lymphocytic leukemia to BCL2
inhibition: results of phase 1 study of navitoclax (ABT-
263) in patients with relapsed or refractory disease. J Clin
Oncol, 2012;30(5):488–496.
Schetelig J, van BA, Brand R, et al. Allogeneic hematopoietic
stem-cell transplantation for chronic lymphocytic
leukemia with 17p deletion: a retrospective European
Group for Blood and Marrow Transplantation analysis. J
Clin Oncol, 2008;26:5094–5100.
Sharman J, de Vos S, Leonard JP, et al. A phase 1 study of the
selective phosphatidylinositol 3-kinase-delta (PI3K)
inhibitor, CAL-101 (GS-1101), in combination with
rituximab and/or bendamustine in patients with relapsed
 Chapter 8: Small lymphocytic lymphoma/chronic lymphocytic leukemia
or refractory chronic lymphocytic leukemia (CLL). Blood
(ASH Annual Meeting Abstracts), 2011;118:1787.
Stilgenbauer S, Cymbalista F, Leblond V, et al. Alemtuzumab
plus oral dexamethasone, followed by alemtuzumab
maintenance or allogeneic transplantation in ultra high-
risk CLL: interim analysis of a phase ii study of the
GCLLSG and fcgcll/MW. Blood (ASH Annual Meeting
Abstracts), 2011;118:2854.
Tam CS, O’Brien S, Wierda W, et al. Long-term results of the
fludarabine, cyclophosphamide, and rituximab regimen
as initial therapy of chronic lymphocytic leukemia. Blood,
2008;112:975–980.
Taylor RP, Lindorfer MA. Antigenic modulation and
rituximab resistance. Semin Hematol,
2010;47(2):124–132.
Tse C, Shoemaker AR, Adickes J, et al. ABT-263: a potent
and orally bioavailable Bcl-2 family inhibitor. Cancer Res,
2008;68:3421–3428.
van Delft MF, Wei AH, Mason KD, et al. The BH3 mimetic
ABT-737 targets selective Bcl-2 proteins and efficiently
induces apoptosis via Bak/Bax if Mcl-1 is neutralized.
Cancer Cell, 2006;10:389–399.
Wang L, Lawrence MS, Wan Y, et al. SF3B1 and other novel
cancer genes in chronic lymphocytic leukemia. N Engl J
Med, 2011;365:2497–2506.
Wierda WG, Padmanabhan S, Chan GW, et al.
Ofatumumab is active in patients with fludarabine-
refractory CLL irrespective of prior rituximab:
results from the phase 2 international study. Blood,
2011;118(19):5126–5129.
Zenz T, Häbe S, Denzel T, et al. Detailed analysis of p53
pathway defects in fludarabine-refractory chronic
lymphocytic leukemia (CLL): dissecting the contribution
of 17p deletion, TP53 mutation, p53-p21 dysfunction,
and miR34a in a prospective clinical trial. Blood,
2009;114(13):2589–2597.
Zenz T, Eichhorst B, Busch R, et al. TP53 mutation and
survival in chronic lymphocytic leukemia. J Clin Oncol,
2010;28(29):4473–4479.
Zenz T, Vollmer D, Trbusek M, et al. TP53 mutation
profile in chronic lymphocytic leukemia: evidence
for a disease specific profile from a comprehensive
analysis of 268 mutations. Leukemia,
2010;24(12):2072–2079.
137
 Chapter

9
Waldenström’s macroglobulinemia/
lymphoplasmacytic lymphoma
Steven P. Treon and Giampaolo Merlini

Introduction
Waldenström’s macroglobulinemia (WM) is a distinct
clinicopathological entity resulting from the accumula-
tion, predominantly in the bone marrow, of clonally
related lymphocytes, lymphoplasmacytic cells, and
plasma cells that secrete a monoclonal IgM protein
(Figure 9.1). This condition is considered to correspond
to lymphoplasmacytic lymphoma (LPL) as defined by
the World Health Organization classification system.
Most cases of LPL are WM, with less than 5% of cases
made up of IgA, IgG, and non-secreting LPL.

Epidemiology and etiology
WM is an uncommon disease, with a geometrical
increase with age. The incidence rate for WM is higher
among Caucasians, with African descendants repre-
senting only 5% of all patients. Genetic factors appear
to be important to the pathogenesis of WM. A com-
mon predisposition for WM with other malignancies
has been raised, and there have been numerous reports
of familial predisposition, including clustering of fam-
ily members with WM and other B-cell lymphoproli-
ferative diseases. In a recent study, 28% of 924 serial
WM patients presenting to a tertiary referral had a first
or second degree relative with either WM or another
B-cell disorder. Frequent familial associations with
other immunological disorders in healthy relatives
have also been reported. The role of environmental
factors in WM remains to be clarified, but chronic
antigenic stimulation from infections, certain drug
and agent orange exposures remain suspect. An etio-
logical role for hepatitis C virus (HCV) infection has
been suggested, though in one study no association
could be established using both serological and
Figure 9.1 Aspirate from a patient with Waldenstrom’s
macroglobulinemia demonstrating excess mature lymphocytes,
lymphoplasmacytic cells, and plasma cells (courtesy of Marvin
Stone M.D.).
molecular diagnostic studies for HCV infection in
100 consecutive WM patients.
Biology
Cytogenetics
Chromosome 6q deletions encompassing 6q21–25 have
been observed in up to half of WM patients, and at a
comparable frequency amongst patients with and with-
out a familial history. The presence of 6q deletions has
been suggested to discern patients with WM from those
with IgM monoclonal gammopathy of unknown signifi-
cance (MGUS), and to have potential prognostic signifi-
cance, including impact on progression-free survival
(PFS) following treatment response, though others
have reported no prognostic significance to the presence
of 6q deletions in WM. Other abnormalities by

138 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
cytogenetic or fluorescent in situ hybridization (FISH)
analyses include deletions in 13q14, TP53, and ATM,
trisomy 4, 12, and 18. Recently, the somatic point muta-
tion MYD 88 L265P (Leucine → Valine) has been iden-
tified in 90% of WM samples but this is only rarely
present in related B-cell lymphoproliferative disorders.
As such, it promises to become a useful diagnostic
marker and to provide new pathogenetic insights.
Nature of the clonal cell
The WM bone marrow B-cell clone shows intraclonal
differentiation from small lymphocytes with large
focal deposits of surface immunoglobulins, to lym-
phoplasmacytic cells, to mature plasma cells that
contain intracytoplasmic immunoglobulins. Clonal
B-cells are detectable among blood B-lymphocytes,
and their number increases in patients who fail to
respond to therapy or who progress. These clonal
blood cells present the peculiar capacity to differentiate
spontaneously, in in vitro culture, to plasma cells. This
is through an interleukin-6 (IL-6)-dependent process in
IgM MGUS and mostly an IL-6-independent process in
WM patients. All these cells express the monoclonal
IgM present in the blood and a variable percentage of
them also express surface IgD. The characteristic
immunophenotypic profile of the lymphoplasmacytic
cells in WM includes the expression of the pan-B-cell
markers CD19, CD20, CD22, CD79, and FMC7.2.
The phenotype of lymphoplasmacytic cells in WM
cells suggests that the clone is a post-germinal center
B-cell. This indication is further strengthened by the
results of the analysis of the nature (silent or amino
acid replacing) and distribution (in framework or CDR
regions) of somatic mutations in immunoglobulin
heavy- and light-chain variable regions performed in
patients with WM. This analysis showed a high rate of
replacement mutations, compared with the closest
germline genes, clustering in the CDR regions and with-
out intraclonal variation. Subsequent studies showed a
strong preferential usage of VH3/JH4 gene families, no
intraclonal variation, and no evidence for any isotype-
switched transcripts. These data indicate that WM may
originate from an IgM+ and/or IgM+ IgD+ memory
B-cell. Normal IgM+ memory B-cells localize in bone
marrow, where they mature to IgM-secreting cells.
Bone marrow microenvironment
Increased numbers of mast cells are found in the bone
marrow of WM patients, wherein they are usually
admixed with tumor aggregates. The role of mast
cells in WM has been investigated in one study
wherein coculture of primary autologous or mast cell
lines with WM LPC resulted in dose-dependent WM
cell proliferation and/or tumor colony formation, pri-
marily through CD40 ligand (CD40L) signaling.
Furthermore, WM cells, through elaboration of solu-
ble CD27 (sCD27), induced the upregulation of
CD40L on mast cells derived from WM patients and
mast cell lines, suggesting a microenvironmental sup-
port system. High levels of CXCR4 and VLA-4 have
also been observed in WM cells. In blocking experi-
mental studies, CXCR4 was shown to support migra-
tion of WM cells, while VLA-4 contributed to
adhesion of WM cells to bone marrow stromal cells.
Clinical features
The clinical and laboratory findings at time of diagnosis
of WM in one large institutional study are presented in
Table 9.1. Unlike most indolent lymphomas, splenome-
galy and lymphadenopathy are prominent in only a
minority of patients (≤15%). Purpura is frequently asso-
ciated with cryoglobulinemia and more rarely with AL
amyloidosis, while hemorrhagic manifestations and
neuropathies are multifactorial (see later). The morbid-
ity associated with WM is caused by the concurrence of
two main components: tissue infiltration by neoplastic
cells and, more importantly, the physicochemical and
immunological properties of the monoclonal IgM. As
shown in Table 9.2, the monoclonal IgM can produce
clinical manifestations through several different mech-
anisms related to its physicochemical properties,
non-specific interactions with other proteins, antibody
activity, and tendency to deposit in tissues.
Morbidity mediated by the effects
of IgM
Hyperviscosity syndrome
Blood hyperviscosity is affected by increased serum IgM
levels, leading to hyperviscosity-related complications.
The main determinants are: (1) a high concentration of
monoclonal IgMs, which may form aggregates and may
bind water through their carbohydrate component; and
(2) their interaction with blood cells. Monoclonal IgMs
increase red cell aggregation (rouleaux formation) and
red cell internal viscosity while also reducing deform- 139
ability. The possible presence of cryoglobulins can
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma

Table 9.1 Clinical and laboratory findings for 149 consecutive newly diagnosed patients with the consensus panel diagnosis of WM
presenting to the Dana Farber Cancer Institute.

Median Range Institutional normal reference range

Age (years) 59 34–84 NA
Gender (male/female) 85/64 NA
Bone marrow involvement 30% 5–95% NA
Adenopathy 16% NA
Splenomegaly 10% NA
IgM (mg/dL) 2870 267–12 400 40–230
IgG (mg/dL) 587 47–2770 700–1600
IgA (mg/dL) 47 8–509 70–400
Serum viscosity (cp) 2.0 1.4–6.6 1.4–1.9
Hct (%) 35.0% 17.2–45.4% 34.8–43.6
9
Plt (× 10 /L) 253 24–649 155–410
WBC (× 109/L) 6.0 0.3–13 3.8–9.2
B2M (mg/dL) 3.0 1.3–13.7 0–2.7
LDH 395 122–1131 313–618
NA (not applicable).

Table 9.2 Physicochemical and immunological properties of the monoclonal IgM protein in Waldenstrom’s macroglobulinemia.

Properties of IgM monoclonal protein Diagnostic Clinical manifestations

condition
Pentameric structure Hyperviscosity Headaches, blurred vision, epistaxis, retinal hemor-
rhages, leg cramps, impaired mentation, intracranial
hemorrhage
Precipitation on cooling Cryoglobulinemia Raynaud’s phenomenon, acrocyanosis, ulcers, purpura,
(type I) cold urticaria
Autoantibody activity to myelin Peripheral Sensorimotor neuropathies, painful neuropathies,
associated glycoprotein (MAG), ganglioside neuropathies ataxic gait, bilateral foot drop
M1 (GM1), sulfatide moieties on peripheral nerve
sheaths
Autoantibody activity to IgG Cryoglobulinemia Purpura, arthralgias, renal failure, sensorimotor
(type II) neuropathies
Autoantibody activity to red blood cell antigens Cold agglutinins Hemolytic anemia, Raynaud’s phenomenon, acrocya-
nosis, livedo reticularis
Tissue deposition as amorphous aggregates Organ Skin: bullous skin disease, papules, Schnitzler’s
dysfunction syndrome
GI: diarrhea, malabsorption, bleeding
Kidney: proteinuria, renal failure (light chain
component)
Tissue deposition as amyloid fibrils (light chain Organ Fatigue, weight loss, edema, hepatomegaly,
component most commonly) dysfunction macroglossia, organ dysfunction of involved organs:
heart, kidney, liver, peripheral sensory, and autonomic
nerves

140
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
contribute to increasing blood viscosity as well as to the
tendency to induce erythrocyte aggregation. Serum
viscosity is proportional to IgM concentration up to
30 g/L, then increases sharply at higher levels. Plasma
viscosity and hematocrit are directly regulated by the
body. Increased plasma viscosity may also contribute to
inappropriately low erythropoietin production, which is
the major reason for anemia in these patients. Clinical
manifestations are related to circulatory disturbances
that can be best appreciated by ophthalmoscopy,
which shows distended and tortuous retinal veins, hem-
orrhages, and papilledema (Figure 9.2). Symptoms
Figure 9.2 Fundoscopic examination of a patient with
Waldenstrom’s macroglobulinemia demonstrating hyperviscosity-
related changes, including dilated retinal vessels, peripheral
hemorrhages, and “venous sausaging” (courtesy of Marvin Stone M.D.).
A Figure 9.3 Cryoglobulinemia
usually occur when the monoclonal IgM concentration
exceeds 50 g/L or when serum viscosity is >4.0 centi-
poises (cp), but there is a great individual variability,
with some patients showing no evidence of hypervis-
cosity even at 10 cp. The most common symptoms are
oronasal bleeding, visual disturbances resulting from
retinal bleeding, and dizziness that may rarely lead to
coma. Heart failure can be aggravated, particularly in
the elderly, owing to increased blood viscosity,
expanded plasma volume, and anemia. Inappropriate
transfusion can exacerbate hyperviscosity and may pre-
cipitate cardiac failure.
Cryoglobulinemia
In up to 20% of WM patients, the monoclonal IgM can
behave as a cryoglobulin (type I), but it is symptomatic
in 5% or less of the cases. Cryoprecipitation is mainly
dependent on the concentration of monoclonal IgM;
for this reason plasmapheresis or plasma exchange are
commonly effective in this condition. Symptoms
result from impaired blood flow in small vessels and
include Raynaud’s phenomenon, acrocyanosis, and
necrosis of the regions most exposed to cold such as
the tips of the nose, ears, fingers, and toes (Figure 9.3),
malleolar ulcers, purpura, and cold urticaria. Renal
manifestations may occur but are infrequent.
Autoantibody activity
Monoclonal IgM may exert its pathogenic effects
through specific recognition of autologous antigens,
the most notable being nerve constituents, immunoglo-
bulin determinants, and red blood cell (RBC) antigens.
manifesting with severe acrocyanosis
in a patient with Waldenstrom’s
macroglobulinemia before (A) and
following warming and
plasmapheresis (B).
141
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
IgM-related neuropathy
The presence of peripheral neuropathy has been esti-
mated to range from 5% to 38% in WM patients. The
nerve damage is mediated by diverse pathogenetic
mechanisms: IgM antibody activity toward nerve con-
stituents causing demyelinating polyneuropathies;
endoneurial granulofibrillar deposits of IgM without
antibody activity, associated with axonal polyneuro-
pathy; occasionally by tubular deposits in the endoneu-
rium associated with IgM cryoglobulin, and, rarely, by
amyloid deposits or by neoplastic cell infiltration of
nerve structures. Half of the patients with IgM neuro-
pathy have a distinctive clinical syndrome that is asso-
ciated with antibodies against a minor 100-kDa
glycoprotein component of nerves, myelin-associated
glycoprotein (MAG). Anti-MAG antibodies are gener-
ally monoclonal IgMκ, and usually also exhibit reactiv-
ity with other glycoproteins or glycolipids that share
antigenic determinants with MAG. The anti-MAG-
related neuropathy is typically distal and symmetrical,
affecting both motor and sensory functions; it is slowly
progressive, with a long period of stability. Most
patients present with sensory complaints (paresthesias,
aching discomfort, dysesthesias, or lancinating pains),
imbalance and gait ataxia, owing to lack of propriocep-
tion, and leg muscle atrophy in advanced stage. Patients
with predominantly demyelinating sensory neuropathy
in association with monoclonal IgM to gangliosides
with disialosyl moieties, such as GD1b, GD3, GD2,
GT1b, and GQ1b, have also been reported. Anti-
GD1b and anti-GQ1b antibodies were significantly
associated with predominantly sensory ataxic neuropa-
thy. These antiganglioside monoclonal IgMs present
core clinical features of chronic ataxic neuropathy
with variably present ophthalmoplegia and/or RBC
cold agglutinating activity. The disialosyl epitope is
also present on RBC glycophorins, thereby accounting
for the red cell cold agglutinin activity of anti-Pr2 spe-
cificity. Monoclonal IgM proteins that bind to ganglio-
sides with a terminal trisaccharide moiety, including
GM2 and GalNac-GD1A, are associated with chronic
demyelinating neuropathy and severe sensory ataxia,
unresponsive to corticosteroids. Antiganglioside IgM
proteins may also cross-react with lipopolysaccharides
of Campylobacter jejuni, whose infection is known to
precipitate the Miller Fisher syndrome, a variant of the
Guillain–Barré syndrome. This finding indicates that
molecular mimicry may play a role in this condition.
Motor neurone disease has been reported in patients
142 with WM, and monoclonal IgM with anti-GM1 and
sulfoglucuronyl paragloboside activity. POEMS (poly-
neuropathy, organomegaly, endocrinopathy, M pro-
tein, and skin changes) syndrome is rarely associated
with WM.
Cold agglutinin hemolytic anemia
Monoclonal IgM may present with cold agglutinin
activity, i.e. it can recognize specific red cell antigens
at temperatures below physiological, producing
chronic hemolytic anemia. This disorder occurs in
<10% of WM patients and is associated with cold
agglutinin titers of >1:1000 in most cases. The mono-
clonal component is usually an IgMκ and reacts most
commonly with I/i antigens, with complement fixation
and activation. Mild chronic hemolytic anemia can be
exacerbated after cold exposure but rarely does hemo-
globin drop below 70 g/L. The hemolysis is usually
extravascular (removal of C3b opsonized cells by the
reticuloendothelial system, primarily in the liver) and
rarely intravascular from complement destruction of
RBC membrane. The agglutination of RBCs in the
cooler peripheral circulation also causes Raynaud’s
syndrome, acrocyanosis, and livedo reticularis.
Tissue deposition
The monoclonal protein can deposit in several tissues
as amorphous aggregates. Linear deposition of mono-
clonal IgM along the skin basement membrane is
associated with bullous skin disease. Amorphous IgM
deposits in the dermis determine the so-called IgM
storage papules on the extensor surface of the extrem-
ities – macroglobulinemia cutis. Deposition of mono-
clonal IgM in the lamina propria and/or submucosa of
the intestine may be associated with diarrhea, malab-
sorption, and gastrointestinal bleeding. It is well
known that kidney involvement is less common and
less severe in WM than in multiple myeloma, probably
because the amount of light chain excreted in the urine
is generally lower in WM than in myeloma and
because of the absence of contributing factors, such
as hypercalcemia, although cast nephropathy has also
been described in WM. The deposition of monoclonal
light chain as fibrillar amyloid deposits (AL amyloi-
dosis) is uncommon in patients with WM.
Manifestations related to tissue
infiltration by neoplastic cells
Tissue infiltration by neoplastic cells is rare and
can involve various organs and tissues, from the
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
bone marrow (described later) to the liver, spleen,
lymph nodes, and possibly the lungs, gastrointestinal
tract, kidneys, skin, eyes, and central nervous system.
In contrast to multiple myeloma, infiltration of
the kidney interstitium with lymphoplasmacytoid
cells has been reported in WM, while renal or peri-
renal masses are not uncommon. The skin can be
the site of dense lymphoplasmacytic infiltrates, sim-
ilar to that seen in the liver, spleen, and lymph nodes,
forming cutaneous plaques and, rarely, nodules.
Chronic urticaria and IgM gammopathy are the two
cardinal features of the Schnitzler syndrome, which
is not usually associated initially with clinical fea-
tures of WM, although evolution to WM is not
uncommon. Thus, close follow-up of these patients
is warranted.
Laboratory investigations
and findings
Hematological abnormalities
Anemia is the most common finding in patients
with symptomatic WM and is caused by a combina-
tion of factors: mild decrease in red cell survival,
impaired erythropoiesis, hemolysis, moderate
plasma volume expansion, and blood loss from the
gastrointestinal tract. Blood smears are usually
normocytic and normochromic, and rouleaux for-
mation is often pronounced. Electronically meas-
ured mean corpuscular volume may be elevated
spuriously owing to erythrocyte aggregation. In
addition, the hemoglobin estimate can be inaccurate,
i.e. falsely high, because of interaction between
the monoclonal protein and the diluent used in
some automated analyzers. Leukocyte and platelet
counts are usually within the reference range at pre-
sentation, although patients may occasionally
present with severe thrombocytopenia. As reported
above, monoclonal B-lymphocytes expressing sur-
face IgM and late-differentiation B-cell markers
are uncommonly detected in blood by flow cytome-
try. A raised erythrocyte sedimentation rate is
almost constantly observed in WM and may be the
first clue to the presence of the macroglobulin.
The clotting abnormality detected most frequently
is prolongation of thrombin time. AL amyloidosis
should be suspected in all patients with nephrotic
syndrome, cardiomyopathy, hepatomegaly, or
peripheral neuropathy.
Biochemical investigations
High-resolution electrophoresis combined with
immunofixation of serum and urine are recommend-
ed for identification and characterization of the IgM
monoclonal protein. The light chain of the monoclo-
nal IgM is κ in 75–80% of patients. A few WM patients
have more than one M-component. The concentration
of the serum monoclonal protein is very variable, but
in most cases lies within the range of 15–45 g/L.
Densitometry should be adopted to determine IgM
levels for serial evaluations because nephelometry is
unreliable and shows large intralaboratory as well as
interlaboratory variation. The presence of cold agglu-
tinins or cryoglobulins may affect determination of
IgM levels and, therefore, testing for cold agglutinins
and cryoglobulins should be performed at diagnosis. If
present, subsequent serum samples should be analyzed
under warm conditions for determination of serum
monoclonal IgM level. Although Bence Jones protein-
uria is frequently present, it exceeds 1 g/24 hours in
only 3% of cases. While IgM levels are elevated in WM
patients, IgA and IgG levels are most often depressed
and do not demonstrate recovery even after successful
treatment, suggesting that patients with WM harbor a
defect that prevents normal plasma cell development
and/or Ig heavy chain rearrangements.
Serum viscosity
Because of its large size (almost 1 000 000 daltons),
most IgM molecules are retained within the intravas-
cular compartment and can exert an undue effect on
serum viscosity. Therefore, serum viscosity should be
measured if the patient has signs or symptoms of
hyperviscosity syndrome. Fundoscopy remains an
excellent indicator of clinically relevant hyperviscosity.
Among the first clinical signs of hyperviscosity is the
appearance of peripheral and mid-peripheral dot- and
blot-like hemorrhages in the retina, which are best
appreciated with indirect ophthalmoscopy and scleral
depression. In more severe cases of hyperviscosity,
dot-, blot-, and flame-shaped hemorrhages can appear
in the macular area along with markedly dilated and
tortuous veins with focal constrictions resulting in
“venous sausaging,” as well as papilledema.
Bone marrow findings
The bone marrow is always involved in WM. Central
to the diagnosis of WM is the demonstration, by
trephine biopsy, of bone marrow infiltration by a 143
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
lymphoplasmacytic cell population constituted by small
lymphocytes with evidence of plasmacytoid/plasma cell
differentiation (Figure 9.1). The pattern of bone mar-
row infiltration may be diffuse, interstitial, or nodular,
showing usually an intertrabecular pattern of infiltra-
tion. A solely paratrabecular pattern of infiltration is
unusual and should raise the possibility of follicular
lymphoma. The bone marrow infiltration should rou-
tinely be confirmed by immunophenotypic studies (flow
cytometry and/or immunohistochemistry) showing the
following profile: sIgM+CD19+CD20+CD22+CD79+.
Up to 20% of cases may express either CD5, CD10, or
CD23. In these cases, care should be taken to exclude
chronic lymphocytic leukemia and mantle cell lym-
phoma satisfactorily. “Intranuclear” periodic acid–
Schiff (PAS)-positive inclusions (Dutcher–Fahey
bodies) consisting of IgM deposits in the perinuclear
space, and sometimes in intranuclear vacuoles, may be
seen occasionally in lymphoid cells in WM. An
increased number of mast cells, usually in association
with the lymphoid aggregates, is commonly found in
WM, and their presence may help in differentiating
WM from other B-cell lymphomas.
Other investigations
Magnetic resonance imaging (MRI) of the spine in
conjunction with computed tomography (CT) of the
abdomen and pelvis are useful in evaluating the disease
status in WM. Bone marrow involvement can be docu-
mented by MRI studies of the spine in over 90% of
patients, while CT of the abdomen and pelvis demon-
strated enlarged nodes in 43% of WM patients. Lymph
node biopsy may show preserved architecture or
replacement by infiltration of neoplastic cells with
lymphoplasmacytoid, lymphoplasmacytic, or poly-
morphous cytological patterns.
Prognosis
WM typically presents as an indolent disease, although
considerable variability in prognosis can be seen. The
median survival reported in several large series has
ranged from 5 to 10 years, though in a recent study
of 436 consecutive patients with WM, the median
overall survival (OS) from time of diagnosis was in
excess of 10 years. The presence of 6q deletions as a
prognostic marker remains controversial. Age is con-
sistently an important prognostic factor (>60–70
144 years), but this factor is often impacted by unrelated
morbidities. Anemia, which reflects both marrow
involvement and the serum level of the IgM monoclo-
nal protein (owing to the impact of IgM on intra-
vascular fluid retention), has emerged as a strong
adverse prognostic factor, with hemoglobin levels of
<9–12 g/dL associated with decreased survival in sev-
eral series. Cytopenias have also been regularly identi-
fied as a significant predictor of survival. However, the
precise level of cytopenias with prognostic significance
remains to be determined. High beta-2 microglobulin
levels (>3–3.5 g/dL), a high serum IgM M-protein
(>7 g/dL), as well as a low serum IgM M-protein
(<4 g/dL), and the presence of cryoglobulins were
shown in several studies to be adverse factors. A few
scoring systems have been proposed based on these
analyses (Table 9.3).
Treatment of Waldenström’s
macroglobulinemia
Treatment indications
Consensus guidelines on indications for treat-
ment initiation were formulated as part of the
Second International Workshop on Waldenström’s
Macroglobulinemia. Initiation of therapy should not
be based on the IgM levels since this may not correlate
with either disease burden or symptomatic status.
Initiation of therapy is appropriate for patients with
constitutional symptoms, such as recurrent fever,
night sweats, fatigue due to anemia, or weight loss.
The presence of progressive, symptomatic lymph-
adenopathy or splenomegaly provides additional rea-
sons to begin therapy. The presence of anemia with a
hemoglobin value of ≤10 g/dL or a platelet count
≤100 × 10/L on this basis of disease is also a reasonable
indication for treatment initiation. Certain complica-
tions of WM, such as hyperviscosity syndrome, symp-
tomatic sensorimotor peripheral neuropathy, systemic
amyloidosis, renal insufficiency, or symptomatic cryo-
globulinemia are also indications for therapy.
Treatment options
A precise therapeutic algorithm for therapy of WM
remains to be defined given the paucity of random-
ized clinical trials. Active agents include alkyla-
tors (chlorambucil, cyclophosphamide), nucleoside
analogs (cladribine, fludarabine), monoclonal
antibodies (rituximab, ofatumumab, alemtuzumab),
bortezomib, thalidomide, everolimus, and bendamus-
tine. Combination therapy, particularly with rituximab,
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma

Table 9.3 Prognostic scoring systems in Waldenstrom’s macroglobulinemia.

Study Adverse prognostic Number of groups Survival

factors
Gobbi et al. Hb <9 g/dL 0–1 prognostic factors Median: 48 months
Age >70 years 2–4 prognostic factors Median: 80 months
Weight loss
Cryoglobulinemia
Morel et al. Age ≥65 years 0–1 prognostic factors 5-year: 87%
Albumin <4 g/dL 2 prognostic factors 5-year: 62%
Number of cytopenias: 3–4 prognostic factors 5-year: 25%
Hb <12 g/dL
Platelets <150 × 109/L
WBC <4 × 109/L
Dhodapkar et al. β2M ≥3 g/dL β2M <3 mg/dL + Hb ≥12 g/dL 5-year: 87%
Hb <12 g/dL β2M <3 mg/dL + Hb <12 g/dL 5-year: 63%
IgM <4 g/dL β2M ≥3 mg/dL + IgM ≥4 g/dL 5-year: 53%
β2M ≥3 mg/dL + IgM <4 g/dL 5-year: 21%
Application of International Albumin ≤3.5 g/dL Albumin ≥3.5 g/dL + β2M Median: NR
Staging System Criteria <3.5 mg/dL
for Myeloma to WM
Dimopoulos et al. β2M ≥3.5 mg/L Albumin ≤3.5 g/dL + β2M Median: 116 months
<3.5 or β2M 3.5–5.5 mg/dL Median: 54 months
β2M >5.5 mg/dL
International Prognostic Age >65 years Prognostic factors* 5-year: 87%
Scoring System for WM (Morel et al.) Hb <11.5 g/dL 2 prognostic factors** 5-year: 68%
Platelets <100 × 109/L 3–5 prognostic factors 5-year: 36%
β2M >3 mg/L *Excluding age
IgM >7 g/dL **Or age >65

has been associated with improved clinical outcomes.
Individual patient considerations, including the pres-
ence of cytopenias, need for more rapid disease control,
age, and candidacy for autologous transplant therapy,
should be taken into account in making the choice of a
first line agent. For patients who are candidates for
autologous transplant therapy, exposure to continuous
chlorambucil or nucleoside analog therapy should be
limited given the potential for stem cell damage. The use
of nucleoside analogs may also increase risk for histo-
logical transformation to diffuse large B-cell lymphoma
(DLBCL), as well as myelodysplasia and acute myelo-
genous leukemia.
Chlorambucil
Oral alkylating drugs, alone and in combination ther-
apy with steroids, have been extensively evaluated in the
upfront treatment of WM. The greatest experience with
oral alkylator therapy has been with chlorambucil,
which has been administered on both a continuous
(i.e. daily dose schedule) as well as an intermittent
schedule. Patients receiving chlorambucil on a contin-
uous schedule typically receive 0.1 mg/kg per day,
whilst on the intermittent schedule patients will typi-
cally receive 0.3 mg/kg for 7 days, every 6 weeks. In a
prospective randomized study, Kyle et al. reported no
significant difference in the overall response rate
between these schedules, although, interestingly, the
median response duration was greater for patients
receiving intermittent versus continuously dosed chlor-
ambucil (46 versus 26 months). Despite the favorable
median response duration in this study for use of the
intermittent schedule, no difference in the median OS
was observed. Moreover, an increased incidence for
development of myelodysplasia and acute myelogenous
leukemia with the intermittent (3 of 22 patients) versus
the continuous (0 of 24 patients) chlorambucil schedule
prompted the authors of this study to express prefer-
ence for use of continuous chlorambucil dosing. The
use of steroids in combination with alkylator therapy
has also been explored. Dimopoulos and Alexanian
evaluated chlorambucil (8 mg/m2) along with predni-
sone (40 mg/m2) given orally for 10 days, every 6 weeks,
and reported a major response (i.e. reduction of IgM by
greater than 50%) in 72% of patients. Additional factors 145
to be taken into account in considering alkylator
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
therapy for patients with WM include necessity for
more rapid disease control given the slow nature of
response to alkylator therapy, as well as consideration
for preserving stem cells in patients who are candidates
for autologous transplant therapy.
Nucleoside analogs
Both cladribine and fludarabine have been extensively
evaluated in untreated as well as previously treated WM
patients. Cladribine administered as a single agent by
continuous intravenous infusion, by 2-hour daily infu-
sion, or by subcutaneous bolus injections for 5–7 days
has resulted in major responses in 40–90% of patients
who received primary therapy, whilst in the salvage
setting responses have ranged from 38% to 54%.
Median time to achievement of response in responding
patients following cladribine ranged from 1.2 to 5
months. The overall response rate with daily infusional
fludarabine therapy administered mainly on 5-day
schedules in previously untreated and treated WM
patients has ranged from 38% to 100% and 30% to
40%, respectively, which are on par with the response
data for cladribine. Median time to achievement of
response for fludarabine was also on par with cladribine
at 3–6 months. In general, response rates and durations
of responses have been greater for patients receiving
nucleoside analogs as first line agents, although in sev-
eral of the above studies wherein both untreated and
previously treated patients were enrolled, no substantial
difference in the overall response rate was reported.
Myelosuppression commonly occurred following pro-
longed exposure to either of the nucleoside analogs, as
did lymphopenia with sustained depletion of both CD4+
and CD8+ T-lymphocytes observed in WM patients
1 year after initiation of therapy. Treatment-related
mortality due to myelosuppression and/or opportunistic
infections attributable to immunosuppression occurred
in up to 5% of all treated patients in some series with
either nucleoside analog. Factors predicting for response
to nucleoside analogs in WM included age at start of
treatment (<70 years), pre-treatment hemoglobin
>95 g/L, platelets >75 000/mm3, disease relapsing off
therapy, patients with resistant disease within the first
year of diagnosis, and a long interval between first line
therapy and initiation of a nucleoside analog in relapsing
patients. The combination of nucleoside analogs with
cyclophosphamide and/or rituximab has been investi-
gated and is discussed below.
The safety of nucleoside analogs has been the
146 subject of investigation in several recent studies.
Autologous stem cell collection may be impaired in
patients who received a nucleoside analog. The long-
term safety of nucleoside analogs in WM was recently
examined by Leleu et al. in a large series of WM
patients. A sevenfold increase in transformation to
an aggressive lymphoma, and a threefold increase in
the development of acute myelogenous leukemia/
myelodysplasia were observed amongst patients
who received a nucleoside analog versus other thera-
pies for their WM. A recent meta-analysis by Leleu
et al. of several trials utilizing nucleoside analogs in
WM patients, which included patients who had pre-
viously received an alkylator agent, showed a crude
incidence of 6.6–10% for development of disease
transformation, and 1.4–8.9% for development of
myelodysplasia or acute myelogenous leukemia.
Monoclonal antibodies
Rituximab is a chimeric monoclonal antibody that
targets CD20, a widely expressed antigen on lympho-
plasmacytic cells in WM. Several retrospective and
prospective studies have indicated that rituximab,
when used at standard dosimetry (i.e. 4 weekly infu-
sions at 375 mg/m2) induced major responses in
approximately 27–35% of previously treated and
untreated patients. Furthermore, it was shown in
some of these studies that patients who achieved
minor responses or even stable disease benefited
from rituximab, as evidenced by improved hemoglo-
bin and platelet counts and reduction of lymphaden-
opathy and/or splenomegaly. The median time to
treatment failure in these studies was found to range
from 8 to 27+ months. Studies evaluating an extended
rituximab schedule consisting of four weekly courses
at 375 mg/m2/week, repeated 3 months later by
another 4-week course, have demonstrated major
response rates of 44–48%, with time to progression
(TTP) estimates of 16+ to 29+ months.
In many WM patients, a transient increase of serum
IgM (IgM flare) may be noted immediately following
initiation of rituximab treatment. The IgM flare may be
related to release of IL-6 by bystander immune cells in
response to binding of rituximab to FcγRIIA receptors,
and also occurs in response to intravenous immuno-
globulin administration in WM patients. The IgM flare
in response to rituximab does not herald treatment
failure, and, while most patients will return to their
baseline serum IgM level by 12 weeks, some patients
may flare for months despite having tumor responses in
their bone marrow. Patients with baseline serum IgM
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
levels of >50 g/dL or serum viscosity of >3.5 cp may be
particularly at risk for a hyperviscosity-related event
and in such patients plasmapheresis should be consid-
ered or rituximab omitted for the first few cycles of
therapy until IgM levels decline to safer levels. Because
of the decreased likelihood of response in patients with
higher IgM levels, as well as the possibility that serum
IgM and viscosity levels may abruptly rise, rituximab
monotherapy should not be used as sole therapy for
the treatment of patients at risk for hyperviscosity
symptoms.
Time to response after rituximab is slow and
exceeds 3 months on average. The time to best response
in one study was 18 months. Patients with baseline
serum IgM levels of <60 g/dL are more likely to
respond, irrespective of the underlying bone marrow
involvement by tumor cells. A recent analysis of 52
patients who were treated with single agent rituximab
has indicated that the objective response rate was sig-
nificantly lower in patients who had either low serum
albumin (<35 g/L) or elevated serum monoclonal pro-
tein (>40 g/L M-spike). Furthermore, the presence of
both adverse prognostic factors was associated with a
short TTP (3.6 months). Moreover, patients who had
normal serum albumin and relatively low serum mono-
clonal protein levels derived a substantial benefit from
rituximab, with a TTP exceeding 40 months.
The genetic background of patients may also be
important for determining response to rituximab. In
particular, a correlation between polymorphisms at
position 158 in the Fc gamma RIIIa receptor (CD16),
an activating Fc receptor on important effector cells
that mediate antibody-dependent cell-mediated cyto-
toxicity (ADCC), and rituximab response was
observed in WM patients. The attainment of better
categorical responses, i.e. very good partial response
or complete response following rituximab-based ther-
apy appears also dependent on the presence of at least
one valine amino acid at FcγRIIIa-158.
Ofatumumab is a fully humanized CD20-directed
monoclonal antibody that targets the small loop of
CD20, a target different to that of rituximab. A 43%
overall response rate was observed in a small series of
symptomatic WM patients following ofatumumab
administration, which included untreated and previ-
ously treated patients. An IgM flare with symptomatic
hyperviscosity was also observed in this series, neces-
sitating plasmapheresis. Additionally, ofatumumab
has been successfully administered to WM patients
who demonstrated intolerance to rituximab.
Bortezomib
Bortezomib is a proteasome inhibitor that has been
extensively investigated in WM. In a multicenter study
of the Waldenstrom’s Macroglobulinemia Clinical
Trials Group (WMCTG), 27 patients received up to
eight cycles of bortezomib at 1.3 mg/m2 on days 1, 4,
8, and 11. All but one patient had relapsed or refractory
disease. Following therapy, median serum IgM levels
declined from 4660 mg/dL to 2092 mg/dL (P < 0.0001).
The overall response rate was 85%, with 10 and 13
patients achieving a minor (<25% decrease in IgM)
and major (<50% decrease in IgM) response.
Responses were prompt, and occurred at a median of
1.4 months. The median TTP for all responding
patients in this study was 7.9 (range 3–21.4+) months,
and the most common grade III/IV toxicities occurring
in ≥5% of patients were sensory neuropathies (22.2%);
leukopenia (18.5%); neutropenia (14.8%); dizziness
(11.1%); and thrombocytopenia (7.4%). Importantly,
sensory neuropathies resolved or improved in nearly
all patients following cessation of therapy. As part of an
NCI-Canada study, Chen et al. treated 27 patients with
both untreated (44%) and previously treated (56%) dis-
ease. Patients in this study received bortezomib utilizing
the standard schedule until they either demonstrated
progressive disease or two cycles beyond a complete
response or stable disease. The overall response rate in
this study was 78%, with major responses observed in
44% of patients. Sensory neuropathy occurred in 20
patients, five with grade >3, and occurred following
two–four cycles of therapy; this resolved in 14 patients.
In addition to the above experiences with bortezomib
monotherapy in WM, Dimopoulos et al. observed
major responses in six of 10 previously treated WM
patients. The combination of bortezomib with steroids
and/or rituximab has also been investigated and is dis-
cussed below.
Immunomodulatory agents
Thalidomide as monotherapy, and in combination
with dexamethasone and/or clarithromycin, has been
examined in WM. Dimopoulos et al. demonstrated a
major response in five of 20 (25%) previously
untreated and treated patients who received single
agent thalidomide. Dose escalation from the thalido-
mide start dose of 200 mg daily was hindered by
development of side effects, including the develop-
ment of peripheral neuropathy in five patients,
obligating discontinuation or dose reduction.
Thalidomide as well as lenalidomide has been 147
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
investigated in combination with rituximab, and these
studies are discussed below.
Bendamustine
Bendamustine is a recently approved agent for the
treatment of relapsed/refractory indolent non-
Hodgkin’s lymphoma (NHL). Bendamustine has
structural similarities to both alkylating agents and
purine analogs. Bendamustine in combination with
rituximab has been investigated in both previously
untreated and relapsed/refractory WM patients and
is discussed below.
Everolimus
Everolimus is an oral inhibitor of the mTOR pathway,
which is approved for the treatment of renal cell car-
cinoma. The Akt-mTOR-p70 pathway is active in
WM, and inhibition of this pathway leads to apoptosis
of primary WM cells, and WM cell lines. Fifty patients
with a median of three prior therapies were treated
with everolimus in a joint Dana Farber/Mayo Clinic
study. The overall response rate was 70%, with 42% of
patients attaining a major response. The PFS at 12
months was estimated to be 62%. Grade 3 or higher
related toxicities were observed in 56% of patients,
with cytopenias constituting the most common tox-
icity. Pulmonary toxicity occurred in 10% of patients.
Dose reductions due to toxicity occurred in 52% of
patients.
A clinical trial examining the activity of everoli-
mus in previously untreated patients with WM has
been initiated by the WMCTG. IgM discordance to
bone marrow tumor burden was common in this
upfront study, and therefore serial bone marrow
biopsies are important in assessing disease response
to everolimus.
Combination strategies
Because rituximab is an active and a non-
myelosuppressive agent, its combination with various
chemotherapeutic agents has been extensively explored
in WM. The combination of CHOP (cyclophospha-
mide, doxorubicin, vincristine, prednisone) with ritux-
imab (CHOP-R) was investigated in a randomized
frontline study by the German Low Grade Lymphoma
Study Group (GLSG) involving 69 patients, most of
whom had WM. The addition of rituximab to CHOP
resulted in a higher overall response rate (94% versus
148 67%) and median TTP (63 versus 22 months) in
comparison to patients treated with CHOP alone.
Dimopoulos et al. investigated the combination of rit-
uximab, dexamethasone, and oral cyclophosphamide
(RCD) as primary therapy in 72 patients with WM. At
least a major response was observed in 74% of patients
in this study, and the 2-year PFS was 67%. Therapy was
well tolerated, though one patient died of interstitial
pneumonia. In the salvage setting, the use of CHOP-R
has been investigated in relapsed/refractory WM
patients. Among 13 evaluable patients, 10 patients
achieved a major response (77%), including three com-
plete response (CR) and seven partial response (PR),
and two patients achieved a minor response. In a retro-
spective study, Ioakimidis et al. examined the outcomes
of symptomatic WM patients who received CHOP-R,
cyclophosphamide, vincristine, prednisone, rituximab
(CVP-R), or cyclophosphamide, prednisone, rituximab
(CP-R). Baseline characteristics for all three cohorts
were similar for age, prior therapies, bone marrow
involvement, hematocrit, platelet count, and serum
beta-2 microglobulin, though serum IgM levels were
higher in patients treated with CHOP-R. The overall
response rates to therapy were comparable among all
three treatment groups: CHOP-R (96%); CVP-R (88%)
and CP-R (95%), though more CR were observed
among patients treated with either CVP-R or CHOP-
R. Comparison of adverse events for these regimens
showed a higher incidence for neutropenic fever as
well as treatment-related neuropathy in patients receiv-
ing CHOP-R and CVP-R versus CP-R. These results
suggest that, in WM, the use of doxorubicin and vin-
cristine may be omitted to minimize treatment-related
complications.
Combination therapy with nucleoside analogs has
been investigated as both first line and salvage ther-
apy in WM. Laszlo et al. evaluated the combination of
subcutaneous cladribine with rituximab in 29 WM
patients with either untreated or previously treated
disease. Intended therapy consisted of rituximab on
day 1 followed by subcutaneous cladribine 0.1 mg/kg
for 5 consecutive days, administered monthly for
four cycles. With a median follow-up of 43 months,
the overall response rate observed was 89.6%, with
seven CR, 16 PR, and three minor responses.
Response activity was similar between untreated
and previously treated patients. No major infec-
tions were observed despite the lack of antimicrobial
prophylaxis.
In a study by the WMCTG, the combination of
rituximab and fludarabine was administered to 43
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
WM patients, 32 (75%) of whom were previously
untreated. The overall response rate was 95.3%, and
83% of patients achieved a major response. The
median TTP was 51.2 months in this series, and was
longer for those patients who were previously
untreated and for those achieving at least a very good
PR. Hematological toxicity was common, particularly
neutropenia and thrombocytopenia. Two deaths
occurred in this study from non-Pneumocystis carinii
pneumonia. Secondary malignancies, including trans-
formation to aggressive lymphoma and development
of myelodysplasia or AML, were observed in six
patients in this series.
Tedeschi et al. reported a multicenter study with
fludarabine, cyclophosphamide, and rituximab (FCR)
in symptomatic WM patients with untreated or
relapsed/refractory disease to one line of chemother-
apy. Treatment consisted of rituximab at 375 mg/m2
on day 1, fludarabine at 25 mg/m2, and cyclophospha-
mide at 250 mg/m2 by intravenous administration on
days 2–4 every 4 weeks. Forty-three patients were
accrued to this study. The overall response rate was
89%, with 83% of patients attaining a major remission,
and 14% a CR. Prolonged neutropenia was observed in
up to a third of patients. With a median follow-up of
15 months, the median PFS for this study has not been
reached.
The combination of bortezomib, dexamethasone,
and rituximab (BDR) has been investigated as primary
therapy in patients with WM by the WMCTG. An
overall response rate of 96%, major response rate of
83%, and CR in 22% was observed with BDR. The
updated median PFS in this study was >56.1 months.
The incidence of grade 3 neuropathy was 30% in this
study, which utilized a twice-a-week schedule for bor-
tezomib administration at 1.3 mg/m2. Peripheral neu-
ropathy from bortezomib was reversible in most
patients in this study following discontinuation of
therapy, and patients benefitted with pregabalin. An
increased incidence of herpes zoster was also observed,
with BDR prompting the use of prophylactic antiviral
therapy. An alternative schedule for bortezomib
administration (i.e. weekly at 1.6 mg/m2) in combina-
tion with rituximab and/or dexamethasone has been
investigated in several studies, with overall response
rates of 80–90%. A lower incidence of peripheral neu-
ropathy was observed in two studies using bortezomib
once a week. The impact of once- versus twice-a-week
bortezomib administration on PFS remains to be clari-
fied, as does the use of subcutaneous bortezomib,
which appears to ameliorate toxicity while maintain-
ing responses similar to intravenous administration.
The combination of immunomodulator agents (tha-
lidomide, lenalidomide) with rituximab was investi-
gated by the WMCTG. Thalidomide was administered
at 200 mg daily for 2 weeks, followed by 400 mg daily
and thereafter for 1 year. Patients received four weekly
infusions of rituximab at 375 mg/m2, beginning 1 week
after initiation of thalidomide, followed by four addi-
tional weekly infusions of rituximab at 375 mg/m2
beginning at week 13. The overall and major response
rate was 72% and 64%, respectively, and the median
TTP was 38 months in this series. Dose reduction and/
or discontinuation of thalidomide was common, and
mainly attributed to treatment-related neuropathy. The
investigators concluded in this study that lower doses of
thalidomide (i.e. 50–100 mg/day) should be considered
in this patient population. The combination of lenali-
domide with rituximab was investigated by the
WMCTG using lenalidomide at 25 mg daily on a syn-
copated schedule wherein therapy was administered for
3 weeks, followed by a 1-week pause for an intended
duration of 48 weeks. Patients received 1 week of ther-
apy with lenalidomide, after which rituximab (375 mg/
m2) was administered weekly on weeks 2–5, then 13–16.
The overall and major response rates in this study were
50% and 25%, respectively, and a median TTP for
responders was 18.9 months. However, an acute
decrease in hematocrit was observed during the first 2
weeks of lenalidomide therapy in 13/16 (81%) patients
with a median absolute decrease in hematocrit of 4.8%,
resulting in anemia-related complications and hospital-
izations. Despite dose reduction, most patients in this
study continued to demonstrate aggravated anemia with
lenalidomide. The mechanism for lenalidomide-related
anemia in WM remains to be determined, and the use of
this agent among WM patients should be avoided.
The use of bendamustine in combination with rit-
uximab was explored by Rummel et al. in the frontline
therapy of WM. As part of a randomized study, patients
received six cycles of bendamustine plus rituximab
(Benda-R) or CHOP-R. A total of 546 patients were
enrolled in this study for patients with indolent NHL,
and included 41 patients with WM. For patients receiv-
ing Benda-R, bendamustine was administered at 90 mg/
m2 on days 1 and 2, and rituximab at 375 mg/m2 on day
1. The overall response rate was 96% for Benda-R-, and
94% for CHOP-R-treated patients. With a median
observation period of 45 months, the median PFS was
69.5 months for Benda-R versus 28.1 months for 149
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
CHOP-R. Importantly, Benda-R was associated with a
lower incidence of grade 3 or 4 neutropenia, infectious
complications, and alopecia. In the salvage setting, the
outcome of 30 WM patients with relapsed/refractory
disease who received bendamustine alone, or with a
CD20-directed antibody, was reported by Treon et al.
An overall response rate of 83.3% was reported. The
median estimated PFS for all patients was 13.2 months.
Overall therapy was well tolerated. Prolonged myelo-
suppression was more common in patients who
received prior nucleoside analogs.
Maintenance therapy
A role for maintenance rituximab in WM patients
following response to a rituximab-containing regimen
was raised in a study examining the outcome of 248
WM rituximab-naïve patients who were either
observed or received maintenance rituximab. In this
retrospective study, categorical responses improved in
16/162 (10%) observed patients, and in 36/86 (42%)
patients who received maintenance rituximab follow-
ing induction therapy. Both PFS (56.3 versus 28.6
months) and OS (>120 versus 116 months) were lon-
ger in patients who received maintenance rituximab.
Improved PFS was evident despite previous treatment
status, induction with rituximab alone, or in combi-
nation therapy. Best serum IgM response was lower,
and hematocrit was higher in those patients receiving
maintenance rituximab. Among patients receiving
maintenance rituximab, an increased number of infec-
tious events, predominantly sinusitis and bronchitis,
were observed, though these were mainly grade 1 or 2.
High-dose therapy and stem cell
transplantation
The use of stem cell transplantation (SCT) therapy has
also been explored in patients with WM. Kyriakou et al.
reported on the outcome of WM patients in the
European Bone Marrow Transplant (EBMT) registry
who received either an autologous or allogeneic SCT.
Among 158 patients receiving an autologous SCT,
which included primarily relapsed or refractory
patients, the 5-year PFS and OS rates were 39.7% and
68.5%, respectively. Non-relapse mortality at 1 year was
3.8%. Chemorefractory disease and the number of prior
lines of therapy at the time of the autologous SCT were
the most important prognostic factors for PFS and OS.
150 The achievement of a negative immunofixation after
autologous SCT had a positive impact on PFS. When
used as consolidation at first response, autologous
transplantation provided a PFS of 44% at 5 years.
In the allogeneic SCT experience from the EBMT,
the long-term outcome of 86 WM patients was reported
by Kyriakou. A total of 86 patients received allograft
by either myeloablative (n = 37) or reduced-intensity
(n = 49) conditioning. The median age of patients in
this series was 49 years, and 47 patients had three or
more previous lines of therapy. Eight patients failed
prior autologous SCT. Fifty-nine patients (68.6%) had
chemotherapy-sensitive disease at the time of allogeneic
SCT. Non-relapse mortality at 3 years was 33% for
patients receiving a myeloablative transplant, and 23%
for those who received reduced-intensity conditioning.
The overall response rate was 75.6%. The relapse rates at
3 years were 11% for myeloablative, and 25% for
reduced-intensity conditioning recipients. Five-year
PFS and OS for WM patients who received a myelo-
ablative allogeneic SCT were 56% and 62%, and for
patients who received reduced-intensity conditioning
were 49% and 64%, respectively. The occurrence of
chronic graft-versus-host disease was associated with
improved PFS, and suggested the existence of a clini-
cally relevant graft-versus-WM effect in this study.
Response criteria in Waldenström’s
macroglobulinemia
As part of the Second and Third International
Workshops on WM, consensus panels developed
guidelines for uniform response criteria in WM. The
category of minor response was adopted at the Third
International Workshop of WM, given that clinically
meaningful responses were observed with newer bio-
logical agents, and this category is based on ≥25 to
< 50% decrease in serum IgM level, which is used as a
surrogate marker of disease in WM. At the Sixth
International Workshop on WM, the categorical
response of very good partial response (VGPR), i.e.
90% reduction in IgM levels, was adopted given
reports of improved clinical outcome associated with
VGPR or better response achievement. In distinction,
the term major response is used to denote a response
of ≥50% in serum IgM levels, and includes partial or
better responses. Response categories and criteria for
progressive disease in WM based on consensus rec-
ommendations are summarized in Table 9.4.
An important concern with the use of IgM as a
surrogate marker of disease is that it can fluctuate,
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma

Table 9.4 Summary of updated response criteria adopted at the Sixth International Workshop on Waldenstrom’s macroglobulinemia.

Complete CR IgM in normal range, and disappearance of monoclonal protein by immunofixation; no histological
response evidence of bone marrow involvement, and resolution of any adenopathy/organomegaly (if present
at baseline), along with no signs or symptoms attributable to WM. Reconfirmation of the CR status is
required by repeat immunofixation studies.
Very good partial VGPR A >90% reduction of serum IgM and decrease in adenopathy/organomegaly (if present at baseline) on
response physical examination or on CT scan. No new symptoms or signs of active disease.
Partial response PR A >50% reduction of serum IgM and decrease in adenopathy/organomegaly (if present at baseline) on
physical examination or on CT scan. No new symptoms or signs of active disease.
Minor response MR A >25% but <50% reduction of serum IgM. No new symptoms or signs of active disease.
Stable disease SD A <25% reduction and <25% increase of serum IgM without progression of adenopathy/organo-
megaly, cytopenias, or clinically significant symptoms due to disease and/or signs of WM.
Progressive PD A >25% increase in serum IgM by protein confirmed by a second measurement or progression of
disease clinically significant findings due to disease (i.e. anemia, thrombocytopenia, leukopenia, bulky aden-
opathy/organomegaly) or symptoms (unexplained recurrent fever >38.4°C, drenching night sweats,
>10% body weight loss, or hyperviscosity, neuropathy, symptomatic cryoglobulinemia or amyloidosis)
attributable to WM.

independent of tumor cell killing, particularly with
newer biologically targeted agents such as rituximab,
bortezomib, and everolimus. Rituximab induces a
spike or flare in serum IgM levels, which can occur
when used as monotherapy or in combination with
other agents, including cyclophosphamide, nucleoside
analogs, thalidomide, and lenalidomide, and last for
several weeks to months. Bortezomib and everolimus
can suppress IgM levels independent of tumor cell
killing in certain patients. Moreover, in patients treat-
ed with selective B-cell-depleting agents such as ritux-
imab and alemtuzumab, residual IgM-producing
plasma cells are spared and continue to persist, thus
potentially skewing the relative response and assess-
ment to treatment. Therefore, in circumstances where
the serum IgM levels appear out of context with the
clinical progress of the patient, a bone marrow biopsy
should be considered to clarify the patient’s underlying
disease burden. Soluble CD27 may serve as an alter-
native surrogate marker in WM, and remains a faithful
marker of disease in patients experiencing a
rituximab-related IgM flare, as well as plasmapheresis.
Further reading
Anagnostopoulos A, Zervas K, Kyrtsonis M, et al. Prognostic
value of serum beta 2-microglobulin in patients with
Waldenstrom’s macroglobulinemia requiring therapy.
Clin Lymph Myeloma, 2006;7:205–209.
Aoki H, Takishita M, Kosaka M, Saito S. Frequent somatic
mutations in D and/or JH segments of Ig gene in
Waldenström’s macroglobulinemia and chronic
lymphocytic leukemia (CLL) with Richter’s syndrome but
not in common CLL. Blood, 1995;85:1913–1919.
Buske C, Hoster E, Dreyling MH, et al. The addition of
rituximab to front-line therapy with CHOP (R-CHOP)
results in a higher response rate and longer time to
treatment failure in patients with lymphoplasmacytic
lymphoma: results of a randomized trial of the German
Low-Grade Lymphoma Study Group (GLSG). Leukemia,
2009;23:153–161.
Chang H, Qi C, Trieu Y, et al. Prognostic relevance of 6q
deletion in Waldenstrom’s macroglobulinemia. Clin
Lymph Myeloma, 2009;9:36–38.
Chassande B, Leger JM, Younes-Chennoufi AB, et al.
Peripheral neuropathy associated with IgM monoclonal
gammopathy: correlations between M-protein antibody
activity and clinical/electrophysiological features in 40
cases. Muscle Nerve, 1998;21:55–62.
Chen CI, Kouroukis CT, White D, et al. Bortezomib is active
in patients with untreated or relapsed Waldenstrom’s
macroglobulinemia: a phase II study of the National
Cancer Institute of Canada Clinical Trials Group. J Clin
Oncol, 2007;25:1570–1575.
Cheson BD, Rummel MJ. Bendamustine: rebirth of an old
drug. J Clin Oncol, 2009;27:1492–1501.
Ciccarelli BT, Yang G, Hatjiharissi E, et al. Soluble CD27
is a faithful marker of disease burden and is
unaffected by the rituximab induced IgM flare, as
well as plasmapheresis in patients with
Waldenstrom’s macroglobulinemia. Clin Lymphoma
Myeloma, 2009;9:56–58.
Daoud MS, Lust JA, Kyle RA, Pittelkow MR. Monoclonal
gammopathies and associated skin disorders. J Am Acad
Dermatol, 1999;40:507–535. 151
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
Dhodapkar MV, Jacobson JL, Gertz MA, et al. Prognostic
factors and response to fludarabine therapy in patients
with Waldenström macroglobulinemia: results of United
States intergroup trial (Southwest Oncology Group
S9003). Blood, 2001;98:41–48.
Dimopoulos MA, Alexanian R. Waldenstrom’s
macroglobulinemia. Blood, 1994;83:1452–1459.
Dimopoulos MA, Kantarjian H, Weber D, et al. Primary
therapy of Waldenström’s macroglobulinemia with 2-
chlorodeoxyadenosine. J Clin Oncol, 1994;12:2694–2698.
Dimopoulos MA, Zomas A, Viniou NA, et al. Treatment of
Waldenström’s macroglobulinemia with thalidomide.
J Clin Oncol, 2001;19:3596–3601.
Dimopoulos MA, Zervas C, Zomas A, et al. Treatment of
Waldenstrom’s macroglobulinemia with rituximab.
J Clin Oncol 2002;20:2327–2333.
Dimopoulos M, Gika D, Zervas K, et al. The international
staging system for multiple myeloma is applicable in
symptomatic Waldenstrom’s macroglobulinemia. Leuk
Lymph, 2004;45:1809–1813.
Dimopoulos MA, Anagnostopoulos A, Kyrtsonis MC, et al.
Treatment of relapsed or refractory Waldenstrom’s
macroglobulinemia with bortezomib. Haematologica,
2005;90:1655–1657.
Dimopoulos MA, Anagnostopoulos A, Zervas C, et al.
Predictive factors for response to rituximab in
Waldenstrom’s macroglobulinemia. Clin Lymphoma,
2005;5:270–272.
Dimopoulos MA, Anagnostopoulos A, Kyrtsonis MC, et al.
Primary treatment of Waldenstrom’s macroglobulinemia
with dexamethasone, rituximab and cyclophosphamide. J
Clin Oncol, 2007;25:3344–3349.
Dimopoulos MA, Gertz MA, Kastritis E, et al. Update on
treatment recommendations from the Fourth
International Workshop on Waldenstrom’s
Macroglobulinemia. J Clin Oncol, 2009;27:120–126.
Eurelings M, Ang CW, Notermans NC, et al. Antiganglioside
antibodies in polyneuropathy associated with
monoclonal gammopathy. Neurology,
2001;57:1909–1912.
Furman RR, Eradat H, Switzky JC, et al. A phase II trial of
ofatumumab in subjects with Waldenstrom’s
macroglobulinemia. Blood, 116:Abstract 1795.
Gad A, Willen R, Carlen B, Gyland F, Wickander M.
Duodenal involvement in Waldenström’s
macroglobulinemia. J Clin Gastroenterol,
1995;20:174–176.
Ghobrial I, Gertz M, LaPlant B, et al. Phase II trial of the oral
mammalian target of rapamycin inhibitor everolimus in
relapsed or refractory Waldenström macroglobulinemia.
J Clin Oncol, 2010;28:1408–1414.
152 Ghobrial IM, Xie W, Padmanabhan S, Badros A, et al. Phase
II trial of weekly bortezomib in combination with
rituximab in untreated patients with Waldenström
macroglobulinemia. Am J Hematol, 2010;85:670–674.
Gobbi PG, Bettini R, Montecucco C, et al. Study of prognosis
in Waldenström’s macroglobulinemia: a proposal for a
simple binary classification with clinical and
investigational utility. Blood, 1994;83:2939–2945.
Gordon PH, Rowland LP, Younger DS, et al.
Lymphoproliferative disorders and motor neuron
disease: an update. Neurology, 1997;48:1671–1678.
Hanzis C, Ojha RP, Hunter Z, et al. Associated malignancies
in patients with Waldenström’s macroglobulinemia and
their kin. Clin Lymphoma Myeloma Leuk, 2011;11:88–92.
Hellmann A, Lewandowski K, Zaucha JM, et al. Effect of a 2-
hour infusion of 2-chlorodeoxyadenosine in the
treatment of refractory or previously untreated
Waldenström’s macroglobulinemia. Eur J Haematol,
1999;63:35–41.
Ho A, Leleu X, Hatjiharissi E, et al. CD27-CD70 interactions
in the pathogenesis of Waldenstrom’s
macroglobulinemia. Blood, 2008;112:4683–4689.
Ioakimidis L, Patterson CJ, Hunter ZR, et al. Comparative
outcomes following CP-R, CVP-R and CHOP-R in
Waldenstrom’s macroglobulinemia. Clin Lymphoma
Myeloma, 2009;9:62–66.
Kimby E, Treon SP, Anagnostopoulos A, et al. Update on
recommendations for assessing response from the Third
International Workshop on Waldenstrom’s
macroglobulinemia. Clin Lymphoma Myeloma,
2006;6:380–383.
Kyle RA, Greipp PR, Gertz MA, et al. Waldenström’s
macroglobulinaemia: a prospective study comparing
daily with intermittent oral chlorambucil. Br J Haematol,
2000;108:737–742.
Kyle RA, Treon SP, Alexanian R, et al. Prognostic markers
and criteria to initiate therapy in Waldenström’s
macroglobulinemia: Consensus Panel Recommendations
from the Second International Workshop on
Waldenström’s macroglobulinemia. Semin Oncol,
2003;30:116–120.
Kyriakou C, Canals C, Cornelissen JJ, et al. Allogeneic stem-
cell transplantation in patients with Waldenström
macroglobulinemia: report from the Lymphoma
Working Party of the European Group for Blood and
Marrow Transplantation. J Clin Oncol,
2010;28:4926–4934.
Kyriakou C, Canals C, Sibon D, et al. High-dose therapy and
autologous stem-cell transplantation in Waldenstrom
macroglobulinemia: the Lymphoma Working Party of
the European Group for Blood and Marrow
Transplantation. J Clin Oncol, 2010;28:2227–2732.
Laszlo D, Andreola G, Rigacci L, et al. Rituximab and
subcutaneous 2-chloro-2’-deoxyadenosine combination
treatment for patients with Waldenstrom
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
macroglobulinemia: clinical and biologic results of a
phase II multicenter study. J Clin Oncol,
2010;28:2233–2238.
Leblond V, Ben Othman T, Deconinck E, et al. Activity of
fludarabine in previously treated Waldenström’s
macroglobulinemia: a report of 71 cases. Groupe
Cooperatif Macroglobulinemie. J Clin Oncol,
1998;16:2060–2064.
Leleu X, Jia X, Runnels J, et al. The Akt pathway regulates
survival and homing in Waldenstrom
macroglobulinemia. Blood, 2007;110:4417–4426.
Leleu X, O’Connor K, Ho A, et al. Hepatitis C viral infection
is not associated with Waldenstrom’s
macroglobulinemia. Am J Hematol, 2007;82:83–84.
Leleu XP, Manning R, Soumerai JD, et al. Increased
incidence of transformation and myelodysplasia/acute
leukemia in patients with Waldenström
macroglobulinemia treated with nucleoside analogs.
J Clin Oncol, 2009;27:250–255.
Leleu X, Tamburini J, Roccaro A, et al. Balancing risk versus
benefit in the treatment of Waldenstrom’s
macroglobulinemia patients with nucleoside analogue
based therapy. Clin Lymph Myeloma, 2009;in press.
Levy Y, Fermand JP, Navarro S, et al. Interleukin 6
dependence of spontaneous in vitro differentiation of B
cells from patients with IgM gammopathy. Proc Natl
Acad Sci USA, 1990;87:3309–3313.
Liu ES, Burian C, Miller WE, Saven A. Bolus administration
of cladribine in the treatment of Waldenström
macroglobulinaemia. Br J Haematol, 1998;103:690–695.
Merlini G, Baldini L, Broglia C, et al. Prognostic factors in
symptomatic Waldenström’s macroglobulinemia. Semin
Oncol, 2003;30:211–215.
Morel P, Monconduit M, Jacomy D, et al. Prognostic factors
in Waldenström macroglobulinemia: a report on 232
patients with the description of a new scoring system and
its validation on 253 other patients. Blood,
2000;96:852–858.
Morel P, Duhamel A, Gobbi P, et al. International prognostic
scoring system for Waldenstrom macroglobulinemia.
Blood, 2009;113:4163–4170.
Ngo HT, Leleu X, Lee J, Jia X, et al. SDF-1/CXCR4 and VLA-
4 interaction regulates homing in Waldenstrom
macroglobulinemia. Blood, 2008;112:150–158.
Nguyen-Khac F, Lejeune J, Chapiro E, et al. Cytogenetic
abnormalities in a cohort of 171 patients with
Waldenström macroglobulinemia before treatment:
clinical and biological correlations. Blood, 2010;116:
Abstract 801.
Ocio EM, Schop RF, Gonzalez B, et al. 6q deletion in
Waldenstrom’s macroglobulinemia is associated with
features of adverse prognosis. Br J Haematol,
2007;136:80–86.
Ogmundsdottir HM, Sveinsdottir S, Sigfusson A, et al.
Enhanced B cell survival in familial macroglobulinaemia
is associated with increased expression of Bcl-2. Clin Exp
Immunol, 1999;117:252–260.
Ogmundsdottir HM, Steingrimsdottir H, Haraldsdottir V.
Familial paraproteinemia: hyper-responsive B-cells as
endophenotype. Clin Lymphoma Myeloma Leukemia,
2011;11:82–84.
Owen RG, Treon SP, Al-Katib A, et al. Clinicopathological
definition of Waldenström’s macroglobulinemia:
Consensus Panel Recommendations from the Second
International Workshop on Waldenström’s
macroglobulinemia. Semin Oncol, 2003;30:110–115.
Rummel MJ, Niederle N, Maschmeyer GA, et al.
Bendamustine plus rituximab versus CHOP plus
rituximab as firstline treatment of patients with
indolent and mantle cell lymphomas: updated
results from the StiL NHL1 study. Proc ASCO, 2012;
abstract #3.
Sahota SS, Forconi F, Ottensmeier CH, et al. Typical
Waldenström macroglobulinemia is derived from a
B-cell arrested after cessation of somatic mutation but
prior to isotype switch events. Blood,
2002;100:1505–1507.
Schop RF, Kuehl WM, Van Wier SA, et al. Waldenström
macroglobulinemia neoplastic cells lack immunoglobulin
heavy chain locus translocations but have frequent 6q
deletions. Blood, 2002;100:2996–3001.
Shiokawa S, Suehiro Y, Uike N, Muta K, Nishimura J.
Sequence and expression analyses of mu and delta
transcripts in patients with Waldenström’s
macroglobulinemia. Am J Hematol, 2001;68:139–143.
Silvestri F, Barillari G, Fanin R, et al. Risk of hepatitis C virus
infection, Waldenström’s macroglobulinemia, and
monoclonal gammopathies. Blood, 1996;88:1125–1126.
Singh A, Eckardt KU, Zimmermann A, et al.
Increased plasma viscosity as a reason for
inappropriate erythropoietin formation. J Clin Invest,
1993;91:251–256.
Strauss SJ, Maharaj L, Hoare S, et al. Bortezomib therapy in
patients with relapsed or refractory lymphoma: potential
correlation of in vitro sensitivity and tumor necrosis
factor alpha response with clinical activity. J Clin Oncol,
2006;24:2105–2112.
Tedeschi A, Benevolo G, Varettoni M, et al. Results of a
phase II multicenter study of immunochemotherapy with
fludarabine, cyclophosphamide and rituximab (FCR) for
symptomatic Waldenstrom’s macroglobulinemia. Blood,
2008;112:Abstract 3692.
Tournilhac O, Santos DD, Xu L, et al. Mast cells in
Waldenstrom’s macroglobulinemia support
lymphoplasmacytic cell growth through CD154/CD40
signaling. Ann Oncol, 2006;17:1275–1282. 153
 Chapter 9: Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma
Treon SP. How I treat Waldenstrom’s macroglobulinemia.
Blood, 2009;114:419–431.
Treon SP, Emmanouilides C, Kimby E, et al. Extended
rituximab therapy in Waldenström’s macroglobulinemia.
Ann Oncol, 2005;16:132–138.
Treon SP, Hansen M, Branagan AR, et al. Polymorphisms in
FcγRIIIA (CD16) receptor expression are associated with
clinical responses to rituximab in Waldenstrom’s
macroglobulinemia. J Clin Oncol, 2005;23:474–481.
Treon SP, Hunter Z, Branagan A. CHOP plus rituximab
therapy in Waldenström’s macroglobulinemia. Clin
Lymphoma Myeloma, 2005;5:273–277.
Treon SP, Hunter ZR, Matous J, et al. Multicenter clinical
trial of bortezomib in relapsed/refractory Waldenstrom’s
macroglobulinemia: results of WMCTG Trial 03–248.
Clin Cancer Res, 2007;13:3320–3325.
Treon SP, Hunter Z, Ciccarelli BT, et al. IgA and IgG
hypogammaglobulinemia is a constitutive feature in most
Waldenstrom’s macroglobulinemia patients and may be
related to mutations associated with common variable
immunodeficiency disorder (CVID). Blood,
2008;112:3749.
Treon SP, Soumerai JD, Branagan AR, et al. Lenalidomide
and rituximab in Waldenström’s macroglobulinemia.
Clin Cancer Res, 2008;15:355–360.
Treon SP, Soumerai JD, Branagan AR, et al. Thalidomide
and rituximab in Waldenstrom’s macroglobulinemia.
Blood, 2008;112:4452–4457.
Treon SP, Branagan AR, Ioakimidis L, et al. Long term
outcomes to fludarabine and rituximab in Waldenstrom’s
macroglobulinemia. Blood, 2009;Epub ahead of print.
Treon SP, Ioakimidis L, Soumerai JD, et al. Primary therapy
of Waldenstrom’s macroglobulinemia with bortezomib,
dexamethasone and rituximab: results of WMCTG
Clinical Trial 05–180. J Clin Oncol, 2009;27:3830–3835.
Treon SP, Hanzis C, Tripsas C, et al. Bendamustine therapy
in patients with relapsed or refractory Waldenström’s
macroglobulinemia. Clin Lymphoma Myeloma Leuk,
2011;11:133–135.
Treon SP, Yang G, Hanzis C, et al. Attainment of complete/
very good partial response following rituximab-based
therapy is an important determinant to progression-free
survival, and is impacted by polymorphisms in FCGR3A
in Waldenstrom macroglobulinaemia. Br J Haematol,
2011;154;223–228.
Treon SP, Hanzis, C, Manning RJ, et al. Maintenance
rituximab is associated with improved clinical outcome
154
in rituximab naïve patients with Waldenstrom’s
macroglobulinemia who respond to a rituximab
containing regimen. Br J Haematol, 2011; May [Epub
ahead of print].
Treon SP, Merlini G, Morra E, et al. Report from the Sixth
International Workshop on Waldenstrom’s
Macroglobulinemia. Clin Lymph Myeloma Leukemia,
2011;11:69–73.
Treon SP, Xu L, Yang G, et al. MYD88 L265P somatic
mutation in Waldenstrom macroglobulinemia. N Engl
J Med, 2012;367:826–833.
Varettoni M, Tedesci A, Arcaini L, et al. Risk of second
cancers in Waldenstrom macroglobulinemia. Ann Oncol,
2011; [Epub ahead of print].
Vital A. Paraproteinemic neuropathies. Brain Pathol,
2001;11:399–407.
Wagner SD, Martinelli V, Luzzatto L. Similar patterns of
V kappa gene usage but different degrees of somatic
mutation in hairy cell leukemia, prolymphocytic
leukemia, Waldenström’s macroglobulinemia, and
myeloma. Blood, 1994;83:3647–3653.
Weber D, Treon SP, Emmanouilides C, et al. Uniform
response criteria in Waldenstrom’s macroglobulinemia:
consensus panel recommendations from the Second
International Workshop on Waldenstrom’s
macroglobulinemia. Semin Oncol, 2003;30:127–131.
Weiss MD, Dalakas MC, Lauter CJ, Willison HJ, Quarles
RH. Variability in the binding of anti-MAG and anti-
SGPG antibodies to target antigens in demyelinating
neuropathy and IgM paraproteinemia. J Neuroimmunol,
1999;95:174–184.
Whittaker SJ, Bhogal BS, Black MM. Acquired
immunobullous disease: a cutaneous manifestation of
IgM macroglobulinaemia. Br J Dermatol,
1996;135:283–286.
World Health Organization. Classification of Tumours of
Haematopoietic and Lymphoid Tissues. In Swerdlow SH,
Campo E, Harris NL, et al. (eds). Lyon: IARC Press,
2008.
Yang G, Xu L, Hunter ZR, et al. The rituximab and IVIG
related IgM flare in Waldenstrom’s macroglobulinemia is
associated with monocytic activation of FCGR2A
signaling, and triggering of IL-6 release by the PI3K/AKT
and MAPK pathways. Blood, 2010;116:2870.
Zinzani PL, Gherlinzoni F, Bendandi M, et al. Fludarabine
treatment in resistant Waldenström’s
macroglobulinemia. Eur J Haematol, 1995;54:120–123.
 Chapter
10
Mantle cell lymphoma
Martin Dreyling and Michael E. Williams
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Introduction
Mantle cell lymphoma (MCL) is a unique subtype of
non-Hodgkin lymphoma (NHL) characterized by the
chromosomal translocation t(11;14)(q13;q32) and
nuclear cyclin D1 overexpression. It was first recog-
nized as a distinct entity in the Kiel Classification in
the 1970s, when it was designated “centrocytic lym-
phoma.” Subsequent descriptors included mantle
zone lymphoma and intermediate lymphocytic lym-
phoma. In 1992, based on the recognition of charac-
teristic morphologies, phenotype, and the t(11;14),
the term “mantle cell lymphoma” was proposed to
reflect the apparent derivation from mantle zone
B-cells. Following this advance, focused clinical
and pathogenesis studies became possible and have
led to an evolving understanding of MCL, its clinical
and biologic spectrum, and special therapeutic
challenges.
Most patients present with advanced stage disease,
often with extranodal dissemination, and, according to
biological and clinical risk factors, may pursue either
an indolent, steadily progressive, or an aggressive clin-
ical course. No standard curative therapy exists aside
from allogeneic transplantation. However, immuno-
chemotherapy regimens with consolidative autolo-
gous stem cell transplantation in younger patients, an
increasing number of effective therapies for sequential
application, novel targeted agents, and maintenance
therapy strategies are improving response durations
and overall survival (OS).
MCL comprises approximately 4–6% of all NHL,
with a preponderance of older males relative to other
lymphoma subtypes. The male:female ratio is 2–3:1
and median age at presentation is 65 years. No specific
etiologic factors have been identified for this disease.
An increased risk of other lymphoid neoplasms is
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
recognized among first-degree relatives of MCL
patients, although MCL occurrence among multiple
family members is quite rare.
Clinical presentation
MCL typically presents in advanced stage; over 90% of
patients are stage III–IV at diagnosis, and frequently
exhibit B-symptoms. Splenomegaly is seen in half or
more of patients, often in association with a leukemic
phase. In some patients disease is largely non-nodal
and confined to the blood, bone marrow, and spleen;
these individuals may experience a more indolent clin-
ical course than those with predominantly nodal dis-
ease (Table 10.1). The most common extranodal site of
disease is the gastrointestinal (GI) tract, including
colonic polyposis. Subclinical involvement of the gas-
tric or colonic mucosa has been reported in most
patients on endoscopic biopsies, even in the absence
of overt polyps or mucosal abnormalities. This trop-
ism for the GI tract is as yet not fully explained,
but may relate to the expression of adhesion molecules
such as the mucosal homing receptor α4β7.
Genitourinary, pulmonary, and soft tissue involve-
ment, including head and neck or periorbital sites of
disease, may also be observed. CNS involvement,
either parenchymal or leptomeningeal, is unusual at
presentation but may subsequently develop in associ-
ation with disease progression or as an isolated site of
relapse in a minority of patients, especially with blas-
toid variants. The occurrence of CNS relapse may be
increasing as patient survivals improve following more
effective systemic therapies. As a result, the role of
CNS prophylaxis is being investigated in clinical trials.
As is true for most cancers, the natural history and
time course from the initial transforming event to clin-
ical presentation is unknown but may be quite long. An
155
 Chapter 10: Mantle cell lymphoma

Table 10.1 Prognostic markers in mantle cell lymphoma.

Better prognosis Poorer prognosis

MIPI score Low Intermediate or high
Clinical presentation Predominant blood, marrow, Predominantly nodal
and splenic disease
Nodal architecture Nodular or mantle zone Diffuse
Cytology Small cleaved cells Blastoid cells
Phenotypic and serologic:
Ki-67 index <10% >30%
SOX-11 Negative Positive
Beta-2 microglobulin Normal Increased
Molecular markers:
p53 Wild-type (WT) Mutated
Chr. 9/CDK p16INK4a WT Deleted or methylated
Proliferation signature Low High
Post-treatment MRD Negative Positive
miR-29 Expressed Loss of expression
miR-17–92 cluster Normal expression Overexpression
MIPI, Mantle cell lymphoma International Prognostic Index; MRD, minimal residual disease.

intriguing insight into this timeline was provided with

the diagnosis of MCL in a patient 12 years after allogen-
eic stem cell transplantation for another disorder.
Identical MCL clonality of donor origin occurring syn-
chronously in both the donor (male) and recipi-
ent (female) was demonstrated; the donor had
likewise been diagnosed 12 years following stem cell
donation. Subsequently, an additional series of MCL
patients were shown to harbor occult MCL with
t(11;14) translocation in nodal biopsies obtained from
unrelated surgical procedures as many as 7–15 years
prior to a clinical diagnosis of MCL. Other reports of
incidentally identified in situ MCL further support a
long latency, with the presence of the t(11;14) in these
early lesions consistent with cyclin D1 overexpression
Figure 10.1 Mantle cell lymphoma (lymph node). A diffuse infiltrate
as an initiating event in MCL pathogenesis. of cells with scanty cytoplasm surrounding a residual germinal center.
The cells have irregular nuclei and indistinct nucleoli.

Pathology
Designations in the older literature used descriptive
terms such as “lymphocytic lymphoma of intermediate
differentiation” and “mantle zone lymphoma.” The first
specific recognition was by Lennert, who described “dif-
fuse germinocytoma” in 1973, renamed “centrocytic
lymphoma” in the 1974 Kiel classification. As immuno-
phenotypic characterization became more refined and,
importantly, the strong correlation with the t(11;14)
156 (q13;q32) chromosomal translocation permitted differ-
entiation from other small cell lymphomas, the current
designation of “mantle cell lymphoma” was proposed
by the International Lymphoma Study Group in 1992
and further delineated in the WHO Classification.
The lymph node architecture is usually effaced by
a diffuse proliferation of lymphoid cells with monot-
onous appearance (Figure 10.1). In some cases there
may be a more nodular growth pattern, or the cells
may be seen surrounding residual reactive germinal
centers in a mantle-zone pattern. In classical cases
(80–90%) the cells are small to intermediate size

with scanty cytoplasm and with nuclei that show
variable irregularity. The nuclear chromatin is
granular and cells may have small, rather indistinct
eosinophilic nucleoli. Rarely the cells may resemble
small lymphocytes with round nuclei or have a
more monocytoid appearance. Large cells resembling
centroblasts or immunoblasts are not a feature.
The background frequently contains hyalinized
small blood vessels and may contain histiocytes.
Proliferation centers are absent. Plasma cells may be
seen in the background but are reactive rather than
part of the neoplastic population.
In a proportion of cases the morphology is not
typical. In the blastoid variant the cells resemble lym-
phoblasts with scanty cytoplasm and fine nuclear chro-
matin associated with a striking mitotic rate. The
pleomorphic variant is composed of more pleomorphic
cells, which are larger than the typical cells of MCL and
have large nuclei with irregular outlines and prominent
nucleoli. These morphological variants are associated
with more aggressive disease.
Immunophenotypically the cells have surface IgM
and IgD. They are positive for CD20 and CD79a, and
typically coexpress CD5, CD43, and FMC7. Although
up to 94% of cases express CD43, as many as 20% of
cases in some series are reported to be CD5-negative.
MCL are negative for CD10 and BCL-2 and are usually
CD23-negative. The cells are positive for BCL-2
protein. There is no staining for MUM1/IRF4.
Staining for nuclear cyclin D1 is characteristically pos-
itive, reflecting the underlying t(11;14)(q13;q32) that is
characteristic of this lymphoma (Figure 10.2). Cyclin
D1 positivity is also present in rare cases that appear to
Figure 10.2 Mantle cell lymphoma (lymph node). Immunostaining
for cyclin D1 gives a characteristic nuclear pattern.
Chapter 10: Mantle cell lymphoma
lack this cytogenetic abnormality; even more rarely,
cyclin D2 or D3 is expressed rather than D1. Such cyclin
D1-negative cases, confirmed by gene expression profil-
ing, may present a considerable diagnostic challenge.
Nuclear expression of the neuronal transcription factor
Sox11 is present in up to 90% of cases of MCL, includ-
ing in some cases that are cyclin D1-negative. Although
expression of Sox11 is not specific to MCL, as it can be
found in a proportion of hairy cell leukemias, lympho-
blastic and Burkitt lymphomas, it may prove a useful
diagnostic tool. Cytoplasmic rather than nuclear
expression of SOX11 expression has been suggested to
be associated with a worse outcome in nodal MCL,
while lack of Sox 11 has been associated with leukemic
presentation of MCL associated with a more indolent
course. Proliferation marker expression is variable, but
a high proliferative index has negative prognostic impli-
cations. Staining for follicular dendritic cells usually
demonstrates an expanded and disrupted meshwork
most obvious in cases with more nodular growth, but
usually present even in those with a diffuse pattern. In
contrast to other B-cell lymphomas, MCL is more fre-
quently associated with lambda rather than kappa light
chain expression.
Detailed examination of reactive appearing lymph
nodes has revealed the infrequent appearance of
small numbers of cyclin D1-positive cells. These
cases are exceptionally rare and are characterized by
the presence of small numbers of cyclin D1-positive
cells in a proportion of the follicles of the affected
nodes with a characteristic localization in the inner
part of the mantle zone adjacent to the germinal
center. These cells contain the classical t(11;14)
which can be demonstrated by fluorescent in situ
hybridization (FISH) analysis. The significance of
the finding of minimal infiltration by mantle cell
lymphoma type cells, currently termed in-situ mantle
cell lymphoma, remains uncertain.
Molecular pathogenesis
The molecular pathogenesis of MCL is a paradigm of
dysregulated control of the cell cycle machinery and the
response to DNA damage in human cancer. Most, if not
all, molecular and cytogenetic alterations described to
date in MCL appear to affect these two crucial cellular
pathways. The hallmark chromosomal translocation,
the t(11;14)(q13;q32), can be detected in the vast major-
ity of MCL cases and results in the overexpression of
cyclin D1, which plays an important role in the regu-
lation of the G1-/S-phase transition of the cell cycle. The 157
 Chapter 10: Mantle cell lymphoma
juxtaposition of the cyclin D1 chromosomal locus on
11q13, designated BCL-1 (B-cell lymphoma/leukemia
1), to the enhancer of the IgH joining region at 14q32
leads to upregulation of this D-type cyclin that is usually
not expressed in B-cells. Cyclin D1 forms a complex
with the cyclin-dependent kinases-4 (CDK4) and -6
(CDK6), and overexpression of cyclin D1 in MCL
leads to increased levels of cyclin D1/CDK complexes
which in turn phosphorylate the retinoblastoma protein
(Rb) and thus directly facilitate cell cycle progression. In
addition, increased levels of cyclin D1/CDK complexes
may sequester the CDK inhibitors p27kip1 and p21 that
are usually bound to cyclin E/CDK2 complexes which,
once activated, promote S-phase entry. These studies
suggest that aberrant expression of cyclin D1 plays a
crucial role in the pathogenesis of MCL; moreover, the
level of cyclin D1 expression is directly correlated with
the proliferation rate of the tumor cells and therefore
with the clinical aggressiveness of this lymphoma.
Besides the deregulation of cyclin D1 as a result of
the translocation t(11;14), other genetic alterations
have been reported that disrupt proper function of
the cell cycle. Specifically, a significant proportion of
patients with MCL harbor hemizygous or homozy-
gous deletions in the chromosomal locus 9p21 (as
detected by conventional karyotyping or FISH),
affecting the CDK inhibitor p16INK4a. p16INK4a serves
as an inhibitor of CDK4 and CDK6 and thus helps to
maintain the Rb protein in its dephosphorylated,
antiproliferative state. Functional inactivation of
this CDK inhibitor may therefore cooperate with
aberrant cyclin D1 expression in MCL and enhance
cell cycle progression. It is noteworthy that MCL
cases with inactivated p16INK4a usually display
increased proliferative activity. Rare cases of MCL
with a wild-type (WT) status of the p16INK4a locus
have been described with overexpression and
genomic amplification of BMI-1 that belongs to the
Polycomb group of genes and acts as a transcriptional
repressor of the p16INK4a locus. BMI-1 overexpres-
sion in a small subset of MCL cases may therefore
represent a pathogenetic alternative to the more fre-
quently observed deletions of its transcriptional tar-
get p16INK4a. Finally, a few MCL cases with genomic
amplification and protein overexpression of CDK4
have been reported which may represent yet another
mechanism of disturbing the G1-/S-phase cell cycle
checkpoint.
As described above, a second major target of genetic
158 alterations in MCL is the DNA damage response
pathway. In particular, the chromosomal region
11q22–23 is frequently deleted in MCL, affecting the
ataxia-telangiectasia mutated (ATM) gene. In addition
to hemizygous ATM deletions, many MCL cases show
concomitant mutations of the ATM gene. ATM encodes
for a kinase that belongs to the PI-3 kinase-related
superfamily and plays a pivotal role in the cellular
response to DNA damage. ATM mutations frequently
affect the kinase domain of the gene or result in a
truncated version of the ATM protein. Besides frequent
ATM alterations, other molecules involved in the DNA
damage response and specifically targets of the ATM
kinase itself are also occasionally altered in MCL.
Specifically, rare alterations of the kinases CHK2 and
CHK1 are observed in MCL. CHK2 stabilizes and acti-
vates p53, which plays a central role in the response to
DNA damage by initiating cell cycle arrest, apoptosis,
or DNA repair. Decreased protein levels and occasional
CHK2 mutations present in a subset of MCL
may compromise p53 function and are associated
with a high number of chromosomal imbalances.
Importantly, p53 is inactivated in approximately 30%
of MCL cases with blastoid morphology and high pro-
liferation rates, while this event is uncommon in MCL
cases with classic morphology and low proliferative
activity.
Several studies suggest that the proliferation rate of
the tumor cells is associated with the clinical aggres-
siveness of the lymphoma. In particular, a low prolif-
eration rate, e.g. as measured immunohistochemically
by the Ki-67 index, corresponds to a more favorable
clinical course, while a high Ki-67 index is associated
with short survival times. A gene expression profiling
study in MCL provided a quantitative measurement of
tumor cell proliferation, termed proliferation signa-
ture, allowing for the definition of prognostic sub-
groups that differ in their median survival by more
than 5 years. Since the gene expression-based meas-
urement of proliferation was superior in predicting the
length of survival than individual molecular param-
eters alone or in combination (e.g. level of cyclin D1
expression, p16INK4a alterations), the proliferation sig-
nature may be viewed as a quantitative integrator of
oncogenic events in MCL, which could be helpful in
the future to test targeted therapeutic approaches and
to guide treatment decisions. More indolent forms of
MCL have attracted attention recently, and these cases
appear to harbor fewer genetic alterations compared to
conventional MCL and are characterized by mutated
IgVH genes in the majority of cases.

Staging
The majority of patients present with advanced stage,
symptomatic disease. Standard lymphoma staging
approaches apply in MCL, including a thorough his-
tory and physical examination, CT scans, and bone
marrow aspirate and biopsy. Flow cytometry and
cytogenetics with FISH for the t(11;14) should be
obtained at the time of marrow biopsy, or on periph-
eral blood if lymphocytosis is present. Careful atten-
tion to assessment of potential sites of extranodal
disease is warranted; colonoscopy and upper endos-
copy should be considered in the staging evaluation,
especially if there is evidence of GI blood loss or
abdominal symptoms. Upper and lower endoscopy
generally is not routinely required in the absence of
relevant signs and symptoms, although some centers
and clinical trials do employ this in both pre- and
post-treatment staging. Overt lymphocytosis is
present in about 25% of patients, although a leukemic
component in virtually all patients can be demon-
strated by sensitive flow cytometric and molecular
detection assays.
The role of positron emission tomography (PET)
scanning in the initial staging and for assessment of
treatment response is not generally recommended,
although MCL is typically PET-avid. PET can be espe-
cially useful for the identification of extranodal dis-
ease, because of the lower sensitivity of routine CT
scans for these sites. The use of this technique to docu-
ment post-therapeutic complete response is reason-
able, especially in the setting of known extranodal
disease and for clinical trials wherein complete remis-
sion is a goal of therapy. Post-treatment PET negativ-
ity and has been shown to correlate with improved
progression-free survival (PFS).
Prognostic factors
MCL displays considerable morphologic, biologic, and
clinical heterogeneity. Most patients pursue a steadily
progressive clinical course and often require therapy at
diagnosis or shortly thereafter. However, up to one-
quarter of patients have a slow pace of disease typical
of indolent B-cell lymphomas and may defer antilym-
phoma therapy for 6–12 months or more. The under-
pinnings of this variability are becoming better
understood at the molecular and cellular levels as dis-
cussed above, and may provide approaches to individ-
ual patient prognosis and risk-adapted treatment
(Table 10.1).
Chapter 10: Mantle cell lymphoma
Morphologic subtype. Both architectural and cyto-
logic variants of MCL have been reported to affect
prognosis. Most patients show a diffuse nodal efface-
ment, although nodular or especially mantle zone pat-
terns are also observed and appear to correlate with a
more indolent pace of disease. A blastoid cell type may
be present at diagnosis or may develop with disease
progression. Blastoid MCL in most series has a poor
survival, averaging 1–2 years as compared with the
non-blastoid MCL survival of at least 4–5 years.
Phenotypic and molecular markers
Immunohistochemical staining of paraffin-embedded
MCL samples for Ki-67 expression, a marker of cellu-
lar proliferation, correlates inversely with clinical out-
comes. Current limitations of this biomarker include
the lack of standardized staining techniques, the sub-
jectivity in interpretation of levels of positivity, and
identifying the relevant cut-offs for the Ki-67 index
(i.e. percentage of positive tumor cells) versus patient
outcomes. Nonetheless, significant differences in OS
were shown for MCL patients treated with CHOP or
R-CHOP stratified by fewer than 10%, 10–29%, and
30% or more Ki-67-positive cells (Figure 10.3), sug-
gesting that the Ki-67 index may be a useful surrogate
for the molecular profile proliferation index.
Molecular markers of “indolent MCL” have been
characterized in a cohort of MCL patients for whom
therapy was not required for months or years. These
include the presence of mutated immunoglobulin
heavy variable chain genes (IgVH), a lack of p53 muta-
tions, limited secondary genetic aberrations, negativity
for expression of the transcription factor Sox11 pro-
tein, and a unique 13-gene molecular expression array
signature. In contrast, “typical MCL” cases display a
high degree of secondary genetic and copy number
alterations that can be detected by standard karyotype,
comparative genomic hybridization, FISH, and
PCR methodologies. The presence of loss-of-function
mutations affecting the 17p13/p53 or 9p21/CDKN2A/
Hippo signaling loci, as well as 3q gain or deletion
13q14, has been associated with poorer survival in
retrospective MCL cohorts.
The role of microRNA aberrations in MCL patho-
genesis is beginning to emerge and to be correlated
with clinical outcomes. Examples include the loss
of expression of miR-29 family members and over-
expression of the miR-17–92 cluster, which have been
associated with a more aggressive clinical course
and poorer outcome. Of interest, truncation of the 159
 Chapter 10: Mantle cell lymphoma

A 1.0 Figure 10.3 Kaplan–Meier plot for overall survival of patients

treated with CHOP (A) and R-CHOP (B) stratified in three
0.9
groups according to the Ki-67 index of less than 10% (<10),
probability of overall survival

0.8 10% to less than 30% (≥10), and 30% or more (≥30) Ki-67-
0.7 positive cells.
0.6
0.5
0.4
0.3
< 10, median = 112
0.2 >= 10, median = 59
0.1 >= 30, median = 35
P = .0002
0.0
0 12 24 36 48 60 72 84 96 108 120
numbers of patients at risk months since registration
< 10 38 37 33 30 24 18 13 9 6 2
>= 10 59 55 49 42 32 21 14 9 2 0
>= 30 19 16 11 8 7 3 2 0

B 1.0
0.9
probability of overall survival

0.8
0.7
0.6
0.5
0.4
0.3 < 10, median not reached
>= 10, median not reached
0.2
>= 30, median = 52
0.1 P = .0126

0.0
0 12 24 36 48 60 72 84 96 108 120
numbers of patients at risk months since registration
< 10 38 27 20 11 7 0
>= 10 50 48 40 19 6 3 1 0
>= 30 16 15 11 7 1 0

30 untranslated region of cyclin D1 mRNA, itself a 10.1. The rapid advances in understanding MCL
marker of poorer prognosis in MCL, leads to loss of pathogenesis and prognostic markers may lead to
miR-16–1 binding sites which in turn impairs normal more individualized treatment approaches in the
cell cycle regulation. Taken together, elucidation of years to come.
the complex physiologic dysregulation in MCL via Clinical prognostic factors. The International
gene mutation and miRNA alterations represents an Prognostic Index (IPI) developed for diffuse large
exciting new avenue of investigation that promises B-cell lymphoma has limited predictive value in
improved pathogenic insights, prognostic stratifica- MCL, as the majority of patients present with
tion, and target identification. advanced-stage disease and frequent leukemic or
Activating mutations have recently been identified extranodal sites of involvement. Patients who present
in NOTCH1, a ligand-activated transcription factor with predominantly peripheral blood, bone marrow,
that is also mutated in some cases of chronic lympho- and splenic involvement without significant lymph-
cytic leukemia (CLL) and T-cell acute lymphoblastic adenopathy often experience an indolent clinical
leukemia. MCL patients with these mutations course; these cases need to be carefully distinguished
appeared to have a poorer OS. Additional biomarkers from CLL, splenic lymphoma with villous lympho-
160 correlated with prognosis are summarized in Table cytes (SLVL), and prolymphocytic leukemia.

The Mantle Cell International Prognostic Index
(MIPI) was developed as a risk assessment tool utiliz-
ing readily available clinical and laboratory para-
meters: patient age, ECOG (Eastern Cooperative
Oncology Group) performance status, total leukocyte
count, and serum lactate dehydrogenase. Based upon
the calculated score, patients with stage III and IV
disease can be stratified into low-, intermediate- and
high-risk groups that directly correlate with OS
following initial chemotherapy. The MIPI has been
validated in other patient cohorts treated with immu-
nochemotherapy and dose-intensive regimens,
including stem cell transplantation. A further refine-
ment of the MIPI incorporates the proliferation
marker Ki-67, described above, to provide a MIPI
biologic index (MIPIb) with improved predictive
power for OS – patients with low-risk scores had
median OS rates following induction chemotherapy
of more than 6 years, whereas high-risk patients had
median survivals of only 3 years.
Minimal residual disease. The presence of low-level
minimal residual disease (MRD) involving the blood
and bone marrow can be determined by sensitive PCR
assays for clonal IgVH and t(11;14) breakpoints using
patient-specific probes. The achievement of molecular
remission following induction therapy as reflected by
MRD-negativity has been associated with prolonged
clinical remission in prospective randomized clinical
trials. Ninety percent of patients had a detectable
molecular marker, and MRD was assayed serially dur-
ing and after therapy. The achievement of molecular
remission strongly and significantly correlated with
response duration, especially among those who
achieved marrow MRD negativity. Further standard-
ization and wider availability of MRD assays will per-
mit prospective assessment of molecular remission as
a therapeutic endpoint and the ability to test early
“pre-emptive” therapeutic intervention prior to overt
clinical relapse.
Initial therapy
In the small number of patients with limited stage
disease, extended-field (EF-RT) or involved-field
radiation (IF-RT) alone achieves only remissions of
limited duration whereas such a radiation consolida-
tion after initial systemic therapy potentially results
in a long-term benefit. In a retrospective study of
stage I–II MCL, 5-year PFS and OS was 68% and
71% after IF-RT, respectively, either alone or in
combination with conventional chemotherapy. In
Chapter 10: Mantle cell lymphoma
advanced-stage disease the benefit of local radiation
therapy is not proven.
Conventional chemotherapy
In advanced-stage disease, conventional immune-che-
motherapy remains a non-curative approach. Because
of the aggressive clinical course in the majority of
patients, a watch and wait strategy generally should
only be pursued in selected cases with low-risk profile
based on clinical manifestation (e.g. GI tract or leuke-
mic disease only), histological features, and/or biology
(Ki-67 <10%). However, those patients presenting a
history consistent with slow-paced and low-tumor
burden advanced-stage disease may be considered for
“watchful waiting”.
Monoclonal antibody and combined
immunochemotherapy
Despite high CD20 expression in MCL, rituximab
monotherapy achieves only moderate response rates
of 20–35%. In the largest dataset available thus far
from the Swiss SAKK study group, four weekly stand-
ard doses of rituximab achieved complete and overall
response rates (ORR) of 2% and 27%, respectively in
104 patients with newly diagnosed or relapsed MCL,
resulting in a median event-free survival of 6 months
only. Thus antibody monotherapy should be applied
only in very low-risk patients with strict contraindica-
tions for systemic chemotherapy.
In contrast, based on a proposed in vitro syner-
gism, a combined immunochemotherapy regimen of
rituximab and CHOP achieved high overall (96%) and
complete remission rates (48%) in a recent phase II
study. However, even patients achieving a molecular
remission displayed a median PFS of only 18.8
months, and there appeared to be no difference in
duration of response between those who did and did
not achieve molecular remission. A randomized trial
confirmed that the addition of rituximab resulted in
a superior ORR of 94% versus 75% with CHOP alone
(P = 0.0054); CR rates were also improved (34% versus
7%; P = 0.00024). After extended follow-up, PFS was
doubled from 14 months to 28 months, but constant
relapses were observed. Thus, additional consolidation
concepts are warranted to translate the high response
rates into long-term benefit for the patient.
In another trial by the German Lymphoma Study
Group, the addition of rituximab not only resulted 161
in significantly improved complete response rates
 Chapter 10: Mantle cell lymphoma

Table 10.2 Combined immunochemotherapy in MCL.

Author n Regimen Disease OR (CR) PFS OS

status
Howard et al. J Clin Oncol 2002; 40 R-CHOP First line 96% (48%) 17 months n.a.
20:1288
Forstpointner et al. Blood 2004; 55 R-FCM Relapse 62% (33%*) 8 months Median not
104:3064 reached
Herold et al. Blood 2004; 104:168a 44 R-MCP First line 71% (32%) 20 months Median not
reached
Hoster et al. Blood 2008; 112:1048 62 R-CHOP First line 92%*(32%) 28 months Median not
reached
Rummel et al. Blood 2009; 114:168 48 R-CHOP First line 95% (35%) 22 months Median not
45 BR 89% (32%) 33 months reached
Median not
reached
n, Number of patients; CR, complete remission; OR, overall response; PFS, progression-free survival; OS, overall survival; n.a.: not available.
R = rituximab, C = cyclophosphamide, H = doxorubicin, O = vincristine, P = prednisone, F = fludarabine, M = mitoxantrone, B = bendamustine.
* Significant improvement in comparison to chemotherapy alone.

(33% versus 0%) and slightly higher overall response
(62% versus 49%, n.s.) in relapsed or refractory MCL,
but, more importantly, an improved OS was observed
after combined immunochemotherapy with R-FCM
(fludarabine, cyclophosphamide, mitoxantrone) in
comparison to FCM chemotherapy alone (Table
10.2). Benefit in duration of response was also dem-
onstrated when this regimen was followed by main-
tenance rituximab therapy.
In a large phase III study of the European MCL
Network, two different regimens of immunochemo-
therapy were compared in 559 first line patients age 60
years and older who were not candidates for stem cell
transplant consolidation therapy. Interestingly, after
the R-FC regimen higher toxicity and significantly
lower 4-year OS rates were observed in comparison
to the standard R-CHOP regimen. A strong benefit for
maintenance rituximab over interferon maintenance
was observed for patients responding to R-CHOP, but
not to R-FC, supporting the role of maintenance rit-
uximab in this setting.
Bendamustine, a compound with characteristics
of both alkylators and purine analogs, was compared
to the R-CHOP regimen in a subset analysis of MCL
patients in another phase III trial. The bendamustine
monochemotherapy plus rituximab (BR) displayed
significantly less acute myelotoxicity, resulting in a
25% reduction of infectious episodes. Remarkably,
162
PFS was prolonged with BR (35 months) in comparison
to R-CHOP (22 months; P = 0.043). Thus, especially in
elderly patients not qualifying for subsequent dose
intensification, the BR regimen represents one of the
new standard approaches.
Radioimmunotherapy (RIT)
Another approach is the application of radio-
(131iodine or 90yttrium) labeled anti-CD20 antibod-
ies. Conventional dose radioimmunotherapy (RIT)
delivered moderate results, with 32% overall response
and rapid progression, resulting in a median PFS of 6
months in relapsed disease. Interestingly, RIT was
not able to overcome either chemotherapy-refrac-
toriness or bulky disease in this phase II study. In
contrast, in first line treatment, RIT consolidation
after systemic induction therapy with rituximab
plus CHOP (similar to radiation in early stages)
increased the CR/CRu rates from 20% to 55% using
standard (non-PET-confirmed) response criteria and
seemed to extend duration of remission, with a
median time to treatment failure of 34 months and
an estimated 5-year OS of 73%. Similarly, myeloabla-
tive regimens incorporating RIT followed by autolo-
gous stem cell transplantation achieved high ORRs of
100% (91% CR) and an estimated 3-year OS of 93%
even in relapsed disease.

Dose-intensified regimens
Various studies on the efficacy of high-dose cytara-
bine (Ara-C) reported promising results. More than
80% of patients obtained a complete remission fol-
lowing a sequential CHOP–DHAP regimen (dexame-
thasone, high-dose cytarabine, and cisplatin). Similar
encouraging results were achieved by several other
study groups. Thus, Lefrere et al. observed a CR rate
of only <10% in previously untreated patients with
MCL following CHOP therapy, which was converted
into an impressive 84% complete remissions after
four additional cycles of the high-dose cytarabine-
containing DHAP regimen.
An even more dose-intensified approach of the
M.D. Anderson Cancer Center group of rituximab
plus Hyper-CVAD (fractionated cyclophosphamide
plus vincristine, doxorubicin, and dexamethasone)
alternating with high-dose methotrexate/cytarabine
achieved again impressive response rates (ORR
97%, CR 87%) and, importantly, prolonged remis-
sions (3-year failure-free survival 64%) similar to
the sequential dose-intensified approach with mye-
loablative consolidation. However, toxicity was sig-
nificant: 29% of patients could not complete
scheduled treatment and there was a therapy-associ-
ated mortality of 8%, including predominantly infec-
tious complications and secondary myelodysplasia
and/or acute myelogenous leukemia in four of 97
patients. Due to toxicity, many MCL patients are
not candidates for this regimen because of age or
comorbid disease and, as such, modifications of the
regimen are under study.
Sequential dose intensification and
autologous transplantation
Encouraging results were obtained in various phase II
studies exploring the potential of consolidation by
myeloablative therapy followed by autologous stem
cell transplantation (ASCT). A randomized trial of
more than 200 patients with previously untreated
MCL compared the addition of myeloablative con-
solidation (12 Gray total body irradiation, cyclophos-
phamide 2 × 60 mg/kg body weight) followed by
ASCT after initial CHOP induction to conventional
chemotherapy and interferon maintenance only
(Table 10.3). This international prospective trial con-
firmed an impressive improvement of complete
remissions (81% versus 37%) and significantly longer
Chapter 10: Mantle cell lymphoma
PFS (median PFS: 39 versus 17 months in the IFNα
study arm; P = 0.01; Figure 10.4). Interestingly, the
benefit of myeloablative consolidation was confirmed
in several subgroup analyses, including patients with
low and intermediate IPI scores as well as those with
CR or PR after chemotherapy induction. After
extended follow-up, a significant improvement of
OS has been observed (P = 0.037), which was inde-
pendent of the addition of rituximab in a pooled
analysis of randomized trials. Thus, myeloablative
radiochemotherapy followed by ASCT represents
one of the standard therapeutic approaches in first
line treatment of younger MCL patients (age <65
years). However, even after such a dose-intensified
consolidation, the majority of patients with MCL
eventually relapse.
One major obstacle of ASCT is the contamination
of the harvested stem cells with circulating lymphoma
cells. Thus, purging procedures have been introduced
to eliminate residual lymphoma cells. Applying such
an in vivo purging with rituximab, various phase II
studies suggested an improved long-term outcome
after antibody-based “in vivo purging” and subsequent
myeloablative therapy. The most favorable results
have been reported by an Italian study group, with a
PFS of 76% after 5 years.
The Scandinavian group explored such an opti-
mized approach of high-dose cytarabine-containing
induction with rituximab followed by ASCT in a
large phase II trial (n = 158). In a historical compar-
ison, this combined approach was significantly
superior to the previous study generation of
CHOP-based induction and myeloablative consoli-
dation only. However, late relapses were observed
after prolonged follow-up and, especially in MIPI
high-risk patients, the outcome still has to be
improved.
The European MCL Network confirmed the ben-
efit of the addition of cytarabine in a large interna-
tional trial. After inclusion of 497 patients,
significantly improved time-to-treatment failure
rates had been observed (88 versus 46 months, after
53 months median follow-up; P = 0.038), thereby
establishing a new standard of care in younger
MCL patients. Despite these encouraging results it
has to be emphasized that the majority of patients in
this and other transplantation trials display low or
intermediate MIPI risk score only, related in part to
the younger age of transplant-eligible patients. A US
phase III Intergroup trial is currently testing the 163
 Chapter 10: Mantle cell lymphoma

Table 10.3 Dose-intensified regimens in the treatment of MCL.

Author Study n Induction Consolidation Response Median Median OS

rate (OR/ PFS/EFS
CR)
Dreyling, Phase III 75 Conventional Intensive (ASCT) 78% (42%) 43 months 90 months
2008b (CHOP/MCP)
Dreger, 2007 Phase II 34 Conventional Intensive (ASCT) 88% (24%) 83% (4 years) 87% (4 years)
(CHOP/ ->R)
LeFrere, Phase II 28 Conventional Intensive (ASCT) 89% (82%) 51 months 81 months
2004 (CHOP/ DHAP)
de Guibert, Phase II 24 Conventional Intensive (ASCT) 96% (92%) 65% (3 years) 69% (3 years)
2006 (R-DHAP)
Delarue, Phase II 60 Conventional Intensive (ASCT) 95% (96%) 83 months 75% (5 years)
2009 (R-CHOP/ R-DHAP)
Dreyling, Phase III 390 Conventional Intensive (ASCT) 91% (51%) 84% (2 years) 77% (2 years)
2008 (R-CHOP)
Romaguera, Phase II 97 Intensive (R-Hyper- – 97% (CR/ 54 months 82% (3 years)
2005 CVAD/MA) CRu: 87%)
Epner, 2007 Phase II 97 Intensive (R-Hyper- – 88% (CR/ 64% (2 years) 74% (3 years)
CVAD/MA) CRu: 58 %)
Magni, 2009 Phase II 28 R-high dose cyclo, Intensive (ASCT) 96% (all CR) 48% in low 76% in low
ara-C, melphalan, risk, 34% in risk, 68% in
mitoxantrone high risk high risk
Tam, 2009* Phase II 42 Intensive (R-Hyper- Intensive (ASCT) 96% (all CR/ 42 months 93 months
CVAD/MA) Cru)
7 Conventional
(R-CHOP)
Ritchie, 2007 Phase II 13 Intensive (R-Hyper- Intensive (ASCT) 100% (92%) 92% (3 years) 92% (3 years)
CVAD/MA)
Till, 2008 Phase II 21 Intensive (R-Hyper- Intensive (ASCT) 100% (CR/ 81% (3 years) 94% (3 years)
CVAD/MA) CRu: 81%)
Vose, 2006 Phase II 32 Intensive (R-Hyper- Intensive (ASCT) 100% (CR/ 78% (3 years) 97% (3 years)
CVAD/MA) CRu: 81%)
Geissler, Phase II 159 Intensive Intensive (ASCT) 96% (55%) 63% (4 years) 81% (4 years)
2008 (R-CHOP-HA)
CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisone; FCM, fludarabine, cyclophosphamide, mitoxantrone; HyperCVAD/MA
(fractionated cyclophosphamide, doxorubicin, vincristine, dexamethasone; alternated with high-dose methotrexate and cytarabine; PR,
partial response rate; CR, complete response rate; PFS, progression-free survival; TTF, time to treatment failure; FFS, failure-free survival; EFS,
event-free survival; OS, overall survival; ASCT, autologous stem cell transplantation. *Only relapsed disease.

R-HyperCVAD/methotrexate–cytarabine regimen remission duration even after effective chemotherapy

versus bendamustine plus rituximab, with respond-
ers undergoing consolidative ASCT.
Rituximab maintenance therapy
Especially after conventionally dosed chemotherapy, a
continuous relapse pattern has been observed in most
series. Thus, there remains an urgent need for effective
164 additional consolidation strategies to prolong
induction.
In a Swiss SAKK trial of rituximab monotherapy
for newly diagnosed and relapsed or refractory MCL,
no benefit of maintenance with four single doses of
rituximab given every 8 weeks was observed following
standard weekly ×4 rituximab induction. Recently, the
results of an international phase III trial have con-
firmed the efficacy of regular postinduction antibody
application in MCL as summarized above. Duration of

1.0
0.9
0.8
survival probability
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 1 2 3
numbers of patients at risk years after end of induction therapy
ASCT 62 38 31 17 10
IFN 60 33 19 9 6
Figure 10.4 Progression-free survival after high-dose
radiochemotherapy followed by autologous stem cell transplantation
(ASCT) or interferon-α maintenance (IFN) in MCL (from Dreyling et al.
Blood, 2005;105:2677).
remission was doubled, with 58% of patients receiving
rituximab maintenance in remission at 4 years
(P = 0.01) in comparison to 29% of those on an
interferon-based maintenance. Additional phase III
studies are ongoing, including the US Intergroup
trial of rituximab versus rituximab plus lenalidomide
maintenance following induction therapy with bend-
amustine/rituximab (BR) or BR plus bortezomib.
Allogeneic stem cell transplantation
Allogeneic bone marrow or stem cell transplantation is
still the only curative approach in patients with
advanced stage MCL. A graft-versus-lymphoma effect
has been suggested to induce long-lasting complete
remissions even in patients with relapsed or refractory
MCL. However, transplantation-related mortality is
relatively high and especially graft-versus-host disease
and infectious complications are common.
Two phase II studies applying a dose-reduced
conditioning reported more encouraging survival
rates in less intensively pre-treated patients. Thus,
with a minimal conditioning regimen (fludarabine
and 2 Gray TBI), disease-free survival and OS in
33 patients with relapsed and refractory MCL was
60% and 65%, respectively, with non-relapse mortal-
ity of 24% at 2 years. However, in a retrospective
survey at the M.D. Anderson Cancer Center, alloge-
neic transplant was superior to autologous trans-
plantation only in relapsed MCL. Thus, despite
Chapter 10: Mantle cell lymphoma
promising results, allogeneic transplantation should
ASCT
IFN be applied only in relapsed disease or selected high-
p = 0.0108 risk patients not appropriately responding to dose-
intensified first line therapy.
Management of relapsed disease
The management of relapsed disease is highly depend-
ent on the initially applied treatment strategies. As a
result, published data can hardly be compared, owing
4 5 6
to different patient risk profiles. Limited data suggest
that the addition of rituximab is reasonable if a remis-
3 sion of at least 6–9 months has been achieved after a
2 rituximab-containing regimen. In younger patients
relapsing after dose-intensified regimens, an initial
reduction of tumor load via chemotherapy or chemo-
immunotherapy and subsequent allogeneic transplan-
tation should always be considered as a potentially
curative option (Figure 10.5).
In elderly patients (>65 years), a non-cross-resist-
ant regimen is recommended which has to be selected
considering patient comorbidity and prior lines of
therapy. In the USA, the proteasome inhibitor borte-
zomib plus dexamethasone, with or without rituxi-
mab, has been registered as second line therapy in
patients relapsing after immunochemotherapy.
Alternatively, patients ineligible or intolerant of bor-
tezomib (e.g. because of peripheral neuropathy) and
those progressing after a response should be consid-
ered for bendamustine alone or in combination with
rituximab. However, based on the expected clinical
course, with the majority of patients relapsing within
1–3 years, the consideration of maintenance with rit-
uximab or the use of RIT consolidation is encouraged.
In medically fit patients, dose intensification, if not
applied in first line, may be re-discussed. Given the
lack of standards of care for treatment of newly diag-
nosed and relapsed disease, clinical trial participation
is always recommended whenever possible.
Bendamustine. Bendamustine is a novel “hybrid”
cytotoxic agent composed of a benzimidazole ring
with an attached nitrogen mustard moiety, which
acts primarily as a bifunctional alkylating agent but
with activity distinct from other alkylators. Cell death
is induced via apoptosis pathways and apoptotic-
independent “mitotic catastrophe.” It has single
agent activity in a variety of hematologic neoplasms,
including NHL, multiple myeloma, and CLL. In a
German study of bendamustine plus rituximab (BR)
165
 Chapter 10: Mantle cell lymphoma

young patient (<65) elderly patient (>65) compromised patient

dose-intensified
immunochemotherapy
(either sequential:
R-CHOP/R-DHAP =>PBSCT
or R-Hyper-CVAD)
First-line treatment
conventional
immunochemotherapy watch & wait ?
(e.g. R-CHOP, BR)
Rituximab maintenance well-tolerated therapy:
radioimmunotherapy ? e.g. R-Chlorambucil, BR

high tumor load:
immunochemotherapy
(e.g. BR, R-DHAP)
allo-transplant ?
radioimmunotherapy ?
Rituximab maintenance
1. relapse
immunochemotherapy
(e.g. BR, R-BAC, R-FC) watch & wait ?
molecular approaches ?
well-tolerated therapy:
e.g. R-Chlorambucil, BR
autologous SCT ?
radioimmunotherapy ? molecular approaches
Rituximab maintenance

higher relapse
molecular approaches:
Bortezomib, Temsirolimus, Thalidomide/Lenalidomide (preferable in combination);
B-cell receptor inhibitors (in studies)
repeat previous therapy (long remissions)

Figure 10.5 Therapeutic approaches for MCL patients (advanced stage). Note: Since no standard therapy has been established for treatment of
newly diagnosed or relapsed disease, treatment on clinical trial should be considered for all patients. (See Table 10.2 for abbreviations.)

in 16 patients with relapsed MCL, including seven
refractory to prior therapy, an ORR of 75% was
observed, including CR in 50%. The median PFS was
18 months. Grade 3 leukopenia was observed in 16% of
all NHL patients treated in this study, which also
included indolent lymphoma. Non-hematologic tox-
icity was mild. A US phase II trial of BR in 12 relapsed
patients with MCL showed objective responses in 11
patients, including CR in over half. The median dura-
tion of response was 19 months. Preliminary results of a
phase III trial of BR versus fludarabine plus rituximab
in patients with relapsed indolent and MCL showed
significant benefit in ORR and CR for BR, with similar
toxicity profiles.
Bortezomib. Bortezomib targets the ubiquitin–pro-
teasome pathway and appears to provide therapeutic
efficacy via effects on multiple cellular mechanisms in
lymphoid neoplasms. The agent has been approved for
the treatment of multiple myeloma as well as relapsed
MCL. Two phase II studies, taken together, showed a
44% ORR with 18% complete responses. Relatively
durable responses were observed among responding
patients, with a median time to next treatment of
about 14 months. Bortezomib activity in MCL is
being further tested in ongoing multicenter studies
166
as a single agent or in combination with rituximab,
and in combination with immunochemotherapy,
including BR, R-CHOP, high-dose cytarabine, and
the modified R-HyperCVAD regimens. Bortezomib
can be administered safely to patients with severe
renal insufficiency, but is associated with peripheral
neuropathy in many patients; this may become dose-
limiting and necessitates dose-modification and vigi-
lant monitoring when given in combination with vin-
cristine-containing regimens. Reactivation of herpes
zoster is also frequently observed in patients treated
with bortezomib, prompting consideration of antiviral
prophylaxis during the course of therapy.
Radioimmunotherapy. RIT delivers a targeted
radiotherapeutic, 90yttrium or 131iodine, via an anti-
CD20 murine monoclonal antibody. Several applica-
tions of these agents are being investigated in MCL, as
described above, including consolidation following
induction immunochemotherapy and as a component
of the conditioning regimen prior to stem cell
transplantation. In a preliminary report of a phase II
single agent study in relapsed and refractory MCL,
responses were observed in eight of 23 patients
(35%), with five patients achieving CR or CR uncon-
firmed. The median response duration was 9.5
months, with delayed myelosuppression the most fre-
quent toxicity.

Novel therapeutic approaches
MCL responds well in most cases to initial therapy.
Improved overall and complete response rates have
been achieved with a number of chemoimmunotherapy
approaches as compared with previous combination
chemotherapy regimens. However, as discussed above,
most patients relapse within 1–5 years even after induc-
tion therapy and/or ASCT consolidation. Second line
regimens may show high therapeutic activity, although
the durability of these responses is often short-lived.
There is a pressing need for better agents for relapsed
and refractory patients, and an increasing number of
novel agents show clinical activity in this setting. These
may be targeted to the dysregulated cell cycle elements
characteristic of this disease, or to other growth and
proliferation or apoptosis pathways (Table 10.4).
Many of these are already being incorporated into front-
line regimens in an effort to improve the complete
response and PFS in MCL, either as a component of
combination therapy or as maintenance or consolida-
tion strategies. The following is a brief summary of
selected agents now undergoing clinical testing.
Immunomodulatory drugs (IMiDs). Lenalidomide
is highly active in multiple myeloma and CLL, with
multiple mechanisms of action, including direct anti-
proliferative effects, downregulation of tumor cell/
stromal cell interactions with disruption of essential
cytokine loops, immunomodulatory and antiangio-
genic effects. Thalidomide, an earlier immunomodu-
latory agent with activity in MCL, led to testing of
lenalidomide in relapsed or refractory patients.
Among heavily pre-treated MCL patients, partial
and complete responses were observed with single
agent oral lenalidomide. These encouraging findings
led to an international phase II trial in which 42% of 57
relapsed MCL patients responded, with a median PFS
of 5.7 months. Toxicity was predominantly reversible
myelosuppression. Responses have been observed in
patients relapsing after stem cell transplantation,
including CRs. A study of 134 patients progressing
after prior R-chemotherapy and bortezomib showed
a 28% ORR with 8% CRs; median response duration
was 16.6 months. Preclinical data demonstrating syn-
ergy with rituximab has led to current testing of so-
called “R2” regimens in CLL and indolent lymphoma,
and has shown responses in over half of patients with
relapsed or refractory MCL. As noted above, lenalido-
mide or the “R2”combination may also find utility as
maintenance in MCL. Thalidomide also has single
agent activity and has been used in a combination
Chapter 10: Mantle cell lymphoma
Table 10.4 Novel agents for treatment of relapsed or
refractory mantle cell lymphoma.
Agent Proposed targets
Lenalidomide Angiogenesis, microenvironment,
and cytokines
Temsirolimus, mTOR (mammalian target of
everolimus rapamycin)
Idelalisib (GS-1101; PI3Kδ (phosphatidylinositol-3-kinase
CAL-101) delta)
Ibrutinib (PCI-32765) BTK (Bruton tyrosine kinase)
Vorinostat HDAC (histone deacetylase)
Flavopiridol Cyclin D1/CDK4 (cyclin-dependent
kinase)
PD 0332991 CDK4/CDK6
Obatoclax, BH3 mimetic, apoptosis (BH = BCL-2
navitoclax homology)
regimen with rituximab plus low-dose oral predni-
sone, etoposide, procarbazine, and cyclophosphamide
(RT-PEPC) in heavily pre-treated MCL patients. An
ORR of 73% and CR rate of 32% were observed, with
antiangiogenic and microenvironmental effects postu-
lated to enhance therapeutic efficacy.
mTOR inhibitors. The mammalian target of rapamy-
cin (mTOR) is a downstream signaling molecule in the
phosphatidylinositol-3 kinase (PI3K)/Akt pathway that
serves a critical role in regulating mRNA translation,
including that of cyclin D1. Temsirolimus, a derivative
of rapamycin, has been tested in phase II single agent
trials in relapsed MCL, based upon the potential ability
to interrupt cyclin D1-dependent pathways. In a phase
III comparison of two doses and schedules of temsiroli-
mus versus investigators’ choice of therapy among 162
relapsed or refractory MCL patients, a schedule of
175 mg weekly for 3 weeks followed by 75 mg weekly
was found to be superior for ORR and PFS.
Temsirolimus is being investigated in frontline combi-
nation regimens and with rituximab and, like the related
mTOR inhibitor everolimus, as consolidation or main-
tenance therapy for MCL high-risk aggressive large cell
lymphomas. Everolimus has also shown single-agent
efficacy in relapsed or refractory MCL, including
patients who are refractory to bortezomib.
PI3K/AKT pathway. PI3K and AKT are upstream
of mTOR and are frequently activated in MCL; indeed,
the delta form of PI3K is expressed in over 90% of B-
cell lymphomas, providing a logical therapeutic target
for small molecule inhibitors. Idelalisib (GS-1101,
167
CAL-101) is an oral PI3K inhibitor with phase I
 Chapter 10: Mantle cell lymphoma
activity in relapsed or refractory CLL and NHL,
including MCL, and is now being studied in a number
of combination and single agent trials.
B-cell receptor (BCR) pathway inhibitors. Normal
humoral immune response to antigen occurs via the
BCR, a signaling pathway constitutively activated in
many B-cell lymphomas that includes PI3K delta.
Signal transduction via this pathway has been thera-
peutically targeted at several additional points in phase
I–II studies with oral small molecule inhibitors in
mantle cell and other lymphomas. These targets and
agents include spleen tyrosine kinase (Syk) with fosta-
matinib, Bruton tyrosine kinase (BTK) with ibrutinib
(PCI-32765), and protein kinase C-beta (PKC-β) with
enzastaurin.
High activity has been identified in relapsed MCL
using ibrutinib.
HDAC inhibitors. Histone acetylation and deacety-
lation (HDAC) are important transcriptional regula-
tory mechanisms that are altered in most cancers,
including lymphoma. Cyclin D1 protein levels can be
downregulated in MCL cells in vitro by treatment with
vorinostat; it also appears to inhibit the PI3K/AKT
pathway. Preliminary studies of vorinostat (already
approved in the USA for treatment of T-cell lym-
phoma) have shown clinical responses in MCL, with
further trials of this and other HDAC inhibitors in
progress. Novel combinations with cytotoxic agents
and bortezomib are also being pursued.
Cell cycle inhibitors. Cell cycle dysregulation is
present in all MCL – cyclin D1 overexpression in
most patients, cyclin D2 or D3 in the remainder –
leading to considerable interest in therapies aimed at
these pathways. Flavopiridol is a synthetic flavone with
mechanisms of action that include downregulation of
cyclin D1 and cyclin D3 as well as competitive inhib-
ition of the cyclin-dependent kinases CDK4 and
CDK6. A Canadian study found only a modest 11%
response rate in MCL for single agent flavopiridol
using a short-infusion protocol, although subsequent
pharmacokinetically driven schedules have shown
much higher response rates in refractory CLL that
may inform better approaches for MCL (Table 10.4).
Directly targeting CDK4/CDK6, an approach which
theoretically would circumvent the upregulation of D2
or D3 that may follow D1 inhibition, is being explored
with PD 0332991. This oral agent has shown activity in
relapsed MCL, and is also being tested in combination
with bortezomib.
168
BCL-2 inhibitors/BH3 mimetics. The regulation of
cell death pathways is highly complex and consists of
both prosurvival and proapoptotic proteins, the latter
characterized by the presence of a BH3 domain. As
prosurvival proteins such as BCL-2 are strongly
expressed in MCL, a current therapeutic strategy is to
utilize agents that mimic the BH3-only proteins which
will in turn promote apoptosis. Several BH3 mimetics
are in clinical trial, including obatoclax (GX 15–070)
and navitoclax (ABT-263).
Future directions
MCL presents rather unique challenges among the
NHLs, displaying an array of clinical behavior ranging
from indolent to aggressive, and typically high
response rates to induction immunochemotherapy
with historically brief durations of response and poor
long-term survival. This situation is steadily improv-
ing, however, with increasing rates of survival being
realized with the advent of improved treatment strat-
ification and new therapeutic agents. Current strat-
egies for management outside clinical trials are
outlined in Figure 10.5, but treatment approaches
continue to evolve rapidly. The use of the MIPI
score, Ki-67 index, and gene expression profiling are
being tested as means to risk-adapt therapy, with fur-
ther refinement possible via monitoring of MRD.
Improved outcomes have been realized for younger
patients via the incorporation of rituximab and high-
dose cytarabine followed by dose-intensive therapy
with ASCT consolidation frontline. Improved survival
is also being achieved via sequential application of
recently available treatments as well as novel agents
for newly diagnosed patients not eligible for intensive
frontline therapy and for those experiencing relapse.
Although challenging, MCL continues to provide
important insights to cancer pathogenesis, including
cell cycle dysregulation, apoptosis, and cell signa-
ling pathways that ultimately may prove useful in a
variety of other hematologic and non-hematologic
malignancies.
Further reading
Andersen NS, Pedersen LB, Laurell A, et al. Pre-emptive
treatment with rituximab of molecular relapse after
autologous stem cell transplantation in mantle cell
lymphoma. J Clin Oncol, 2009;27:4365–4370.

Ansel SM, Tang H, Kurtin PJ, et al. Temsirolimus and
rituximab in patients with relapsed or refractory mantle
cell lymphoma: a phase 2 study. Lancet Oncol,
2011;12:361–368.
Determan O, Hoster E, Ott G, et al. Ki-67 predicts outcome
in advanced-stage mantle cell lymphoma patients treated
with anti-CD20 immunotherapy: results from
randomized trials of the European MCL Network and the
German Low Grade Lymphoma Study group. Blood,
2008;111:2385–2387.
Dreyling M, Lenz G, Hoster E, et al. Early consolidation by
myeloablative radiochemotherapy followed by
autologous stem cell transplantation in first remission
significantly prolongs progression-free survival in mantle
cell lymphoma – results of a prospective randomized trial
of the European MCL Network. Blood,
2005;105:2677–2684.
Ek S, Dictor M, Jerkeman M, et al. Nuclear expression of the
non B-cell lineage Sox11 transcription factor identifies
mantle cell lymphoma. Blood, 2008;111:800–805.
Fernandez V, Salamero O, Espinet B, et al. Genomic and
gene expression profiling defines indolent forms of
mantle cell lymphoma. Cancer Res, 2010;70:1408–1418.
Forstpointner R, Dreyling M, Repp R, et al. The addition of
rituximab to a combination of fludarabine,
cyclophosphamide, mitoxantrone (FCM) significantly
increases the response rate and prolongs survival as
compared to FCM alone in patients with relapsed and
refractory follicular and mantle cell lymphomas – results
of a prospective randomized study of the German low
grade lymphoma study group (GLSG). Blood,
2004;104:3064–3071.
Forstpointner R, Unterhalt M, Dreyling M, et al.
Maintenance therapy with rituximab leads to a
significant prolongation of response duration after
salvage therapy with a combination of rituximab,
fludarabine, cyclophosphamide and mitoxantrone
(R-FCM) in patients with relapsed and refractory
follicular and mantle cell lymphomas – results of a
prospective randomized study of the German Low Grade
Lymphoma Study Group (GLSG). Blood,
2006;108:4003–4008.
Geisler CH, Kolstad A, Laurell A, et al. The Mantle Cell
Lymphoma International Prognostic Index (MIPI)
is superior to the International Prognostic Index (IPI)
in predicting survival following intensive first-line
immunochemotherapy and autologous stem
cell transplantation (ASCT). Blood, 2010;115:
1530–1533.
Ghielmini M, Schmitz Hsu S-F, Cogliatti S, et al. Effect of
single-agent rituximab given at the standard schedule or
as prolonged treatment in patients with mantle cell
lymphoma: a study of the Swiss Group for Clinical
Cancer Research (SAKK). J Clin Oncol, 2005;23:705–711.
Chapter 10: Mantle cell lymphoma
Goy A, Bernstein S, Kahl B, et al. Bortezomib in patients with
relapsed or refractory mantle cell lymphoma: updated
time-to-event analyses of the multicenter phase 2
PINNACLE study. Annals Oncol, 2008;20:520–525.
Goy A, Sinha R, Williams ME, et al. Phase II multicenter study
of single-agent lenalidomide in subjects with mantle cell
lymphoma who relapsed or progressed after or were
refractory to bortezomib: the MCL-001 Study (EMERGE™
Trial). Blood, 2012;120(Suppl.):abstract #905.
Hermine O, Hoster E, Szymczyk M, et al. Alternating courses
of 3x chop and 3x dhap plus rituximab followed by a high
dose ara-c containing myeloablative regimen and
autologous stem cell transplantation (ASCT) is superior
to 6 courses CHOP plus rituximab followed by
myeloablative radiochemotherapy and ASCT in mantle
cell lymphoma: final results of the MCL Younger Trial of
the European Mantle Cell Lymphoma Network. Blood,
2012;120(Suppl.): abstract #151.
Hess G, Herbrecht R, Romaguera J, et al. Phase III study to
evaluate temsirolimus compared with investigator’s
choice therapy for the treatment of relapsed or refractory
mantle cell lymphoma. J Clin Oncol, 2009;27:3822–3829.
Hoster E, Dreyling M, Klapper W, et al. A new prognostic
index (MIPI) for patients with advanced-stage mantle cell
lymphoma. Blood, 2008;111:558–565.
Howard OM, Gribben JG, Neuberg DS, et al. Rituximab and
CHOP induction therapy for newly diagnosed mantle-
cell lymphoma: Molecular complete responses are not
predictive of progression-free survival. J Clin Oncol,
2002;20:1288–1294.
Iqbal J, Shen Y, Liu Y, et al. Genome-wide miRNA profiling
of mantle cell lymphoma reveals a distinct subgroup with
poor prognosis. Blood, 2012;119:4939–4948.
Kirschbaum M, Frankel P, Popplewell L, et al. Phase II study
of vorinostat for treatment of relapsed or refractory
indolent non-Hodgkin’s lymphoma and mantle cell
lymphoma. J Clin Oncol, 2011;28:1198–1203.
Kluin-Nelemans HC, Hoster E, Hermine O, et al. Treatment
of older patients with mantle-cell lymphoma. N Engl J
Med, 2012;367:520–531.
Kridel R, Meissner B, Rogic S, et al. Whole transcriptome
sequencing reveals recurrent NOTCH1 mutations in
mantle cell lymphoma. Blood, 2012;119:1963–1971.
Lefrere F, Delmer A, Levy V, et al. Sequential
chemotherapy regimens followed by high-dose therapy
with stem cell transplantation in mantle cell lymphoma:
an update of a prospective study. Haematologica,
2004;89:1275–1276.
Lenz G, Dreyling M, Hoster E, et al. Immunochemotherapy
with rituximab and CHOP significantly improves
response and time to treatment failure, but not long-term
outcome in patients with previously untreated mantle cell
lymphoma: results of a prospective randomized trial of 169
 Chapter 10: Mantle cell lymphoma
the German Low Grade Lymphoma Study Group
(GLSG). J Clin Oncol, 2005;23:1984–1992.
Leonard JP, LaCasce AS, Smith MR, et al. Selective CDK4/6
inhibition with tumor responses by PD0332991 in
patients with mantle cell lymphoma. Blood,
2012;119:4597–4607.
Magni M, Di Nicola M, Carlo-Stella C et al. High-dose
sequential chemotherapy and in vivo rituximab-purged
stem cell autografting in mantle cell lymphoma: a 10-year
update of the R-HDS regimen. BMT, 2009;43:509–511.
Martin P, Chadburn, A, Christos P, et al. Outcome of
deferred initial therapy in mantle cell lymphoma. J Clin
Oncol, 2009;27:1209–1213.
Mato AR, Svoboda J, Feldman T, et al. Post-treatment (not
interim) positron emission tomography–computed
tomography scan status is highly predictive of outcome in
mantle cell lymphoma patients treated with R-
HyperCVAD. Cancer, 2012;118:3565–3570.
Ondrejka SL, Lai R, Smith SD, His ED. Indolent mantle
cell leukaemia: a clinicopathological variant
characterized by isolated lymphocytosis, interstitial
bone marrow involvement, kappa light chain
restriction and good prognosis. Haematologica,
2011;96:1121–1127.
Perez-Galan P, Dreyling M, Wiestner A. Mantle cell
lymphoma: biology, pathogenesis, and the molecular basis
of treatment in the genomic era. Blood, 2011;117:26–38.
Pott C, Hoster E, Beldjord K, et al. R-CHOP/R-DHAP
compared to R-CHOP induction followed by high dose
therapy with autologous stem cell transplantation induces
higher rates of molecular remission in MCL: results of the
MCL Younger Intergroup Trial of the European MCL
Network. Blood, 2010;116(21):425.
Pott C, Hoster E, Delfau-Larue M-H, et al. Molecular
remission is an independent predictor of clinical outcome
in patients with mantle cell lymphoma after combined
immunochemotherapy: a European MCL intergroup
study. Blood, 2010;115:3215–3223.
Racke F, Simpson S, Christian B, et al. Evidence of long
latency periods prior to development of mantle cell
lymphoma. Blood, 116(21):147.
Renner C, Zinzani P, Gressin R, et al. A multicenter
phase II trial (SAKK 36/06) of single-agent
everolimus (RAD001) in patients with relapsed or
refractory mantle cell lymphoma. Haematologica,
2012;97:1085–1091.
Robinson KS, Williams ME, van der Jagt RH, et al. Phase II
multicenter study of bendamustine plus rituximab in
patients with relapsed indolent B-cell and mantle cell
non-Hodgkin’s lymphoma. J Clin Oncol,
2008;26:4473–4479.
Romaguera JE, Fayad LE, Feng L, et al. Ten-year follow-
170 up after intense chemoimmunotherapy with
rituximab-Hyper-CVAD alternating with rituximab-
high dose methotrexate/cytarabine (R-MA) and
without stem cell transplantation in patients with
untreated aggressive mantle cell lymphoma. Br J
Haematol, 2010;150:200–208.
Ruan J, Martin P, Coleman M, et al. Durable responses with
the metronomic rituximab and thalidomide plus
prednisone, etoposide, procarbazine, and
cyclophosphamide regimen in elderly patients with
recurrent mantle cell lymphoma. Cancer,
2010;116:2655–2664.
Rummel MJ, Niederle N, Maschmeyer G, et al.
Bendamustine plus rituximab is superior in respect of
progression free survival and CR rate when compared
to CHOP plus rituximab as first-line treatment of
patients with advanced follicular, indolent, and
mantle cell lymphomas: final results of a
randomized phase III study of the StiL. Blood,
2009;114(22):168.
Rummel MJ, Kaiser U, Balser C, et al. Bendamustine
plus rituximab versus fludarabine plus rituximab in
patients with relapsed follicular, indolent and mantle
cell lymphomas–final results of the randomized phase
III study NHL 2–2003 on behalf of the StiL group.
Blood, 2010;116(21):373.
Rummel MJ, Niederle N, Maschmeyer G, et al.
Bendamustine plus rituximab versus chop plus
rituximab as first-line treatment in patients with
indolent and mantle cell lymphomas: updated results
from the stil nhl1 study on behalf of the StiL (Study
Group indolent Lymphomas, Germany). Proc ASCO
2012; abstract #3.
Schaffel R, Hedvat CV, Teruya-Feldstein J, et al. Prognostic
impact of proliferative index determined by quantitative
image analysis and the International Prognostic Index in
patients with mantle cell lymphoma. Annals Oncol,
2010;21:133–139.
Smith MR, Li H, Gordon L, et al. Phase II study of
rituximab plus CHOP immunochemotherapy
followed by yttrium-90-ibritumomab tiuxetan in
untreated mantle-cell lymphoma: ECOG Study E1499.
J Clin Oncol, 2012;30:3119–3126.
Swerdlow SH, Williams ME. From centrocytic to
mantle cell lymphoma: a clinicopathologic and
molecular review of 3 decades. Human Pathol,
2002;33:7–20.
Swerdlow SH, Campo E, Harris NL, et al. WHO
Classification of Tumours of Haematopoietic and
Lymphoid Tissues. 4th edn. WHO Press, 2008.
Tam CS, Bassett R, Ledesma C, et al. Mature results of the M.
D. Anderson Cancer Center risk-adapted transplantation
strategy in mantle cell lymphoma. Blood,
2009;113:4144–4152.

Wang M, Fayad L, Wagner-Bartak N, et al. Lenalidomide
in combination with rituximab for patients
with relapsed or refractory mantle-cell
lymphoma: a phase 1/2 clinical trial. Lancet Oncol,
Zhao J-J, Lin J, Lwin T, et al. MicroRNA expression
2012;13:716–723.
Wang M, Rule SA, Martin P, et al. Interim results of an
international, multicenter, phase 2 study of Bruton’s
tyrosine kinase inhibitor, ibrutinib (PCI-32765), in
Chapter 10: Mantle cell lymphoma
relapsed or refractory mantle cell lymphoma: durable
efficacy and tolerability with longer follow-up. Blood,
2012;120(Suppl.):abstract #904.
profile and identification of miR-29 as a prognostic
marker and pathogenetic factor by targeting
CDK6 in mantle cell lymphoma. Blood,
2010;115:2630–2639.
171
 Chapter
11
Burkitt and lymphoblastic lymphoma:
clinical therapy and outcome
Nirali N. Shah, Alan S. Wayne, and Wyndham H. Wilson
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Introduction and presentation
Burkitt (BL) and lymphoblastic lymphomas (LBL) are
highly aggressive diseases with distinct natural histories
and clinical presentations. BL mostly occurs in the first
two decades of life and accounts for 1–2% of all lympho-
mas. Three clinical variants are recognized: (1) endemic
BL, which is primarily found in equatorial Africa;
(2) sporadic BL, which presents worldwide but is
the most common type in western countries; and
(3) immunodeficiency-associated BL, which is associated
with HIV infection. There are important clinical differ-
ences in these variants (Table 11.1), with endemic BL
involving the jaw, orbit, and paraspinal regions in half of
the cases as well as the mesentery and gonads, while
sporadic BL mostly involves the distal ileum, cecum,
and/or mesentery, and rarely the jaw. When bulky or
disseminated disease is present, extranodal involvement
of the ovaries, kidney, breasts, and/or CNS may be seen.
Clinical presentation in a Berlin–Frankfurt–Munster
Group (BFM) series of 152 pediatric patients included
advanced stage (III/IV) disease in 38%, bone marrow
involvement in 33%, and central nervous system (CNS)
disease in 4%. Of the patients in this series, 27% pre-
sented as acute leukemia, referred to as the L3 subtype of
acute lymphoblastic leukemia (ALL) within the French–
American–British (FAB) classification, or B-ALL. BL
infrequently presents in adults, but does occur with
increased frequency in patients with HIV infection.
LBL is most commonly a malignancy of T-cell
precursor cells, and, as such, it is identical to T-cell
acute lymphoblastic leukemia (T-ALL). The World
Health Organization (WHO) classifies these as a single
entity, with the exception that blasts make up <25% of
nucleated bone marrow cells in LBL. T-ALL makes up
approximately 15% of pediatric and 25% of adult ALL.
172 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
Most individuals without frank leukemia present with
advanced-stage disease (III/IV), and bulky adeno-
pathy, anterior mediastinal mass, pleural effusion,
and/or massive organomegaly are common. Patients
should be closely evaluated for emergent complica-
tions such as airway obstruction, superior vena cava
syndrome, and pericardial tamponade. Marrow
involvement may result in cytopenias and/or hyper-
leukocytosis, and CNS involvement is frequent
(5–15%) at diagnosis. Rare individuals with B-lineage
LBL present with more confined disease (e.g. bone,
skin). LBL/T-ALL has historically fared poorer than
B-precursor ALL; however, intensified treatment has
minimized this difference. Currently, the outcome for
children with LBL is excellent, although older individ-
uals have lower relapse-free survival rates.
Pathology
Burkitt lymphoma
BL is characterized by a diffuse proliferation of rather
monotonous intermediate-sized cells with little pleo-
morphism. Within the sheet of neoplastic cells are dis-
persed macrophages containing apoptotic nuclear
debris that give the so-called “starry sky” to the low
power appearance of this lymphoma (Figure 11.1). The
cells have deeply basophilic cytoplasm that contains
small vacuoles (best seen on imprint/cytology sections,
but which can sometimes be seen at the edge of tissue
sections) that contain fat. The nuclei are not usually
larger than those of the associated macrophages and are
characteristically round with a smooth contour and
little variation in size or shape. The nuclear chromatin
is granular and there are 2–5 rather indistinct small
nucleoli. In some BLs associated with HIV infection

Table 11.1 Comparison of endemic, sporadic, and HIV-associated Burkitt lymphoma.
Endemic
Epidemiology Equatorial Africa and Papua,
New Guinea. Geographic asso-
ciation with malaria
Incidence 5–10 cases per 100 000
Age and Malignancy of childhood
gender Peak incidence: 4–7 years
Male:female of 2:1
Clinical Jaw and facial bones in ≈50%.
presentation Also involves mesentery and
gonads. Increased risk of CNS
dissemination
Figure 11.1 Burkitt’s lymphoma. A diffuse proliferation of cells and
scattered macrophages with apoptotic debris giving a “starry sky”
appearance.
the cells may have slightly eccentric nuclei and a plas-
macytoid appearance. Mitoses are frequent.
Immunophenotypically the cells are positive for
CD45 and express the B-cell markers CD20 and
CD79a. They are positive for markers of germinal
center origin CD10 and BCL-6 but, with exceptionally
rare exceptions, lack BCL-2 protein expression. They
are negative for CD5 and CD23. The proliferation
marker Ki-67 stains >98% of tumor cell nuclei
(Figure 11.2). The cells are negative for terminal de-
oxynucleotidyl transferase (TdT).
Previous classifications (Revised European
American Lymphoma [REAL] classification and the
third edition of WHO) recognized a group of lympho-
mas designated Burkitt-like or atypical BL. In general,
these had a morphology and immunophenotype sim-
ilar to classical BL but showed greater pleomorphism.
Chapter 11: Burkitt and lymphoblastic lymphoma
Sporadic HIV-associated
United States and Europe United States and Europe
2–3 cases per million 6 per 1000 AIDS cases
Malignancy of childhood and young adults Malignancy of adults
Median age: 30 years Associated with higher CD4
Male:female of 2–3:1 counts >100/mm3
Abdomen most common presentation, Nodal presentation most
often involving the ileocecal region. Other common with occasional
extranodal sites include bone marrow, ova- bone marrow. Increased risk of
ries, kidneys, and breasts. Increased risk of CNS dissemination
CNS dissemination
Figure 11.2 Burkitt’s lymphoma. The proliferation marker Ki-67
stains >98% of tumor cell nuclei.
This variation in morphological appearance is more
common in adults and in cases with nodal rather than
extranodal sites of involvement. In some cases the
atypical appearance may be due to poor fixation.
The current 2008 fourth edition of the WHO clas-
sification of tumors of the hemopoietic and lymphoid
tissues recognizes an overlap between cases of BL and
diffuse large B-cell lymphoma (DLBCL), referring to
such cases as “B cell lymphoma, unclassifiable, with
features intermediate between diffuse large B cell lym-
phoma and Burkitt lymphoma.” This group consists of
aggressive lymphomas that either have a cellular mor-
phology resembling BL but with a high proliferation,
starry sky appearance, and immunophenotype consis-
tent with classical BL or have a morphology more
reminiscent of true BL but atypical immunophenotype 173
or genotype.
 Chapter 11: Burkitt and lymphoblastic lymphoma
Precursor B-lymphoblastic lymphoma/
lymphoblastic leukemia
This is predominantly a disease of bone marrow,
where there is effacement of the intertrabecular space
by a sheet-like diffuse proliferation of cells. In lymph
nodes the infiltrate usually effaces the nodal architec-
ture with a diffuse pattern but there are usually occa-
sional residual reactive germinal centers within the
diffuse blastic infiltrate. Scattered histiocytes may be
seen within the infiltrate.
Morphologically the lymphoma cells are medium-
sized blast cells with scanty cytoplasm. The nuclei are
either monotonous or may show a degree of pleomor-
phism. They are usually round with indentations but
may be convoluted. The chromatin is classically finely
granular (“salt and pepper” chromatin). Nucleoli are
not usually prominent but may be seen in some cases.
Mitoses are easily seen. Distinction of reactive hema-
togones from subtle marrow infiltration by lympho-
blastic lymphoma is essential and may be problematic
as hematogones may be numerous in children and
following chemotherapy.
Immunophenotypically the cells frequently do not
express CD45 and expression of CD20 is variable and
often negative. There is expression of CD19 and CD79a,
and the majority of cases are positive for CD10, with
expression of BCL-2 protein. The precursor nature of
the lymphoid cells is indicated by expression of TdT,
with expression of other markers of immaturity such as
CD34, CD117, and CD99. Proliferation is high, but does
not reach 95% (Figure 11.3).
Figure 11.3 Precursor B-lymphoblastic lymphoma. Expression of
174 TdT and other markers of immaturity such as CD34, CD117, and CD99.
Proliferation is high but does not reach 95%.
Precursor T-lymphoblastic lymphoma/
lymphoblastic leukemia
Infiltration is usually as a sheet with linear infiltration
of cells into surrounding soft tissue at the periphery.
Morphologically T-lymphoblastic lymphoma is
indistinguishable from the B-cell counterpart. The
cells are medium-sized blasts with scanty cytoplasm
and nuclei that may be round, indented, or more obvi-
ously convoluted. Chromatin is finely granular (“salt
and pepper”). Nucleoli are usually indistinct or small.
Mitoses are frequent and there may be scattered histio-
cytes in the infiltrate. In mediastinal lesions remnants of
normal thymus may be seen. Distinction from normal
thymus and lymphocyte-rich thymomas is essential and
may be difficult morphologically, particularly when the
biopsies are small. This is helped by the demonstration
of intimately admixed epithelial proliferations in thy-
momas using antigens to high molecular weight cyto-
keratin, while the T-cells of lymphoblastic lymphoma
extend well away from any residual epithelial structures
and into fatty connective tissue.
Immunophenotypically the cells are usually nega-
tive for CD45. There is variable expression of T-cell
antigens CD2, CD3, CD5, and CD7. Cells may express
CD4 or CD8 but frequently express both and may
express neither. The cells are usually positive for
BCL-2 protein. Expression of CD79a (usually consid-
ered a marker of B-cell phenotype) is frequent, and
expression of CD10 may also be seen. The immature
nature of the cells is confirmed by expression of TdT,
CD34, CD117, and CD99. There is variable expression
of CD1a. Occasionally, myeloid associated antigens
(CD13, CD33) may be expressed. Proliferation is
high but does not exceed 95%.
Molecular pathology and cytogenetics
Burkitt lymphoma
The genetic hallmark of BL is translocations targeting
the MYC gene in chromosome band 8q24, most fre-
quently via the t(8;14)(q24;q32) translocation and
involving the immunoglobulin heavy chain (IGH)
locus in 85% of cases, or, alternatively, the IG light
chain loci in 2p12 and 22q11. MYC controls a wide
variety of processes involved in proliferation, differ-
entiation, and apoptosis. In endemic BL, the molecular
breakpoints occur 5' upstream of MYC, and IGH
breakage is found in the VDJ region of IGH. In con-
trast, in sporadic BL, the breakpoints are found either

directly 5' upstream, or within the first intron or exon
of MYC and in the switch region of IGH. Therefore,
the two tumor types obviously are distinguished by
distinct differentiation stages, in which the transloca-
tions occur, namely either during VDJ rearrangement
and class switch recombination.
Other genes frequently altered in BL comprise
TP53 and CDKN2A. Also, MYC itself is frequently
targeted by mutations. BL usually carry few secondary
aberrations, most frequently gains in 1q, 7q, 12q, 20q,
and Xq, and losses in 13q. Two landmark studies
revealed that BL has a distinct gene expression signa-
ture, allowing for construction of a molecular classi-
fier. As would have been expected, MYC and its target
genes are commonly upregulated in BL, but also ger-
minal center B-cell-related genes were expressed at
higher levels. In contrast, MHC class I and NF-κB-
related genes were expressed at lower levels in BL in
comparison to DLBCL. Applying this molecular clas-
sifier to aggressive B-cell lymphomas, BL were clearly
set apart from DLBCL, although in both studies, an
“intermediate” or “gray zone” group of cases was iden-
tified, not readily falling into one of the clearly defined
core groups. Most strikingly, both groups identified a
group of aggressive tumors, morphologically indistin-
guishable from DLBCL but presenting with a BL gene
expression profile – the so-called “molecular BL.” This
finding clearly raises the question whether these cases
should be treated with intensified BL-like chemother-
apy regimens. Another finding substantiated in these
studies was that up to 10% of DLBCL identified by
morphology and molecular gene expression profile
carry MYC translocations. Preliminary data would
suggest that these cases are likely to pursue a worse
clinical course than MYC-negative DLBCL.
B-cell lymphoma, unclassified, with
features intermediate between DLBCL
and Burkitt lymphoma
These cases harbor, both from a morphological as well
as from a genetic view, intermediate features between
BL and DLBCL. Judging from available data, 35–50% of
cases carry MYC translocations. In contrast to classical
BL, however, MYC in these cases is frequently trans-
located to non-IG genes. Most cases with dual trans-
locations, the so-called “double hit lymphomas,”
involving mainly MYC and BCL-2, and, to a lesser
extent, MYC and BCL-6, fall into this category. In
contrast to classical BL, secondary genetic abnormalities
Chapter 11: Burkitt and lymphoblastic lymphoma
are frequent, with complex karyotypes as evidenced by
array comparative genomic hybridization (CGH)
studies.
Precursor B-lymphoblastic leukemia/
lymphoma
The translocation t(12;21)(p13;q22), resulting in a
TEL–AML1 fusion transcript, is detected in up to 25%
of childhood ALL. Both TEL (also known as ETV6) and
AML1 (also known as RUNX1) genes are involved in
other leukemia-associated translocations as well, e.g. by
generating fusion transcripts with PDGFRβ, MN1, ABL,
EVI1, JAK2, and CBFβ. In contrast to childhood ALL,
the TEL–AML1 translocation represents a relatively
rare genetic aberration in adult ALL patients (3% of
adult B-ALL). In both age groups, however, the trans-
location confers a highly favorable prognosis, with
event-free survival (EFS) rates close to 90%.
Microarray-based gene expression profiling has
revealed a distinctly homogeneous expression pattern
in ALL cases with TEL–AML1 translocations, suggest-
ing that tumors harboring this aberration represent a
unique leukemia subtype. Interestingly, the tumor cells
have been shown to overexpress the erythropoietin
receptor, which potentially affects proliferation and
survival signaling of the tumor cells.
Hyperdiploidy (more than 50 chromosomes per
leukemic cell) is also a common genetic alteration
detected in about 30% of pediatric and 9% of adult
B-ALL cases. This condition also defines a patient
subgroup with an excellent clinical outcome. In addi-
tion, pediatric cases with trisomies 4, 10, and 17 may
have a specifically favorable prognosis. In about 20%
of patients with hyperdiploid chromosome clones,
activating mutations in the receptor tyrosine kinase
gene FLT3 have been demonstrated, suggesting a pos-
sible therapeutic benefit from the administration of
small tyrosine kinase inhibitor molecules in this
group.
The translocation t(9;22), encoding the BCR–ABL
fusion protein, can be detected in about 30% of adult
and 5% of pediatric B-ALL tumors, and confers an
adverse prognosis in both age groups. The different
incidence of the t(9;22) among pediatric and adult
B-ALL may, at least in part, account for the overall
difference in clinical outcome between the two age
groups, and bone marrow transplantation during
first remission is often recommended in B-ALL 175
patients with a BCR–ABL fusion transcript.
 Chapter 11: Burkitt and lymphoblastic lymphoma
In contrast to hyperdiploidy, hypodiploidy (fewer
than 45 chromosomes) is an infrequent finding in
B-ALL patients (<2%), and has been associated with
poor outcome.
The tumors in about 6% of B-ALL patients may
harbor the t(1;19)(q23;p13) chromosome transloca-
tion that generates the E2A–PBX1 fusion transcript.
This chimeric protein interferes with hematopoietic
differentiation by disrupting the HOX gene-dependent
regulatory network.
Aberrations involving the chromosomal band
11q23 are detected in lymphoblastic and acute myeloid
leukemias. The mixed-lineage leukemia (MLL) gene in
11q23 is involved in translocations with more than 40
different translocation partners, generating fusion
proteins harboring the NH2-terminus of the MLL
gene and COOH-terminus of the partner genes.
Translocations involving MLL probably play a role in
leukemogenesis via deregulation of the HOX pathway.
The most common MLL translocation, the t(4;11)
(q21;q23), results in an AF4/MLL fusion protein. It is
present in almost 50% of infant ALL cases and, less
frequently, in other age groups, and is associated with
poor prognosis. Approximately 20% of the cases with
MLL translocations show a high expression of the
FLT3 receptor tyrosine kinase due to a mutation in
the activation loop region. This finding suggests that
MLL fusion proteins may cooperate with activated
tyrosine kinases in promoting leukemogenesis. In
keeping with this concept, FLT3 inhibitors are anti-
tumor effective in vitro and in mouse models.
Precursor T-lymphoblastic leukemia/
lymphoma
Chromosomal rearrangements in precursor
T-lymphoblastic leukemia/lymphoma frequently result
in proto-oncogene deregulation by juxtaposing them to
T-cell receptor (TCR) regulatory elements. Important
genes involved in such rearrangements are TAL1, MYC,
LYL1, LMO1, LMO2, HOX11, and HOX11L2. TAL1
overexpression is induced by the translocation t(1;14)
(1p32;q11) or by deletion of upstream regulatory
elements of the gene, and occurs in about 25% of pedia-
tric T-ALL. The deregulated expression of TAL1 prob-
ably activates a set of genes that are normally quiescent
in T-cell progenitors or, alternatively, may inactivate
E2A homodimers and heterodimers. The translocations
t(11;14)(p15;q11) and t(11;14)(p13;q11) are thought to
176 affect similar pathways via the ectopic expression of the
LMO1 and LMO2 genes. LMO1 and LMO2 proteins are
highly expressed in the central nervous system but
nearly inactive in T-cells and their progenitors. The
deregulated expression of HOX11 in T-cells, as a con-
sequence of either the t(10;14)(q24;q11) or the t(7;10)
(q35;q24), promotes transformation via disruption of
normal cellular regulatory networks. These aberrations
occur in approximately 30% of adult T-ALL patients
and in only 3% of childhood ALL cases. In both age
groups, abundant ectopic HOX11 protein expression is
associated with a favorable clinical outcome. The
HOX11L2 gene located in the chromosomal band 5q35
is deregulated in a subset of T-ALL patients by either the
t(5;14)(q35;q11) or t(5;14)(q35;q32). The latter trans-
location results in HOX11L2 overexpression by juxta-
posing it to the BCL11b gene that is differentially
expressed during T-cell development. Using molecular
techniques, HOX11L2 translocations are detected in
about 20% of T-ALL patients, but are an infrequent
finding in adults. While the NOTCH1 gene is rarely
targeted by chromosomal translocations, activating
mutations in this gene have been detected in approx-
imately 50% of T-ALL tumors of all different molecular
subtypes. The NOTCH1 gene encodes a transmembrane
receptor involved in the regulation of normal T-cell
development, and chromosomal aberrations involving
NOTCH1 have been shown to induce T-ALL in mouse
models. New ABL1 fusion genes have been identified
that provide proliferation and survival advantage to
lymphoblasts. Recently, microarray based gene expres-
sion studies revealed that T-ALL cases overexpressing
HOX11 and HOX11L2 as well as T-ALL with MLL
fusion genes share the same expression pattern, charac-
terized by a global HOXA gene deregulation signature.
The translocation t(11;19)(q23;p13) resulting in an
MLL–ENL fusion transcript occurs in acute myeloid
leukemia patients, but also in patients with B-ALL and
also T-ALL. In T-ALL patients, the translocation is
associated with a favorable prognosis and long-term
survival. T-ALL with a CALM–AF10 fusion, generated
by the t(10;11)(p13;q14), characteristically display
immature phenotypic features, either lacking TCR
expression or expressing TCRγ/δ. Using genome-wide
profiling strategies in T-ALL, multiple genetic abnor-
malities have been identified, including deletions and/or
mutations targeting TAL1, RB1, PTEN, and others.
Analysis of copy-neutral loss of heterozygosity (CN-
LOH) revealed frequent CN-LOH in the vicinity of the
CDKN2A/B gene in 9p that, however, also usually har-
bors focal deletions.

Staging
Staging is a critical component of treatment for lym-
phomas. In particular, the recognition that early-stage
BL and LBL require less treatment than advanced-stage
disease has led to risk-adaptive strategies that are
dependent on the accuracy of the staging classification
systems. The Ann Arbor staging system, which is the
most commonly used system for lymphomas, was ini-
tially developed for radiation treatment of Hodgkin’s
lymphoma. As such, it is relatively useful for identifying
nodal and extranodal regions involved by disease but
much less so for disease bulk or specific disease sites.
The occurrence of BL and LBL in pediatrics led to the
St. Jude staging system, which was developed to reflect
more accurately tumor bulk and high-risk disease sites
(Table 11.2). The St. Jude staging classification recog-
nizes early-stage (I and II) disease as low risk. Unlike the
Ann Arbor system, resection of abdominal disease,
which is an adverse site in BL, is considered stage II in
the St. Jude system. Stage III disease is identified by
disease on both sides of the diaphragm, similar to the
Ann Arbor classification, but also includes primary
intrathoracic or unresected abdominal disease. Stage
IV in both classifications involves the bone marrow
and/or central nervous system (CNS). Overall, the
St. Jude staging system is most commonly used for
risk-adaptive treatment of BL and LBL.
Table 11.2 Staging systems.
Stage St. Jude staging
I Single site (excluding abdomen or mediastinum)
II Single extranodal site with regional
nodes
Two or more nodal sites, same side of
diaphragm
Two extranodal sites, same side of diaphragm
Primary gastrointestinal, completely
resected (IIR)
III Two extranodal sites, both sides of diaphragm
Two or more nodal sites, both sides of
diaphragm
Primary thoracic
Primary gastrointestinal, extensive
Paraspinal, epidural
IV CNS and/or bone marrow (<25%)
Chapter 11: Burkitt and lymphoblastic lymphoma
Standard staging entails the use of whole body com-
puterized tomography (CT) scans and FDG–PET (posi-
tron emission tomography) scans, with particular
attention to the bowel in the case of BL where involve-
ment may not be readily apparent. Evaluation of the
CNS with cerebral spinal fluid analysis for cytology,
and if clinical symptoms warrant with magnetic reso-
nance imaging (MRI), should be performed to rule out
active involvement by BL or LBL. All patients should also
undergo bone marrow aspiration and biopsy. Serologies
for HIV and hepatitis B are also indicated given the
increased incidence of BL in the former and the risk of
hepatitis B reactivation in patients with a carrier state.
Treatment principles of Burkitt
lymphoma
BL is a systemic disease and requires chemotherapy for
all disease stages. Importantly, locoregional radiation
does not improve survival and should be avoided.
While older studies showed that surgical resection of
abdominal disease improves outcome, indicating the
importance of tumor volume, more effective and risk-
adapted treatments have made surgical resection
unnecessary except for specific complications like
intestinal obstruction associated with intussusception,
perforation, fistula, or bleeding.
Ann Arbor staging
Single node region or extranodal site (IE)
Two or more node regions, same side of diaphragm +/–
• Localized contiguous extranodal site (IIE)
Two or more node regions, both sides of diaphragm +/–
• Localized contiguous extranodal site (IIIE)
• Spleen (IIIS) or
• Both (IIIES)
Bone marrow or liver
• Diffuse extranodal disease not encompassed in a single radiation
field
E: Single extranodal site contiguous with a known nodal site
A: No symptoms
B: Fever, weight loss, night sweats
177
 Chapter 11: Burkitt and lymphoblastic lymphoma
Early treatment strategies for BL were modeled on
ALL regimens which employed dose-intense and pro-
longed treatment with induction, consolidation, and
maintenance phases. These approaches stand in con-
trast to the significantly less dose-intense regimens
used in adults with “intermediate-grade” lymphoma,
such as CHOP and CHOP-based regimens, that only
produced a 50–60% event free survival (EFS). While
dose intensity and dose density are important treat-
ment components for BL, later studies indicated that
shorter treatment durations were equally successful
and less toxic. Furthermore, the recognition that
tumor volume is an important prognostic feature led
to the use of risk-adaptive approaches and a further
reduction in treatment for early stage patients.
Several biological characteristics of BL have helped
guide treatment strategies, including its high prolifera-
tive fraction. It has been recognized for years that BL is
sensitive to multiple chemotherapy classes and, in
endemic BL, apparent cures were occasionally achieved
with single agent cyclophosphamide. Despite initial sen-
sitivity, however, patients frequently relapsed, particu-
larly those with higher volume disease. This apparent
dichotomy can potentially be explained by the high
tumor proliferative rate. The role of tumor cell kinetics
was raised some 30 years ago by Skipper et al., who
observed that the fraction of cells undergoing DNA
replication, termed the “growth fraction,” greatly influ-
enced drug sensitivity, a finding that likely reflects the
greater sensitivity during S-phase to many drug classes.
Although a high growth fraction would predict greater
drug sensitivity, it could also lead to greater tumor
proliferation between cycles. Depending on the relative
impact of these two effects, cure could increase owing to
higher fractional cell kill or decrease if tumor prolifer-
ation between cycles leads to a “kinetic failure.” One
strategy to overcome such “kinetic failure” is to increase
dose density through frequent chemotherapy adminis-
tration, a strategy employed by most BL regimens.
Another strategy is to increase the fractional cell kill or
efficacy of chemotherapy, thereby reducing the number
of tumor cells that can survive and proliferate between
cycles. Hence, common BL regimens are relatively high
dose and employ multiple chemotherapy agents admin-
istered in alternating cycles. They typically include
anthracylines, epipodophyllotoxins, vinca alkaloids,
and alkylators, as well as methotrexate and cytarabine,
which are cell cycle active agents and take advantage of
the high tumor proliferation. These agents, however, are
178 administered in a variety of combinations and
schedules, indicating the empiric nature of the actual
combinations (Table 11.3). Indeed, the optimal dose
and schedule of cyclophosphamide, methotrexate, and
cytarabine remain unknown and vary among the major
regimens (Table 11.4).
The risks of tumor lysis syndrome and propensity
for CNS dissemination in BL both have important
treatment implications. To reduce tumor lysis, which
can produce life-threatening electrolyte imbalances
and renal failure, many regimens employ a pre-phase
whereby relatively low-dose cyclophosphamide and
prednisone are administered. This strategy has been
incorporated into the regimens of the French Society
of Pediatric Oncology (SFOP), German Multicenter
ALL Group (GMALL), and BFM, but not CODOX-M
or hyper-CVAD (Table 11.4). The high risk of CNS
involvement is handled by the use of relatively high-
dose intravenous methotrexate and cytarabine, both of
which have CNS penetration, and intrathecal chemo-
therapy administration. An important advancement
has been to reduce intrathecal treatment and eliminate
whole brain radiation for prophylaxis, which has sig-
nificantly reduced CNS toxicity.
Treatment of Burkitt lymphoma
Beginning in 1981, the SFOP conducted a series of
protocols to refine the treatment of BL and B-ALL
(LMB). These studies demonstrated that short dose-
intensive treatment was effective in patients without
CNS involvement, and that dose intensification of
methotrexate, cytarabine, and etoposide with cranial
irradiation improved the EFS of patients with CNS
involvement to 75%. Based on these findings, a risk-
adapted protocol (LMB 89 protocol) was developed
in which treatment was based on tumor burden and
early response to chemotherapy (Table 11.4). Using
the St. Jude (Table 11.2) staging, group A included
stage I or II with abdominal resection; group B
included unresected stage I, non-abdominal stage II,
any stage III or IV disease, and/or B-ALL (CNS-
negative and <70% marrow blasts); and group C
patients had CNS involvement and/or >70% marrow
blasts. Based on these risks, group A only received
induction, group B received pre-phase, induction,
consolidation, and limited maintenance, and group
C also received extended maintenance and cranial
irradiation if the CNS was involved. If a CR was not
achieved in groups B and C after the third or fourth
induction-consolidation course, patients underwent
 Chapter 11: Burkitt and lymphoblastic lymphoma

Table 11.3 Comparison of dose and dose intensity among Burkitt lymphoma regimens.

CODOX-M/ IVAC LMB 89 BFM 90 Dose range

Doxorubicin
Dose* 40 60 50 40–50
Dose intensity* 13 20 25 13–25
Cyclophosphamide
Dose 1600 1500 1000 1000–1600
Dose intensity 533 500 500 500–533
Vincristine
Dose 3 2 1.5 1.4–2
Dose intensity 1.0 0.63 0.75 0.63–0.93
Etoposide
Dose 300 800 200 200–800
Dose intensity 100 266 100 100–266
Methotrexate
Dose 6720 8000 5000 5000–8000
Dose intensity 2240 2666 2500 2240–2666
Cytarabine
Dose 8000 9000 600 600–9000
Dose intensity 2666 3000 200 200–3000
Ifosfamide
Dose 7500 NA 4000 7500–4000
Dose intensity 2500 NA 2000 2500–2000
Other drugs NA Prednisone Dexamethasone NA
2
* Dose is expressed as mg/m . Dose intensity is expressed as planned dose intensity and calculated as
mg/m2/week.

autologous transplant. This strategy was highly success-
ful in pediatric patients, with 5-year EFS and OS of 92%
(Table 11.4). The success of LMB 89 in pediatrics led to
its testing in adults with minor modifications
(Table 11.4). The outcome in 72 adult patients, mostly
with advanced disease, was favorable, with EFS and
overall survival (OS) of 65% and 70%, respectively, at
2 years. Toxicity remains a problem for advanced stage
patients because of the treatment intensity. Treatment-
related deaths from infection were higher in adults
compared to pediatrics, with incidences of 5% and
1.6%, respectively, among patients in groups B and
C. Myelosuppression was the primary treatment com-
plication, with over 40% of adults experiencing febrile
neutropenia. Tumor lysis syndrome requiring dialysis
was prevented by the pre-phase treatment and associ-
ated supportive care.
The GMALL reported excellent results with an
intensive 18-week regimen in adult patients with
B-ALL. In an effort to reduce CNS toxicity, the Cancer
and Leukemia Group B (CALGB) studied a modified
version in which intrathecal methotrexate was reduced
and cranial radiation was only administered to high-risk
patients (Table 11.4). These patients were compared to
another cohort who received the standard CNS prophy-
laxis. Overall, 92 patients enrolled on the two cohorts,
and there were no differences in outcome with EFS and
OS of 45–52% and 50–54%, respectively, at 3 years. This
study suggested that short-duration treatment with less
intensive CNS prophylaxis was equally effective, with
significantly less neurotoxicity in comparison to more
intensive prophylaxis.
Another highly effective regimen for BL was devel-
oped by the BFM (Table 11.4). Like other regimens for
BL, the BFM approach was based on short, intensive
cycles and, over the course of several protocols, led to a
reduction in the number of cycles based on risk strat-
ification. The BFM 90 protocol continued to refine the
risk stratification further and to improve the outcome
of patients who had an incomplete initial response with
further treatment intensification. Among 322 pediatric 179
patients with BL or B-ALL treated with the BFM 90
 Chapter 11: Burkitt and lymphoblastic lymphoma

Table 11.4 Selected regimens for Burkitt lymphoma.

Regimen Patients Histology Median age Stage (%) CR (%) EFS (%) OS (%)
(No.) (No.) (years)
(range)
LMB 89 561 Burkitt and 8 (0.17–18) III–IV 97% 92% (5-year) 92%
B-ALL (420) 79% (5-year)
Modified LMB 72 Burkitt and 33 (18–76) III–IV 72% 65% (2-year) 70%
B-ALL 67% (2-year)
GMALL 35 B-ALL 36 (18–65) N/A 74% 71% (4-year) 51%
B-NHL 86 (DFS) (4-year)
Modified GMALL 92 Burkitt and 47 (17–78) III–IV 74% 45–52% 50–54%
B-ALL 89% (3-year) (3-year)
BFM 90 413 Burkitt and 9 (1.2–17.9) III–IV N/A 89% (6-year) 14
B-ALL (322) 60% deaths
CODOX-M/ IVAC 21 peds Burkitt 12 (3–17) III–IV 95% 85% (peds) 2 deaths
20 adult 25 (18–59) 78% 100% (adults)
(2-year)
CODOX-M/ 52 Burkitt 35 (15–60) III–IV 77% 65% (2-year) 73%
IVAC 61% (2-year)
CODOX-M/IVAC 53 Burkitt 37 (17–76) III–IV 64% (2-year) 67%
(lower dose 76% (PFS) (2-year)
methotrexate 3 g/m2)
Hyper-CVAD 26 B-ALL 58 (17–79) N/A 81% 61% (3-year) 49%
(DFS) (3-year)
Hyper-CVAD/R 31 Burkitt / 46 (17–77) IV 45% 86% 80% (3-year) 89%
B-ALL >60 (29%) (3-year)
DA-EPOCH-R 25 Burkitt 30 (18–66) III–IV 56% 100% 96% (3-year) 100%
(3-year)
5-drug induction 27 Burkitt 36 (15–64) III–IV 44% 81% 73% (5-year) 81%
followed by BEAM (5-year)
and autologous SCT

regimen, EFS was 89% at 6 years. Importantly, this
represented a significant advance for advanced-stage
patients compared to the previous BFM 86 protocol.
Two clinical trials conducted at the National Cancer
Institute (NCI protocols 77–04 and 89-C-41) diverged
from the leukemia model of treatment and investigated
the role of more limited treatment (Table 11.4). In the
77–04 protocol, three cycles of intravenous cyclophos-
phamide, vincristine, doxorubicin, methotrexate, and
leucovorin rescue with cytarabine intrathecal prophy-
laxis (CODOX-M) showed that intensive combination
chemotherapy produced an EFS of 66% in patients with
low-risk disease. However, patients with high-risk dis-
ease (i.e. St. Jude stage IV) fared poorly, with a 19% EFS.
Based on the observation that some patients were sal-
vaged with combination ifosfamide, etoposide, and
cytarabine (IVAC), a successor protocol 89-C-41 was
180
developed in which CODOX-M was alternated with
IVAC and administered for four cycles in high-risk
patients. This approach yielded an overall EFS of 85%
compared to 55% for all patients treated on CODOX-M
alone, with no difference in low- and high- risk disease.
When the CODOX-M/IVAC regimen was tested in an
adult cooperative group trial, however, only a 65%
2-year EFS was achieved. A recent prospective trial
incorporated risk stratification and dose-modification
of CODOX-M in adults with or without IVAC based
on staging. The methotrexate dose was reduced from 6.7
g/m2 to 3 g/m2. EFS and OS at 2 years were 64% and
67%, respectively. Although there was no significant
reduction in toxicity, as outcomes in this trial were
similar to results with higher dose methotrexate, the
authors suggested that the lower dose methotrexate
should be the new standard of care in this treatment arm.

The hyper-CVAD regimen was based on a mod-
ification of a regimen developed by Murphy et al. for
pediatric B-ALL (Table 11.4). In this regimen, hyper-
fractionated cyclophosphamide, vincristine, doxoru-
bicin, and dexamethasone were alternated with
methotrexate and cytarabine for a total of eight cycles.
The results in 26 adults with B-ALL were favorable,
with an EFS of 73% at 5 years. Based on these results in
B-ALL, this regimen is also used for BL.
In an effort to improve the outcome of BL, inves-
tigators have explored the role of autologous stem
cell transplant (SCT) consolidation. One study
investigated the role of two courses of intensive
chemotherapy, without high-dose methotrexate or
cytarabine, followed by BEAM (carmustine (BCNU),
etoposide, cytarabine, and melphalan) and autologous
SCT (Table 11.3). In a multiple center study of 27
adult patients with BL without CNS or extensive bone
marrow involvement, the EFS was 73% at 5 years.
While these results are encouraging, they are not
significantly different from other regimens, and
question the added benefit of autologous SCT over
current approaches.
While current treatments are effective in BL,
improvements are needed in patients in advanced-
stage disease. Furthermore, toxicity of the regimens for
advanced-stage disease is high, especially in older adults,
and treatment-related mortality is excessive. Because all
BL regimens are very aggressive, there are no specific
recommendations regarding which ones to use in adults.
However, the treating physician must be cognizant of
their toxicity, particularly in patients over 50 years of
age. Hence, new strategies are needed to improve the
therapeutic index of treatment and to increase efficacy.
The success of rituximab in DLBCL has prompted its
testing in BL. Rituximab has been incorporated in the
hyper-CVAD regimen with a 3-year EFS of 80% and OS
of 89% in 31 patients, most of whom had advanced-
stage disease. In this study, favorable outcomes were
seen even in the elderly subgroup (age >60 years),
which represented approximately 30% of the study pop-
ulation. Similarly, rituximab is being studied in pediatric
BL. A phase II window study evaluated the activity and
safety of single agent rituximab in children <19 years of
age with newly diagnosed BL and other mature B-cell
NHLs. Of evaluable patients with BL, 27 of 67 (40%)
were considered “responders” to rituximab alone
(defined as ≥25% decrease of at least one lesion or blasts
and no disease progression at other sites 5 days after
receiving rituximab and prior to receiving systemic
Chapter 11: Burkitt and lymphoblastic lymphoma
chemotherapy). The Children’s Oncology Group
(COG) has also recently completed a prospective trial
that incorporated rituximab into the LMB 96 regimen.
Analysis of 45 patients (56% with BL) revealed that the
addition of rituximab appeared to be safe, and that the
3-year EFS and OS were both 95%. Preliminary results
with the NCI dose-adjusted pharmacodynamic-based
infusional regimen with etoposide, prednisone, vincris-
tine, cyclophosphamide, doxorubicin, and rituximab
(DA-EPOCH-R) suggest that optimizing the schedule
may significantly improve the therapeutic index of che-
motherapy and reduce toxicity. Preliminary results in 25
patients (17 HIV-negative, 8 HIV-positive) with
untreated BL have shown a 3-year EFS and OS of 96
and 100%, respectively. Such results suggest that DA-
EPOCH-R may be an excellent and well-tolerated alter-
native to dose-intense regimens, especially for HIV-
positive patients and older adults with BL where toxicity
is a major concern. Multiple new targeted agents are also
currently under development and may have utility in
BL. However, these agents are in early clinical
development.
Management of relapse in Burkitt
lymphoma
The salvage treatment of relapse or refractory BL is
usually unsuccessful. Early-stage patients who relapse
after limited treatment, however, may still be curable,
whereas advanced-stage patients who have received
intensive treatment or are primary induction failures
are rarely cured. A COG prospective trial evaluating
the use of rituximab, ifosfamide, carboplatin, and eto-
poside in pediatric relapsed/refractory BL and other
mature B-cell lymphomas was recently conducted.
Twenty patients with DLBCL and BL were treated
with an objective response rate for the entire group
of 60% with a median follow-up of 10.5 months. The
CR and PR rates for the 14 patients with BL and B-ALL
were 29% (four patients) and 36% (five patients),
respectively. A review from the European Group for
Blood and Marrow Transplantation (EBMT) by
Sweetenham et al. reported an OS of 37% at 3 years
for patients with chemosensitive relapse, but only 7%
for those with resistant disease following autologous
SCT. An analysis of allogeneic transplantation in BL
did not find any relationship between survival and
graft-versus-host disease, suggesting no graft-versus-
tumor effect, and the OS was lower than patients
treated with autologous transplant. 181
 Chapter 11: Burkitt and lymphoblastic lymphoma
HIV-associated Burkitt lymphoma
BL comprises 13–18% of HIV-associated lymphomas,
a significant decrease since the development of highly
active antiretroviral therapy (HAART). Clinically, BL
in HIV-infected patients is similar to sporadic BL and
typically occurs before the development of severe
immunodeficiency. Unlike other HIV-associated lym-
phomas, there has been no improvement in the out-
come of BL since the development of HAART. A
retrospective analysis of HIV-associated BL diagnosed
between 1982 and 2003 from the University of
Southern California showed no change in median
survivals of 6.4 and 5.7 months, respectively, in the
pre- and post-HAART eras. This contrasts with the
outcome of DLBCL during these periods, when
the median survival improved from 8.3 to 43.2
months, respectively. This improvement is likely
because of a more favorable DLBCL pathobiology
that occurs at higher CD4 cell counts in the HAART
era. This does not seem to be the case with BL, which
occurs at a high median CD4 cell count and does not
have such a favorable pathobiology.
The relatively poor outcome of HIV-associated BL
is likely due to the use of CHOP-based regimens, which
are known to have a poor outcome in this disease. In
contrast, dose-intense regimens like hyper-CVAD have
shown encouraging results in HIV-associated BL, with
92% complete remission, but are quite toxic. The LMB
86 regimen was prospectively evaluated in the treatment
of 63 HIV-infected patients with stage IV BL, with a
CR rate of 70% and 2-year OS of 47%. Like other
intensive therapies, this was associated with significant
treatment-related toxicities, including pancytopenia
and deaths from neutropenia-associated sepsis.
Patients with higher CD4 counts and better perform-
ance status tolerated this therapy better and had
improved outcomes. Until recently, the use of rituxi-
mab in HIV-related malignancies has been somewhat
controversial. However, recent prospective trials have
demonstrated both safety and efficacy, with the incor-
poration of rituximab in HIV-related BL. A prospective
trial of rituximab with a GMALL chemotherapy back-
bone in 19 patients with HIV-related BL showed higher
incidences of grade 3–4 mucositis and more severe
infectious episodes and outcomes comparable to the
non-HIV cohort, with a 2-year OS of 82% and disease
free survival (DFS) of 93%. These studies highlight the
necessity to balance treatment efficacy and toxicity in
patients with HIV-associated NHL. Dose-intense BL
182
regimens are considered too toxic to warrant their gen-
eral use in HIV-positive patients, and have led to the use
of CHOP-based treatment. The excellent results
achieved with the DA-EPOCH-R regimen in HIV-
associated lymphoma overall and BL in HIV-negative
patients suggest it may be an excellent regimen in HIV-
positive patients because of its relatively low toxicity.
Preliminary results from a trial at the NCI of DA-
EPOCH-R in 25 patients (eight HIV-positive) with
untreated BL have shown a 3-year EFS and OS of 96
and 100%, respectively. Similar studies conducted by
other groups have revealed tolerability and efficacy, but
with increased infection-related deaths in patients with
CD4 counts <50/μL.
Treatment recommendations in
Burkitt lyphoma
Multiple regimens, as discussed above, have shown
equivalent efficacy in BL and may be used. Current
treatment plans should include risk-adaptive strategies
and reduced CNS prophylaxis to minimize toxicity.
Adults may be effectively treated with pediatric regi-
mens, although the CODOX-M/IVAC, CALGB and
hyper-CVAD regimens have been specifically investi-
gated in adults and shown to be effective and relatively
well tolerated. Treatment with DA-EPOCH-R is prom-
ising in both HIV-negative and HIV-positive patients.
Treatment principles of
lymphoblastic lymphoma
Therapy consists predominantly of combination che-
motherapy. The most commonly employed regimens
include the following phases: induction, consolidation,
CNS treatment, and maintenance. Induction with ≥4
drugs is often given as a 28-day course, although there
are alternative approaches. A variety of consolidation,
also known as intensification, blocks have been shown
to improve long-term outcome for high-risk pediatric
subgroups, including those with slow or incomplete
responses to induction. All patients with LBL require
CNS-directed therapy to decrease the risk of menin-
geal relapse, and a prolonged maintenance phase.
Shortening therapy to <2 years appears to compromise
outcome for both adults and children.
Most patients with LBL present with advanced-
stage (III/IV) disease. Although less intense treatment
is commonly employed for patients with low-stage

(I/II) LBL, randomized clinical trials confirm the need
to treat with aggressive, prolonged ALL-type therapy.
For example, an early Pediatric Oncology Group
(POG) trial demonstrated an advantage of the addition
of a 24-week maintenance phase for pediatric patients
with low-stage disease. Reducing therapy for low-stage
LBL increases the risk of relapse, although salvage rates
are better for such patients in comparison to those
who relapse after more intensive regimens. No appa-
rent difference in outcome has been seen within low-
stage (i.e. I versus II) or advanced-stage (i.e. III versus
IV) LBL in pediatric series. Bone marrow involvement
was associated with poor outcome in the LMT 89 trial
in adults. However, subgroup numbers are small in all
series, preventing firm conclusions. Rare individuals
with B-precursor phenotype LBL are treated according
to B-precursor ALL-specific regimens, which are not
considered in detail in this chapter.
Treatment of lymphoblastic lymphoma
Sequential pediatric cooperative group trials in the
setting of LBL/T-ALL have led to the development of
highly effective specific regimens (Table 11.5). Initial
studies in the 1970s demonstrated the advantage of
ALL-type therapy over CHOP-like regimens. In a
trial conducted by the Children’s Cancer Group
(CCG) in North America, the Memorial Sloan
Kettering Cancer Center (MSKCC) LSA2L2 regimen
was shown to be superior to COMP (cyclophospha-
mide, vincristine, methotrexate, and prednisone),
with 64% versus 35% EFS, respectively. An early
study conducted by the BFM group also demonstra-
ted efficacy of ALL-type therapy for children with
LBL (70% EFS). Serial BFM trials have led to steady
improvements, with EFS exceeding 90% on the most
current pediatric protocols. The BFM regimen “back-
bone” has been incorporated into North American
studies through the CCG and COG.
Clinical trials have also demonstrated the advant-
age of ALL-type therapy for adults with LBL. The
pediatric BFM regimen has been applied successfully
to adult patients in serial trials of the GMALL.
Investigators at the MD Anderson Cancer Center
have achieved good results with the hyper-CVAD reg-
imen. However, although remission induction rates
on these trials exceed 90%, relapse-free survival rates
(44–62%) are lower than in pediatric series. In the
case of GMALL studies, this is likely due at least in
part to the use of less intensive and shorter duration
Chapter 11: Burkitt and lymphoblastic lymphoma
chemotherapy in comparison to the most effective
pediatric BFM regimen.
A number of specific chemotherapy agents, doses,
and schedules are particularly effective against LBL/T-
ALL. Increasing the intensity of induction above and
beyond the standard four-drug regimen (corticoste-
roids, vincristine, anthracyclines, asparaginase) by the
addition of other agents (e.g. cyclophosphamide)
appears to improve CR rates in children and adults.
High doses of methotrexate are required to achieve
adequate intracellular concentrations of methotrexate
polyglutamates in T-lymphoblasts. Early French stud-
ies (LMT 81) added high-dose methotrexate to the
LSA2L2 regimen with improved outcome (80% EFS).
The value of intensive L-asparaginase in LBL/T-ALL
has been demonstrated in POG and Dana Farber
Cancer Institute (DFCI) cooperative group studies.
High-dose cytarabine has also been shown to have
specific activity against T-NHL/ALL. Results of series
with the best outcomes are summarized in Table 11.5.
Currently, treatment can be expected to result in EFS
rates of approximately 80–90% for children and
50–60% for adults. Hyper-CVAD and variants have
also been evaluated in a predominantly adult group
with LBL. Forty-nine patients (median age 31, range
17–59), with de novo LBL received hyper-CVAD-
based regimens, CNS prophylaxis, radiation therapy
as indicated for bulky mediastinal disease, and main-
tenance therapy. A CR rate of 96% was seen in 46
evaluable patients with 5-year DFS and OS of 72%
and 68%, respectively.
To maximize therapeutic benefits, and in the
absence of prohibitive toxicity, attempts should be
made to deliver specified doses of all prescribed chemo-
therapy agents. Furthermore, doses of methotrexate
and 6-mercaptopurine (6-MP) during maintenance
should be dose-escalated to achieve a targeted degree
of myelosuppression. In the event of significant
chemotherapy-related toxicity, specific agents should
be dose-reduced or discontinued as clinically indicated.
Those individuals with thiopurine S-methyltransferase
deficiency (~10% incidence) require dose reduction of
6-MP to avoid toxicity. Results from EORTC 58951, a
randomized phase III trial with a BFM backbone for
pediatric patients with LBL and ALL, demonstrated
improved OS and DFS in patients who were treated
with pulses of vincristine and corticosteroids during
maintenance.
Recent studies have been designed to eliminate local
radiation to bulky sites of disease. A randomized trial 183
Table 11.5 Selected regimens for lymphoblastic lymphoma.

Pediatric trials
Regimen Treatment Patients (no.) Outcomes by stage
All I/II III/IV III IV
BFM 86 7-drug induction, 54 83% EFS (7-year)
consolidation, reinduction,
maintenance; 24 months
BFM 90 7-drug induction, 105 90% EFS (5-year) 90% EFS (5-year) 90% EFS 95% EFS (5-year)
consolidation, reinduction, (5-year)
maintenance; 24 months
BFM 95 7-drug induction, 156 82% EFS
consolidation, reinduction, 88% DFS 85% OS
maintenance; 24 months (5-year)
CCG 123 – New 4–6 drug induction, 371 67% EFS (7-year)
York I and BFM consolidation, reinduction,
arms maintenance; 24 months
COG A5971 4–5 drug induction, 254 80–84%
consolidation EFS (5-year)
DFCI 87/91/95 4–5-drug induction, 15 93% OS 87% EFS
consolidation, maintenance; 24 71% CCR (5-year)
months (5-year)
EORTC 58881 4-drug induction, 119 86% OS Stage I: 86% OS 84% OS 88% OS
consolidation, reinduction, 78% EFS (6-year) 86% EFS 78% EFS 77% EFS (6-year)
maintenance; 24 months Stage II: (6-year)
100% OS
64% EFS
(6-year)
EORTC 58951 4-drug induction, 411 94% OS 91% DFS
consolidation, reinduction, 83% DFS (6-year)
maintenance; 24 months (6-year) intervention intervention
arm arm
LMT 81 5-drug induction, 82 76% OS 73% EFS 79% EFS 72% EFS (5-year)
consolidation, maintenance; 24 75% EFS (5-year) (5-year) (5-year)
months
LSA2L2 4-drug induction, 95 79% OS 87% EFS 87% 90% OS IVA: 79% OS;
consolidation, maintenance; 24– 75% EFS (5-year) OS (5-year) 85% EFS 73% EFS (5-year)
36 months (5-year) IVB: 74% OS
70% EFS (5-year)
POG 8704 - 6-drug induction, 84 78% CCR (4-year)
Intensive consolidation, maintenance; 24
L-asparaginase months
arm
NHL 13 7-drug induction, 41 90% OS
consolidation, maintenance; 24 83% EFS
months (5-year)
Adult trials
GMALL 89/93 7-drug induction, 45 51% OS 56% OS (7-year) 48% OS (7-year) 49% OS (7-year)
consolidation, reinduction, 62% DFS (7-year)
maintenance; 6–12 months
Hyper-CVAD 4-drug induction, 26 62% PFS (3-year)
consolidation, repeated × 4,
maintenance; 24–36 months
LMT 89 5-drug induction, 27 63% OS 85% OS 37% OS
consolidation, maintenance; 12– 44% PFS (5-year) 67% PFS 21% PFS (5-year)
24 months (5-year)
CCR, Continuous complete remission; DFS, disease-free survival; EFS, event-free survival; OS, overall survival; PFS, progression-free survival.
 Chapter 11: Burkitt and lymphoblastic lymphoma
conducted by the POG failed to show benefit for
involved field radiation for pediatric patients with low-
stage disease. Mediastinal radiation was eliminated
from the treatment of advanced-stage disease on the
pediatric BFM 90 study without an apparent increase
in the local recurrence rate (7%). Notably, the media-
stinum was the primary site of relapse (17%) in adults
treated on recent GMALL trials despite the fact that
most had received 2400 cGy of local radiation, suggest-
ing that chemotherapy is the critical determinant of
bulky site disease control. In contrast, adjuvant radia-
tion did appear to decrease the risk of mediastinal
recurrence for adults treated with the hyper-CVAD
regimen. Although males with testicular ALL/LBL
have historically been treated with radiation, recent
results with regimens containing intermediate- or
high-dose methotrexate suggest that radiation may not
be needed. Thus, in the absence of a protocol-specified
role, radiation should be reserved for emergency man-
agement of life- or organ-threatening complications.
Even in that setting, however, corticosteroids are com-
monly adequate to induce rapid disease reduction.
There have been few studies designed to evaluate the
relative efficacy of high-dose therapy with autologous
SCT for LBL. Most published series represent single-
institution or registry studies that often include mixed
lymphoma subtypes and/or patients with high-risk fea-
tures. The European Group for Blood and Bone
Marrow Transplantation and the United Kingdom
Lymphoma Group conducted a randomized trial of
autologous SCT versus conventional chemotherapy as
post-remission treatment for adults with LBL. Although
there was a trend towards improved relapse-free sur-
vival on the autologous SCT arm, no OS benefit was
demonstrated. Thus, there is no current evidence to
suggest that autologous SCT is more effective than
standard LBL-specific chemotherapy regimens.
Studies of allogeneic SCT are also limited in the
setting of LBL. Most prospective trials compare allo-
geneic to autologous stem cell rescue. In general,
relapse rates are lower after allogeneic SCT; however,
much of this advantage is offset by increased
treatment-related mortality. Consequently, allogeneic
SCT is usually reserved for those with specific high-
risk features. A recent prospective trial based on the
BFM 90 and BFM 95 regimens employed matched
sibling donor SCT in CR1 for pediatric patients with
high-risk T-cell ALL. High-risk was defined as a poor
response to prednisone, non-response on day 33,
186 or chromosomal translocations t(9;22) or t(4;11).
The 5-year DFS was 67% for 36 patients who received
SCT versus 42% for the 120 patients treated with
chemotherapy alone, with OS rates of 67% and 47%,
respectively. This study demonstrates that, for high-
risk T-cell ALL, SCT in CR1 results in excellent out-
comes. In light of the excellent outcome for children
and adolescents with LBL with current chemotherapy
regimens, allogeneic SCT is rarely used for pediatric
patients except for those with poor response to or
relapse after standard treatment.
CNS treatment of lymphoblastic
lymphoma
LBL has a high rate of CNS involvement compared to
other subtypes of lymphoma and leukemia.
Approximately 5–15% of patients have meningeal dis-
ease at presentation; additionally, recent studies from
COG A5971 suggest that these patients have the worst
prognosis. CNS-directed prophylaxis is required for all
patients with LBL because of the high risk of CNS
relapse, even in those without overt meningeal involve-
ment. Historically, craniospinal irradiation has been
relied upon for CNS prophylaxis for patients with
advanced-stage LBL/T-ALL. To reduce the neurocogni-
tive dysfunction encountered in survivors of childhood
ALL, regimens that limit radiation exposure have been
designed over the past two decades. Initial trials led to
reductions in the cranial radiation dose and replacement
of spinal irradiation with intensive intrathecal chemo-
therapy. More recently, cranial radiation has been elim-
inated in all but the highest risk patients. Thus, CNS
prophylaxis for patients with ALL now consists primar-
ily of intrathecal chemotherapy in combination with
systemic agents that have good CNS penetration, most
notably dexamethasone and high-dose methotrexate. It
must be emphasized that most of these data are derived
from patients with B-precursor ALL. However, initial
results from EORTC 58881, BFM 95, and COG A5971
suggest that this approach may be applied effectively to
LBL/T-ALL. EORTC 58951 incorporated the use of tri-
ple intrathecal therapy without any craniospinal irradi-
ation and no CNS relapses have occurred to date. Other
more recent studies, including the St. Jude NHL 13 trial,
also eliminated prophylactic cranial irradiation and had
excellent outcomes and OS comparable to other regi-
mens. Nonetheless, follow-up remains relatively short
for radiation-free regimens and cranial irradiation is
still considered the standard treatment for individuals
with active meningeal leukemia. It is notable that in the

BFM 95 study, the elimination of cranial radiation pro-
phylaxis was associated with an increase in bone marrow
relapse, the cause of which is not completely understood.
Finally, in light of both the poorer overall outcome and
the lower risk of neurologic toxicity in older individuals,
cranial irradiation is still commonly employed for adults
with ALL and LBL.
A number of practical aspects of intrathecal
administration are worthy of comment. To minimize
the risk of meningeal contamination from traumatic
lumbar puncture, spinal taps should be performed by
experienced practitioners. Furthermore, it is recom-
mended that intrathecal chemotherapy be adminis-
tered at the time of the initial diagnostic lumbar
puncture, especially when there are circulating blasts
in the peripheral blood. To improve chemotherapy
distribution, the volume of cerebrospinal fluid (CSF)
removed should equal the volume administered and
patients should remain prone for approximately 30
minutes after administration. The use of an Ommaya
reservoir may improve delivery to the ventricles,
although this approach is not routinely employed.
Finally, intrathecal chemotherapy should be dosed
according to age owing to age-related CSF volume
changes.
Management of relapse in lymphoblastic
lymphoma
In general, the outcome after relapse is poor for chil-
dren and adults with LBL. In a retrospective analysis of
pediatric patients with LBL treated by the BFM
between 1990 and 2003, the salvage rate for those
with progressive disease or relapse was poor, with a
14% OS rate. Long-term survival was only achieved in
those few patients who were able to undergo an allo-
geneic SCT. Those patients with low-stage disease who
relapse after abbreviated therapy have higher salvage
rates. Allogeneic SCT is recommended for patients
with HLA-matched related donors. In addition,
matched unrelated donor transplantation should be
considered for young individuals. Agents targeted to
T-cell malignancies hold promise. Nelarabine (com-
pound 506U78), a water-soluble prodrug of ara-G,
which was approved by the FDA for treatment of
relapsed T-LBL/ALL in 2005, has single agent activity
in pediatric and adult patients with relapsed/refractory
LBL/T-ALL. An objective response rate of 33% was
observed in pediatric patients (CR 26%; PR 7%) and a
31% CR rate in adults.
Chapter 11: Burkitt and lymphoblastic lymphoma
Treatment recommendations for
lymphoblastic lymphoma
Therapy should be instituted as soon as possible after
diagnosis and should be directed by practitioners
experienced with the specific regimen and the manage-
ment of expected complications. It is recommended
that patients be treated on clinical trials whenever
possible. The best reported outcome for LBL is with
aggressive ALL-type regimens (Table 11.5), and recent
BFM regimens appear superior. In general, treatment
is similar for adult and pediatric patients, although, as
detailed above, therapy is often stratified based on risk
of relapse and organ toxicity.
Supportive care of Burkitt and
lymphoblastic lymphoma
Aggressive supportive care and surveillance for com-
plications during and after treatment is critical to
successful outcome.
Antiemetics: routine prophylaxis and treatment of
nausea and vomiting are required during induction,
consolidation, and CNS-directed therapy phases.
Tumor lysis syndrome: as LBL and BL frequently
present with a large tumor burden and cell turnover is
extremely rapid, tumor lysis syndrome is commonly
encountered during initiation of therapy. Some regi-
mens include a “pre-phase” of treatment to allow grad-
ual induction, for example with a few day courses of
corticosteroids alone. Tumor lysis prophylaxis should
be started as soon as possible after diagnosis and at least
12 hours prior to the initiation of induction chemo-
therapy. Prophylaxis and monitoring should continue
until disease burden is reduced and it is apparent that
no complications of lysis have developed (usually 3–7
days). The following is recommended:
* Allopurinol: 100 mg/m2/dose PO TID. Urate
oxidase (Rasburicase) is an alternative for
management of extreme hyperuricemia.
* Hydration: IV fluids at a rate of ≥2 times
maintenance (≥120 ml/m2/hour) adjusted to
maintain urine specific gravity ≤1.010 and normal
urine output. Because of the risk of hyperkalemia,
potassium should be avoided.
* Serum potassium, phosphorus, calcium, creatinine,
blood urea nitrogen (BUN), and uric acid should be
assayed every 4–6 hours for the first 24–48 hours,
then less frequently once stable. 187
 Chapter 11: Burkitt and lymphoblastic lymphoma
Transfusions: blood transfusion therapy is usually
required to counter severe cytopenias associated with
induction and consolidation phases of therapy.
Specialized blood products should be employed in
attempt to decrease the risk of transfusion-associated
complications.
* Platelets: to prevent bleeding, platelet counts
should routinely be maintained above 10 000/μL.
Higher levels may be needed to manage active
bleeding, prior to invasive procedures, and to
reduce the risk of leukostasis-induced CNS
hemorrhage in the setting of hyperleukocytosis.
Single donor platelets are recommended whenever
possible to decrease donor exposure and the risk of
HLA-alloimmunization.
* Red blood cells: concomitant anemia often partially
offsets the hyperviscosity associated with severe
hyperleukocytosis. Thus, red blood cell transfusion
should be avoided if possible when the white blood
count is >100 000/μL. If red cell transfusion is
needed, the hemoglobin should be increased slowly
using small aliquots of packed red cells until the
peripheral blast count is reduced.
* Irradiation: all cellular blood products should be
irradiated to prevent transfusion-associated graft-
versus-host disease.
* Leukodepletion: platelets and red cells should be
leukocyte-reduced to decrease the risk of febrile
reactions and platelet-refractoriness due to HLA-
alloimmunization.
Infectious prophylaxis: aggressive surveillance, prophy-
laxis, and treatment for bacterial, fungal, viral, and
opportunistic infections is required throughout
therapy.
* Pneumocystis jiroveci pneumonia (PCP): all
patients should receive prophylaxis against PCP
until approximately 6 months after completion of
chemotherapy.
* Neutropenic fever: patients with fever in the setting
of an absolute neutrophil count (ANC) <500/μL
require emergent evaluation and management for
possible infection. Immediate, empiric, broad-
spectrum, intravenous antibiotics are indicated.
Antifungal therapy should be initiated for
persistent neutropenic fever beyond 5–7 days. In
general, antibiotics should be continued until the
ANC rises to >500/μL, fever resolves, cultures are
188 negative, and any infection is fully treated.
* Myeloid growth factors: myeloid growth factor
support (e.g. G-CSF) is commonly employed as
part of treatment of BL in both children and adults.
However, the benefits of such have not been proved
in the setting of LBL therapy.
Chemotherapy-prophylaxis: agent-specific prophylaxis
should be utilized as clinically indicated (e.g. gastritis
prophylaxis during corticosteroids).
Nutritional support: nutritional status should be
monitored and supplementation provided as needed.
Routine folic acid use should be avoided around the
time of methotrexate administration as this may coun-
teract the therapeutic efficacy of folate antagonism.
In contrast, leucovorin rescue is required after inter-
mediate- and high-dose methotrexate.
Response and toxicity evaluations: serial monitor-
ing for response- and therapy-associated toxicity
should be conducted during and after treatment.
Life-long follow-up for possible late effects is indicated
for long-term survivors, including the following:
* Cardiomyopathy: to decrease the risk of
cardiotoxicity, cumulative anthracycline doses are
usually limited to <400 mg/m2. Left ventricular
function should be monitored.
* Neurologic toxicity: children are at high risk of
neurotoxicity from CNS-directed therapy and
neurodevelopmental assessment is required.
* Endocrinologic dysfunction: patients should be
monitored for endocrinopathies, including growth
retardation, infertility, and hormone deficiencies.
* Osteonecrosis: corticosteroids are associated with a
high incidence of osteonecrosis.
* Secondary malignancy: long-term survivors are at
risk of secondary malignancies even beyond the
first decade after treatment.
Further reading
Amylon MD, Shuster J, Pullen J, et al. Intensive high-dose
asparaginase consolidation improves survival for
pediatric patients with T cell acute lymphoblastic
leukemia and advanced stage lymphoblastic lymphoma: a
Pediatric Oncology Group study. Leukemia,
1999;13:335–342.
Burkhardt B, Woessmann W, Zimmermann M, et al.
Impact of cranial radiotherapy on central nervous
system prophylaxis in children and adolescents
with central nervous system-negative stage III or
IV lymphoblastic lymphoma. J Clin Oncol,
2006;24:491–499.

Dave SS, Fu K, Wright GW, et al. Molecular diagnosis
of Burkitt’s lymphoma. N Engl J Med,
2006;354:2431–2442.
Divine M, Casassus P, Koscielny S, et al. Burkitt lymphoma
in adults: a prospective study of 72 patients treated with
an adapted pediatric LMB protocol. Ann Oncol,
2005;16:1928–1935.
Gaynon PS, Trigg ME, Heerema NA, et al. Children’s Cancer
Group trials in childhood acute lymphoblastic leukemia:
1983–1995. Leukemia, 2000;14:2223–2233.
Goldberg JM, Silverman LB, Levy DE, et al. Childhood T-cell
acute lymphoblastic leukemia: the Dana-Farber Cancer
Institute acute lymphoblastic leukemia consortium
experience. J Clin Oncol, 2003;21:3616–3622.
Goldman S, Smith L, Anderson JR, et al. Rituximab and
FAB/LMB96 chemotherapy in children with Stage III/IV
B-cell non-Hodgkin lymphoma: a Children’s Oncology
Group report. Leukemia, 2013;27:1174–1177.
Hoelzer D, Ludwig WD, Thiel E, et al. Improved outcome in
adult B-cell acute lymphoblastic leukemia. Blood,
1996;87:495–508.
Hoelzer D, Gokbuget N. Treatment of lymphoblastic
lymphoma in adults. Best Pract Res Clin Haematol,
2002;15:713–728.
Kluin PM, Harris NL, Stein H, et al. B-cell lymphoma,
unclassifiable, with features intermediate between diffuse
large B-cell lymphoma and Burkitt lymphoma. In
Swerdlow SH, Campo E, Harris NL, et al. (eds). WHO
Classification of Tumours of Haemopoietic and Lymphoid
Tissues. Lyon: IARC, 2008;265–266.
Hummel M, Bentink S, Berger H, et al. A biologic definition
of Burkitt’s lymphoma from transcriptional and genomic
profiling. N Engl J Med, 2006;354:2419–2430.
Jabbour E, Koscielny S, Sebban C, et al. High survival rate
with the LMT-89 regimen in lymphoblastic lymphoma
(LL), but not in T-cell acute lymphoblastic leukemia
(T-ALL). Leukemia, 2006;20:814–819.
Jaffe ES, Harris NL, Stein H, Vardiman JW. World Health
Organization Classification of Tumours. Pathology and
Genetics of Tumours of Hematopoietic and Lymphoid
Tissues. Lyon: IARC Press, 2001.
Leoncini L, Raphael M, Stein H, et al. Burkitt lymphoma. In
Swerdlow SH, Campo E, Harris NL, et al. (eds). WHO
Classification of Tumours of Haemopoietic and Lymphoid
Tissues. Lyon: IARC, 2008; 262–264.
Magrath I, Adde M, Shad A, et al. Adults and children with
small non-cleaved-cell lymphoma have a similar excellent
outcome when treated with the same chemotherapy
regimen. J Clin Oncol, 1996;14:925–934.
Mead GM, Sydes MR, Walewski J, et al. An international
evaluation of CODOX-M and CODOX-M alternating
with IVAC in adult Burkitt’s lymphoma: results of United
Chapter 11: Burkitt and lymphoblastic lymphoma
Kingdom Lymphoma Group LY06 study. Ann Oncol,
2002;13:1264–1274.
Millot F, Suciu S, Philippe N, et al. Value of high-dose
cytarabine during interval therapy of a Berlin–Frankfurt–
Munster-based protocol in increased-risk children with
acute lymphoblastic leukemia and
lymphoblastic lymphoma: results of the European
Organization for Research and Treatment of Cancer
58881 randomized phase III trial. J Clin Oncol,
2001;19:1935–1942.
Mora J, Filippa DA, Qin J, Wollner N. Lymphoblastic
lymphoma of childhood and the LSA2-L2
protocol: the 30-year experience at Memorial-
Sloan-Kettering Cancer Center. Cancer,
2003;98:1283–1291.
Mullighan CG, Downing JR. Global genomic
characterization of acute lymphoblastic leukemia.
Semin Hematol, 2009;46:3–15
Murphy SB, Bowman WP, Abromowitch M, et al. Results of
treatment of advanced-stage Burkitt’s lymphoma and B
cell (SIg+) acute lymphoblastic leukemia with high-dose
fractionated cyclophosphamide and coordinated high-
dose methotrexate and cytarabine. J Clin Oncol,
1986;4:1732–1739.
Patte C, Kalifa C, Flamant F, et al. Results of the LMT81
protocol, a modified LSA2L2 protocol with high dose
methotrexate, on 84 children with non-B-cell
(lymphoblastic) lymphoma. Med Pediatr Oncol,
1992;20:105–113.
Patte C, Auperin A, Michon J, et al. The Societe Francaise
d’Oncologie Pediatrique LMB89 protocol: highly
effective multiagent chemotherapy tailored to the tumor
burden and initial response in 561 unselected children
with B-cell lymphomas and L3 leukemia. Blood,
2001;97:3370–3379.
Pilozzi E, Pulford K, Jones M et al. Co-expression of CD79a
(JCB117) and CD3 by lymphoblast lymphoma. J Pathol,
1998;186:140–143.
Reiter A, Schrappe M, Ludwig WD, et al. Intensive
ALL-type therapy without local radiotherapy provides a
90% event-free survival for children with T-cell
lymphoblastic lymphoma: a BFM group report. Blood,
2000;95:416–421.
Reiter A, Schrappe M, Parwaresch R, et al.
Non-Hodgkin’s lymphomas of childhood and adolescence:
results of a treatment stratified for biologic subtypes
and stage: a report of the Berlin–Frankfurt–Munster Group. J
Clin Oncol, 1995;13:359–372.
Reiter A, Schrappe M, Tiemann M, et al. Improved
treatment results in childhood B-cell neoplasms with
tailored intensification of therapy: a report of the Berlin–
Frankfurt–Munster Group Trial NHL-BFM 90. Blood,
1999;94:3294–3306.
189
 Chapter 11: Burkitt and lymphoblastic lymphoma
Rizzieri DA, Johnson JL, Niedzwiecki D, et al. Intensive
chemotherapy with and without cranial radiation for
Burkitt leukemia and lymphoma: final results of Cancer
and Leukemia Group B Study 9251. Cancer,
2004;100:1438–1448.
Schrappe M, Reiter A, Ludwig W, et al. Improved
outcome in childhood acute lymphoblastic leukemia
despite reduced use of anthracyclines and cranial
radiotherapy: results of trial ALL-BFM 90. Blood,
2000;95:3310–3322.
Steinherz PG, Gaynon PS, Breneman JC, et al. Treatment of
patients with acute lymphoblastic leukemia with bulky
extramedullary disease and T-cell phenotype or other
poor prognostic features: randomized controlled trial
from the Children’s Cancer Group. Cancer,
1998;82:600–612.
Termuhlen AM, Smith LM, Perkins SL, et al. Br J Hematol.
2013;ePub ahead of print.
190
Thomas DA, Cortes J, O’Brien S, et al. Hyper-CVAD program
in Burkitt’s-type adult acute lymphoblastic leukemia. J Clin
Oncol, 1999;17:2461–2470.
Thomas DA, O’Brien S, Cortes J, et al. Outcome with the
hyper-CVAD regimens in lymphoblastic lymphoma.
Blood, 2004;104:1624–1630.
Uyttebroeck AS, Suciu S, Laureys G, et al.
Treatment of childhood T-cell lymphoblastic
lymphoma according to the strategy for acute
lymphoblastic leukaemia, without radiotherapy:
long term results of the EORTC CLG 58881 trial. Eur J
Cancer, 2008;44(6):840–846.
van Imhoff GW, van der Holt B, MacKenzie MA, et al.
Short intensive sequential therapy followed by
autologous stem cell transplantation in adult Burkitt,
Burkitt-like and lymphoblastic lymphoma. Leukemia,
2005;19:945–952.
 Chapter
12
Therapy of diffuse large B-cell lymphoma
John W. Sweetenham
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Introduction
Diffuse large B-cell lymphoma (DLBCL) is the most
common type of non-Hodgkin’s lymphoma (NHL),
accounting for 35–40% of all cases of NHL. Despite
high reported response rates to anthracycline-based
combination chemotherapy regimens, only 50–65%
of patients with this disease have achieved long-term
disease-free survival (DFS) with this approach. The
emergence of new strategies, including monoclonal
antibodies, dose-dense chemotherapy approaches,
and the identification of new rational therapeutic tar-
gets by gene expression profiling, has resulted in
improvements in outcome for patients with this dis-
ease in recent years.
Involvement of extranodal sites, either as a primary
site of disease or as sites of dissemination, is relatively
common in DLBCL. With the exception of primary
central nervous system (CNS) lymphoma (see
Chapter 13) and some other specific anatomic sites,
treatment recommendations for DLBCL are generally
identical for nodal and extranodal disease, with the
exception of single site non-lower limb DLBCL of the
skin (see Chapter 15).
Clinical presentation
The clinical presentation of DLBCL, as with other types
of NHL, is most commonly with painless lymphaden-
opathy. Approximately 25% of patients present with
anatomically limited stage disease (clinical stage I or
II), with the remaining 75% having more advanced
(bulky stage II or stage III or IV) disease. Many patients
will also experience constitutional (“B”) symptoms,
including drenching night sweats, unexplained fevers,
and unexplained weight loss of more than 10% of body
weight. Since extranodal disease is relatively common in
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
DLBCL, and since almost any organ can be affected by
this disease, presenting symptoms may mimic many
other diseases. For example, patients with primary
mediastinal large B-cell lymphoma may present with
chest discomfort, respiratory obstruction, and symp-
toms and signs of superior vena caval obstruction.
Patients with extranodal disease in the gastrointestinal
(GI) tract may present with abdominal discomfort, GI
bleeding, or evidence of intestinal obstruction. Patients
with DLBCL may therefore be diagnosed through many
different subspecialties.
Pathology
Diffuse large B-cell lymphoma not
otherwise specified (NOS)
DLBCL replaces the nodal architecture, or that of
extranodal sites of involvement, by a proliferation of
large cells with a diffuse growth pattern. Several mor-
phological variants have been described based on the
morphological appearance of the cells, the cellular
background, and the immunophenotype (Figure 12.1).
Several morphological variants are recognized, but
subclassification remains of uncertain clinical signifi-
cance, although a recent study has suggested that immu-
noblastic lymphomas have a worse clinical outcome.
Centroblastic lymphoma is characterized by sheets of
large cells with large vesicular nuclei that contain prom-
inent nucleoli that are mainly associated with the
nuclear membrane. The nuclei are usually round or
oval but in some cases may show prominent nuclear
lobulation (polylobated centroblastic lymphoma-type)
(Figure 12.2). In the immunoblastic variant the vast
majority of cells (>90%) are immunoblasts, with a mod-
erate amount of basophilic cytoplasm and a round
nucleus that contains a single, central eosinophilic
191
 Chapter 12: Diffuse large B-cell lymphoma
Figure 12.1 Diffuse large B-cell lymphoma. A sheet of large
lymphoid cells with pale cytoplasm and large pleomorphic nuclei
containing prominent nucleoli.
nucleolus. The cells of this variant of DLBCL have
abundant cytoplasm. Their nuclei are pleomorphic
and may resemble Reed–Sternberg cells or cells of ana-
plastic T-cell lymphoma. Multinucleate giant cells may
be seen. There may be a sinusoidal pattern of infiltration
within nodes.
Characteristically, DLBCL express CD45 with the
pan-B-cell markers CD19, CD20, CD22, CD79a, and
PAX5. Monotypic surface and/or cytoplasmic immu-
noglobulin is demonstrable in a majority of cases.
Combinations of staining for CD10, BCL-6, and
MUM1 have been used as an immunocytochemical
surrogate for the identification of germinal center-
like phenotype (CD10+ or CD10–, BCL-6+ and
MUM1–) as opposed to non-germinal center (acti-
vated) B-cell-like DLBCL recapitulating the separa-
tion seen by studies using microarray techniques. The
separation of germinal center-like and activated
B-cell-like categories can be enhanced by the addition
of staining for FOXP1 and GCET. Expression of
BCL-2 protein is present in 30–50%. A proportion
of cases express CD5 but these are negative for cyclin
D1. Expression of CD30 is seen in occasional cases
with non-anaplastic morphology. Lymphomas with
anaplastic morphology stain for CD30 but are
negative for CD15 and ALK1, distinguishing them
from classical Hodgkin and anaplastic large cell
lymphoma. Staining for CD23 may also be seen in
some cases (up to 16% in some series). Expression
of CD138 may be seen in cases showing plasmablastic
192 morphology but is rarely seen in other cases.
Figure 12.2 Diffuse large B-cell lymphoma. A sheet of large cells
with abundant pale eosinophilic cytoplasm and nuclei that show
marked lobulation.
T-cell/histiocyte-rich B-cell lymphoma
(TCHRBCL)
In this variant the neoplastic B-cells form a small per-
centage (usually <10%) of the cellular population while
the majority of the cells are T-lymphocytes and/or
histiocytes. The large cells may have a variety of appear-
ances, including centroblastic or immunoblastic, and in
some cases they may resemble Reed–Sternberg cells or
the cells of lymphocyte-predominant Hodgkin lym-
phoma. Small B-cells are essentially absent. The cells
express the pan-B-cell markers (CD19, CD20, CD22,
CD79a, and PAX5). Staining for BCL-6 is frequently
seen but the cells are usually negative for CD10.
Staining for BCL-2 is seen in a proportion of cases.
Infrequently there is expression of CD30 but the cells
lack CD15 and there is expression of epithelial mem-
brane antigen (EMA) in about one-third of cases.
These cases need to be distinguished from classical
and lymphocyte-predominant Hodgkin lymphoma.
Immunophenotypic studies help distinguish these
cases from classical Hodgkin lymphoma, as the latter
express CD30 and CD15 and lack the full B-cell-
related antigen profile, while TCHRBCL express
CD20 with CD79a with staining for OCT2 and
BOB1 and they are positive for CD45 with staining
for EMA in a proportion of cases. The presence of
nodular areas with large cells within areas containing
small B-cells and follicular dendritic cell meshworks
raises the possibility of lymphocyte-predominant
Hodgkin lymphoma.

Diffuse large B-cell lymphoma with
expression of full length anaplastic
lymphoma kinase (ALK)
The cells of this variant resemble centroblasts, immu-
noblasts, or plasmablasts and in some cases may have a
more anaplastic appearance, but in general they have
more abundant cytoplasm than these more classical
variants. The cells are weakly positive for CD45 and
lack CD20 and CD79a. They are frequently positive for
EMA but are negative for CD30. The cells contain
cytoplasmic IgA. Staining for ALK kinase protein
shows a granular cytoplasmic pattern. There is no
expression of T-cell-related antigens.
Intravascular large B-cell lymphoma
These lymphomas can be difficult to diagnose, with the
infiltrate being quite subtle. Biopsies are frequently from
extranodal sites, especially skin. The infiltrate is essen-
tially confined to the lumen of small-/intermediate-sized
vessels although minimal extravascular infiltration may
be seen. The cells are medium to large, morphologically
resembling centroblasts or immunoblasts, with some
showing more anaplastic features. They may be cohesive
within the lumen or show a palisaded pattern in the
periphery. Occlusion of the vessel lumen either by
tumor alone or in association with fibrin thrombus
may induce tissue necrosis.
The cells are positive for CD45 and express pan-
B-cell markers (CD19, CD20, CD79, and PAX5) and
stain for BCL-2 protein. The majority stain for IRF4/
MUM1 but a small proportion of cases (10–20%) is
positive for CD10 and BCL-6. Cases with CD5 expres-
sion are seen but these are negative for CD23 and
cyclin D1 (Figure 12.3).
Diffuse large B-cell lymphoma associated
with inflammation
These lymphomas usually occur in the context of long-
term pyothorax but can occur associated with other
forms of prolonged inflammatory disorder. The mor-
phology of the cells is indistinguishable from those of
DLBCL-NOS although some cases can have plasma-
blastic features. Necrosis is common and an angiocen-
tric distribution can be seen. The cells usually express
pan-B-cell markers (CD19, CD20, CD79a, and PAX5),
although these may be lost in cases with more plasma-
blastic differentiation. The majority express IRF4/
MUM1 and there is variable expression of CD30
Chapter 12: Diffuse large B-cell lymphoma
Figure 12.3 Intravascular large B-cell lymphoma. Clusters of large
cells confined within vascular spaces. The cells have pleomorphic
nuclei with prominent nucleoli.
(CD15 is negative). A small minority of cases has
been associated with aberrant expression of T-cell-
related antigens. The cells are positive for EBER and
LMP1.
EBV-positive diffuse large B-cell
lymphoma of the elderly
There is effacement of the lymph node architecture
frequently associated with geographic areas of
necrosis. The infiltrate may be monotonous and com-
posed of large cells or there may be an associated
polymorphic background, but there is no clinical sig-
nificance to the monomorphic or polymorphic var-
iants. The cells are positive for B-cell markers (CD19,
CD20, CD79a, and PAX5) and usually express IRF4/
MUM1 while staining for CD10 and BCL-6 are rare.
There is variable expression of CD30 without CD15.
CD20 may be lost in cases with plasmablastic mor-
phology. In the vast majority of cases the cells are
positive for EBER and LMP1.
Mediastinal (thymic) large B-cell
lymphoma
There is diffuse infiltration of the mediastinum (or
other extranodal sites with dissemination) by a pop-
ulation of cells with variable size. Most of the cells are
large and in many cases the nuclei are lobulated.
Characteristically the cytoplasm is abundant and
pale. Sclerosis is usually present but the pattern is 193
variable, ranging from pericellular to broad sclerotic
 Chapter 12: Diffuse large B-cell lymphoma
Figure 12.4 Mediastinal (thymic) large B-cell lymphoma
(mediastinal mass). Sheets of large cells with pale/clear cytoplasm
and large lobulated nuclei separated by fibrous bands with focal
necrosis.
bands separating the infiltrate into cellular nodules.
Eosinophils may be associated with the neoplastic
cell infiltrate and some cases may morphologically
resemble Hodgkin lymphoma of nodular sclerosis
subtype (Figure 12.4).
Immunophenotypically the cells express CD45 and
the B-cell antigens CD19, CD20, CD22, CD79a, and
PAX5. Staining for BCL-6 is most often positive but
CD10 is infrequently expressed. Staining for CD5 is
usually negative. There may be staining for EMA.
CD30 is frequently expressed either focally or in all
cells, but this is usually weaker than is seen in classical
Hodgkin lymphoma. Up to 70% of cases express CD23.
Cytoplasmic and surface immunoglobulin is usually
absent. Staining for MAL is positive in about 70% of
cases and is rare in other types of large B-cell lymphoma.
Molecular pathology and cytogenetics
DLBCL is a heterogeneous disease clinically and, not
surprisingly, this is also reflected on the molecular
level. No unifying genetic alteration has been uncov-
ered so far in this aggressive lymphoma; however,
several chromosomal translocations, genomic copy
number changes, and molecular alterations occur in
DLBCL with increased frequency. The hallmark trans-
location in follicular lymphoma (FL), the t(14;18)(q32;
q21), is present in approximately 20% of DLBCL and
this genetic alteration is strongly associated with the
germinal-center type of DLBCL (GCB DLBCL), as
194 defined by gene expression profiling. DLBCL carrying
the t(14;18) may arise by transformation of a pre-
existing FL, but this translocation presumably occurs
also in de novo DLBCL. The BCL-6 oncogene, located
in the chromosomal band 3q27, is rearranged by trans-
location in approximately 30–40% of DLBCL cases.
Translocations of BCL-6 involve immunoglobulin
heavy and light chain gene loci (in 14q32, 2p12, and
22q11) in half of the cases, whereas a number of alter-
native, non-immunoglobulin BCL-6 translocation
partners have also been described. BCL-6 is typically
expressed in germinal-center B-cells, and BCL-6 trans-
locations may exert an oncogenic effect by preventing
BCL-6 downregulation, which occurs in normal
B-cells upon terminal differentiation. As an alternative
way of BCL-6 deregulation in DLBCL, the regulatory
region of BCL-6 can harbor mutations, some of which
may significantly deregulate BCL-6 expression. The
Burkitt translocation t(8;14)(q24;q32) involving the
c-myc oncogene is also encountered in a small subset
of DLBCL (5–10%), and occasionally the distinction
between a DLBCL carrying a c-myc translocation and a
bona fide Burkitt lymphoma can be difficult (see
Chapter 11).
An important aspect of the molecular pathogenesis
of DLBCL may be the observation that 50% of these
tumors show aberrant somatic mutation of various
oncogenes and proto-oncogenes. Physiological
somatic hypermutation occurs in normal germinal-
center B-cells and affects predominantly immunoglo-
bulin genes in a process termed affinity maturation. In
DLBCL, however, the hypermutation machinery
appears aberrantly activated, leading to mutational
events in additional genes such as C-MYC, PIM1,
PAX5, and RhoH/TTF, potentially resulting in their
oncogenic activation.
In recent years, deeper insights into the molecular
heterogeneity of DLBCL were gleaned by gene expres-
sion profiling studies using the DNA microarray tech-
nology. In particular, two major DLBCL subgroups
could be discerned that differ in their underlying
molecular features and in their clinical behavior.
GCB DLBCL shows similarity in its global gene
expression pattern to normal germinal-center B-cells,
while the other subgroup, termed activated B-cell-like
DLBCL (ABC DLBCL), displays gene expression fea-
tures of mitogenically in vitro-activated B-cells. In a
retrospective analysis, the clinical outcome of GCB
and ABC DLBCL patients varied considerably, with
5-year survival times of 60% (GCB DLBCL) and 35%
(ABC DLBCL), respectively. Beyond major differences

in their global gene expression profiles, additional
molecular features of the tumors point to a profound
biological difference between GCB and ABC DLBCL.
In particular, as mentioned above, translocations of
the BCL-2 oncogene are strongly associated with the
GCB-type of DLBCL, as are amplifications of a chro-
mosomal region in 2p that harbors the REL and BCL-
11A loci. Additional genetic differences between GCB
and ABC DLBCL include frequent chromosomal gains
of 12q in GCB DLBCL, whereas chromosomal gains/
amplifications of 3q and 18q are predominantly
detected in the ABC DLBCL subtype. In contrast to
GCB DLBCL, ABC DLBCL is characterized by con-
stitutive activation of the oncogenic NF-κB pathway.
Cell lines representative of this DLBCL subtype can be
killed by blocking NF-κB, suggesting that this pathway
may represent a potential new therapeutic target, e.g.
for proteasome inhibitors (PS341).
Functional studies of GCB and ABC DLBCL cell
lines have revealed additional molecular differences.
For example, phosphodiesterase 4B (PDE4B), an inac-
tivator of cyclic AMP (cAMP) which mediates apop-
tosis in B-cells via AKT inactivation, is highly
expressed in ABC DLBCL. As a result, activation of
the cAMP pathway in GCB DLBCL cell lines leads to
apoptotic cell death whereas there is no effect in ABC
DLBCL lines that have PDE4B expression. Likewise,
the response to interleukin-4 stimulation appears to be
different between the two subtypes.
The transcriptional heterogeneity of DLBCL is fur-
ther highlighted by the description of additional gene
expression signatures that are variably expressed
between these tumors. Monti and colleagues, for
example, defined “oxidative phosphorylation,” “B-cell
receptor proliferation,” and “host response” clusters
that show variable expression in DLBCL tumors. In
the clinical setting, the gene expression profiling
approach may also be used to predict survival of
DLBCL at the time of diagnosis. By combining gene
expression signatures that capture the germinal-center
phenotype, the proliferative activity, MHC class II
expression, and the host response, a powerful mathe-
matical predictor can be constructed that identifies
DLBCL patients with particularly favorable or poor
survival. Genetic alterations, e.g. gains of 3p, may
improve such survival predictors, indicating that
genetic alterations may not be generally inherent in
global gene expression profiles.
The more favorable prognosis of patients with the
GCB DLBCL phenotype is still evident when current
Chapter 12: Diffuse large B-cell lymphoma
treatment regimens, specifically R-CHOP, are used.
Moreover, the composition of the microenvironment
(“stromal signature”) appears to be of prognostic
importance. In the prognostically less favorable ABC
DLBCL subtype, activating or inactivating mutations of
genes involved in the NF-κB pathway (A20, CARD11,
TRAF2, TRAF5, RANK) have been described in a sig-
nificant proportion of cases. Additional mutations were
discovered in B-cell receptor-associated molecules
(CD79B), leading to “chronic active” B-cell receptor
signaling and in MYD88, an adaptor protein involved
in toll/interleukin-1 receptor signaling.
Staging investigations
Recommended staging investigations for newly pre-
senting patients with DLBCL are summarized in
Table 12.1.
Present treatment approaches to DLBCL are still
largely based upon anatomic stage, and, particularly in
the context of clinical trials, on the risk factors
described by the International Prognostic Index
(IPI), as described below. Emerging data on the poten-
tial value of 18fluoro-deoxyglucose positron emission
tomography (FDG–PET) as a staging and response
evaluation technique in this disease is likely to modify
determination of disease extent in the future. It is
therefore important to emphasize that new treatment
strategies may emerge as new data on the value of
FDG–PET appear. This subject is discussed more
fully in Chapter 3.
Table 12.1 Standard staging investigations in DLBCL.
Complete history and physical examination
Determination of performance status
Complete blood count and differential white cell count
Serum biochemistry profile
Lactate dehydrogenase
Serum β2 microglobulin
Serum protein electrophoresis and immunofixation
Chest X-ray
CT scan neck, chest, abdomen, pelvis
Bone marrow aspirate and core biopsy for routine
histopathology, flow cytometry, cytogenetics
Other imaging techniques, including magnetic resonance
imaging as indicated clinically
Potential future role for FDG–PET
195
 Chapter 12: Diffuse large B-cell lymphoma
Prognostic factors
Since the description of the IPI, described in detail in
Chapter 2, most clinical trials in DLBCL have been
stratified according to this index, or have included
patients with specific risk groups as defined by the
IPI. For patients aged 60 years or less, in whom
intensive treatment strategies have been widely
used, the age-adjusted IPI (aaIPI) is a frequently
used predictive model. Treatment strategies for clin-
ically limited (stage I and II) disease have emerged
separately from those for patients with more
advanced-stage disease, and a “stage-adjusted” IPI
has also been described in patients with stage I and
II DLBCL.
The utility of the IPI for patients treated since the
introduction of rituximab has been assessed in several
studies, with varying conclusions. This has resulted in
the description of a revised IPI, in which three rather
than four major prognostic groups are recognized. At
present, the original IPI remains the most widely used
clinical model.
Although the IPI has proved to be a useful model
for risk stratification in clinical trials in DLBCL, there
is marked variability in outcome within each of the
risk groups identified in the IPI, reflecting the under-
lying biological heterogeneity of this disease. This has
limited the clinical utility of the IPI, and has led many
investigators to study “biological” prognostic factors
in DLBCL by the use of techniques such as gene
expression profiling (GEP) and tissue microarray
(TMA) studies. These techniques have allowed iden-
tification of distinct biological entities within
DLBCLs, with possible clinical and prognostic sig-
nificance, particularly related to the cell of origin of
the lymphoma. Emerging data suggest that the effec-
tiveness of some treatments may vary according to
the cell of origin of DLBCL, as defined by GEPs or
immunohistochemistry. For example, preliminary
data suggest that the activity of bortezomib in
DLBCL may be restricted to patients with a non-
germinal-center phenotype. Overexpression of ther-
apeutic targets such as Bruton’s tyrosine kinase (Btk)
appear to be restricted to non-germinal-center
DLBCL, and clinical trials are being developed to
investigate targeted therapies in this group.
At present, although some evidence suggests that
distinct treatment approaches should be used in
patients with early-stage compared with advanced-
stage disease, there is no clear evidence that risk-
196 adapted treatment of this disease has improved
outcomes for specific risk groups. As a result, standard
treatment approaches are still based primarily on ana-
tomic stage of disease.
Treatment of early-stage DLBCL
Interpretation of clinical trials in patients with early-
stage DLBCL has been complicated by the variable
definition of limited disease. Many of the major clin-
ical trials that have established treatment approaches
for this disease were designed before the description of
the WHO classification and the IPI, and are heteroge-
neous with respect to inclusion of patients with vari-
ous diffuse aggressive lymphomas. Since DLBCL is
likely to be the predominant histological subtype in
these trials, the results have generally been considered
to be relevant to the management of DLBCL.
Early studies of the treatment of limited stage
DLBCL used involved-field radiation therapy
(IF-RT), with most patients relapsing at sites distant
from the irradiated area. Combination chemotherapy
was therefore introduced in addition to radiation ther-
apy in an attempt to control clinically undetected dis-
ease at distant sites. The use of combined modality
therapy with three cycles of CHOP (cyclophospha-
mide, doxorubicin, vincristine, prednisone) followed
by IF-RT was compared with standard chemotherapy
using eight cycles of CHOP in a study from the
Southwest Oncology Group (SWOG). Localized dis-
ease was defined as non-bulky stage I or II disease.
Disease bulk was defined as any mass measuring more
than 10 cm in maximum diameter, or a mediastinal
mass measuring more than one-third of the trans-
thoracic diameter on a standard PA chest X-ray. The
5-year progression-free survival (PFS) was 77% for the
combined modality arm, versus 64% for chemother-
apy alone (P = 0.03). The corresponding figures for
overall survival (OS) were 82% and 72% (P = 0.02).
This study established combined modality therapy
with three cycles of CHOP chemotherapy followed
by IF-RT as the standard of care for most patients
with localized DLBCL. However, longer term follow-
up of this study has shown that the early survival
advantage associated with combined modality therapy
has not been maintained, primarily because of late
relapses in the combined modality group, most of
which occurred outside the radiation field.
Other trials have investigated the need for IF-RT in
this patient population.
A recent study from the Groupe d’Etude des
Lymphomes de l’Adulte (GELA) included 647 patients

with localized aggressive NHL (81% of whom had
DLBCL) who were randomized to receive either
three cycles of CHOP chemotherapy followed by IF-
RT, or dose-intensive chemotherapy with ACVBP
(doxorubicin, cyclophosphamide, vindesine, bleomy-
cin, prednisone) followed by sequential consolidation
therapy with methotrexate, etoposide, ifosfamide, and
cytarabine. With a median follow-up of 7.7 years, the
5-year event-free survival (EFS) was 82% in the che-
motherapy arm, compared with 74% in the combined
modality arm (P < 0.001). The corresponding figures
for OS were 90% and 81% (P = 0.001).
The outcomes for patients with non-bulky disease
were analyzed separately, and the difference in EFS and
OS in favor of the chemotherapy-only arm was main-
tained. The upper age limit for the GELA study was 61
years, with a median age of 46 years. By comparison, the
median age in the SWOG was 59 years, with approx-
imately 50% of the patients being over 60 years. It is
therefore not clear whether an unselected patient pop-
ulation with limited-stage disease would tolerate or
benefit from the dose-intensive chemotherapy used in
the GELA study, particularly since this regimen was
associated with a 25% hospitalization rate with each of
the first three cycles of ACVBP chemotherapy.
Although the results of these studies suggest that
chemotherapy alone may be adequate treatment for
bulky disease, this was not confirmed by a study from
the Eastern Co-operative Oncology Group (ECOG), in
which 352 patients with clinical stage I or II disease
(including bulky disease) were initially treated with
eight cycles of CHOP chemotherapy. Patients in com-
plete response after chemotherapy were randomized
between 30 Gy of IF-RT or no further treatment.
Patients in partial remission after chemotherapy
received 40 Gy of IF-RT. For the 172 randomized
patients, the 6-year DFS was 73% for the radiation
therapy arm compared with 56% for the observation
arm (P = 0.05). No OS difference was observed.
The apparent differences in outcomes between the
studies described above are, in part, related to patient
selection for the various studies. Since patients with
limited-stage disease represent a heterogeneous
patient group, a stage-adjusted IPI has recently been
proposed, to facilitate risk stratification in future stud-
ies in early-stage disease (Table 12.2). This prognostic
model has been validated in other patient populations
with early-stage disease. According to this model,
patients with no adverse risk factors have a projected
5-year OS of approximately 95% when treated with
Chapter 12: Diffuse large B-cell lymphoma
Table 12.2 Stage-adjusted International Prognostic Index.
Adverse factor Stage-adjusted IPI
Stage Bulky stage II
Age >60
LDH >Normal
Performance status ≥2
Extranodal sites Not applicable
brief duration chemotherapy (CHOP × 3) and IF-RT.
Such combined modality therapy should therefore be
considered the standard approach for this patient
group, since future studies are very unlikely to dem-
onstrate further improvements in outcome.
In contrast, patients with one adverse factor have a
projected 5-year OS rate of around 70%; only 50–60%
of those with three or four adverse factors survive for 5
years. New approaches are therefore needed for these
groups, although combined modality therapy as
described above remains the current standard
treatment.
Several studies in advanced DLBCL have now dem-
onstrated improvements in DFS and OS by the addi-
tion of the anti-CD20 monoclonal antibody,
rituximab, to combination chemotherapy such as
CHOP. No randomized studies have been reported
to date in limited-stage disease. A recent phase II
study from the SWOG has tested the addition of ritux-
imab to three cycles of CHOP chemotherapy plus IF-
RT for diffuse aggressive B-cell NHL. The 2-year PFS
was 94%. These results compared favorably with an
historical series of patients treated with CHOP × 3 plus
IF-RT using identical selection criteria. Randomized
studies will be required to evaluate the benefit of addi-
tion of rituximab to chemotherapy in this context
prospectively.
The potential use of FDG–PET to determine which
patients with early stage DLBCL can be managed with-
out IF-RT has been investigated at the British
Columbia Cancer Agency. Patients with early-stage
disease who are PET-negative after three cycles of
R-CHOP therapy receive one further cycle, while
PET-positive patients receive IF-RT. Of 47 patients
who were PET-negative and in whom IF-RT was omit-
ted, only one has relapsed to date, suggesting that
FDG–PET may be a useful tool for directing
de-escalation of therapy, although this will require
prospective evaluation. 197
 Chapter 12: Diffuse large B-cell lymphoma
Primary extranodal disease
Extranodal involvement in the context of widespread
disease is a frequent finding in DLBCL. In addition,
approximately 40% of all cases of DLBCL present with
primary extranodal involvement. The most common
site for primary extranodal disease is the stomach, but
other common sites include bone (which may be mul-
tifocal), thyroid, kidney, spleen, and testis. Primary
CNS and skin DLBCL are specific entities covered
elsewhere in this book. In general, the treatment of
localized primary extranodal DLBCL is exactly as for
nodal disease – many of the randomized trials for
patients with limited-stage disease have included
patients with nodal and extranodal primaries, with
no obvious differences in outcome observed. Primary
involvement of certain anatomic sites, notably the
testis, paranasal sinuses, and the nasopharnyx, are
associated with an increased risk of CNS relapse, and
CNS prophylaxis is typically given to patients with
these primary sites of involvement.
Early-stage DLBCL – conclusions
A treatment algorithm for early-stage DLBCL is
shown in Figure 12.5.
* Combined modality therapy with brief duration
chemotherapy and IF-RT is the standard of care,
irrespective of risk group according to the stage-
adjusted IPI.
* The addition of rituximab to initial chemotherapy
has not been formally evaluated in a randomized
Clinical stage I or II disease
Disease bulk >10 cm
Treat as for advanced DLBCL
IPI = 0
Chemotherapy (rituximab*) x 3 and Chemotherapy (rituximab*) x 3 and
involved field radiation therapy involved field radiation therapy
198 Consider clinical trial
trial. Preliminary data suggests that it may reduce
the need for IF-RT, but this requires prospective
evaluation.
* Novel approaches include the addition of
radiolabeled monoclonal antibodies to standard
combined modality therapy. Cooperative group
trials testing this approach are under way at
present.
* Preliminary data suggest that FDG–PET may be
used to determine the requirement for radiation
therapy after abbreviated chemotherapy.
Treatment of advanced-stage DLBCL
Approximately 75% of patients with DLBCL present
with bulky stage II, or stage III–IV disease, and should
therefore be regarded as having advanced-stage dis-
ease. Until recently, CHOP was considered the stand-
ard first line therapy for all patients with DLBCL,
mainly on the basis of the SWOG randomized trial
which compared this regimen with three more inten-
sive regimens, showing no difference in response rates,
PFS, or OS, but a higher toxicity rate in the regimens
other than CHOP.
In recent years, several approaches have been
investigated to improve the outcome for patients
with advanced-stage disease.
Dose-intensified and dose-dense therapy
Intensification of the CHOP regimen, either by redu-
cing the treatment duration from 21 to 14 days, or by
Figure 12.5 Treatment algorithm for
early stage DLBCL. The role of rituximab-
containing chemotherapy in early-stage
disease has not been evaluated
prospectively. Use of rituximab is
recommended based on results in
advanced-stage disease.
Disease bulk ≤10 cm
IPI ≥1

the addition of etoposide to the standard regimen, has
been investigated in two studies from the German non-
Hodgkin’s Lymphoma Study Group. Parallel studies
were conducted in young and elderly patients, and in
each study the comparison of CHOP21 with CHOP14
and CHOP with the same regimen plus etoposide
(CHOEP) was performed using a 2 × 2 factorial design.
The trial in older patients included 689 patients, 71% of
whom had DLBCL. In this study, CHOP14 was shown
to be significantly superior to CHOP21. The 5-year
event-free rate for CHOP21 was 33% compared with
44% for CHOP14 (P = 0.003). The corresponding rates
for OS were 41% versus 53% (P < 0.001). The addition
of etoposide had no effect in this study.
The corresponding study included 710 younger
patients, aged 18–60 years, with stage I–IV disease,
with a normal lactate dehydrogenase (LDH) level. No
EFS or OS benefit was seen in this group from reduction
of the treatment interval from CHOP21 to CHOP14.
However, an EFS benefit was seen for patients receiving
CHOEP14 or 21 (5-year EFS = 69%) compared with
those receiving CHOP14 or 21 (% year EFS = 58%;
P = 0.004). No OS benefit was observed, possibly due
to the ability to “salvage” patients relapsing after first
line therapy.
The clinical relevance of these findings is unclear,
and they must be evaluated in the context of
rituximab-containing therapy.
Based on the results of these studies, the German
non-Hodgkin’s Lymphoma Study Group has con-
ducted a randomized trial comparing CHOP chemo-
therapy given at 14 day intervals with R-CHOP, also
given at 14 day intervals in elderly patients with
advanced DLBCL. This was a four-arm study which
also compared six versus eight cycles of therapy. The
rituximab-containing arms of this study were both
shown to have superior EFS and PFS when compared
to the arms using CHOP only. Only the patients
receiving R-CHOP for six cycles showed an OS
advantage in this study, primarily because of late
toxic deaths observed in patients receiving eight cycles
of the same therapy. This trial, known as the
“RICOVER 60” trial, established R-CHOP for six
cycles at 14 day intervals as the standard approach in
Germany and parts of the rest of Europe.
This approach has not been adopted more widely
because of a lack of randomized trials directly compar-
ing the R-CHOP regimen given at 21 and 14 day
intervals. Two such studies have now been conducted
(see below).
Chapter 12: Diffuse large B-cell lymphoma
The dose-adjusted EPOCH-R (etoposide, predni-
sone, vincristine, cyclophosphamide, doxorubicin, rit-
uximab) [DA-EPOCH-R] regimen represents another
approach to intensification of chemotherapy dose in
DLBCL. This regimen requires close monitoring of
hematologic toxicity, and reduction or increase of
chemotherapy doses according to the neutrophil
nadir. In the initial single institution phase II study
of this regimen in DLBCL, 5-year PFS and OS rates
were 79% and 80%, respectively, and did not differ
according to tumor proliferation rates or BCL-2
expression. This regimen is now being compared
with R-CHOP in an ongoing intergroup randomized
trial in the USA.
High-dose therapy and autologous stem
cell transplantation as a component of
first line therapy
The reported effectiveness of high-dose therapy and
autologous stem cell transplantation (ASCT) as a sal-
vage treatment in aggressive NHL has prompted many
groups to investigate the use of ASCT as a component
of first line therapy, particularly for patients identified
as having “poor-risk” disease at presentation. Many
trials are now published, although comparison of these
studies is complicated by the variable inclusion crite-
ria, and variable definition of poor-risk disease. Some
of these studies are retrospective, subset analyses of
clinical trials that were not initially stratified according
to risk groups and not statistically powered to detect
differences in subgroup analysis.
Many prospective studies have subsequently been
conducted, using the aaIPI to define poor-risk patients
(Table 12.3). Results have been variable, although
most studies have failed to show an advantage for
high-dose versus conventional dose remission
consolidation.
A randomized study compared eight cycles of
CHOP chemotherapy with two cycles of a more
dose-intensive first line regimen, followed by high-
dose therapy (HDT), and ASCT for responding
patients with advanced diffuse aggressive NHL, has
recently been reported by Milpied et al. (Table 12.3).
Eligible patients were aged 15–60 years with low, low–
intermediate, or high–intermediate risk disease
according to the aaIPI, and therefore represented a
relatively favorable group compared with many other
studies. 199
 Chapter 12: Diffuse large B-cell lymphoma

Table 12.3 Results of high-dose therapy and ASCT in first remission for DLBCL and other aggressive NHLs.

Reference n Randomization DFS (conventional OS (conventional

chemotherapy chemotherapy
versus SCT) versus SCT)
Haioun et al. 464 Sequential chemotherapy versus 52% versus 59% 71% versus 69%
J Clin Oncol, HDT and ASCT in patients in at 3 years at 3 years
2000;18:3025–3030 CR after induction (P = 0.46) (P = 0.6)
chemotherapy
Santini et al. 124 VACOP-B versus VACOP-B plus 60% versus 80% 65% versus 65%
J Clin Oncol, HDT and ASCT for responding at 6 years at 6 years
1998;16:2796–2802 patients (P = 0.1) (P = 0.5)
Kluin-Nelemans et al. 194 CHVmP/BV versus CHVmP/BV 56% versus 61% 77% versus 68%
JNCI, 2001;93:22–30 plus HDT and ASCT for at 5 years at 5 years
responding patients (P = 0.712) (P = 0.336)
Gianni et al. 98 MACOP-B versus high dose 49% versus 76% 55% versus 81%
NEJM, 1997;336:1290–1297 sequential therapy including at 7 years at 7 years
HDT and ASCT (P = 0.004) (P = 0.09)
Gisselbrecht et al. 370 ACVBP and sequential consolida- 76% versus 58% 60% versus 46%
J Clin Oncol, 2002;20:2472–2479 tion versus intensive induction at 5 years at 5 years
chemotherapy plus HDT and ASCT (P = 0.004) (P = 0.007)
Milpied et al. 207 CHOP versus intensive induction 37% versus 55% 44% versus 74%)
NEJM 2004;350:1287–1295 chemotherapy plus HDT and ASCT at 5 years at 5 years
(P = 0.037) (P = 0.001)
SC, Stem cell transplant; CR, complete remission; HDT, high-dose therapy; ASCT, autologous stem cell transplantation.

An intent-to-treat analysis was performed, demon-
strating a significantly higher EFS in the HDT arm
(55% versus 37%; P = 0.037), although there was no
difference in OS. Subset analysis showed a significant
difference in OS in patients with high–intermediate
risk disease (P = 0.001).
In view of the conflicting results that have been
reported in studies of first remission transplantation in
aggressive NHL, a meta-analysis has recently been
conducted. This study included data from 2018
patients from 13 randomized trials who were evaluable
for outcome data. Although a significantly higher
complete response (CR) rate was reported for HDT
and ASCT, no differences in EFS or OS were observed.
No difference in outcome was seen according to IPI
group, and analyses according to the number of
patients with DLBCL, transplant conditioning regi-
men, and response status prior to ASCT also failed to
clearly identify a group with superior outcome after
HDT. Data regarding the benefit of ASCT for patients
with IPI poor risk disease was inconclusive and there
was some evidence that patients with favorable risk
disease might have an inferior outcome after ASCT.
Based on these results, HDT and ASCT should not
200 be considered a component of first line therapy in
patients with diffuse aggressive lymphoma (including
DLBCL), irrespective of IPI risk group, and the addi-
tion of ASCT to standard induction therapy should
still be considered experimental, and used only in the
context of clinical trials.
Since most of the studies reported above were
conducted prior to the introduction of rituximab (see
below), or dose-dense chemotherapy regimens, their
relevance to current management of diffuse aggressive
NHL is unclear. The SWOG 9704 study may help to
clarify the role of remission consolidation with ASCT
in patients receiving R-CHOP as first line therapy.
This study is now completed although results are not
yet available.
Addition of rituximab to chemotherapy
The benefit of adding rituximab to CHOP chemother-
apy for DLBCL was initially demonstrated in a random-
ized trial from the GELA. In this study, 399 previously
untreated patients aged between 60 and 80 years with
DLBCL were randomized to receive eight cycles of
CHOP chemotherapy at 21-day intervals, or the same
chemotherapy plus rituximab given on day 1 of each
cycle. The R-CHOP arm was shown to be superior to
CHOP in terms of complete response rate (76% versus
63%; P = 0.005), 2-year EFS (61% versus 43%; P = 0.002)

and 2-year OS (70% versus 57%; P = 0.007). The survival
advantage for R-CHOP in this trial was observed in all
IPI risk groups. The results of this trial have been
updated recently. The 10-year PFS rates were 20% and
36.5% for CHOP and R-CHOP, respectively. The cor-
responding OS rates were 27.6% and 43.5%.
The addition of rituximab to CHOP in elderly
patients with DLBCL has also been investigated in an
intergroup study in the USA, including 632 patients
aged over 60 years with previously untreated, advanced
DLBCL. Patients were randomized to receive either six
or eight cycles of CHOP, according to response, or the
same chemotherapy plus rituximab. A second random-
ization was included for responding patients between
observation only, and maintenance rituximab, given
once per week for 4 weeks at 6-month intervals for a
total of 2 years. In all, 632 patients were randomized, of
whom 415 responders were subsequently randomized
to maintenance rituximab or observation only. There
was a significant improvement in 3-year failure-free
survival (FFS) in the R-CHOP arm compared with
CHOP (53% versus 46%; P = 0.04) and in the main-
tenance rituximab arm compared with observation
alone. The advantage of maintenance rituximab
appeared to be limited to patients who did not receive
this agent as part of their induction regimen. No OS
differences were observed in the study, possibly because
around 40% of patients randomized to receive CHOP
alone received rituximab in the second randomization.
Further evidence for the benefit of addition of
rituximab to chemotherapy has been reported from a
retrospective, population-based study from British
Columbia, Canada, which has demonstrated higher
EFS and OS rates for patients with DLBCL since the
introduction of rituximab.
A benefit for rituximab in younger, low-risk
patients has also been shown in the MInT
(MabThera International Trial). This study included
patients with DLBCL with bulky stage I or stage
II–IV disease, aged 18–60 years with IPI scores of 0
or 1. Treatment comprised six cycles of CHOP,
CHOEP, or a comparable chemotherapy regimen
with or without rituximab administered on day 1 of
each chemotherapy cycle. Preliminary results have
been reported for the first 326 randomized patients.
The 2-year time to treatment failure (TTF) was 76%
for patients receiving chemotherapy plus rituximab,
compared with 60% for those receiving chemother-
apy alone (P < 0.001). The corresponding rates for OS
were 94% versus 87% (P = 0.001).
Chapter 12: Diffuse large B-cell lymphoma
These results suggest that the addition of rituxi-
mab to chemotherapy is likely to benefit all risk
groups. Further studies will be required to determine
whether biological markers such as BCL-2 protein
expression will reliably predict those patients likely
to benefit from the addition of rituximab to
chemotherapy.
The potential role of first remission HDT and
ASCT is uncertain for patients receiving rituximab as
part of their induction regimen. The SWOG 9704
study described above may help to clarify its benefit
in this context.
Other ongoing trials are comparing the R-CHOP
combination given according to a standard 21 day
or accelerated 14 day schedule. Preliminary data
have been presented from two studies. Neither
study yet shows a clear difference in response rates,
or PFS or OS endpoints, but further follow-up is
required for both.
Addition of other new agents
to chemotherapy
Early results from phase II trials incorporating other
novel agents with first line therapy for DLBCL have
been reported recently. The anti-CD22 monoclonal
antibody, epratuzumab, has been combined with the
R-CHOP regimen in a phase II study in 107 patients
with advanced DLBCL. The 12-month EFS and OS
rates were 79% and 89%, respectively, and no signifi-
cant additional toxicity was observed compared
with historical experience with R-CHOP alone.
Comparison with historical controls suggested that
there may be a benefit for the epratuzumab rituximab
(ER)-CHOP regimen in patients with poor-risk dis-
ease by IPI, although this requires confirmation in a
randomized trial. Bortezomib has also been combined
with first line chemotherapy in DLBCL. Initial results
suggest that this combination can be delivered safely,
although the incidence and severity of peripheral neu-
ropathy appears to vary according to the dose of bor-
tezomib. Early data also suggest that the benefit of
adding bortezomib may be limited to patients with
non-germinal-center DLBCL.
Use of involved-field radiation therapy
in advanced DLBCL
The role of IF-RT in limited-stage DLBCL has been
well documented (see above). In the setting of 201
 Chapter 12: Diffuse large B-cell lymphoma

advanced-stage disease, its role is less well defined.

Many prospective studies have included an option
for the use of IF-RT to sites of prior bulk disease,
although bulk has been variably defined and
adherence to the protocols in these studies has also
been variable. As a result, the benefit of IF-RT in stages
III and IV DLBCL is unclear. Recent data have
also been conflicting. In a retrospective study, which
included a matched pair analysis, Phan et al. reported
the impact of consolidative IF-RT in patients
receiving R-CHOP chemotherapy for all stages of
DLBCL. The use of IF-RT was associated with higher
rates of 5-year DFS and OS in patients with
advanced-stage disease and was an independent prog-
nostic factor in multivariate analysis. By contrast,
analysis of results from the MInT study according
to disease bulk showed that bulk was, as expected, an
adverse prognostic factor, but the addition of radiation
therapy to R-CHOP in this study did not appear to
affect DFS or OS.
As discussed above, the potential role of functional
imaging to identify patients who may benefit from
consolidative IF-RT after chemotherapy requires pro-
spective evaluation.
Advanced stage DLBCL – conclusions
A treatment algorithm for advanced and relapsed/
refractory DLBCL is shown in Figure 12.6.
* Rituximab–CHOP 21 should be regarded as the
standard of care therapy for patients with
advanced-stage DLBCL regardless of risk group.
* The role of maintenance rituximab after
rituximab-containing induction is unknown. It
should only be used in the context of clinical trials.
* There is no current evidence to support the use of
HDT and stem cell transplantation as a component
of first line therapy, even for poor-risk patients –
this approach should still be considered only in the
context of prospective clinical trials.
Primary mediastinal (thymic) large
B-cell lymphoma
Mediastinal large B-cell lymphoma (MLBCL) is a dis-
tinct subtype of DLBCL, most likely arising from the
thymus gland and typically presenting with a clinically
localized, often bulky, anterior mediastinal mass which

Figure 12.6 Treatment algorithm for advanced

All stages, all IPI risk groups and relapsed/refractory DLBCL. NR, no response;
PD, progressive disease; CR, complete remission;
PR, partial remission; SCT, stem cell transplant.

Rituximab-CHOP x 8
CR/PR

observe NR/PD
NR/PD Not eligible for
Eligible for ASCT ASCT
Relapse/PD

Intensive 2nd line regimen

response

NR/PD
High-dose therapy/autologous
SCT

NR/PD Consider clinical trial of novel

CR/PR
therapy including allogeneic SCT

202 observe Symptomatic management


frequently invades surrounding structures such as the
pleura and pericardium. It typically affects young
adults, with a female predominance. Disseminated dis-
ease beyond the chest is relatively uncommon. Gene
expression profiling studies have demonstrated similar-
ities with Hodgkin lymphoma.
Recent clinical studies have demonstrated that this
entity has a favorable prognosis, although optimal
treatment strategies remain unclear. Several single
center phase II studies have reported favorable treat-
ment results for relatively intensive chemotherapy
regimens such as MACOP-B (methotrexate, doxoru-
bicin, cyclophosphamide, vincristine, prednisone,
bleomycin) compared with CHOP chemotherapy.
However, these studies were conducted prior to the
introduction of rituximab. Recent data have suggested
that R-CHOP is an effective regimen for this condi-
tion. The role of radiation therapy is also uncertain.
Although consolidative radiation therapy is widely
used in this disease, its benefit has been questioned in
an analysis from the MInT trial and in a retrospective
study from British Columbia. The potential use of
FDG–PET after induction chemotherapy to determine
which patients may benefit from consolidative IF-RT
will be investigated in a prospective study in Europe.
Management of relapsed DLBCL
Second line therapy prior to ASCT
Although HDT and ASCT remains the standard of
care for patients with relapsed DLBCL, a survival
benefit for this approach has only been demonstrated
convincingly in patients with disease that is responsive
to second line salvage chemotherapy. Commonly used
second line regimens, including DHAP (dexametha-
sone, high-dose cytosine arabinoside, cisplatin), ICE
(ifosfamide, carboplatin, etoposide) and mini-BEAM
(carmustine, etoposide, cytarabine, melphalan) pro-
duce overall response rates (ORR) of 40–60%, and
CR rates of only 25–35%. Analysis of the effectiveness
of ASCT from the date of transplantation therefore
overestimates the effectiveness of this approach in the
entire population of relapsed patients. When analyzed
by intent to treat, the EFS for relapsing patients with
DLBCL treated with second line salvage therapy fol-
lowed by ASCT is between 20% and 35%.
The development of more effective second line
regimens has the potential advantage of increasing
response rates and therefore increasing the number
of patients eligible for ASCT. Two randomized studies
Chapter 12: Diffuse large B-cell lymphoma
in which rituximab-based second line regimens are
compared are in progress. Results from the CORAL
study, comparing R-DHAP with R-ICE in patients
with relapsed DLBCL, have been reported recently.
No difference was observed in response rates or
3-year EFS according to second line regimen.
The National Cancer Institute of Canada is leading
a similar study comparing R-DHAP with another sec-
ond line regimen, R-GDP (rituximab, gemcitabine,
dexamethasone, cisplatin). This study is still in
progress.
Patients who do not achieve at least a partial remis-
sion to second line therapy have a poor outcome when
treated with HDT and ASCT in most series. These
patients should be considered for trials of novel treat-
ment approaches. A recent retrospective analysis from
the UK has demonstrated that patients who do not
respond to one conventional dose salvage regimen are
not rescued by subsequent conventional dose salvage
therapy.
High-dose therapy and autologous
stem cell transplantation
The use of HDT and ASCT has been regarded as the
standard of care for patients with relapsed DLBCL
for over a decade. This is based on the results of
the PARMA randomized trial. This study included
215 patients with relapsed aggressive NHL (mostly
DLBCL), initially treated with two cycles of salvage
chemotherapy with DHAP. Responding patients were
randomized to receive further DHAP chemotherapy,
or to proceed to HDT using BEAC (carmustine,
etoposide, cytarabine, cyclophosphamide) and autolo-
gous bone marrow transplantation. Significantly supe-
rior 5-year EFS (46% versus 12%; P = 0.0001) and OS
(53% versus 32%; P = 0.038) rates were observed for
the transplant arm compared with the conventional
chemotherapy arm. No formal follow-up analysis of
the PARMA study has been published. Although this
study established ASCT as the standard approach
for patients with relapsed chemosensitive DLBCL,
the results should be interpreted cautiously. Of the
215 patients entered onto the study, only 109 were
randomized, most commonly because the patients
did not achieve an adequate response to second line
therapy with DHAP (only 56% of patients responded
to this chemotherapy). All subsequent survival anal-
yses were restricted to randomized patients only, and 203
no intent-to-treat analysis was performed.
 Chapter 12: Diffuse large B-cell lymphoma
In addition to these limitations, the relevance
of the study in the present context is unclear.
Improved supportive care, including the use of
peripheral blood progenitor cells, has reduced the
morbidity associated with HDT and extended its
use to older patient groups, typically up to 70–75
years old. Most centers will now accept patients for
transplantation if they achieved partial response (PR)
to prior therapy, unlike the PARMA study, in which
a previous CR was required for eligibility. The
population of patients now receiving ASCT is there-
fore less defined than that in the original PARMA
study, raising questions concerning the current rele-
vance of this trial. The addition of rituximab to
combination chemotherapy regimens, and the
advent of accelerated 14 day regimens for first line
treatment, have improved DFS and OS in DLBCL.
It is not clear whether patients whose disease relapses
after one of these regimens will have the same
salvage rates as those treated without monoclonal
antibodies as part of their initial treatment.
Retrospective comparisons of outcomes following
ASCT for patients receiving first line CHOP or
CHOP-R have recently been reported. No difference
in EFS or OS following ASCT was observed accord-
ing to initial therapy in these studies. By contrast,
the CORAL study identified prior therapy with rit-
uximab as an adverse prognostic factor. The 3-year
EFS for patients previously treated with rituximab
was 21% compared with 47% for those without
previous rituximab exposure.
A recent report of dose-adjusted EPOCH-R as
primary therapy in DLBCL has shown a 2-year PFS
rate of 83%, with a very similar OS, indicating the high
activity of this regimen, and the apparent inability to
salvage relapsed patients with a standard transplant
approach.
Although HDT and ASCT remains the standard of
care for patients with relapsed DLBCL which is still
sensitive to second line chemotherapy, the true benefit
of this approach in the context of modern first line
therapies is unclear and requires re-evaluation.
The potential benefit of rituximab maintenance
after HDT and ASCT has also been investigated in
the CORAL study, in which a second randomization
to maintenance rituximab versus no further therapy
was included for patients undergoing ASCT. Results
for this second randomization have not yet been
reported and this approach should be considered
204 experimental.
Allogeneic stem cell transplantation in DLBCL
Current data on the role of allogeneic SCT in aggres-
sive lymphoma, either using myeloablative or non-
myeloablative conditioning regimens, are limited.
Comparative studies of allogeneic and autologous
SCT in aggressive NHL have not shown a survival
advantage for allogeneic SCT, despite the lower relapse
rate in allogeneic recipients. The lower relapse rate has
been offset by the increased regimen-related mortality
associated with allogeneic transplantation. In the
absence of clear evidence of a clinically relevant
graft-versus-lymphoma effect in aggressive NHL, the
use of allogeneic SCT should be restricted to research
protocols.
Allogeneic transplantation for patients who relapse
after autologous transplantation is increasing in use,
although there are few data to confirm its benefit. A
recent retrospective study from the International Bone
Marrow Transplant Registry analyzed results for 114
patients with various subtypes of NHL who received
allogeneic SCT after relapse following autologous
SCT. All patients underwent myeloablative condition-
ing. The regimen-related mortality was 22% at 3 years,
and the 5-year OS and PFS were 24% and 5%, respec-
tively. No analysis was performed according to NHL
subtype, but the study suggests that the curative poten-
tial for this approach is low and that its use should be
restricted to patients in prospective trials.
Prognostic factors for relapsed DLBCL
Multiple early single institution and registry studies of
ASCT in aggressive NHL demonstrated the impor-
tance of sensitivity of disease to second line therapy
as a predictive factor for outcome after transplanta-
tion. Short (less than 1 year) remission duration and
disease bulk at the time of ASCT were also identified as
adverse factors in many studies.
The aaIPI has been shown to have predictive value
in a follow-up report of the PARMA study. It proved
highly predictive of response to DHAP. Patients with
an aaIPI score of 0 had an ORR of 77% compared to
only 42% for those with three adverse factors. The
aaIPI was predictive of OS for the entire patient
cohort.
When randomized patients were analyzed sepa-
rately, the aaIPI was predictive for those receiving
DHAP, but not in those undergoing ASCT. In a subset
analysis, there was no difference in OS or PFS accord-
ing to the randomized arm for patients with an aaIPI

score of 0, although a significant difference remained
for those with scores of 1–3. Similar results have been
reported from other centers.
In the CORAL study (see above), in addition to
prior rituximab exposure, other adverse factors for
EFS were a prior remission duration of less than 12
months after diagnosis, and an IPI score of 2 or higher.
These subset analyses should be interpreted cau-
tiously, but question whether ASCT offers a survival
advantage to low-risk patients.
One recent study has also addressed the potential
value of cell of origin, as defined by tissue microarrays,
as a prognostic factor for patients undergoing salvage
therapy with HDT and ASCT. In this study, no differ-
ence in OS was observed when patients with germinal-
center or non-germinal-center phenotypes were
compared.
The potential value of functional imaging using
FDG–PET following salvage therapy but prior to
HDT and ASCT has also been assessed. In patients
with Hodgkin’s lymphoma, PET scan positivity prior
to ASCT has been shown to be highly predictive of
outcome. In the case of DLBCL, limited published data
are available from some small series, and the predictive
value of PET in this context therefore remains unclear.
Other treatment approaches for relapsed
disease
Radiolabeled monoclonal antibodies
Both 131I tositumomab (Bexxar) and 90Y-ibritumomab
tiuxetan (Zevalin) are active agents in indolent and
transformed CD20-positive B-cell lymphomas. Few
data are available for these agents in DLBCL. A phase
II study of 90Y-ibritumomab tiuxetan in 104 patients
with relapsed or refractory DLBCL, not eligible for
HDT and ASCT, has shown an ORR of 44%. The
response rate was higher in patients who had not had
prior therapy with rituximab, compared with those who
had previously been treated with rituximab and chemo-
therapy as their primary treatment. Median PFS was
around 6 months for patients who were rituximab-
naïve, compared with only about 2 months for those
previously treated with rituximab.
These agents have also been studied in high-dose
regimens used with ASCT. Most studies to date have
included patients with mixed histologic subtypes of
NHL. Early results indicate that both 131I tositumo-
mab and 90Y-ibritumomab tiuxetan can be combined
with standard high-dose chemotherapy regimens
Chapter 12: Diffuse large B-cell lymphoma
without additional toxicity or prolongation of engraft-
ment times. In view of the emerging data on the use of
rituximab as maintenance therapy in DLBCL, new
studies are currently being planned to assess the use
of radioimmunotherapy to consolidate remission after
induction therapy with R-CHOP and other rituximab-
chemotherapy combinations in DLBCL.
Relapsed DLBCL – conclusions
* HDT and ASCT is the standard of care for patients
with chemosensitive relapse.
* The true impact of SCT in patients relapsing after
rituximab-containing therapy is unclear.
* Allogeneic SCT has no proven benefit in relapsed
DLBCL and should be used only in the context of
prospective clinical trials.
* Patients with refractory disease do not benefit from
SCT – other, experimental, strategies should be
considered.
* New treatment approaches are required for
patients who relapse after SCT or who are ineligible
for transplantation.
New therapeutic targets in DLBCL
The use of molecular techniques, including GEP, has
identified many potential rational therapeutic targets
which are now under investigation in DLBCL. Some of
these are listed in Table 12.4.
Histone deacetylase (HDAC) inhibitors may have
several potential mechanisms of action in DLBCL.
Acetylation of histones in the nucleosome is a major
determinant of the regulation of many genes.
Deacetylation of histones results in condensed chro-
matin structure and repression of gene transcription.
Inhibitors of HDAC such as suberoylanilide
Table 12.4 Potential therapeutic targets in DLBCL.
CD22
Histone deacetylase
Proteasome
BCL-6
BCL-2
mTOR/AKT
PKC-β
Syk
205
Btk
 Chapter 12: Diffuse large B-cell lymphoma
hydroxamic acid (SAHA) have been shown to induce
differentiation and/or apoptosis in various tumor cell
lines, and this agent has demonstrated clinical activity
in heavily pre-treated patients with NHL.
HDAC inhibitors may exert some of their activity
in DLBCL through BCL-6. This gene is thought to be
important in the pathogenesis of DLBCL and has been
shown to be antiapoptotic in tumor cells, through
inhibition of transcription of the p21WAF-1 gene. In
the presence of HDAC inhibitors, p21WAF-1 tran-
scription results in growth inhibition, apoptosis, and
differentiation. Various HDAC inhibitors are now in
early-phase clinical trials in DLBCL with confirmed
activity.
Mammalian target of rapamycin (mTOR) is a ser-
ine/threonine protein kinase involved in the regula-
tion of cell growth in response to multiple nutrients. It
is phosphorylated via the PI3/AKT pathway and plays
a role in the regulation of apoptosis. It also appears to
be involved in the regulation of angiogenesis through
effects on production of vascular endothelial growth
factor (VEGF). Phase I and II studies of various agents
which target mTOR, including everolimus and temsir-
olimus, have confirmed that these agents have clinical
activity in relapsed and refractory DLBCL.
The expression of protein kinase C β (PKCβ) by
immunohistochemistry has been shown to be an
adverse prognostic factor in DLBCL, and in vitro
studies of inhibitors of PKCβ, including bryostatin,
have validated this molecule as a target in DLBCL.
A phase II study of the PKCβ inhibitor, enzastaurin,
has confirmed the activity of this agent in DLBCL.
A phase III study of this agent as a maintenance
therapy after completion of R-CHOP is currently in
progress.
Data have also been reported recently for the clin-
ical activity of agents which target the B-cell receptor
and downstream signaling molecules in DLBCL.
Fostamatinib, an agent that targets spleen tyrosine
kinase (syk), has been shown to produce responses in
patients with relapsed and refractory DLBCL.
Similarly, PCI-37265, an inhibitor of Bruton’s tyrosine
kinase (Btk), has been shown to produce responses in a
phase I study, including in patients with relapsed
aggressive NHL, including DLBCL.
Immunomodulatory drugs, including lenalido-
mide, have demonstrated activity in DLBCL in phase
I and II clinical trials, and these are now being eval-
uated in several clinical contexts, including as main-
206 tenance therapy.
Many of these agents have shown promising clin-
ical activity, with non-overlapping toxicities compared
with conventional chemotherapy. As a result, phase II
studies adding these agents to standard induction reg-
imens such as R-CHOP for previously untreated
patients with DLBCL are now being developed.
Further reading
Calaminici M, Piper K, Lee AM, Norton AJ. CD23
expression in mediastinal large B-cell lymphomas.
Histopathology, 2004;45:619–624.
Choi WWL, Weisenburger DD, Greiner TC, et al. A new
immunostain algorithm classifies diffuse large B-cell
lymphoma into molecular subtypes with high accuracy.
Clin Cancer Res, 2009;15:5494–5502.
Coiffier B, Thieblemont C, Van Den Neste E, et al. Long-
term outcome of patients in the LNH-98.5 trial, the first
randomized study comparing rituximab-CHOP to
standard CHOP chemotherapy in DLBCL patients: a
study by the Groupe d’Etudes des Lymphomes de
l’Adulte. Blood, 2010;116:2040–2045.
Compagno M, Lim WK, Grunn A, et al. Mutations of
multiple genes cause deregulation of NF-kappaB in
diffuse large B-cell lymphoma. Nature,
2009;459:717–721.
Copie-Bergman C, Plonquet A, Alonso MA, et al. MAL
expression in lymphoid cells: further evidence for MAL as
a distinct molecular marker of primary mediastinal large
B-cell lymphomas. Mod Pathol, 2002;15:1172–1180.
Davis RE, Ngo VN, Lenz G, et al. Chronic active B-cell-
receptor signalling in diffuse large B-cell lymphoma.
Nature, 2010;463:88–92.
Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a
standard regimen (CHOP) with three intensive
chemotherapy regimens for advanced non-Hodgkin’s
lymphoma. N Engl J Med, 1993;328:102–106.
Greb A, Bohlius J, Trelle S, et al. High-dose chemotherapy
with autologous stem cell support in first-line treatment
of aggressive non-Hodgkin lymphoma – results of a
comprehensive meta-analysis. Cancer Treat Rev,
2007;33:338–346.
Gisselbrecht C, Glass B, Mounier N, et al. Salvage regimens
with autologous transplantation for relapsed large B-cell
lymphoma in the rituximab era. J Clin Oncol,
2010;28:4184–4190.
Habermann TM, Weller EA, Morrison VA, et al.
Rituximab-CHOP versus CHOP alone or with
maintenance rituximab in older patients with diffuse
large B-cell lymphoma. J Clin Oncol, 2006;24:3121–3127.
Hans CP, Weisenberger DD, Greiner TC, et al.
Conformation of the molecular classification of diffuse

large B-cell lymphoma by immunohistochemistry using a
tissue microarray. Blood, 2004;103:275–282.
Horning SJ, Weller E, Kim K, et al. Chemotherapy with or
without radiotherapy in limited-stage diffuse aggressive
non-Hodgkin’s lymphoma: Eastern Cooperative
Oncology Group Study 1484. J Clin Oncol,
2004;22:3032–3038.
Linderoth J, Jerkemen M, Cavallin-Stahl E, et al.
Immunohistochemical expression of CD23 and CD40
may identify prognostically favorable subgroups of
diffuse large B-cell lymphoma: a Nordic lymphoma
group study. Clin Cancer Res, 2003;9:722–728.
Lossos IS. Molecular pathogenesis of diffuse large B-cell
lymphoma. J Clin Oncol, 2005;23:6351–6357.
Miller TP, Dahlberg S, Cassady JR, et al. Chemotherapy
alone compared to chemotherapy plus radiotherapy for
localized intermediate- and high-grade non-Hodgkin’s
lymphoma. N Engl J Med, 1998;339:21–26.
Ngo VN, Young RM, Schmitz R, et al. Oncogenically active
MYD88 mutations in human lymphoma. Nature,
2011;470:115–119.
Ott G, Ziepert M, Klapper W, et al. Immunoblastic
morphology but not the immunohistochemical GCB/
nonGCB classification predicts outcome in diffuse large
B-cell lymphoma in the RICOVER-60 trial of the
DSHNHL. Blood, 2010;116:4916–4925.
Persky DO, Unger JM, Spier CM, et al. Phase II study of
rituximab plus three cycles of CHOP and involved-field
radiotherapy for patients with limited-stage aggressive
B-cell lymphoma: Southwest Oncology Group study
0014. J Clin Oncol, 2008;26:2258–2263.
Pfreundschuh M, Trumper L, Kloess M, et al. Two-weekly
or 3-weekly CHOP chemotherapy with or without
etoposide for the treatment of young patients with
good-prognosis (normal LDH) aggressive lymphomas:
results of the NHL-B1 trial of the DSHNHL. Blood,
2004;104:626–633.
Pfreundschuh M, Trümper L, Osterborg A, et al.
CHOP-like chemotherapy plus rituximab versus
CHOP-like chemotherapy alone in young patients with
good-prognosis diffuse large-B-cell lymphoma: a
randomised controlled trial by the MabThera
International Trial (MInT) Group. Lancet Oncol,
2006;7:379–391.
Pfreundschuh M, Ho A, Cavallin-Stahl E, et al. Prognostic
significance of maximum tumour (bulk) diameter in
young patients with good-prognosis diffuse large B-cell
lymphoma treated with CHOP-like chemotherapy with
Chapter 12: Diffuse large B-cell lymphoma
or without rituximab: an exploratory analysis of the
MabThera International Trial Group (MInT) study.
Lancet Oncology, 2008;9:435–444.
Pfreundschuh M, Schubert J, Ziepert M, et al. Six versus
eight cycles of bi-weekly CHOP-14 with or without
rituximab in elderly patients with aggressive CD20+
B-cell lymphomas: a randomised controlled trial
(RICOVER-60). Lancet Oncol, 2008:9:105–116.
Phan J, Mazloom A, Medeiros JL, et al. Benefit of
consolidative radiation therapy in patients with diffuse
large B-cell lymphoma treated with R-CHOP
chemotherapy. J Clin Oncol, 2010;28:4170–4176.
Philip T, Guglielmi C, Hagenbeek A, et al. Autologous bone
marrow transplantation as compared with salvage
chemotherapy in relapses of chemotherapy-sensitive
non-Hodgkin’s lymphoma. N Engl J Med,
1995;333:1540–1545.
Reyes F, Lepage E, Ganem G, et al. ACVBP versus CHOP
plus radiotherapy for localized aggressive lymphoma. N
Engl J Med, 2005;352:1197–1205.
Rieger M, Osterborg A, Pettengell R, et al. Primary
mediastinal B-cell lymphoma treated with CHOP-like
chemotherapy with or without rituximab: results of the
Mabthera International Trial Group study. Ann Oncol,
2011;22:664–670.
Rosenwald A, Wright G, Chan WC, et al. The use of
molecular profiling to predict survival after
chemotherapy for diffuse large B-cell lymphoma. N Engl J
Med, 2002;346:1937–1947.
Savage KJ, Al-Rajhi N, Voss N, et al. Favorable outcome of
primary mediastinal large B-cell lymphoma in a single
institution: the British Columbia experience. Ann Oncol,
2006;17:123–130.
Sehn LH, Donaldson J, Chhanabhai M, et al. Introduction
of combined CHOP plus rituximab therapy
dramatically improved outcome of diffuse large B-cell
lymphoma in British Columbia. J Clin Oncol,
2005;23:5027–5033.
Sehn LH, Savage KJ, Hoskins P, et al. Limited stage DLBCL
patients with a negative PET scan following 3 cycles of
R-CHOP have an excellent outcome following
abbreviated immunochemotherapy alone. Ann Oncol,
2008;19 (suppl): (abstract 052).
Wilson WH, Dunleavy K, Pittaluga S, et al. Phase II study of
dose-adjusted EPOCH and rituximab in untreated diffuse
large B-cell lymphoma with analysis of germinal center
and post-germinal center biomarkers. J Clin Oncol,
2008;26:2717–2724.
207
 Chapter
13
Central nervous system lymphomas
Andrés J. M. Ferreri and Lisa M. DeAngelis
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
A variety of lymphomas can involve the central nervous
system (CNS) at different phases of their evolution,
both in immunocompetent and immunocompromised
individuals. They represent a heterogeneous group
of malignancies, with variable clinical and behavioral
characteristics, and require different therapeutic
approaches. In this chapter, the therapeutic manage-
ment of these malignancies will be analyzed separately
in three main entities: primary CNS lymphomas
(PCNSL), secondary CNS lymphomas (SCNSL), and
other, less common, forms of CNS lymphomas.
Histopathology
Primary diffuse large B-cell lymphoma (DLBCL) of
the CNS shows diffuse parenchymal growth with
dense cuffing in perivascular spaces. The latter is best
appreciated towards the periphery of the tumor. Large
areas of necrosis may be present, especially in those
treated with steroids prior to biopsy. Towards the
periphery of the tumor the neoplastic cells are often
admixed with astrocyte gliosis. Cytologically, the neo-
plastic cells resemble those of DLBCL encountered
elsewhere. Similarly, the immunophenotype of pri-
mary CNS DLBCL is similar to those outside the
CNS and is positive for B-cell markers (CD20,
CD79a, and PAX-5). Strong IRF4/MUM1 staining is
seen in 90% of cases. CD10 is expressed by a minority
of cases, while many express BCL-6. BCL-2 is often
expressed. As these lymphomas (by definition) occur
in immunocompetent patients, Epstein–Barr virus
(EBV) is generally absent.
Many lymphomas involving the CNS in immu-
nocompromised patients are EBV-positive and show
features of DLBCL, many with an immunoblastic
morphology. Burkitt lymphoma, a common lym-
phoma in the immunocompromised patient, may
involve the CNS.
208 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
Rare cases of small lymphocytic lymphoma, dural-
based mucosa-associated lymphoid tissue (MALT)
lymphoma, and T-/NK-cell lymphoma similar to
those seen in tissue outside the CNS have been
described. Secondary involvement by lymphoma orig-
inating elsewhere is also encountered.
Conventional therapeutic strategies
PCNSLs are aggressive malignancies arising within
and confined to the CNS. They comprise 3% of intra-
cranial neoplasms, and in the past two decades their
incidence has risen both in immunocompromised and
immunocompetent individuals, especially in those
over 50 years of age. The main patient characteristics
at presentation are summarized in Table 13.1. Current
therapeutic knowledge in PCNSL results from three
randomized trials, multiple single arm phase II trials,
meta-analyses, and large retrospective, multicenter
series. The few randomized trials limit comparison
between new approaches, but therapeutic progress
has been made by these studies as well as single arm
phase II trials. However, the use of divergent study
designs and entry criteria in all prospective trials, as
well as the presence of some methodological pitfalls,
make comparisons difficult. Despite the improved
outcome reported in prospective trials, progress in
the treatment of PCNSL has not been reflected in
studies of population-based cohorts, and survival has
not improved consistently in the past three decades.
This finding highlights one of the most important
limitations of all prospective trials, which is potential
patient selection bias.
The optimal treatment of PCNSL requires a multi-
disciplinary approach. If neuroimaging suggests the
possibility of PCNSL (Figure 13.1), then a stereotactic
biopsy is the most appropriate surgical approach.
Aggressive surgical resection should be avoided
 Chapter 13: Central nervous system lymphomas

Table 13.1 Characteristics in immunocompetent patients

with PCNSL.

Median age (range) 61 (14–85) years

Age >70 years 14%
Male:female ratio 1.4:1
Performance status (ECOG score)
0–1 35%
2–3 50%
4 15%
Prior cancer 4%
Histology
Indolent 5%
Diffuse large B-cell lymphoma 60%
Highly aggressive 15%
Unclassified 20%
Figure 13.1 Gadolinium-enhanced MRI lesion involving the basal
Presenting symptoms
ganglia and periventricular area (arrows). These are common
Motor and/or sensory focal deficit 50%
radiological features of primary CNS diffuse large B-cell lymphoma in
Cognitive and personality changes 40%
immunocompromised patients.
Headache 25%
Intracranial hypertension (nausea,
vomiting, papilledema) 30%
Seizures 15% three risk groups. Both scoring systems could result
Ocular symptoms 15% in the application of risk-adjusted therapeutic strat-
T-phenotype 2% egies in future studies.
Elevated LDH serum level 35% Even if diverse therapeutic strategies for PCNSL
Intraocular disease 13%
are now available, some of these strategies are associ-
ated with an increased risk of severe treatment-related
Positive CSF cytology examination 16%
neurotoxicity, especially among elderly patients.
High CSF protein concentration 61% Therefore, a dilemma in PCNSL treatment is the
Multiple lesions 34% choice between strategies designed to intensify therapy
Involvement of deep structures of the 36% to improve the cure rate, versus strategies of treatment
brain* de-escalation to avoid severe neurotoxicity.
* Involvement of deep structures of the brain involves basal
ganglia and/or brainstem and/or cerebellum. ECOG, Eastern
Cooperative Oncology Group; LDH, lactate dehydrogenase;
CSF, cerebrospinal fluid.
Chemotherapy
Chemotherapy plays a central role in the management
of patients with PCNSL. However, efficacy is limited by
because it does not improve survival and may result in several factors, including the biology and microenviron-
neurological deterioration and chemotherapy delay. ment of this malignancy that is protected by the blood–
Complete staging work-up and prognostic factor brain barrier (BBB). Thus, the capability of crossing the
assessment should be performed (Table 13.2). The BBB should be taken into account with drug selection to
International Prognostic Index (IPI) does not discrim- treat patients with PCNSL. From this perspective, avail-
inate between risk groups in PCNSL, but the combi- able drugs can be divided into three groups: (1) drugs
nation of five independent predictors of response and with poor BBB penetration that cannot be administered
survival, i.e. age, performance status (PS), serum lac- at high doses because they are associated with dose-
tate dehydrogenase (LDH) level, cerebrospinal fluid limiting toxicity (i.e. anthracyclines, vinca-alkaloids);
(CSF) protein concentration, and the involvement of (2) drugs exhibiting low to moderate capability to
deep structures of the brain (Table 13.2), distinguishes cross the BBB that can be safely administered at high
three different risk groups based on the presence of doses to obtain therapeutic concentrations in the CNS
0–1, 2–3, or 4–5 unfavorable features; this prognostic (i.e. methotrexate [MTX] and cytarabine [araC]); and
index is named the I.E.L.S.G. score. An alternative (3) drugs able to cross the BBB and reach therapeutic
scoring system employing only age and performance concentrations in the CNS even when administered at 209
status can also stratify the PCNSL population into conventional doses (e.g. steroids, some alkylating
 Chapter 13: Central nervous system lymphomas
Table 13.2 Staging work-up and pre-treatment evaluations
in PCNSL.
Staging
Physical examination
Routine blood studies
Whole-brain MRI
Contrast total body CT scan
Ophthalmologic evaluation (including slit-lamp
examination)
Cerebrospinal fluid cytology
Cerebrospinal fluid biochemical examination
Bone marrow biopsy
Testicular ultrasonography (older men)
18
FDG–PET (investigational role)
Suspicion of vitreal infiltration may require confirmation
by vitrectomy
Prognostic factors
Age
Performance status
Lactate dehydrogenase serum level
Cerebrospinal fluid protein concentration
Involvement of deep regions of the brain
Pre-treatment assessment
Neurological examination
Biochemical serum profile
Baseline neuropsychiatric tests
Renal and hepatic functionality tests (creatinine clearance)
Cardiac functionality tests*
HIV, hepatitis B and C virus evaluation
* Echocardiography with left ventricular ejection fraction
evaluation is advisable in elderly patients eligible for HD-MTX-
based chemotherapy considering that the administration of
this drug requires adequate hydration (near 3000 mL/day on
the days before and after MTX). HIV, Human immunodeficiency
virus.
agents). The combination of drugs from the first group
(i.e. the CHOP regimen) represents the backbone of
treatment of extra-CNS DLBCL but exhibits negligible
activity in PCNSL. While many patients have an imme-
diate radiographic response, most experience progres-
sion after two to three cycles. This may be explained by
positron emission tomography (PET) studies that dem-
onstrate normalization of the disrupted BBB 3–4 weeks
after initial chemotherapy, suggesting that the bulky
tumor not protected by the BBB responds, but micro-
scopic tumor is not adequately treated and progresses.
Importantly, the addition of CHOP chemotherapy both
to high-dose methotrexate (HD-MTX) and whole-brain
radiotherapy (WBRT) did not improve outcome with
respect to the use of these strategies as an exclusive
approach. As a consequence, CHOP and CHOP-like
regimens have been abandoned, and chemotherapy
combinations for PCNSL are currently designed by
210 using drugs from the other two subgroups.
Methotrexate and cytarabine
Antimetabolites such as MTX and araC constitute the
backbone of most chemotherapy combinations used in
PCNSL, with demonstrated efficacy in prospective trials
(Table 13.3). Different regimens have been used in an
attempt to improve efficacy; MTX doses from 1 to 8 g/
m2 are feasible in PCNSL; both total dose and infusion
rate are important for MTX delivery to the brain paren-
chyma and CSF. MTX doses ≥1 g/m2 result in tumor-
icidal levels in the brain parenchyma and doses ≥3 g/m2
yield tumoricidal levels in the CSF. Conversely, doses
up to 8 g/m2 delivered in a 24-hour continuous infusion
do not reliably achieve an adequate CSF level. HD-
MTX, when used alone, is a safe treatment even in
patients older than 70. Calculated or measured creati-
nine clearance and glomerular filtration rates have been
proposed to tailor MTX dose. Individualized dosing of
HD-MTX might have the potential to improve outcome
in patients with PCNSL, even when administered con-
currently with high-dose (HD)-araC. In the future, this
could be carried out by using first-cycle pharmacoki-
netic modeling with determination of potential dose
adaptations for later cycles using Bayesian analysis.
HD-MTX (8 g/m2) monochemotherapy yielded an
overall response rate (ORR) of 51–68%, with a 3-year
OS of 35%, which is similar to the 3-year OS of 32–
47% recorded in trials evaluating polychemotherapy
combinations with a MTX dose of 3.5 g/m2
(Table 13.3). Noteworthy, dose reduction because of
impaired creatinine clearance was needed in 45% of
patients in trials using MTX 8 g/m2, whereas in the 3.5
g/m2 trials few patients needed reduction in MTX
dose. Thus, tolerability and activity of 3.5 g/m2 suggest
that this might be a good compromise between safety
and efficacy for combination regimens.
A large meta-analysis and a retrospective series
have suggested a survival improvement when araC is
added to MTX. In a recently published randomized
trial, 79 patients with newly diagnosed PCNSL were
assigned to four cycles of MTX (3.5 g/m2, day 1) alone
or combined with araC (2 g/m2/12 h, days 2 and 3);
both arms were followed by WBRT. The addition of
araC resulted in significantly improved complete
remission rate and OS compared with MTX alone.
Hematologic toxicity was higher in the MTX-araC
arm, while non-hematologic toxicities were uncom-
mon. This MTX-araC combination may become the
current standard chemotherapeutic approach for de
novo PCNSL since it is supported by the only random-
ized study thus far.
Table 13.3 Published prospective trials on PCNSL in immunocompetent patients treated with chemotherapy alone or combined treatment.

Primary chemotherapy2 Median OS Neuro-

1 3 4
No. TS Drugs M dose itCHT ORR CRR follow-up 2-year 5-year toxicity
Chemotherapy alone
31 C M 8 g/m2/14 d – 100% 31 months 63% NR 0%
2
50 C M L P MP 1 g/m /10 d M 48% 42% 36 months 45% NR 8%
25 C M 8 g/m2/14 d – 74% 52% 23 months 70% NR 5%
2
65 C MVICA 5 g/m /28 d ivM/a 71% 61% 26 months 69% 43% 3%
37 C M 2
8 g/m /14 d – 35% 30% 56 months 51% 25% 20%
HD-MTX plus RT
25 CR M 3.5 g/m2/21 d – 88–92% 56–88% 60 months 58% 38% 8%
46 CR M 1 g/m2/7 d a* NR–95% NR–82% 36 months 62% 37% 22%#
31 CRC M 1 g/m2/7 d M 64–87% NR–87% 97 months 72% 22% 32%
HD-MTX-containing chemotherapy plus RT
25 CR AaCMOP 3 g/m2/21 d M/a/P 72–72% 67–78% 24 months 70% 56% 0%
57 CRC aBnMO±CHOP 1.5–3 g/m2/14 d – 68–71% 62–64% 59 months 60% 36% NS
56 CR Bn M N P 1.5 g/m2/28 d M 71–100% 54–61% 8 months 86% NS 29%
2
31 CR ABCMOP 2 g/m /15 d M/a/P* 89–67% 24 months 48% 36% 7%
52 CRC MNO 3.5 g/m2/7 d M 90–94% 56–87% 60 months 75% 40% 25%
102 CR MNO 2.5 g/m2/14 d M 94%–NR 58–NR 56 months 64% 32% 15%
52 CR Bn M O P 3 g/m2/14 d M NR–81% 33–9% 27 months 69% NR 12%
41 CR AIMT 3.5 g/m2/21 d – 76–83% 44–6% 49 months 50% 41% NR
30 CRC MNOR 3.5 g/m2/14 d M* 93%–NR 44–77% 37 months 67% NR NR
Randomized trials
79 CR M 3.5 g/m2/21 d – 40–40% 18–30% 30 months 39% 26% 20%
Ma 3.5 g/m2/21 d – 69–74% 46–64% 56% 46% 6%
Only trials on 25 patients or more and published as original articles are considered. Trials on high-dose chemotherapy supported by ASCT are excluded.
No. = number of enrolled patients. NS, Not specified; NR, not reported. # 5-year risk rate. (Updated from The Lancet 2009;374:1512–1520.)
1
Treatment sequence: C, chemotherapy alone; CR, chemotherapy followed by radiotherapy; CRC, chemotherapy followed by radiotherapy and further chemotherapy.
2
Primary chemotherapy: A or H, adriamycin; a, cytarabine; B, bleomycin; Bn, BCNU; C, cyclophosphamide; Cn, CCNU; E, etoposide; M, methotrexate; m, nitrogen mustard; N, procarbazine; O,
vincristine; P, prednisone or other corticoids; R, rituximab; T, thiotepa; V, teniposide. Series using intrathecal chemotherapy (itCHT) exclusively in patients with positive CSF cytology at diagnosis
are marked with an asterisk.
3
ORR = Overall response rate; in series treated with combined modality, response rate after chemotherapy and (–) after the entire planned treatment is reported.
4
CRR = Complete remission rate; in series treated with combined modality, response rate after chemotherapy and (–) after the entire planned treatment is reported.
 Chapter 13: Central nervous system lymphomas
Corticosteroids
Corticosteroids, which are routinely started in any
patient with a new intracranial mass, have a potent
and rapid oncolytic effect, causing radiographic
regression in up to 40% of patients. This condition,
named “ghost” or “vanishing” tumor, is strongly sug-
gestive for PCNSL. However, it is important to under-
line that only one-half of “vanishing tumors” are
PCNSL, while sarcoidosis, multiple sclerosis, acute
disseminated encephalomyelitis, and other malignan-
cies can exhibit a dramatic response to steroids.
Although some authors reported that patients with
PCNSL treated with corticosteroids before diagnosis
can be successfully diagnosed without stopping the
steroids, there is a large consensus that these drugs
should be withheld in any patient with a presumptive
diagnosis of PCNSL until tissue has been obtained to
avoid the risk of unreliable biopsies. A clinical or
radiologic response to corticosteroid administration,
even if transient, was associated with better prognosis,
but confirmatory studies are not available.
Maintenance steroid therapy (1 g of methylpredniso-
lone every month until progression) after response to
definitive chemotherapy has been associated with a 6-
month OS of 33% in elderly patients with PCNSL,
which deserves further study.
Other active drugs
The small number of available active drugs limits
further improvements in chemotherapy efficacy. It
is important that patients with relapsed or refrac-
tory PCNSL be entered into phase I/II trials assess-
ing new drugs and combinations. Some drugs have
emerged recently from prospective and retrospec-
tive studies and are now being incorporated into
ongoing phase II trials assessing new HD-MTX-
based combinations.
Temozolomide is an oral alkylating agent that spon-
taneously undergoes chemical conversion to MTIC
(5-(3-methyl-1-triazeno) imidaxole-4-carboxamide),
resulting in 0–6 methylguanine-DNA methyltransferase
depletion. This drug displayed excellent tolerability and
a 26% response rate, mostly complete remissions, in a
multicenter phase II trial on PCNSL relapsed or refrac-
tory to HD-MTX. As upfront monotherapy, temozolo-
mide has been associated with a complete response rate
(CRR) of 47%, and median OS of 21 months in elderly
patients with PCNSL; O(6)-methylguanine-DNA meth-
212 yltransferase promoter methylation seems to predict
a positive response. Temozolomide can be safely
combined with MTX in elderly patients, which is a
relevant issue since about 50% of PCNSL patients are
older than 60 years.
Preliminary data suggest that a rituximab–temo-
zolomide combination is well tolerated and active. The
use of rituximab, a chimeric monoclonal antibody
directed against the B-cell-specific antigen CD20, in
patients with PCNSL is based on its positive effect
when added to CHOP in patients with DLBCL, the
most common lymphoma category arising in the CNS.
However, as a large protein it has poor penetration
into the CNS as measured by CSF levels, and the
capability of rituximab to prevent CNS dissemination
of DLBCL remains a matter of debate. Promising
effects of rituximab, as single agent or in combination
with HD-MTX, have been reported in a few studies,
and the precise role of this monoclonal antibody in
PCNSL remains to be defined in randomized trials.
Preliminary results suggest that anti-CD20 radioim-
munotherapy with 90Y-ibritumomab tiuxetan
®
(Zevalin ) is feasible in PCNSL patients, and able to
target brain lymphoma. However, this radioimmuno-
conjugate is unable to treat microscopic lesions with
an intact BBB adequately.
Topotecan, a camptothecin derivative that inhibits
the enzyme topoisomerase I, produced an objective
response in one-third of patients with refractory or
relapsed PCNSL, with a 1-year progression-free sur-
vival (PFS) of 13%. The use of this drug was frequently
followed by progressive disease and was associated
with grade 3–4 leukopenia in 26% of patients and
clinically evident neurological deterioration in two-
thirds of survivors.
Radiotherapy
PCNSL is a radiosensitive tumor and WBRT was the
standard treatment for many years; however, WBRT
alone is rarely curative in PCNSL patients since
response is usually short-lived, with median survival
ranging from 10 to 18 months. In PCNSL patients
treated with radiotherapy, the whole brain should be
irradiated because of the diffuse infiltrative nature of
this lymphoma; attempts to treat with focal brain
radiotherapy have resulted in high rates of recurrences
within and outside of the irradiated port. Although
microscopic CSF dissemination is common, more
extensive craniospinal irradiation does not confer
additional survival benefit and is associated with sig-
nificant morbidity; this strategy plays only a palliative
role in leptomeningeal lymphoma.

The optimal dose of WBRT is controversial, but
the results of several studies suggest a dose between 40
and 50 Gy, while WBRT doses greater than 50 Gy are
associated with an increased risk of neurotoxicity. A
prospective Radiation Therapy Oncology Group
(RTOG) study of 40-Gy WBRT followed by a 20-Gy
focal boost demonstrated a radiographic response in
62% and a CRR in 19%. The addition of a boost did not
improve local tumor control or survival. A more
recent RTOG study failed to show a clear benefit in
terms of disease control or reduced neurotoxicity
when hyperfractionated WBRT was used. WBRT
40–50 Gy as exclusive treatment appears advisable in
PCNSL patients who cannot be treated with primary
chemotherapy, whereas consolidation after chemo-
therapy represents the best role for radiotherapy.
Combined chemoradiotherapy
Even though not confirmed in a randomized trial, there
is a consensus that combined chemoradiotherapy is
superior to radiotherapy alone (Figure 13.2). Several
single arm phase II trials and a recently reported
randomized trial used chemoradiotherapy as an upfront
approach to patients with PCNSL (Table 13.3). With
variable selection criteria, chemotherapy combinations
and radiation parameters, CRRs oscillate between 30%
and 87%, with a 5-year OS of 30–50%.
Late neurotoxicity is a major complication in
patients treated with HD-MTX-based chemotherapy
followed by WBRT, with a cumulative incidence
varying from 25% to 35% at 5 years. Long-term
impairment in the areas of attention, executive func-
tion, memory, and psychomotor speed are the most
commonly reported. One-third of these patients died
of complications related to the neurotoxicity. The
mechanisms by which treatment produces CNS dam-
age are unknown, but demyelination, tissue necrosis,
and microcavity changes have been described; con-
tributing factors are advanced age, neurological
comorbidity, genetic predisposition, and treatment
characteristics. Treatment-related neurotoxicity in
PCNSL patients has not been defined clearly because
prospective assessment of neurotoxicity has been per-
formed rarely, assessed series are small and exposed
to selection biases, and there is no agreement about
neurotoxicity definition, tests and scores useful to
measure neurocognitive functioning as well as
approaches to radiographic monitoring of treatment
effect in PCNSL. A group of investigators of the
Chapter 13: Central nervous system lymphomas
Figure 13.2 MRI showing dural lymphoma of the left
parietotemporal region (arrows) infiltrating the skull and the surface
of the brain. The lesion was entirely resected because the main
clinical suspicion was meningioma. Histological diagnosis was diffuse
large B-cell lymphoma.
International PCNSL Collaborative Group has pro-
posed a panel of neuropsychological tests to assess,
quantify, and follow up treatment-related neurologic
deterioration. A wide use of this panel will allow
better definition of this severe complication both in
prospective trials and ordinary clinical practice.
Authorities have proposed different therapeutic
strategies to reduce the risk of severe iatrogenic neuro-
toxicity in PCNSL patients. The first one is to avoid the
use of consolidation WBRT in patients in CR after
primary chemotherapy. Some small experiences sug-
gest that this approach is feasible in older patients,
with an OS similar to those obtained with chemora-
diotherapy combinations. In the largest randomized
phase III trial, 551 patients treated with HD-MTX-
based chemotherapy were randomly allocated to
receive complementary WBRT 45 Gy versus observa-
tion in the case of CR after chemotherapy or versus
HD-araC in the case of partial or no response. The
data indicate that consolidation WBRT is associated
with a significantly better PFS, but does not change
survival. Unfortunately, this trial is fraught with
design and execution flaws that could result in impor-
tant interpretation biases.
The second strategy to reduce neurotoxicity is to
optimize radiation parameters; in particular, whole-
brain dose, which is directly correlated to severity of
neurotoxicity. A retrospective study and a single arm
phase II trial suggest that WBRT dose can be reduced 213
 Chapter 13: Central nervous system lymphomas
to 23–30 Gy in patients in CR after primary chemo-
therapy obtaining similar outcome with respect to
higher doses, with better neurotolerability. In fact,
these patients complained only of difficulties in verbal
memory and motor speed.
The third strategy to minimize neurotoxicity is to
replace WBRT with an experimental approach such as
high-dose chemotherapy-supported autologous stem
cell transplantation (HDC/ASCT) or BBB disruption
(see below). These three main strategies will remain
the foremost instruments to fight neurotoxicity risk
until biological, molecular, and technological research
furnishes new relevant information on how to maintain
neuropsychological performance in these patients.
Experimental therapeutic strategies
High-dose chemotherapy supported by
autologous stem cell transplantation
HDC/ASCT can be used to dose-intensify chemother-
apy as well as to replace WBRT in an effort to avoid
treatment-related neurotoxicity. The rationale for the
use of HDC/ASCT is multiple and includes the admin-
istration of high doses of cytostatics to overcome drug
resistance and to achieve therapeutic concentrations in
the lymphoma tissue and other chemotherapy sanctua-
ries, like CSF, meninges, and eyes, where lymphoma
cells usually grow. Initially, HDC/ASCT was used in
patients with relapsed or refractory PCNSL, obtaining
a CRR of 60%, treatment-related mortality of 16%,
severe neurotoxicity in 12%, and a 2-year OS of 45%.
These encouraging results prompted investigators to
include HDC/ASCT as part of first line treatment of
PCNSL, but published prospective experience is limited
to a few small single arm phase II trials (Table 13.4).
Activity data are controversial because of the different
induction and conditioning combinations used. These
trials included induction chemotherapy with two to five
cycles of MTX from 3 to 8 g/m2, with or without
intensification and followed by conditioning chemo-
therapy supported by ASCT, followed or not by
WBRT. Cumulative data from these few trials seem to
suggest that HD-MTX-based polychemotherapy is
more active than monochemotherapy, while intensifi-
cation with araC, alone or in combination with thiotepa
or ifosfamide, was useful as a stem cell mobilizing
regimen, but did not further improve response rates.
Overall outcome seems to be strongly related to the type
214 of ASCT conditioning regimen (Table 13.4), with a
busulfan–thiotepa combination appearing particularly
toxic, and a thiotepa-containing regimen being more
effective. However, use of a BEAM regimen (BCNU,
etoposide, araC, and melphalan) without WBRT,
chosen because these drugs could penetrate the BBB
and the regimen is well tolerated in older patients, was
ineffective with an overall, median event-free survival
(EFS) of 9.3 months for those patients who completed
transplant (Table 13.4, trial 5). However, a more recent
small prospective study suggests that HDC/ASCT could
replace consolidation radiotherapy in patients with
newly diagnosed PCNSL (Table 13.4, trial 9). The real
impact of HDC/ASCT on neuropsychological functions
has not been defined since treated patients were not
prospectively assessed with adequate neuropsychologi-
cal tests, and only clinical criteria were used to detect this
severe complication. Thus, the comparison between
WBRT and HDC/ASCT in two ongoing randomized
trials will establish the most effective and best tolerated
strategy to consolidate response obtained with primary
chemotherapy.
Blood–brain barrier disruption
Reversible BBB disruption (BBBD) by intra-arterial
infusion of hypertonic mannitol followed by intra-
arterial chemotherapy is a strategy that leads to
increased drug concentrations in the lymphoma-
infiltrated brain and thus may improve survival. In
institutions with adequate expertise, BBBD plus HD-
MTX has been associated with acceptable morbidity,
high tumor response and survival rates, and only a
14% loss of cognitive function at 1 year. In a recently
reported retrospective series of 149 patients with newly
diagnosed PCNSL treated with this strategy at four
institutions, the CRR was 58%, with a 5-year PFS of
31%, and a median OS of 3 years. In relapsed patients,
carboplatin-based chemotherapy plus BBBD pro-
duced a 36% response rate, with a median duration
of 7 months. BBBD may prove most useful in the
delivery of agents unlikely to traverse an intact BBB.
However, this strategy is a procedurally intensive
treatment, requiring monthly intravascular interven-
tions under general anesthesia over 1 year.
Particular clinical conditions
Leptomeningeal lymphoma
PCNSL tends to infiltrate the subependymal tissues,
disseminating through the CSF to the meninges;
Table 13.4 Prospective phase II trials on high-dose chemotherapy supported by autologous stem cell transplantation in PCNSL.

Ref. No. Median Treatment Therapy Conditioning WBRT Outcome Neurotoxicity (%) Median TRM
of age line (induction → regimen follow-up (%)
pts. (range) intensification) (months)
1 22 53 (27–64) Salvage araC + VP16 TT/Bu/Cy No 3-year 32 41 4
OS 64%
2 43 52 (23–65) Salvage araC + VP16 TT/Bu/Cy No 2-year 5 36 7
OS 45%
3 6 53 (30–66) First line MBVP → IFO + araC BEAM Yes 2-year 33 41 0
OS 40%
4 25 52 (21–60) First line MBVP → IFO + araC BEAM Yes 4-year 8 34 4
OS 64%
5 28 53 (25–71) First line HD-MTX → araC BEAM No 2-year 0 28 4
OS 55%
6 23 55 (18–69) First line HD-MTX → – Bu/TT Yes§ 2-year 39 15 13
OS 48%
7 7 56 (41–64) First line* HD-MTX → araC Bu/TT/Cy No 3-year 0 28 14
OS 50%
8 30 54 (27–64) First line HD-MTX → araC + TT BCNU/TT Yes 5-year 17 63 3
OS 69%
9 13 54 (38–67) First line HD-MTX → araC + TT BCNU/TT Yes§ 3-year 0 23 0
OS 77%
araC, cytarabine; BCNU, carmustine; BEAM, carmustine, etoposide, cytarabine, and melphalan; Bu, busulfan; Cy, cyclophosphamide; IFO, ifosfamide; MBVP (regimen), methotrexate,
carmustine, etoposide, and methylprednisolone; OS, overall survival; TRM, treatment-related mortality; TT, thiotepa; VP16, etoposide; WBRT, whole-brain irradiation.
* One patient received the treatment as salvage therapy.
§ Only for patients not achieving a complete remission.
References: (1) Soussain C, et al. J Clin Oncol, 2001;19(3):742–749; (2) Soussain C, et al. J Clin Oncol, 2008;26(15):2512–2518; (3) Brevet M, et al. Eur J Haematol, 2005;75(4):288–292; (4) Colombat P,
et al. Bone Marrow Transplant, 2006;38(6):417–420; (5) Abrey LE, et al. J Clin Oncol, 2003;21(22):4151–4156; (6) Montemurro M, et al. Ann Oncol, 2007;18(4):665–671; (7) Cheng T, et al. Bone Marrow
Transplant, 2003;31(8):679–685; (8) Illerhaus G, et al. J Clin Oncol, 2006;24(24):3865–3870; (9) Illerhaus G, et al. Haematologica, 2008;93(1):147–148.
 Chapter 13: Central nervous system lymphomas
leptomeningeal lymphoma in the absence of a paren-
chymal mass represents less than 5% of all PCNSL
(Figure 13.3). Meningeal dissemination is detected by
conventional CSF cytology examination in 16% of
patients, by polymerase chain reaction (PCR) in 11%,
and by magnetic resonance imaging (MRI) in 4%.
However, the rate of discordant PCR and cytomor-
phologic results is high, and cytology and PCR should
be regarded as complementary methods. In effect, an
autopsy study demonstrating meningeal involvement
in 100% of cases suggests that leptomeningeal dissem-
ination is underestimated with current methods.
Prognosis of leptomeningeal lymphoma is variable,
owing to the difficulty of delivering effective treatment
to the subarachnoid space, which could be optimized by
using an Ommaya reservoir. However, indications for
intrathecal/intraventricular chemotherapy are debat-
able considering that its efficacy has not been studied
prospectively, while a large PCNSL retrospective study
showed no benefit from intrathecal therapy. Conversely,
the comparison of two phase II trials seems to provide
indirect evidence of the benefit of intraventricular drug
delivery in PCNSL. In fact, the exclusion of intraven-
tricular chemotherapy was associated with worse results
with respect to the same chemotherapy plus intraven-
tricular drug delivery. However, the theoretical advant-
age of intraventricular drug administration may not
outweigh the additional risk of infective complications
of the reservoir, and the increased risk of neurotoxicity
216
and chemical meningitis associated with this strategy.
Figure 13.3 Flow diagram of
recommended therapy for PCNSL
according to eligibility for chemotherapy,
trial availability, and age.
a
The disputes on the impact of survival
and neurotoxicity of consolidation
radiotherapy in elderly patients in
complete remission after primary
chemotherapy play a central role in the
design of a risk-tailored treatment against
PCNSL. The best cut-off of age to
distinguish subgroups of patients with
different risks of neurotoxicity after
consolidation radiotherapy remains to be
defined. HD-MTX, High-dose
methotrexate; CR, complete response; PR,
partial response; SD, stable disease; PD,
progressive disease.
Finally, therapeutic MTX concentrations can be
achieved in CSF using intravenous doses ≥3 g/m2, and
preliminary data suggest that systemic HD-MTX is able
to clear the CSF of neoplastic cells.
Intraocular lymphoma
In 5–20% of patients, PCNSL involves the eyes.
Intraocular lymphoma (IOL) can occur in isolation
(primary ocular lymphoma) or as a component of
more extensive PCNSL. Since the eye is an extension
of the CNS, its involvement is not considered systemic
dissemination, even if bilateral. The neoplastic cells
can infiltrate the vitreous humor, the retina, the cho-
roid, and, less frequently, the optic nerve. Patients
typically present with floaters and blurred vision and
are often misdiagnosed as having a benign ophthal-
mologic condition. Patients treated for isolated IOL
have an 80% risk of developing cerebral involvement
up to 10 years or more after initial diagnosis.
The prognosis of IOL is similar to that of PCNSL
without ocular involvement. There is no standard treat-
ment for isolated IOL. Systemic administration of MTX
and cytarabine can yield therapeutic levels of drug in the
intraocular fluids and clinical responses have been
documented; however, relapse is common. The efficacy
of cytostatics is dependent on intraocular pharmacoki-
netics, which are not well understood, and preliminary
data suggest that intraocular concentrations of MTX are
achieved when the drug is given at 8 g/m2, with a high

rate of persistent ocular disease. Better disease control
has been reported by combining ocular irradiation with
HD-MTX-based chemotherapy. Irradiation of the pos-
terior two-thirds of the globes with 35–45 Gy has been
recommended, and more recent experience suggests
using radiotherapy of the entire orbit up to 20 Gy,
followed by an additional 10 Gy after shielding the
anterior chamber of the eyes. Even in the presence of
unilateral disease, both eyes should be irradiated
because microscopic bilateral involvement occurs in
80% of patients. The actual incidence of ocular compli-
cations (cataract, dry eyes, punctate keratopathy, retin-
opathy, optic atrophy) is not reported because of the
short follow-up of published series; however, cataract
occurs in virtually all patients.
The poor results obtained with conventional strat-
egies have induced investigators to identify new ther-
apeutic approaches. Intravitreal injections of MTX
have been associated with better intraocular disease
control (ORR 100%) in small series of patients with
relapsed IOL, but this benefit did not affect OS.
Moreover, important side effects have been reported
in up to 73% of treated eyes, with a significant deteri-
oration of visual acuity in 27% of patients. Drugs other
than MTX have been administered by this route in
IOL patients. The administration of rituximab appears
safe and possibly efficacious, whereas intravitreal thi-
otepa requires further investigation.
Spinal cord lymphoma
Primary lymphomas of the spinal cord are the rarest
manifestation of PCNSL. These lymphomas often arise
in the upper thoracic or lower cervical regions of the
spinal cord. Presenting symptoms depend on the spinal
cord level involved. Myelography shows a widened
spinal cord. MRI reveals hyperintensity on T2-images
and homogeneous enhancement after gadolinium
administration on T1-weighted images. Increased CSF
protein concentration is common. Lymphomas arising
in the spinal nerves and ganglia (“neurolymphomato-
sis”), cauda equina, and the sciatic nerve are extremely
rare and should be distinguished from neural infiltra-
tion by a systemic lymphoma. Prognosis of patients
with spinal cord lymphoma is poor, mainly because of
delayed diagnosis. Therapeutic results are strongly
affected by pre-treatment neurological status. Similar
to other PCNSL, corticosteroids and radiotherapy are
associated with lymphoma regression and clinical
improvement. Only sparse data on chemotherapy effi-
cacy are available; however, spinal cord lymphomas
Chapter 13: Central nervous system lymphomas
should be treated similarly to other PCNSL. Because
neurolymphomatosis is a condition outside the CNS,
there may be a rationale to incorporate rituximab (in
CD20-positive lymphomas) and CHOP chemotherapy
in combination with agents that penetrate the blood–
nerve barrier in the treatment of this entity.
Relapsed disease
Median survival for patients with recurrent or refrac-
tory PCNSL without treatment is 2 months, but sal-
vage treatment may substantially prolong survival.
Unfortunately, the optimal salvage regimen has not
been defined and a variety of possible strategies have
been reported (Table 13.5). The choice of the optimal
salvage strategy should take into consideration the
patient’s age, PS, site of relapse, and prior therapy.
Radiotherapy may be used at recurrence in previously
non-irradiated patients; salvage WBRT has been
associated with an ORR of 79% and a median survival
after relapse of 16 months in patients who experi-
enced failure after initial HD-MTX. Salvage chemo-
therapy is often used in an effort to improve disease
control while reducing neurotoxicity. Reinduction
with HD-MTX resulted in a response in approxi-
mately 50% of patients, with a median PFS of 10
months. Preliminary experiences showed encourag-
ing but unconfirmed results with polychemotherapy
combinations (Table 13.5). Patients with relapsed or
refractory PCNSL constitute the optimal context to
identify potentially new active drugs.
Isolated systemic (extra-CNS) dissemination is
observed in 3–7% of failures among patients with
PCNSL; this rare condition seems to be associated
with a significantly better outcome with respect to
CNS relapses, in particular, if treated with conven-
tional antilymphoma chemotherapy.
CNS involvement in HIV-related
lymphomas
PCNSL in immunocompromised patients has differ-
ent characteristics and behavior compared to the
immunocompetent population (Table 13.6). EBV
infection and c-myc translocation result in the prolif-
eration of malignant lymphocytes in HIV patients; a
CD4 count <50 cells/μL and high peripheral HIV viral
load are the most significant risk factors for PCNSL
development. Since the introduction of highly active
antiretroviral therapy (HAART), the incidence of 217
AIDS PCNSL has declined.
Table 13.5 Salvage regimens for PCNSL.

Treatment Study No. Median Prior PS CR + mPFS mOS 1-yr N PLT Other
age RT 0–1 PR OS toxicities
Topotecan P 27 51 52% 60% 19+14% 2.0 8.4 39% 26% 15% 11%
Temozolomide P 36 60 86% 28% 25+6% 2.8 4.0 31% 6% 3% 3%
Methotrexate R 22 58 14% – 73+19% 26 26 70% 5% 5% 36%
Temozolomide + rituximab R 15 69 13% 67% 40+13% 2.2 10.5 58% 7% 27% 7%
VP16 + ifosfamide + araC R 16 54 100% 37% 37+0% 4.5 6.0 41% 69% 50% 37%
i. a. carboplatin ± VP16 ± CTX ± RT R 37 57 24% 76% 24+11% 3.0 6.8 25% 22% 19% >30%
Radiotherapy R 27 67 – – 37+37% 9.7 10.9 49% – – 15% neuro
Radiotherapy R 20 – – – 60 + NR – 19.0 – – – 58% neuro
%
HD-araC + VP16 → TTP + busulfan P 43 52 33% – 50 + 0% – 18 – – – TRM: 1%
+ CTX
Intrathecal rituximab P 10 56 80% – 0 + 60% – 5.2 30%
Study: P, prospective; R, retrospective; RT, radiotherapy; PS, performance status; CR, complete response; PR, partial response; mPFS, median progression-free survival; mOS, median overall
survival; 1-year OS, 1-year overall survival; N, grade 3–4 neutropenia; PLT, grade 3–4 thrombocytopenia; tox, toxicity; i.a., intra-arterial; TRM, treatment-related mortality; VP16, etoposide; HD-
araC, high-dose cytarabine; TTP, thiotepa; CTX, cyclophosphamide.
 Chapter 13: Central nervous system lymphomas

AIDS PCNSL is typically ring-enhancing on MRI.
Cerebral toxoplasmosis, abscesses, and progressive
multifocal leukoencephalopathy are the main differ-
ential diagnoses. Single-photon emission computed
tomography (SPECT), PET, and proton MR spectro-
scopy may distinguish tumor from infectious pro-
cesses, but are not always definitive. Given the variety
of intracranial diseases indistinguishable by imaging, a
brain biopsy is often required for diagnosis, particu-
larly if the patient has rapid neurologic deterioration,
negative CSF cytology, negative toxoplasma serology,
radiographic features atypical for toxoplasmosis, and
progressing symptoms within the first 1–2 weeks of
antitoxoplasmosis therapy. Surgical biopsy could be
avoided in AIDS patients with brain lesions that have
both increased thallium uptake on SPECT scanning
and a positive CSF for EBV DNA. The combination of
these procedures is associated with 100% sensitivity
and specificity. SPECT is insufficient as a sole diag-
nostic modality and may give a false-negative result in
lesions <0.6 cm, located near the skull or ependyma, or
after steroid therapy.
Because they are profoundly immunocompromised,
these patients have a much worse prognosis than non-
AIDS patients. They tolerate cytotoxic chemotherapies
poorly and develop more infectious complications.
Bone marrow toxicity and opportunistic infections fre-
quently complicate treatment regimens. In some series,
up to 40% of patients did not receive any treatment.
Before HAART, the mean survival was 1–2 months
without treatment. WBRT improved OS to 3–4 months.
Since the advent of HAART, the ability to deliver
adequate chemotherapy to AIDS patients with PCNSL
has improved. However, only 10% of AIDS patients
with PCNSL are eligible for chemotherapy because of
comorbidity or poor neurological condition. A few cases
of HIV-positive patients with PCNSL successfully treat-
ed with HAART, with or without other specific antitu-
mor therapy, have been published. Therefore, HIV-
positive patients with newly diagnosed PCNSL should
be started on HAART (or have their HAART regimen
optimized) and receive appropriate prophylaxis for
opportunistic infections. It is possible that this approach
will be adequate in patients who are antiretroviral drug-
naïve. However, most patients will require additional
specific antitumor therapy. Given the likelihood that
WBRT may exacerbate or accelerate the risk of HIV-
related dementia, it seems most appropriate to initiate
therapy with MTX-based chemotherapy, similar to the
approach in immunocompetent patients. HD-MTX,

Table 13.6 Clinical, radiological, and histological features of PCNSL in HIV-infected patients.

Incidence Decreasing since HAART

Pathogenesis EBV infection, c-myc translocation
Male:female ratio 7.4:1
Median age (years) 30
Interval from symptoms to diagnosis <2 months
Clinical presentation Mental status changes in >50%
Number of lesions Multiple in >50%
CSF cytological examination Positive: 25%
Histotype DLBCL: 75%
Small non-cleaved: 25%
T-cell phenotype Very rare
EBV genomic DNA Positive
MRI appearance Multifocal and heterogeneous lesions
Small perilesional edema and mass effect
Ring enhancement common§
Sites of disease Peripheral or cortically based
Rarely involve the corpus callosum
EBV, Epstein–Barr virus; CSF, cerebrospinal fluid; DLBCL, diffuse large B-cell lymphoma.
§ Ring enhancement results from central necrosis and subacute hemorrhage.
219
 Chapter 13: Central nervous system lymphomas
Table 13.7 Relevant clinical aspects regarding secondary CNS lymphomas.
Incidence of CNS recurrence according to histotype
Indolent lymphomas
Aggressive lymphomas (DLBCL and PTCL)
Highly aggressive lymphomas (Burkitt and lymphoblastic)
Interval between NHL diagnosis and CNS recurrence
Median
Range
Later than 1 year of follow-up
Site of recurrence (% of relapses)
CNS + systemic recurrence
CNS recurrence followed by systemic progression
Isolated CNS recurrence
Survival data
1-year survival after CNS recurrence
Median survival after relapse
DLBCL, Diffuse large B-cell lymphoma; PTCL, peripheral T-cell lymphomas.
alone or in combination, with or without intrathecal
drug delivery, has been associated with a response rate
of 33–57% and a median survival of 3–4 months. The
combination of oral zidovudine and HD-MTX admin-
istered weekly for three to six cycles resulted in a 46%
CRR and a median survival of 12 months. In contrast,
the use of CHOP followed by WBRT has been associ-
ated with disappointing results. In a small group of
highly selected patients (lesions limited to the brain
and high CD4 counts), WBRT associated with hydroxy-
urea followed by PCV (procarbazine, lomustine, and
vincristine) chemotherapy was associated with a 100%
response rate and a median survival of 13 months.
Experimental strategies that directly target EBV to
induce lymphoma remission are under investigation.
These approaches include direct antiviral therapy (zido-
vudine, ganciclovir, and interleukin-2), monoclonal
antibodies, cytotoxic T-lymphocytes, episomal replica-
tion inhibition by hydroxyurea or antisense oligonu-
cleotides, pro-drug activation by tumor cells, and
kinase expression upregulation and lytic induction.
However, clinical experience is still limited and successes
are exceptional. Depleting the viral reservoir (i.e.
B-lymphocytes) in high-risk patients by intravenous
rituximab could play a relevant role in this setting.
Secondary CNS lymphomas
Prophylaxis against CNS relapse
CNS dissemination is an almost uniformly fatal
220 complication of hematological malignancies. The
<3%
5%
24%
3–6 months
0–44 months
4%
20%
30%
5%
25%
4–5 months
risk of CNS relapse ranges between 5% and 30% in
all NHL subtypes. The most relevant clinical aspects
of this heterogeneous condition are summarized in
Table 13.7. Given the high morbidity and mortality
associated with CNS dissemination of NHL, a pro-
phylactic strategy, analogous to that employed suc-
cessfully for acute lymphoblastic leukemia (ALL), is
indicated. However, the incidence of CNS recurrence
in NHL is not sufficiently high to warrant the use of
CNS prophylaxis in all patients. The analysis of risk
factors for CNS relapse in NHL is biased by differ-
ences in the definition criteria and retrospective
nature of reported studies. Nevertheless, advanced
disease, increased LDH serum levels, certain extra-
nodal sites of disease, and highly aggressive lympho-
mas (Burkitt and lymphoblastic lymphomas) have
been associated with an increased risk for CNS
recurrence. The risk is increased for patients with a
high IPI score, particularly those with a high serum
LDH level, or involvement of more than one extra-
nodal site, but CNS recurrence occurs in all the IPI
risk groups. Patients displaying involvement of mul-
tiple extranodal organs and increased LDH serum
levels have a 1-year risk for CNS recurrence of 20%.
These patients, as well as those with testicular or
paranasal sinus involvement, should be assessed for
CNS involvement at diagnosis.
In highly aggressive lymphomas, the 5-year CNS
recurrence rate was substantially higher among
patients who did not receive CNS prophylaxis
(32–78%) in comparison to patients who did (19%).
This wide variation may be explained by a variable

proportion of T-cell lymphoblastic lymphoma (LbL)
patients in the study cohorts. In fact, patients with
T-cell LbL or T-cell ALL have a higher risk for CNS
relapse. The inclusion of CNS-directed prophylaxis
led to a substantial reduction in CNS relapse rates in
LbL patients; 0–36% with intrathecal chemotherapy
alone, 3–21% with intrathecal chemotherapy and
WBRT, and <5% with intrathecal chemotherapy,
WBRT, and systemic HD-MTX and HD-araC.
For aggressive lymphomas, CNS prophylaxis is
currently used in 10–15% of cases; however, there is
still no uniform practice, which reflects the fact that
published data are open to different interpretations.
Intrathecal chemotherapy and the addition of
systemic HD-MTX or HD-araC to conventional che-
motherapy reduces the CNS relapse rate and appears
to be as effective as and less toxic than WBRT.
Incorporating both intrathecal MTX and consolida-
tion with systemic HD-MTX, ifosfamide, and cytar-
abine significantly reduced CNS recurrence from
8.3% to 2.7% in NHL patients aged 61–69 years
with IPI ≥1, in comparison with standard CHOP.
The use of conventional-dose rituximab to prevent
CNS relapse in patients with DLBCL treated with
CHOP has been associated with conflicting results
in two randomized trials.
Treatment of SCNSL
In a large proportion of patients, CNS relapse is
accompanied or rapidly followed by systemic relapse,
and the patient’s outcome is determined equally by
the control of systemic and CNS disease. Systemic
high-dose chemotherapy is the most effective strategy
that meets these requirements. Similar to PCNSL, the
choice of drugs is based on the ability to cross the
BBB and have antilymphoma activity. Most first line
combinations against SCNSL contain HD-MTX or
HD-araC, but the most effective administration
schedules for these drugs in SCNSL remain to be
defined. Most lymphoma patients with a high risk
of CNS relapse may receive these drugs as first line
treatment, so salvage options may be limited because
few other drugs cross the BBB. Some anecdotal expe-
rience with cisplatin-based regimens has been
reported, and autologous or allogeneic transplanta-
tion may prove beneficial for selected patients.
Patients with leptomeningeal involvement without
focal neurological deficits are usually treated with
intrathecal chemotherapy (MTX, cytarabine and
Chapter 13: Central nervous system lymphomas
hydrocortisone) twice a week using an Ommaya res-
ervoir. Although this strategy invariably results in a
decrease in the number of tumor cells in the CSF,
symptomatic improvement occurs in <20% of cases.
Slow-release formulation of cytarabine, injected once
every 2 weeks, produces a higher response rate and a
better quality of life relative to that produced by free
cytarabine injected twice a week except for the high
incidence of chemical meningitis associated with lip-
osomal araC. Patients with focal neurological deficits
or intraparenchymal lesions in the brain, cranial
nerves, or spinal cord are commonly treated with
radiation therapy, which results in symptomatic
improvement in >65% of cases; however, CNS or
systemic progression is the rule. The optimal approach
to managing overt CNS disease at diagnosis in chil-
dren with large-cell lymphoma is controversial
because of the low frequency and marked heterogen-
eity of this group of malignancies. In children with
Burkitt lymphoma, HD-MTX and cytarabine and
intrathecal chemotherapy are currently used for both
CNS prophylaxis and treatment.
Some of the most successful treatment regimens
against highly aggressive lymphomas have excluded
WBRT. In children, there is a trend toward reducing
the use of radiotherapy for CNS disease and prophy-
laxis to reduce late sequelae such as second cancer,
endocrinopathy, and neuropsychologic defects. In
recent clinical trials, the prophylactic WBRT dose
has been reduced to 12 Gy, and the therapeutic
dose to 18 Gy, while the safety of WBRT omission
is being assessed in current trials on LbL in children.
In recent trials, the elimination of WBRT has not
been associated with an excessive rate of CNS
relapse.
Autologous transplantation
Some reports conclude that 20–40% of adults with
NHL and CNS disease can achieve durable remissions
after HDC/ASCT, primarily patients with highly
aggressive lymphomas in first remission at the time
of transplantation. On the other hand, attempts to
transplant adults with active CNS disease at the
moment of autologous transplantation have had dis-
appointing results, with a PFS of 9% at 71 months.
Moreover, it is difficult to estimate the exact contri-
bution of autologous transplantation to survival in
patients with highly aggressive lymphomas and CNS
involvement at diagnosis, which may be curable with
conventional-dose treatment. In some series, overall 221
 Chapter 13: Central nervous system lymphomas
results with HDC/ASCT are disappointing, which
could be partially explained by a high incidence of
fatal CNS complications in previously irradiated
patients.
Conditioning regimens are not specifically
designed for patients with CNS disease. Ideally,
drugs should be chosen on the basis of antilymphoma
activity and capacity to cross the BBB. In fact, encour-
aging results with a conditioning regimen based on
total body irradiation (TBI), busulfan, cyclophospha-
mide, thiotepa, and nitrosourea have been reported.
TBI is highly immunosuppressive, has good antitu-
mor activity, and is not affected by the BBB.
However, this strategy is associated with an increased
risk of severe neurotoxicity, especially when com-
bined with intrathecal chemotherapy or systemic
HD-MTX or araC. Busulfan and cyclophosphamide,
with or without thiotepa, can be curative in children
with highly aggressive lymphomas and CNS involve-
ment, but novel conditioning combinations need to
be investigated considering the disappointing results
with this strategy.
Allogeneic transplantation
There is some evidence to suggest that allogeneic
transplantation is superior to ASCT for the prevention
or treatment of CNS recurrence in highly aggressive
lymphomas and ALL, both in children and adults.
Moreover, consolidation with allogeneic transplanta-
tion is associated with significantly improved outcome
in children with early CNS recurrence of ALL. It is
conceivable that graft-versus-lymphoma effects may
extend to the CNS and contribute to eradication
of CNS disease. However, survival benefit with this
strategy is obscured by increased treatment-related
mortality. Moreover, the real impact of allogeneic
transplantation in CNS recurrence is difficult to define
considering the interpretation bias related to the tox-
icity of immunosuppressive therapy used for preven-
tion or treatment of graft-versus-host disease. Better
results with allogeneic transplantation can be obtained
in patients whose CNS disease is in remission at
transplant.
Rare forms of CNS lymphoma
Rare histopathological variants of CNS lymphomas
exist (Table 13.8). Diagnostic, staging, and therapeutic
approaches to these malignancies are variable, and
222 clinical evidence supporting therapeutic decisions is
often anecdotal.
Indolent lymphomas
Low-grade marginal zone B-cell lymphoma of MALT-
type primarily involving the CNS usually arises in the
dura (Figure 13.4). These lymphomas have excellent
long-term prognosis with local therapy alone, and
consolidation radiotherapy may be unnecessary after
complete resection. Immunocytoma has been listed as
accounting for up to 13% of all PCNSL. However,
many CNS immunocytoma cases reported before the
WHO classification likely represent examples of
MALT lymphomas. This indolent malignancy can be
treated with radiotherapy with acceptable disease con-
trol; experience with HD-MTX is anecdotal and sug-
gests modest activity.
Aggressive lymphomas
PCNSL of T-cell phenotype is more common in
Japan (8% of PCNSL) compared to Western coun-
tries (Table 13.8). The diagnosis of T-cell PCNSL can
be difficult, and is possibly overestimated, owing to
the frequent infiltration of reactive perivascular
T-cells, which could interfere with the interpretation
of immunophenotyping, particularly after steroid
administration. T-cell PCNSL should be treated
in an identical fashion to standard PCNSL with the
expectation of similar results, at least in Western
countries. The overall therapeutic approach to
primary CNS anaplastic large-cell lymphoma, usu-
ally displaying T-cell immunophenotype, is similar
to the current therapy of PCNSL; meningeal
prophylaxis appears advisable. Young patients and
ALK-1-positive lymphomas seem to have a better
outcome.
Intravascular large B-cell lymphoma is an aggres-
sive and disseminated malignancy, characterized
by rapidly progressing manifestations of multiorgan
failure caused by multiple infarcts, with CNS
involvement in 34% of cases. Intravascular lym-
phoma involving the CNS has a worse prognosis,
requiring conventional R-CHOP chemotherapy
in combination with agents with high CNS
bioavailability.
CNS involvement by plasmacytoma is rare
(Table 13.8), being the sole site of disease in one-
half of cases. Intracranial plasmacytoma is unlikely to
develop systemic disease in patients with initial
negative staging. Involved-field irradiation with 50
Gy has been reported as an acceptable strategy
in patients with solitary intracranial plasmacytoma;
 Chapter 13: Central nervous system lymphomas

Table 13.8 Rare forms of CNS lymphoproliferative disorders.

Category % of Age/ Stage Clinical features Therapeutic

PCNSL gender (usually) management
T-cell 1–4% M>F IEA Systemic symptoms: 11% Similar to B-cell PCNSL
lymphomas 60 years Similar outcome
Anaplastic <1% M>F IEA T-cell; common meningeal Similar to PCNSL + CSF prophy-
large-cell 29 years disease; normal LDH levels; laxis; better outcome in young
lymphomas ↑CSF protein ALK-1+ patients
Intravascular <1% M>F IVA Poor PS, B symptoms, Anthracycline-based
lymphomas 70 years anemia, chemotherapy
↑LDH levels; multiorgan ± rituximab ± drugs with high
failure CNS bioavailability; 3-year OS of
Exclusive CNS involvement 25%
rare
Low-grade 5% M<F IEA Usually arises in dura Excellent prognosis with local
marginal young Favorable clinical outcomes therapy; RT may be
zone B-cell Meningioma is differential unnecessary after complete
lymphoma diagnosis resection
of MALT-type
Immunocytoma 4–13% M<F IEA Indolent course HD-MTX has poor (anecdotal)
Young Radiological suspicion with activity; good control with WBRT
MRI
Plasmacytoma <1% M>F IEA/ Trend to remain within CNS RT 50 Gy alone if solitary
elderly IVA Favorable outcome if Chemotherapy in large lesions§
solitary
Hodgkin 1.4% M=F IEA/ Isolated brain or meningeal WBRT has acceptable DFS; intra-
lymphoma 50–85 IVA lesion; nodular sclerosis in thecal therapy in CSF+ patients
years 50% of cases
M, male; F, female; LDH, lactate dehydrogenase; PS, performance status; CSF, cerebrospinal fluid; ↑, elevated.
§
Disseminated plasmacytoma with CNS disease should be treated like multiple myeloma with the addition of drugs with high CNS
bioavailability.

patients with large lesions have obtained clinical

Figure 13.4 MRI showing lymphoma (arrows) of the
parasellar region. This well-limited lesion was entirely resected;
histological diagnosis was low-grade marginal zone B-cell
lymphoma.
benefit from chemotherapy.
Primary CNS Hodgkin lymphoma (HL) may
present as an isolated brain lesion, meningeal dis-
semination with subdural infiltration, or as a calva-
rial lesion with meningeal or cerebral infiltration.
One-half of reported cases had nodular sclerosis.
The detection of Reed–Sternberg cells or eosinophilia
in the CSF is uncommon. WBRT (35–45 Gy) fol-
lowed by a tumor-bed boost (5–15 Gy) has provided
acceptable outcome, while the role of systemic che-
motherapy remains to be defined. In a large retro-
spective series, the median OS was 61 months from
first diagnosis of HL and 44 months from diagnosis
of CNS involvement. Patients who achieve a CR to
treatment, particularly those who present with CNS
involvement as the sole site of relapse and who
received RT, exhibit the best outcome.
223
 Chapter 13: Central nervous system lymphomas
Further reading
Abrey LE, Batchelor TT, Ferreri AJ, et al. Report of
an international workshop to standardize
baseline evaluation and response criteria for
primary CNS lymphoma. J Clin Oncol,
2005;23(22):5034–5043.
Abrey LE, Ben-Porat L, Panageas KS, et al. Primary central
nervous system lymphoma: the Memorial Sloan-
Kettering Cancer Center prognostic model. J Clin Oncol,
2006;24:5711–5715.
Angelov L, Doolittle ND, Kraemer DF, et al. Blood–brain
barrier disruption and intra-arterial methotrexate-based
therapy for newly diagnosed primary CNS lymphoma: a
multi-institutional experience. J Clin Oncol,
2009;27(21):3503–3509.
Beral V, Peterman T, Berkelman R, et al. AIDS-associated
non-Hodgkin lymphoma. Lancet,
1991;337(8745):805–809.
Brannan SO, Matthews BN, Savant V, Brown RD, Matthews
TD. Solitary intracranial extra-osseous plasmacytoma
presenting with ophthalmic signs. J Neuroophthalmol,
2003;23(4):268–271.
Bromberg JE, Siemers MD, Taphoorn MJ. Is a “vanishing
tumor” always a lymphoma? Neurology,
2002;59(5):762–764.
Camilleri-Boet S, Davi F, Feuillard J, et al. AIDS-related
primary brain lymphomas: histopathologic and
imunohistochemical study of 51 cases. French Study
Group for HIV-associated Tumours. Human Pathol,
1997;28(3):367–374.
Camilleri-Broet S, Criniere E, Broet P, et al. A uniform
activated B-cell-like immunophenotype might explain
the poor prognosis of primary central nervous system
lymphomas: analysis of 83 cases. Blood,
2006;107(1):190–196.
Correa DD, Maron L, Harder H, et al. Cognitive functions in
primary central nervous system lymphoma: literature
review and assessment guidelines. Ann Oncol,
2007;18(7):1145–1151.
Correa DD, Rocco-Donovan M, DeAngelis LM, et al.
Prospective cognitive follow-up in primary CNS
lymphoma patients treated with chemotherapy and
reduced-dose radiotherapy. J Neurooncol,
2009;91(3):315–321.
Ferreri AJ, Reni M, Pasini F, et al. A multicenter study of
treatment of primary CNS lymphoma. Neurology,
2002;58(10):1513–1520.
Ferreri AJ, Abrey LE, Blay JY, et al. Summary statement on
primary central nervous system lymphomas from the
Eighth International Conference on Malignant
Lymphoma, Lugano, Switzerland, June 12 to 15, 2002. J
224 Clin Oncol, 2003;21(12):2407–2414.
Ferreri AJ, Blay JY, Reni M, et al. Prognostic scoring system
for primary CNS lymphomas: the International
Extranodal Lymphoma Study Group experience. J Clin
Oncol, 2003;21(2):266–272.
Ferreri AJ, Crocchiolo R, Assanelli A, Govi S, Reni M. High-
dose chemotherapy supported by autologous stem cell
transplantation in patients with primary central nervous
system lymphoma: facts and opinions. Leuk Lymphoma,
2008;49(11):2042–2047.
Ferreri AJ, Assanelli A, Crocchiolo R, Ciceri F. Central
nervous system dissemination in immunocompetent
patients with aggressive lymphomas: incidence, risk
factors and therapeutic options. Hematol Oncol,
2009;27 (2):61–70.
Ferreri AJ, Reni M, Foppoli M, et al. High-dose cytarabine
plus high-dose methotrexate versus high-dose
methotrexate alone in patients with primary CNS
lymphoma: a randomised phase 2 trial. Lancet, 2009;
374 (9700):1512–1520.
Ferreri AJ, Verona C, Politi LS, et al. Consolidation
radiotherapy in primary central nervous system
lymphomas: impact on outcome of different fields and
doses in patients in complete remission after upfront
chemotherapy. Int J Radiat Oncol Biol Phys, 2010; (in
press).
Fischer L, Thiel E, Klasen HA, et al. Response of relapsed or
refractory primary central nervous system lymphoma
(PCNSL) to topotecan. Neurology,
2004;62(10):1885–1887.
Gerstner ER, Abrey LE, Schiff D, et al. CNS Hodgkin
lymphoma. Blood, 2008;112(5):1658–1661.
Giordano A, Perrone T, Guarini A, et al. Primary
intracranial dural B cell small lymphocytic lymphoma.
Leuk Lymphoma, 2007;48(7):1437–1443.
Iwamoto FM, DeAngelis LM, Abrey LE. Primary dural
lymphomas: a clinicopathologic study of treatment and
outcome in eight patients. Neurology,
2006;66:1763–1765.
Joerger M, Huitema AD, Krähenbühl S, et al. Methotrexate
area under the curve is an important outcome predictor
in patients with primary CNS lymphoma: a
pharmacokinetic-pharmacodynamic analysis from the
IELSG no. 20 trial. Br J Cancer, 2010;102(4):673–677.
Kasamon YL, Ambinder RF. AIDS-related primary central
nervous system lymphoma. Hematol Oncol Clin North
Am, 2005;19(4):665–vii.
Kluin P, Deckert M, Ferry JA. Primary diffuse large B-cell
lymphoma of the CNS. In: Swerdlow SH, Campo E,
Harris NL et al. (eds). WHO Classification of Tumours of
Haematopoietic and Lymphoid Tissues. 4th ed. Lyon:
IARC; 2008:240–241.
Mead GM, Bleehen NM, Gregor A, et al. A medical research
council randomized trial in patients with primary

cerebral non-Hodgkin lymphoma: cerebral radiotherapy
with and without cyclophosphamide, doxorubicin,
vincristine, and prednisone chemotherapy. Cancer,
2000;89(6):1359–1370.
Ponzoni M, Terreni MR, Ciceri F, et al. Primary brain CD30
+ ALK1+ anaplastic large cell lymphoma (‘ALKoma’): the
first case with a combination of ‘not common’ variants.
Ann Oncol, 2002;13(11):1827–1832.
Ponzoni M, Ferreri AJ, Campo E, et al. Definition, diagnosis,
and management of intravascular large B-cell lymphoma:
proposals and perspectives from an international
consensus meeting. J Clin Oncol, 2007;25(21):3168–3173.
Reni M, Ferreri AJ, Villa E. Second-line treatment for
primary central nervous system lymphoma. Br J Cancer,
1999;79(3–4):530–534.
Reni M, Ferreri AJ, Guha-Thakurta N, et al.
Clinical relevance of consolidation radiotherapy and
other main therapeutic issues in primary central nervous
system lymphomas treated with upfront high-dose
methotrexate. Int J Radiat Oncol Biol Phys,
2001;51(2):419–425.
Chapter 13: Central nervous system lymphomas
Reni M, Mazza E, Foppoli M, Ferreri AJ. Primary central
nervous system lymphomas: salvage treatment after
failure to high-dose methotrexate. Cancer Lett,
2007;258(2):165–170.
Reni M, Zaja F, Mason W, et al. Temozolomide as salvage
treatment in primary brain lymphomas. Br J Cancer,
2007;96(6):864–867.
Shenkier TN, Blay JY, O’Neill BP, et al. Primary CNS
lymphoma of T-cell origin: a descriptive analysis from the
international primary CNS lymphoma collaborative
group. J Clin Oncol, 2005;23(10):2233–2239.
Smith JR, Rosenbaum JT, Wilson DJ, et al. Role of
intravitreal methotrexate in the management of
primary central nervous system lymphoma with
ocular involvement. Ophthalmology,
2002;109(9):1709–1716.
Thiel E, Korfel A, Martus P, et al. High-dose methotrexate
with or without whole brain radiotherapy for primary
CNS lymphoma (G-PCNSL-SG-1): a phase 3,
randomised, non-inferiority trial. Lancet Oncol,
2010;11(11):1036–1047.
225
 Chapter
14
T-cell non-Hodgkin lymphoma
Sheetal M. Kircher, Beverly P. Nelson, Steven T. Rosen,
and Andrew M. Evens
Introduction
T-cell non-Hodgkin lymphomas (NHLs) are uncom-
mon malignancies, representing approximately 10–12%
of all lymphomas in the United States. Varied geographic
frequencies of T-cell NHL have been documented, rang-
ing from 18% of all NHLs diagnosed in Hong Kong to
1.5% in Vancouver, British Columbia. This may in part
reflect increased exposure to pathogenic factors such as
human T-cell leukemia virus I (HTLVI) and Epstein–
Barr virus (EBV) in Asian nations. Immunophenotypic,
cytogenetic, and molecular analyses have significantly
enhanced the diagnostic capabilities as well as improved
classification and prognostication for T-cell NHL. It is
advocated that all cases of T-cell NHL be reviewed by an
expert hematopathologist.
The current World Health Organization (WHO)/
European Organization for Research and Treatment of
Cancer (EORTC) classification recognizes seven dis-
tinct clinicopathologic non-cutaneous peripheral
T-cell NHLs (Table 14.1). They include peripheral T-
cell lymphoma (PTCL), not otherwise specified (NOS);
angioimmunoblastic T-cell lymphoma (AITL); adult T-
cell leukemia/lymphoma (ATLL); NK-/T-cell lym-
phoma (NK/TCL), nasal type; enteropathy-associated
T-cell lymphoma; hepatosplenic T-cell lymphoma; and
anaplastic large-cell lymphoma (ALCL), anaplastic lym-
phoma kinase-positive (ALK+) subtype; additionally,
ALCL ALK– subtype is a provisional subtype in the
most recent WHO classification. Subcutaneous
panniculitis-like T-cell lymphoma and cutaneous γδ
T-cell lymphoma are also discussed herein since these
diagnoses portend similar prognoses to many of the
above non-cutaneous T-cell NHLs and treatment para-
digms are similar. Cutaneous T-cell lymphoma (CTCL)
and the other primary cutaneous lymphomas are
reviewed elsewhere.
226 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
The most common subtypes are PTCL-NOS
(26%), AITL (19%), NK/TCL (10%), and ALCL
(12%: ALK+ 6.5%, ALK– 5.5%). The broad spectrum
of pathologic subtypes with varied clinical behavior
poses a challenge to the systematic study of these
diseases. Further, these distinct T-cell NHL subtypes
have unique characteristics and often warrant individ-
ualized diagnostic and therapeutic treatment strat-
egies. We review herein the etiology, prognosis,
pathology, diagnosis, clinical characteristics, and avail-
able treatment strategies.
Etiology
Genetic alterations involved in lymphoma oncogen-
esis include chromosome rearrangements, disruption
of tumor suppressor genes, and increase in number of
copies of genes (gene amplification). Moreover, infec-
tion of cells by viruses and bacteria such as HTLVI,
EBV, human herpes virus 8 (HHV8), hepatitis C, and
Helicobacter pylori may also contribute to lymphogen-
esis. The data related to the risk of lymphoma with the
use of tumor necrosis factor alpha (TNF-α) inhibitors
is conflicting, but there is evidence that the use of these
drugs for inflammatory conditions may increase the
risk of developing lymphoma.
Chromosome rearrangements contribute to
altered gene function through varied mechanisms,
such as proto-oncogene activation and deregulation
of gene expression. The primary mechanism of proto-
oncogene activation in lymphoma is reciprocal and
balanced chromosomal translocations. These trans-
locations are mostly recurrent and non-random in
NHL. The majority of chromosome translocations
in NHL involve the juxtaposition of a proto-oncogene
from one chromosome next to regulatory sequences
of a partner chromosome. This contributes to control

Table 14.1 WHO classification of T-cell and NK-cell neoplasms
(2008).
Mature T-cell and NK-cell neoplasms
Peripheral T-cell lymphoma, NOS
Angioimmunoblastic T-cell lymphoma
Anaplastic large-cell lymphoma, ALK+
Anaplastic large-cell lymphoma, ALK–*
Adult T-cell leukemia/lymphoma
Extranodal NK-/T-cell lymphoma, nasal type
Enteropathy-associated T-cell lymphoma
Subcutaneous panniculitis-like T-cell lymphoma
Primary cutaneous CD8+ gamma-delta T-cell lymphoma*
Systemic EBV+ T-cell lymphoproliferative diseases of childhood
Hydroa vacciniforme-like lymphoma
Aggressive NK-cell leukemia
T-cell prolymphocytic leukemia
T-cell large granular lymphocytic leukemia
Chronic lymphoproliferative disorder of NK-cells*
Mycosis fungoides
Sézary syndrome
Primary cutaneous CD30+ T-cell lymphoproliferative disorders*
Primary cutaneous CD4+ small/medium T-cell lymphoma
* Provisional entities for which the WHO Working Group felt
there was insufficient evidence to recognize as distinct diseases
at this time.
NOS, not otherwise specified; ALK, anaplastic lymphoma kinase;
NK, natural killer; EBV, Epstein–Barr virus.
of the proto-oncogene by a promoter associated with
an immunoglobulin (Ig) or T-cell receptor (TCR)
gene. The two subtypes of TCRs that T-cell lympho-
cytes express are gamma-delta (γδ) or alpha-beta
(αβ). Approximately 95% of normal T-lymphocytes
express the αβ heterodimer, while the minority of
T-lymphocytes express the γδ heterodimer. Alpha-
beta T-cells develop predominantly in the thymus,
while γδ T-cells may develop in extrathymic locations
such as the skin, intestinal epithelium, and
spleen. The four TCR genes are arranged in germline
configuration in non-continuous segments of varia-
ble (V), diversity (D), joining (J), and constant (C)
regions. The precise mechanism by which transloca-
tion of TCR and Ig genes occur is not known, but it
appears to in part involve dysfunctional gene remod-
eling, including V–D–J recombination, isotype
Chapter 14: T-cell non-Hodgkin lymphoma
switching, and somatic hypermutation. T-cell neo-
plasms may have rearrangements involving the site
of TCR α and δ genes on chromosome 14, or, more
rarely, chromosome 7 (7q34–36 and 7p15), the site of
TCR β and γ genes. Many of the genes located at the
breakpoints of recurring chromosome translocations
have been identified. The majority of translocated
genes encode transcription factors. Transcription fac-
tors are involved in the initiation of gene transcrip-
tion and cell differentiation. The most common result
of chromosome translocations in lymphomas that
involve TCR and/or Ig genes is deregulation of gene
expression with irregular expression or overexpres-
sion in cells that normally do not express this gene.
Few lymphomas have been recognized to contain
translocations that produce a fusion protein, such as
the t(2;5)(p23;q35) translocation in ALCL that results
in expression of the nucleophosmin (NPM)–ALK
protein.
Inactivation of tumor suppressor genes may also
play a role in lymphogenesis. The most common
mechanism of tumor suppressor inactivation occurs
through the Knudson two-hit model, where a reduc-
tion of homozygosity leads to tumor formation (for
example, following germline deletion of one allele and
somatic mutation of the other). Tumor suppressor
genes identified with NHL include p53, p15, and p16.
Moreover, specific chromosomal deletions that have
been detected in NHL (including some T-cell lympho-
mas), such as 3p, 6q, 13q, and 17p, may represent sites
of yet to be identified tumor suppressor loci. Other
mediators that may be involved in lymphogenesis
include the cyclin-dependent kinase (cdk) inhibitors,
such as p21(Waf1). A function of the p21(Waf1) protein
includes the arrest of cells in G1-phase checkpoint by
associating with cyclin–CDK complexes, but the exact
factors critical for apoptosis have not been clearly
defined. The gene for the p21(Waf1) protein has been
identified as a downstream target of p53 in regulating
cell cycle progression through G1. Induction of
p21(Waf1) has also been demonstrated to occur through
a p53-independent pathway.
Gene amplification leads to an increase in the
number of copies of a gene in the genome of a cell,
which may contribute to lymphogenesis. Gene
amplification has been identified mostly in B-cell
lymphomas (e.g. REL gene), although amplification
of TCR genes in varied T-cell lymphomas has been
described. Random genomic instability, as seen in
many epithelial cancers, is not a characteristic of 227
 Chapter 14: T-cell non-Hodgkin lymphoma
the more stable lymphoma genome. Defects in DNA
mismatch repair that manifest as genomic microsa-
tellite instability are also less frequently recognized in
lymphoma, as compared to various hereditary solid
tumor syndromes and rare sporadic cancers.
Recently, gene expression profiling has helped to
identify emerging oncogenic pathways and unique
molecular subgroups within PTCL. Iqbal et al. pub-
lished results from 144 cases of PTCL and NK/TCLs,
including AITL (n = 36), ALK + ALCL (n = 20), ALK–
ALCL (n = 6), ATLL (n = 12), NK/TCL (n = 14), PTCL-
NOS (n = 44), and others (n = 10), who had gene
expression profiling. They were able to identify a unique
molecular subgroup of PTCL-NOS. This subgroup
showed marked upregulation in normal activated CD8
+ T-cells as well as a distinct set of cytokines associated
with CD8+ T-cells and NK-cells. AITL cases had more
frequent activation of the NF-κβ pathway as compared
with PTCL-NOS (P < 0.001). Immunosuppressive path-
ways were enriched with increased IL-6 signaling as well
as other gene signatures linked with angiogenesis. In
ALK+ ALCL, the expression of IL-17-associated mole-
cules was present. ATLL had a homogeneous molecular
signature with high expression of HTLV type I genes.
In this study, they also constructed a molecular
prognosticator of AITL in which they identified genes
that were expressed differently between AITL cases
with less than a 3-year overall survival (OS) (poor
prognosticator) and those with a greater than 3-year
OS (good prognosticator). Genes upregulated in the
poor prognosis group included two gene signatures
associated with immunosuppressive functions: tol-
erogenic DCs43 and CD31 stromal cells. Cases with
a poor prognosis also showed high expression of
receptors or cell adhesion molecules that mediate
proliferative signals, including PDGFR. In contrast,
transcripts that have inhibitory effects on myeloid cell
functions (CD200, MIF, SERPINB1), associated with
B-cells (SpiB, BTLA4, SYK), or encoded members of
the ribosomal protein synthesis pathway were highly
expressed in the good prognostic group. Huang et al.
published results whereby biopsies and cell lines of
NK/TCL, nasal type, underwent combined gene
expression profiling and array-based comparative
genomic hybridization (CGH) analysis. NK/TCL
samples were compared to both normal NK-cells
and PTCL-NOS. As compared to PTCL-NOS, NK/
TCL had greater transcripts for NK-cell and cytotoxic
228 molecules, especially granzyme H. As compared with
normal NK-cells, tumors more closely resembled
activated cells as opposed to resting cells. Many
features of NK/TCL indicated perturbation of angio-
genic pathways, including overexpression of platelet-
derived growth factor receptor α. In addition, they
identified that there is evidence of dysregulation of
the tumor suppressor HACE1 in the frequently
deleted 6q21 region. These molecular signatures of
T-cell lymphomas will need to be validated in larger
populations.
Prognosis
Even in the pre-rituximab era, peripheral T-cell lym-
phomas had inferior OS rates compared with their
aggressive B-cell counterparts (e.g. diffuse large B-cell
lymphoma [DLBCL]). As noted before, T-cell lympho-
mas are relatively uncommon and many clinicopatho-
logic subtypes exist. A number of studies have shown
that the T-cell immunophenotype alone, with the
exception of ALK+ ALCL, is an independent adverse
prognostic indicator. Melnyk et al. found that T-cell
lymphomas had a significantly inferior complete remis-
sion, failure-free survival (FFS), and OS when com-
pared with DLBCL. Over the years, treatment
paradigms and prognostic indicators have frequently
been extrapolated from aggressive B-cell NHLs. These
aforementioned findings suggest that there may need to
be different prognostic models and treatment regimens
for T-cell lymphomas. There have been several studies
attempting to validate previous prognostic models and
to create new systems of risk stratification (Table 14.2).
The International Prognostic Index (IPI) was ini-
tially used to predict survival of patients with aggres-
sive B-cell lymphomas but has since been used to
risk-stratify T-cell lymphomas as well. The IPI is a
prognostic indicator based on disease stage, lactate
dehydrogenase (LDH), age, extranodal sites, and per-
formance status (PS). Sonnen et al. conducted a ret-
rospective study of 125 patients diagnosed with
ALCL, PTCL-NOS, and AITL in an effort to deter-
mine whether IPI can be used to predict survival in
T-cell lymphoma patients. It confirmed that patients
with T-cell lymphomas markedly differ in their pre-
sentation of disease, their response to treatment, and
OS. In multivariate analysis, the IPI, not the histo-
logic subtype, was the main factor in determining
overall prognosis. Furthermore, other studies have
shown that higher IPI scores, elevated LDH,
B-symptoms, and elevated LDH were all individually
associated with inferior OS. Most recently, it was
shown that tumor score, elevated beta-2
 Chapter 14: T-cell non-Hodgkin lymphoma

Table 14.2 Prognostic models for T-cell lymphoma.

Patients Factors Prognosis

IPI PTCL-NOS Disease stage III–IV 5-year OS
Elevated LDH Low risk (0–1 factor): 73%
Age >60 Low–intermediate (2 factors): 51%
>1 extranodal site High–intermediate (3 factors): 43%
PS 2–4 High risk (4–5 factors): 26%
PIT PTCL-NOS Elevated LDH 5-year OS
Age >60 0 factors: 62%
PS 2–4 1 factor: 53%
Bone marrow involvement 2 factors: 33%
3 or 4 factors: 18.3%
mPIT PTCL-NOS and AITL Elevated LDH Median survival
Age >60 0–1 factor: 37 months
PS 2–4 2 factors: 23 months
Ki67 ≥80% 3–4 factors: 6 months
IPTCL PTCL-NOS and AITL Age >60 5-year OS
PS 2–4 Low risk (0 factor): 58%
Platelet count <150 Low–intermediate (1 factors): 15%
High–intermediate (2 factors): 5%
High-risk (3 factors): 0%
PTCL, peripheral T-cell lymphoma; NOS, not otherwise specified; AITL, angioimmunoblastic lymphoma; IPI,
International Prognostic Index; PIT, Prognostic Index for PTCL-u; mPIT, modified Prognostic Index for T-cell
lymphoma; IPTCL, International Peripheral T-Cell Lymphoma Project score; TCL, T-cell lymphoma; LDH, lactate
dehydrogenase; PS, performance status; OS, overall survival.

microglobulin, and bulky disease were associated
with inferior OS.
The prognostic index for PTCL (PIT) is a prog-
nostic model used to risk-stratify patients with
PTCL-NOS (excluding other clinicopathologic
T-cell lymphoma subtypes). It takes into considera-
tion four variables: age >60 years, PS (2 or higher),
elevated LDH (>1× upper limit of normal [ULN]),
and bone marrow involvement, compiling these fac-
tors into a score. A score of 1 is assigned for no
adverse factors, 2 for patients with one factor, 3 for
patients with two factors, and 4 for patients with
three or four adverse factors. A score of 1 correlated
with a 63% 5-year survival rate while a score of 3
corresponded with a 53% survival rate. Scores 3 and
4 corresponded with 32% and 18% survival rates,
respectively. Gallamini et al. found this model to be
more predictive compared with the IPI.
The marker expression in PTCL (mPIT) is a prog-
nostic model proposed by Went et al. that integrates
biologic factors into the model when assigning risk for
patients with PTCL-NOS and AITL. The model is
composed of PS > 2, age >60 years, elevated LDH,
and Ki67 ≥80%. Group 1 has zero or one of the
above factors, group 2 are those patients with two
factors, and group 3 are those patients with three or
four factors. Median survival for patients in group 1
was 37 months, 23 months for group 2, and 6 months
for group 3.
The International Peripheral T-cell Lymphoma
Project (IPTCLP) recently proposed another model
for risk stratification of PTCL and AITL which takes
into consideration age >60, PS > 2, and platelet count
<150 as adverse indicators. Patients were classified
into low risk with no risk factors, low–intermediate
if they had one risk factor, high–intermediate or
high risk if they had two or three risk factors, respec-
tively. Corresponding 5-year OS rates were 58%
for low-risk patients, 15% for low–intermediate
risk, 5% for high–intermediate, and 0% for high-risk
patients.
Recently, a comparison of these four prognostic
scores was completed by Garcia et al. Concordance
among IPI, PIT, and IPTCLP was 52% for the low-risk
group, 27% for the low–intermediate group, 20% for
the high–intermediate group, and 14% for the high-
risk group. They found that all the scores demonstra-
ted usefulness in assessing the outcomes of patients
with PTCL, with IPTCLP being most significant for
prediction of OS in multivariate analysis. 229
 Chapter 14: T-cell non-Hodgkin lymphoma
Specific disease types
Specific disease types are summarized in Table 14.3.
Adult T-cell leukemia/lymphoma (ATL)
The retrovirus HTLVI is critical to the development of
ATL. In endemic areas in Japan, approximately 6–37%
of the population is infected with HTLVI. The USA and
Europe are considered low-risk areas, with less than 1%
of the population who are seropositive. Approximately
2–4% of HTLVI carriers develop ATL. HTLVI is trans-
mitted through sexual intercourse, blood products con-
taining white blood cells, shared needles, breast milk,
and vertically. Transfusion of HTLVI-contaminated
blood products results in seroconversion in approxi-
mately 30–50% of patients at a median of 51 days. The
median age at presentation is 55 years. Patients present
with lymphadenopathy (72%), skin lesions (53%), hep-
atomegaly (47%), splenomegaly (25%), and hypercalce-
mia (28%). Cellular immune suppression is common; a
significant minority of patients may have concurrent
Strongyloides infection.
Pathology
The neoplastic T-cells are characteristically medium to
large, with pleomorphic nuclei that show convolutions
or lobulations that may be pronounced (flower cells)
(Figure 14.1). In the most common acute variant, these
abnormal lymphocytes are seen in the peripheral blood
smear. In chronic and smouldering variants the cells
may be predominantly small, with round nuclei. The
cytoplasm is basophilic. Giant cells may be seen. The
skin shows nodular, diffuse, or perivascular dermal
infiltration, frequently with extension of clusters of
atypical cells into the epidermis. Bone marrow infiltra-
tion is often patchy. Increased osteoclastic activity may
be seen in bone marrow biopsies even in the absence of
marrow infiltration. In some cases large cells resem-
bling the Reed–Sternberg cells of Hodgkin lymphoma
may be seen in involved lymph nodes. In most cases
these are CD30-positive B-cells that may also express
CD15 and they show evidence of EBV infection. These
are likely to be reactive and a reflection of the immu-
nosuppression associated with the lymphoma.
Immunophenotypically the cells are mature T-cells
with no expression of terminal deoxynucleotidyl trans-
ferase (TdT) or CD1a. There is usually expression of
CD2, CD3, and CD5, but lack of CD7 is common. Most
230 cases are CD4+/CD8– but CD4+/CD8+ and CD4–/
CD8+ cases are rarely seen. The cells are
characteristically positive for CD25. Some cases may
show expression of CD30, particularly in the large
cells. Stains for TIA-1 and granzyme are usually
negative.
Molecular pathology and cytogenetics
The prerequisite for the diagnosis is the demonstration
of integrated HTLVI genomes in virtually all cases.
TCR genes are clonally rearranged. Some recurring
structural and numerical chromosome aberrations
have been reported, among them trisomy for chromo-
somes 3p, 7q, and 14q, and losses in 6q. These changes
are predominantly encountered in aggressive variants.
Subtypes
ATL is classified into four subtypes based on clinicopa-
thologic features and prognosis: acute, lymphoma,
chronic and smoldering. In 1991, Shimoyama and col-
leagues reported on the characteristics of 818 ATL
patients. Patients with the acute type present with hyper-
calcemia, leukemic manifestations, and tumor lesions,
and had the worst prognosis, with a median survival of
approximately 6 months. Lymphoma-type patients
present with low circulating abnormal lymphocytes
(<1%) and nodal, liver, splenic, central nervous system,
bone, and gastrointestinal disease, and had a median
survival of 10 months. Chronic-type patients present
with >5% abnormal circulating lymphocytes and had a
median survival of 24 months, while the median survival
of smoldering type had not yet been reached.
Treatment
ATL, acute/leukemic, and lymphoma subtypes are dif-
ficult malignancies to treat. Patients may initially
respond to combination chemotherapy, but OS is
poor, with a median of 8 months. Response rates of
70–90% to combination IFN-α and zidovudine therapy
have been demonstrated in both the leukemic and lym-
phoma subtypes of ATL, with associated median sur-
vival rates of 11–18 months. A clinical trial that
investigated initial cytoreductive therapy with cyclo-
phosphamide, doxorubicin, vincristine, and prednisone
(CHOP) followed by antinucleoside (zidovudine or zal-
citabine), IFN-α, and oral etoposide therapy demon-
strated encouraging results. A Japanese phase III trial
of 118 patients compared biweekly CHOP versus dose-
intensified multiagent chemotherapy: VCAP (vincris-
tine, cyclophosphamide, doxorubicin, prednisolone),
AMP (doxorubicin, ranimustine, prednisolone), and
VECP (vindesine, etoposide, carboplatin, prednisolone)
Table 14.3 Immunophenotype, EBV status, and genetic features of peripheral T-cell NHL.

Neoplasm CD3 CD5 CD7 CD4 CD8 CD25 CD30 CD52 CD103 TCR EBV Genetic T-receptor
s; c abnormality genes
Adult T-cell leukemia/ + + − + − + −/+ + − NA − Multiple Rearranged
lymphoma
Peripheral T-cell +/− +/− +/− +/− −/+ NA −/+ −/+ − αβ>γδ − Often Rearranged
lymphoma, NOS complex
Angioimmunoblastic + + + +/− −/+ NA −* −/+ − αβ Present Trisomy 3 and Rearranged
T-cell lymphoma in 5; additional X
lymph
nodes
Anaplastic large-cell +/− +/− NA −/+ −/+ + + − − αβ − Table 14.5 Rearranged
lymphoma, primary
systemic type (ALK+
and ALK–)
Subcutaneous + + + − + NA − NA NA αβ – − Rearranged
panniculitis-like T-cell
lymphoma
Cutaneous γδ T-cell + − +/− − − NA −/+ NA NA γδ − − Rearranged
lymphomas
Hepatosplenic γδ T- + − + − − NA − NA NA γδ>αβ − i(7q) and tris- Rearranged
cell lymphoma omy 8
Extranodal NK-/T-cell −;+ − −/+ − − NA − NA NA − + del 6 and 13 −
lymphoma, nasal type
Enteropathy-type T- + + + − +/− NA +/− NA +/− αβ>γδ − *LOH 9p21 Rearranged
cell lymphoma
* T-cells are CD30-negative; however, B-cells are typically CD30+.
NOS, not otherwise specified; ALK, anaplastic lymphoma kinase; NK, natural killer; EBV, Epstein–Barr virus; αβ, alpha-beta; γδ, gamma-delta; LOH, loss of heterozygosity; del, deletion; I, isochrome;
NA, not available; +, ≥90%; −, ≤90%; +/−, ≥50% positive; −/+, ≤50% negative; s, surface; c, cytoplasmic.
 Chapter 14: T-cell non-Hodgkin lymphoma
Figure 14.1 (a) Adult T-cell lymphoma/leukemia cells in the blood. The cells have condensed chromatin with folded nuclei that evokes the
term “flower cells.” Both cells have an ample amount of vacuolated cytoplasm. (b) Adult T-cell lymphoma/leukemia involving lymph node. The
large lymphoma cells have several visible nucleoli and abundant cytoplasm. The lymphoma shows a high proliferative rate, as evidenced by
many mitoses and histiocytes that impart a “starry sky” appearance.
and intrathecal prophylaxis. The complete response
(CR) rate was higher in the VCAP–AMP–VECP arm
compared with biweekly CHOP arm, with 40% versus
25%, respectively (P = 0.02). The OS at 3 years was 24%
in the VCAP–AMP–VECP arm and 13% in the CHOP
arm. Subgroup analysis suggested that younger patients
(age <56 years) and patients with a poorer PS (1–3) had
more benefit with VCAP–AMP–VECP therapy,
although small numbers precluded a definitive
conclusion. Toxicity was significantly increased with
VCAP–AMP–VECP compared with biweekly CHOP,
with rates of grade 4 neutropenia, grade 4 thrombocy-
topenia, and grade 3 or 4 infection of 98% versus 83%,
74% versus 17%, and 32% versus 15%, respectively. In
addition, three treatment-related deaths were reported
in the VCAP–AMP–VECP arm.
There is also a role for combination arsenic triox-
ide (As2O3) with IFN-α therapy, which induces cell
cycle arrest and apoptosis both in HTLVI transformed
cell lines and primary ATL cells in culture. Arsenic
trioxide and IFN-α in combination with zidovudine
(AZT) was studied in a phase II trial with ten patients
with newly diagnosed chronic ATL with response seen
in all patients, including seven CRs. A retrospective
study examined CD52 expression rates in various TCL
subtypes and found that approximately 35% expressed
CD52, which could have clinical implications for use
of alemtuzumab, a humanized anti-CD52 antibody.
Future research should continue to explore novel
232 agents, including proteasome inhibitors, all-trans ret-
inoic acid (ATRA), and recombinant toxins, including
denileukin diftitox (target CD25), all of which have
been shown to have an effect on ATL cell cycle
progression.
Peripheral T-cell lymphoma, not
otherwise specified (PTCL-NOS)
PTCL-NOS is predominantly a nodal lymphoma that
represents the most common T-cell lymphoma subtype
in western countries, comprising approximately 50–60%
of T-cell lymphomas and 5–7% of all NHL. PTCL-NOS
usually affects male adults (1.5:1 male:female ratio), with
a median age of 61 years (range 17–90) in a large study
with 25% of patients presenting in stage I or IIE, 12%
stage III, and 63% stage IV. PTCL-NOS patients from
this study commonly presented with unfavorable char-
acteristics, including B-symptoms (40%), elevated LDH
(66%), bulky tumor ≥10 cm (11%), non-ambulatory PS
(29%), and extranodal disease (56%), leading to the
majority of patients (53%) falling into the unfavorable
IPI category (score of 3–5). Other smaller studies of
PTCL-NOS patients have corroborated male preponder-
ance, B-symptoms, and extranodal disease (bone mar-
row > liver > skin > lung and bone).
Pathology
This is a polymorphous group of tumors that includes
many subtypes that were previously identified as spe-
cific entities in the Kiel classification of lymphomas.
The morphology is varied, but usually consists of a
pleomorphic mix of medium and large cells that

A B
Figure 14.2 (a) Peripheral T-cell lymphoma, unspecified. The lymphoma cells are small with condensed chromatin and are associated with
many small blood vessels and eosinophils. (b) Peripheral T-cell lymphoma, unspecified, CD3 stain. Numerous lymphoma cells are highlighted.
diffusely efface the lymph node architecture. High
endothelial venules are prominent and may have an
arborizing pattern. The background population
includes a mixture of small lymphocytes, plasma
cells, histiocytes, and eosinophils (Figure 14.2).
In some cases, the infiltrate is perifollicular and the
neoplastic cells are often small with little nuclear pleo-
morphism (T-zone variant). In the lymphoepithelioid
variant (Lennert lymphoma), the neoplastic cells are
usually small with clusters of epithelioid macrophages
within the background infiltrate. Scattered EBV-
positive Reed–Sternberg-like cells may also be
seen. The recently described follicular variant is char-
acterized by intrafollicular/perifollicular clusters of
atypical clear cells. Although a t(5;9) involving ITK
and SYK genes has been described in a subset of
cases, this variant is not well defined and shows over-
lapping features with AITL.
Immunophenotypically the cells are mature T-cells
that do not express TdT or CD1a but are positive for
CD2, CD3, CD5, and CD7, although partial loss of T-cell
antigen expression is frequent. The majority of cases are
CD4-positive and negative for CD8, but other combina-
tions can be seen. In some cases the cells express CD30
and these must be distinguished from ALCL.
Molecular pathology and cytogenetics
The cytogenetic and molecular genetic constitution of
peripheral T-cell lymphoma, unspecified (PTCL-NOS),
is still poorly defined and – in keeping with its variable
morphologic features – no unifying cytogenetic aberra-
tion exists. Classical banding studies are rare, but
have revealed recurring chromosomal aberrations.
Chapter 14: T-cell non-Hodgkin lymphoma
Recurring deletions have been described in the short
arm of chromosome 1, the long arm of chromosome 6,
and the short arm of chromosome 17. Gains are com-
mon in chromosomes 3, 7, and X. PTCL-NOS of the
large-cell type are frequently characterized by a tetra-
ploid karyotype. Only one recurring translocation, t
(5;9)(q33;q22), has been identified so far, fusing ITK
and SYK genes. Interestingly, this translocation may be
indicative of a particular PTCL-NOS subtype morpho-
logically characterized by a perifollicular and intrafol-
licular growth pattern. Studies using conventional CGH
as a platform resulted in the revelation of a plethora of
complex chromosomal alterations resulting in recur-
ring imbalances and in the definition of clearly non-
random minimal overlapping target regions. The most
frequent recurrent genetic losses in PTCL-NOS were
encountered in 13q, 6q, 9p, 10q, 12q, and 5q. Gains were
observed in 7q, 17q, 16p, 8q, 9q, and others. Some of
these imbalances seem to occur simultaneously and
may define particular PTCL-NOS subgroups.
Treatment
A standard therapeutic regimen does not exist for
PTCL-NOS. Most adult treatment series have been
small, retrospective, and/or were reported in an era
where patients may now be classified as a different
histology (e.g. Hodgkin disease or DLBCL).
Traditionally, owing in part to the lack of clinical
trial data and the heterogeneity in existing reports,
chemotherapy for T-cell lymphomas has been
extrapolated from aggressive B-cell lymphoma data/
trials. In many of the prognostic studies discussed 233
above, T-cell NHL patients were typically treated in
 Chapter 14: T-cell non-Hodgkin lymphoma
the same manner as intermediate-grade B-cell patients
with anthracycline-based combination chemotherapy,
including regimens such as CHOP alone or CHOP
alternating with DHAP (dexamethasone, cisplatin,
and cytarabine), mBACOD (adriamycin, cyclophos-
phamide, vincristine, bleomycin, methotrexate, and
dexamethasone) or ACVB (adriamycin, cyclophos-
phamide, vindesine, bleomycin, and prednisone) or
VIM-3 (mitoxantrone, ifosfamide, methyl-GAG, teni-
poside, prednisone, and methotrexate) or CVP (cyclo-
phosphamide, vincristine, and prednisone), CHOP or
CHOP with etoposide or COPBLAM (cyclophospha-
mide, vincristine, prednisone, bleomycin, adriamycin,
and procarbazine).
With conventional-dose systemic anthracycline-
containing chemotherapy, the overall response rate
(ORR) is approximately 60%, but relapses are frequent
and the 5-year OS is only approximately 20–30%.
Immediately prior to the rituximab era, randomized
clinical trials in “aggressive NHL,” including aggres-
sive B-cell and T-cell lymphomas, showed that inten-
sified CHOP was associated with superior survival
compared with classic CHOP-21. Schmitz et al. ana-
lyzed 343 patients treated within the German High-
Grade Non-Hodgkins Lymphoma Study Group
(DSHNHL). The histologic subtypes included ALCL,
ALK+ (n = 78), ALCL, ALK– (n = 113), PTCL-NOS
(n = 70), ATCL (n = 28), and other subtypes (n = 54).
Patients were treated with 6–8 courses of CHOP every
14 and every 21 days or CHOEP (cyclophosphamide,
doxorubicin, vincristine, etoposide, and prednisone)
every 14 and every 21 days. Three-year event-free
survival (EFS) and OS were 76% and 90% (ALK+
ALCL), 50% and 68% (AITL), 46% and 62% (ALK–
ALCL), and 41% and 54% (PTCL-NOS), respectively.
For patients <60 years old with normal LDH levels,
etoposide improved 3-year EFS to 75% versus 51%
(P = 0.003). In patients over 60 years old, however,
there were no significant differences between any
treatment group. Thus, for this patient population,
standard CHOP21 remains the standard of care. In a
retrospective study evaluating clinical outcome in
T-cell NHLs, intensive treatments such as hyper-
CVAD/MA (fractionated cyclophosphamide, doxoru-
bicin, vincristine, dexamethasone/methotrexate and
cytarabine) and/or early transplant produced no better
results than CHOP; however, numbers were small and
the patient population was mixed.
In relapsed or refractory disease, there is no standard
234 of care. For younger, good PS patients, therapy with
salvage non-cross-resistant combination chemotherapy
is a valid option (e.g. ESHAP, ICE, DHAP, etc).
Furthermore, patients who have chemotherapy-
sensitive disease to salvage therapy may benefit from
subsequent high-dose chemotherapy and autologous
stem cell transplant (SCT), but long-term survival
rates for this strategy remain modest, as reviewed later.
There are small series that have showed the feasibility of
allogeneic SCT for relapsed/refractory PTCL-NOS.
The pyrimidine antimetabolite, gemcitabine, has
activity in relapsed/refractory T-cell NHL as a single
agent, with reported response rates of 60% in small
single institution studies. Gemcitabine has also been
studied in combination with cisplatin and methylpred-
nisone; among studies using this combination, ORR
(both CR and partial response [PR]) is approximately
70–75%. Newer and novel therapeutic agents, such as
pralatrexate, are discussed later.
Angioimmunoblastic T-cell
lymphoma (AITL)
AITL, also known as angioimmunoblastic lymph-
adenopathy with dysproteinemia, is one of the more
common T-cell lymphomas, accounting for 15–20%
of cases and 2–3% of all lymphomas. Mean age of
presentation is 57–65, with a slight male predomi-
nance, and the majority of patients present with stage
III or IV disease. AITL is commonly a systemic dis-
ease with nodal involvement with various associated
disease features such as organomegaly, B-symptoms
(50–70%), skin rash, pruritis, pleural effusions,
arthritis, eosinophilia, and varied immunologic
abnormalities (positive Coombs’ test, cold agglutins,
hemolytic anemia, antinuclear antibodies, rheuma-
toid factors, cryoglobulins, and polyclonal hyper-
gammaglobulinemia). In a retrospective study by
Mourad et al., 157 AITL patients were retrieved
from the GELA LNH87–LNH93 randomized clinical
trials. Of these, 147 patients received a CHOP-like
regimen with intensified courses in half of them.
Overall 7-year survival was 30%. In multivariate anal-
ysis, only male sex (P <0.004), mediastinal lymph-
adenopathy (P <0.041), and anemia (P <0.042)
adversely affected OS. IPI was not found to be a
predictor of survival in this study.
Pathology
The pathology of AITL is defined by a polymorphous
infiltrate involving lymph nodes that show prominent
 Chapter 14: T-cell non-Hodgkin lymphoma

A B

Figure 14.3 (a) Angioimmunoblastic T-cell lymphoma involving lymph node. The pale staining lymphoma cells expand the interfollicular area,
and replace a significant portion of the lymph node. (b) Angioimmunoblastic T-cell lymphoma involving lymph node. The lymphoma cells range
from small with condensed chromatin to large cells with distinct visible nucleoli. Histiocytes, eosinophils, and blood vessels are present in the
infiltrate. (c) Angioimmunoblastic T-cell lymphoma involving lymph node; CD20 stain. Clusters of CD20+ B-cells, including some in residual
follicles, are evident. Many of the large lymphocytes are CD20+, indicating that they are B-cells.

proliferation of high endothelial venules and follicular
dendritic cell (FDC) meshworks. In established cases
lymph node architecture is effaced, with perinodal
infiltration, but the peripheral sinus is often patent.
Clusters of clear cells previously thought to be essential
for diagnosis are present in less than 50% of cases. The
polymorphous infiltrate comprises small, medium,
and large lymphoid cells, eosinophils, plasma cells,
and histiocytes (Figure 14.3). In early cases, there is
preservation of lymph node architecture with infiltra-
tion of the perifollicular area surrounding hyperplastic
follicles. Unless clusters of clear cells are present,
this early nodal involvement by AITL is often misdiag-
nosed as reactive and only appreciated retrospectively
when the disease relapses with more typical histology.
Furthermore, in the early cases, the FDC proliferation,
a hallmark feature of the disease, is subtle or
absent. With more extensive infiltration, there is para-
cortical expansion with scattered residual follicles
that are atrophic or depleted, resembling follicles of
Castleman disease. In established cases the FDC
proliferation is prominent and tends to encircle
vessels. Although best appreciated by immunohisto-
chemistry, when pronounced the FDC proliferation
may be seen on conventional H&E stain.
The neoplastic cells of AITL are mature follicular
helper T-cells that do not express CD1a or TdT.
They are positive for pan-T-cell antigens (CD2,
CD3, CD5, and CD7) with rare antigen loss. In keep-
ing with the follicular T-helper cell phenotype they
are CD4-positive and negative for CD8, usually
express PD1 and inducible costimulator (ICOS),
and, in the majority of cases, also express CXCL13 235
and CD10.
 Chapter 14: T-cell non-Hodgkin lymphoma
The background population contains a mixed pop-
ulation that includes many small CD8-positive T-cells
and plasma cells. The expanded proliferation of FDC
can be highlighted, with antibodies against antigens
expressed on these cells (CD21, CD23, and CD35).
Residual mantle zone B-cells may be present in rela-
tion to expanded FDC meshworks. The immunosup-
pression that is associated with the disease results in
increased numbers of large EBV-positive B-cells in the
vast majority of cases, and in approximately 25% of
cases this may lead to clonal expansion, at times result-
ing in an EBV-positive DLBCL that may obscure the
underlying T-cell neoplasm.
Molecular pathology and cytogenetics
Angioimmunoblastic T-cell lymphoma (AITL) rep-
resents an entity in which recurring chromosome
abnormalities have been defined. TCR genes are
rearranged in the majority of cases, but clonal rear-
rangements of IgH genes may also be detected in
20–30% of tumors. By in situ hybridization, EBV
early repeat (EBER) transcripts have been noted in
varying numbers of infected cells (60–95% of cases).
Recurring chromosomal aberrations in AITL are
trisomies 3 and 5 and, less frequently, an additional
X chromosome. Some of these aberrations may be of
prognostic importance.
Treatment
AITL typically follows an aggressive clinical course,
although spontaneous regression occurs on rare occa-
sions. Treatment with anthracycline-based combina-
tion chemotherapy results in complete remission rates
of 50–70%; however, only 10–30% of patients are long-
term survivors. In one prospective, non-randomized
multicenter study, newly diagnosed “stable” AITL
patients were treated with single agent prednisone
whereas patients presenting with “life-threatening”
disease or relapsed/refractory disease received combi-
nation chemotherapy. The CR rate was 29% with
single agent prednisone while CR rates for relapsed/
refractory patients or patients treated initially with
combination chemotherapy were 56% and 64%,
respectively. With a median follow-up of 28 months,
the OS and disease-free survival (DFS) was 40.5% and
32.3%, respectively, although median OS was 15
months. There are anecdotal reports of relapsed
AITL patients who have responded to immunosup-
pressive therapy, such as low-dose methotrexate/pred-
236 nisone and cyclosporine, purine analogs, and
denileukin diftitox. A French study evaluated 25 eld-
erly AITL patients treated with CHOP plus rituximab.
The ORR was 80%, with 44% achieving a CR and 36%
a PR. With a median follow-up of 24 months, the
2-year progression-free survival (PFS) was 42%, and
the 2-year OS was 62%. This study demonstrated that
the addition of rituximab to CHOP did not seem to
improve outcomes over CHOP alone.
Anaplastic large-cell lymphoma (ALCL),
systemic type, ALK+ and ALK–
ALCL, primary systemic type, accounts for approx-
imately 1–3% of all NHL. This disease mainly
involves lymph nodes, although extranodal sites
may be involved. Additionally, it is critical to separate
primary cutaneous ALCL, which is associated with
excellent long-term survival rates (5-year DFS >90–
95%), from systemic ALCL (Table 14.4). Systemic
ALCL may be divided in part based on the expression
of the tyrosine kinase anaplastic lymphoma kinase
(ALK). When heterogeneous patient populations
are analyzed, the prevalence of ALK-positivity in
primary systemic ALCL cases is 50–60%. ALK-
positive ALCL is typically diagnosed in men prior to
age 35 (male:female ratio 3:0) with frequent systemic
symptoms, extranodal, and advanced-stage disease.
ALK-negative patients are usually older (median
age, 61 years) with a male:female ratio of 0.9, and
are less likely to present with extranodal disease. In a
more recent series, Savage et al. confirmed the
younger median age for ALK-positive versus ALK-
negative ALCL (34 years versus 58 years, respectively;
P < 0.001), while the male:female ratio was similar
between ALK groups.
The determination of ALK-positivity is important
as it connotes a significant favorable prognosis, with
reported 5-year OS rates of 70% versus 49% for ALK-
negative ALCL cases. Moreover, the prognosis for
ALK-positive and ALK-negative ALCL groups may
be further divided based on CD56 positivity (neural
cell-adhesion molecule), which portends a signifi-
cantly inferior outcome when it is expressed in either
ALCL subgroup (Table 14.5).
Pathology – ALK-positive
Although the mix of cells may be highly variable, all
cases contain “hallmark” cells that have the character-
istic appearance of abundant cytoplasm with large
eccentric, horse-shoe or kidney-shaped nuclei, with a
 Chapter 14: T-cell non-Hodgkin lymphoma

Table 14.4 Comparison of clinicopathologic features of primary cutaneous ALCL and systemic ALCL.

Criteria Cutaneous ACLC Systemic ALCL Systemic ALCL

Pathology Cohesive sheets of CD30+
large cells with an
anaplastic, pleomorphic,
or immunoblastic
morphology
Express CLA, but typically
lack EMA
Few inflammatory cells
(neutrophils,
eosinophils, histiocytes)
Reed–Sternberg-like cells;
neutrophil-rich,
sarcomatoid,
keratoacanthoma-like
variants
ALK expression rare
Clinical Older patients, median age
features 61 years
Localized or generalized
cutaneous
manifestations
Extracutaneous
involvement in 5–10%
Spontaneous resolution
5-year survival >95%
CLA, cutaneous lymphocyte antigen; EMA, epithelial membrane antigen; yr, years; M, male; F, female; ALK, anaplastic lymphoma kinase; OS,
overall survival; IPI, International Prognostic Index.
ALK-positive
Cohesive sheets of CD30+ large cells
(hallmark cells)
Broad morphologic spectrum,
including classic, small-cell,
lymphohistiocytic, sarcomatoid,
Hodgkin-like, neutrophil-rich,
eosinophil-rich, and signet ring
variants
EMA more often expressed (compared
with ALK-negative)
Reed–Sternberg cells absent
ALK expressed
Children, young adults (median age third
decade)
M>F
Lymph nodes and extranodal sites (skin,
lung, liver, bone, and bone marrow)
5-year OS 70%
ALK-negative
Cohesive sheets of CD30+ large cells
(hallmark cells)
Broad morphologic spectrum,
including classic,
lymphohistiocytic, sarcomatoid,
Hodgkin-like, neutrophil-rich,
eosinophil-rich, and signet ring
variants
More often CD2+ and CD3+
(compared with ALK+ cases)
Reed–Sternberg cells absent
ALK negative
Adults (median age fifth decade)
M=F
Lymph nodes and extranodal sites
(bone, subcutaneous tissue, bone
marrow, and spleen)
Skin manifestations in 20%
5-year OS 49%

Table 14.5 Characteristics of fusion proteins associated with ALK-positive ALCL.

Genetic aberration Frequency Fusion protein Staining pattern

connected with TK
domain of ALK 2p23
t(2;5) 75% NPM Cytoplasmic and nuclear
t(1;2) 10–20% TPM3 Cytoplasmic
t(2;3) 2.5% TFG Cytoplasmic
t(2;22) 2.5% CLTC Granular cytoplasmic
inv(2) 2.5% ATIC Cytoplasmic
t(2;17) Rare AL017 Cytoplasmic
t(2;19) Rare TPM4 Cytoplasmic
t(2;22) Rare MYH9 Cytoplasmic
t(2;X) Rare MSN Membrane
TK, Tyrosine kinase; ALK, anaplastic lymphoma kinase; ALCL, anaplastic large-cell lymphoma; NPM,
nucleophosmin gene; TPM3, non-muscle tropomyosin; TFG, tropomyosin receptor kinase-fused gene;
CLTC, clathrin heavy chain; ATIC, 5-aminoimidazole-4-carboxamine-1-beta-D-ribonucleotide
transformylase/inosine monophosphate cyclohydrolase; inv, inversion; AL017, ALK lymphoma
oligomerization partner on chromosome 17; TPM4, tropomyosin 4; MYH, myosin heavy chain; MSN, moesin.

237
 Chapter 14: T-cell non-Hodgkin lymphoma

A B

C D
Figure 14.4 (a) Anaplastic large-cell lymphoma involving lymph node; H&E stain. Clusters of lymphoma cells are visible within the subcapsular
sinuses. Less visible lymphoma cells extend into the parenchyma and into capsule which is thickened. (b) ALCL involving lymph node; ALK-1
stain. The ALK+ lymphoma cells within the sinuses and throughout the parenchyma are highlighted. The staining pattern is both nuclear and
cytoplasmic associated with t(2;5). (c) ALCL involving lymph node; H&E stain. The lymphoma cells are large with an abundant amount of
cytoplasm. (d) ALCL involving lymph node; CD30 stain. The ALCL cells show characteristic bright staining for CD30 with a membranous and Golgi
pattern.

perinuclear area of eosinophilia in the cytoplasm.
Some cells show striking cytoplasmic nuclear pseu-
doinclusions (doughnut cells). The nuclei usually
have clumped or dispersed chromatin with basophilic
nucleoli. While infiltration of the lymph node may be
diffuse in cases with partial infiltration of the node, the
cells are frequently seen packing sinusoids mimicking
metastatic tumor (Figure 14.4).
Several morphological variants have been
described. A proportion of cases show mixed patterns
and, in cases that relapse, the second lesion may be a
morphological variant that defers from the first.
with sheets of more monotonous large cells, either as
the sole finding or admixed with polymorphic areas,
has also been described.
Lymphohistiocytic variant (10%)
In this variant the lymph node is effaced by a population
of histiocytes with pale cytoplasm that may show
erythrophagocytosis. There is usually an associated
proliferation of small lymphocytes. The neoplastic cells
may be scanty and usually have a perivascular
distribution. They may be smaller than those of the com-
mon type and are best highlighted by staining for CD30.

Common variant Small-cell variant (5–10%)

238 This is the most frequent variant (70%). There are In these cases the predominant cell is small- to
sheets of large cells as described above. A variant medium-sized and has irregular cerebriform nuclei.

Large cells with a classical morphology are always
present but may be scanty with a predominantly peri-
vascular location.
Other morphological patterns
Several other morphological patterns of infiltration
have been described, including lesions rich in giant
cells and subtypes with sarcomatoid or signet ring cells.
Immunophenotypically the neoplastic cells stain
for CD30 with a membranous and perinuclear
(Golgi) pattern. The small cells of the small-cell variant
are, however, frequently negative for CD30, but the
scattered large cells express this antigen. The lym-
phoma cells characteristically show expression of one
or more T-cell-related antigens, but loss of T-cell anti-
gens may result in some cases showing an apparent
“null” phenotype while showing T-cell lineage at the
molecular level. Staining for CD3 is negative in up to
75% and loss of CD5 and CD7 is also frequent. A
higher proportion of cases are positive for CD2 and
CD4, and CD43 is positive in about 75%. Staining for
CD8 is usually negative. There is staining for cytotoxic
granule-associated antigens (TIA-1, granzyme, and
perforin) in the majority of cases. Expression of
CD45 is variable. Staining for epithelial membrane
antigen (EMA) may be demonstrated in a proportion
of cells in most cases. Expression of CD15 is very rare
and, if present, is usually seen in only a small propor-
tion of the neoplastic cells.
Staining for ALK is characteristic of this lym-
phoma. The pattern of staining is variable and appears
to be related to the underlying cytogenetic aberration,
with the most common t(2;5) showing nuclear and
cytoplasmic staining, while cytoplasmic staining
alone is seen with t(1;2), t(2;3), and inv2. The t(2;17)
is associated with granular cytoplasmic staining.
Pathology – ALK-negative
The neoplastic infiltrate, which is typically sheet-like
or cohesive, may efface the node or show partial nodal
involvement, often with characteristic involvement of
sinuses. The morphology of the neoplastic cells is
similar to ALCL ALK+, but is often larger and more
pleomorphic.
CD30 expression, which is strong and uniform,
usually shows a membrane and Golgi pattern of stain-
ing, but cytoplasmic staining may also be seen. Similar
to ALCL ALK+, these cases show loss of T-cell anti-
gens; however, many cases express at least one T-cell
marker. They are usually CD4-positive and express
Chapter 14: T-cell non-Hodgkin lymphoma
cytotoxic granules. Unlike ALCL ALK+, EMA is pos-
itive in only a proportion of cases.
Molecular pathology and cytogenetics
ALCL is clearly divided into two subgroups defined by
genetics, with tremendous clinical and prognostic dif-
ferences. In roughly 60–80% of the cases, a reciprocal
chromosome translocation, t(2;5)(p23;q35) or variants,
has been detected, leading to the generation of a chi-
meric protein that consists of the Nucleophosmin
(NPM) gene in 5q35 and the Anaplastic Lymphoma
Kinase (ALK) gene in 2p23. The new protein has prom-
inent phosphokinase activity, and the overexpression of
the constitutively overexpressed ALK kinase can be
detected in tissue sections by immunohistochemistry.
In cases with the t(2;5), ALK is expressed both in the
nucleus and the cytoplasm, while in the rarer variant
translocations also involving ALK, but different fusion
partners, it is seen in the cytoplasm exclusively. Patients
with ALK+ ALCL are generally younger and predom-
inantly of male gender. In comparison to ALK– ALCL,
they follow a distinctly favorable clinical course (with
10-year survival rates of 70–90%). ALCL, therefore, is
an example of the importance of genetic markers in the
delineation of particular risk groups. Secondary chro-
mosomal alterations have been less thoroughly defined
in ALCL, but the meticulous comparison to other
PTCLs reveals distinct differences in their genetic
constitution.
Treatment
Therapy for adult ALCL, systemic type, has commonly
included anthracycline-based polychemotherapy regi-
mens. Traditionally, because of lack of clinical trial
data and heterogeneity of existing series as noted before,
treatment regimens have been extrapolated from data in
aggressive B-cell lymphoma. Thus, CHOP has been
commonly employed. The response to therapy depends
in part on ALK expression and prognostic factors such
as the IPI score. Patients with IPI scores of 0 or 1 have
survival rates above 90%, while patients with unfavor-
able factors have survival rates of 41%. The 5-year OS for
ALK+ patients was 70% as compared to 49% for patients
who were ALK–. It is important to recognize, however,
that patients with ALK+ ALCL may present with high-
risk features and poorer prognosis, especially with IPI
≥3, which has associated 4-year OS rates of 25–40%.
Incorporation of autologous SCT in first complete 239
remission has been advocated by some groups, although
 Chapter 14: T-cell non-Hodgkin lymphoma
this approach warrants prospective validation (see autol-
ogous SCT and allogeneic SCT sections below).
Several novel therapies have been studied as single
agents for relapsed/refractory disease or in combina-
tion with CHOP for newly diagnosed patients. Agents
that have been combined with CHOP include denileu-
kin diftitox (Ontak), alemtuzumab (Campath), beva-
cizumab (Avastin), and bortezomib (Velcade). In a
phase II multicenter trial, 24 patients with T-cell
NHL (12% ALCL) received alemtuzumab with
CHOP as first line therapy. The ORR was 75% (CR
68%) and all ALCL patients (3/3) achieved a CR that
appeared to be durable.
For relapsed/refractory disease, a targeted therapeu-
tic agent with impressive single agent activity is brentux-
imab vedotin (SGN-35). As described before, CD30 is
uniformly expressed in cutaneous and systemic ALCL.
Several “cold” anti-CD30 antibodies were studied for the
treatment of relapsed/refractory Hodgkin disease and
systemic ALCL. MDX-060 (Medarex) is a fully human
anti-CD30 IgG1k monoclonal antibody that has been
shown to inhibit growth of CD-30-positive tumor cells
in vitro and tumor xenograft models. In a phase I/II trial,
MDX-060 was well tolerated, but clinical activity was
modest with an 8% response rate (6/72), with 35%
maintaining stable disease. MDX-1401 is a second gen-
eration antibody with an absent fucose, thereby increas-
ing the antibody’s antibody-dependent cellular
cytotoxicity. In preclinical studies, MDX-1401 had
improved antibody effector function over the fucose-
containing parental anti-CD30 antibody (MDX-060).
Early phase MDX-1401 trials are ongoing.
SGN-30 (Seattle Genetics) is a chimeric anti-CD30
monoclonal antibody that was evaluated in a phase II
study of 38 relapsed/refractory Hodgkin disease
patients and 41 systemic ALCL. In the ALCL group,
two patients achieved CR and five additional patients
achieved a PR (ORR 17%), with response durations
ranging from 27 to 1460+ days. Modifications to this
antibody have also been made; SGN-35 is an antibody–
drug conjugate, which was formed by coupling the anti-
CD30 antibody, cAC10, to monomethyl auristatin E
(MMAE), an antitubulin agent. In preclinical mouse
xenograft models, SGN-35 also induced durable dose-
dependent tumor regression compared to either
untreated mice or another control group receiving
IgG-vcMMAE. In two phase I studies, SGN-35
appeared to have significant clinical activity in
relapsed/refractory systemic ALCL and relapsed/refrac-
240 tory Hodgkin disease. Seven systemic ALCL patients
were treated on these relapsed/refractory CD30+ two
SGN-35 phase I trials (one q 3 week and one trial weekly
dosing); 86% of patients (6/7) had documented CR, all
of which were durable. Preliminary results were
recently presented from the pivotal phase II systemic
ALCL trial. Shustov et al. recently presented interim
results of a phase II, single arm, multicenter study of the
antibody–drug conjugate (ADC) in 30 patients with
relapsed or refractory ALCL. Brentuximab vedotin
1.8 mg/kg was administered every 3 weeks as a 30-
minute outpatient IV infusion for up to 16 cycles of
treatment. The objective response rate was 87%, with
57% of patients achieving a CR (n = 17) and 30% of
patients achieving a partial remission (PR; n = 9). The
remaining patients had stable disease (n = 3) or were
not evaluable for response (n = 1). Similar proportions
of ALK-1-negative and ALK-1-positive patients
achieved CR and PR. Reduction in tumor burden was
observed in 97% of patients. In August 2001, brentux-
imab vedotin was FDAA approved for patients whose
disease has progressed after a prior chemotherapy treat-
ment. Please see the Emerging and novel therapies
section below for additional new therapies being
studied in relapsed/refractory ALCL.
Subcutaneous panniculitis-like T-cell
lymphoma (SCPTL)
SCPTL is a rare T-cell lymphoma that primarily infil-
trates the subcutaneous fat without dermal and
epidermal involvement, causing erythematous to vio-
laceous nodules and/or plaques. It may be associated
with a systemic hemophagocytic syndrome – charac-
terized by high fever, skin lesions, lung infiltrates,
jaundice, hepatosplenomegaly, liver dysfunction,
coagulation abnormalities, pancytopenia, and a
benign prominent histiocytic proliferation with hemo-
phagocytosis. The hemophagocytic syndrome (HPS)
associated with T-cell neoplasms can occur preceding,
during, or while the disease is in remission. Moreover,
a controversial entity known as cytophagic histiocytic
panniculitis (CHP) has been described as an inflam-
matory disease with a possible association to SCPTL.
CHP is a disease that has been recognized to have
diverse outcomes, ranging from indolent to aggres-
sive/fatal clinical courses. There is increasing evidence
that, within the group of SCPTL, there is a distinction
between cases with an αβ T-cell phenotype and cases
with a γδ phenotype. In the recent WHO–EORTC
classification, only the cases with an αβ phenotype

are classified as SCPTL. Cases previously classified as
SCPTL with a γδ phenotype, which comprised 25% of
all cases, are now classified as cutaneous γδ T-cell
lymphomas. SCPTL has a more indolent course and
is less likely to be associated with the HPS, as com-
pared to cutaneous γδ T-cell lymphomas.
To better define the features and prognosis the αβ
phenotype and γδ phenotype, 63 patients with the αβ
phenotype and 20 patients with the γδ phenotype were
analyzed at a workshop of the EORTC Cutaneous
Lymphoma Group. SCPTL αβ had a CD4–, CD8+,
CD56–, βF1+ phenotype and had a favorable prognosis,
with 5-year OS 82%. SCPTL αβ was associated with HPS
in only 17%. SCPTL αβ patients without HPS had a
significantly better survival than patients with HPS, with
5-year OS of 91% versus 46% (P < 0.001). SCPTL γδ had
a CD4–, CD8–, CD56+/–, βF1– phenotype and carried
a poor prognosis, with 5-year OS of 11%. This poor
outcome was not affected by the presence of HPS or the
type of treatment. The results of this study indicate that
SCPTL αβ and SCPTL γδ are distinct entities.
Pathology
The neoplastic T-cell infiltrate involves fat lobules and
shows characteristic rimming of individual fat cells.
Septa are typically spared. Unlike cutaneous γδ T-cell
lymphoma, SCPTL does not usually involve the der-
mis and epidermis. The neoplastic cells may vary in
size from case to case, but are monomorphic in a given
case and usually have irregular hyperchromatic nuclei.
Rimming of individual fat spaces is a characteristic
feature. Necrosis, apoptosis, and associated fat
necrosis are common features. The neoplastic T-cells
which are of αβ type are usually CD8-positive and
express cytotoxic granules.
Treatment
The clinical course of SCPTL is variable, ranging from
indolent disease to rapidly fatal fulminant hemophago-
cytosis. When treatment is warranted, treatment varies
from only surgery or radiotherapy to doxorubicin-
based chemotherapy or high-dose chemotherapy
followed by autologous SCT. Because of the data sup-
porting the excellent prognosis of SCPTL αβ without
associated HPS, some authors question the use of
aggressive multiagent chemotherapy. A small case series
has been reported where patients have achieved durable
responses with the combination of cortosteroids and
methotrexate. There have also been anecdotal reports
of denileukin diftitox having activity in this disease.
Chapter 14: T-cell non-Hodgkin lymphoma
Cutaneous gamma delta (γδ) T-cell
lymphoma (CGD-TCL)
Cutaneous γδ T-cell lymphomas (CGD-TCL) include
lymphomas previously classified as SCPTL with a γδ
phenotype. CGD-TCL have a more aggressive course
as compared with SCPTL. Patients present with skin
lesions similar to SCPTL; however, they may also have
epidermal and dermal involvement. Patients with sub-
cutaneous disease do worse as compared to those with
only epidermotropic or dermal disease. Patients with
CGD-TCL may also have mucosal involvement. There
is still debate as to whether the cutaneous and mucosal
γδ lymphomas are two different diseases or different
presentations of the same disease. HPS can occur with
CGD-TCL but generally patients do not have evidence
of systemic disease on presentation.
Pathology
The fourth edition of the WHO Classification of
Tumors of Hematopoietic and Lymphoid Tissues sepa-
rates primary cutaneous T-cell lymphomas that
express the γδ T-cell receptor from SCPTL. Unlike
the latter, primary cutaneous γδ T-cell lymphoma is
not confined to the subcutis and can show three major
overlapping histological patterns: epidermotropic,
dermal, and subcutaneous. There may be necrosis
and vascular invasion, and the subcutaneous pattern
shows fat rimming by neoplastic T-cells, but, unlike
SCPTL, it usually also shows dermal and/or epidermal
involvement.
Treatment
Patients are usually treated with aggressive regimens.
Toro et al. treated patients with local therapies, such as
topical steroids, psoralen and PUVA radiation, as well
as systemic therapies like IFN-α, IFN-γ, CHOP, radi-
ation therapy, and bone marrow transplant. Patients
with dermal and epidermotropic involvement had a
median survival of 29 months. Patients with subcuta-
neous disease had a less favorable median survival of
13 months. In those with aggressive disease, there may
be a potential role for early allogeneic SCT.
Hepatosplenic T-cell lymphoma (HSTCL)
HSTCL is a rare form of PTCL that represents only 1.4%
of cases of peripheral T-cell and NK-cell lymphomas
worldwide. Regionally, it represents a slightly higher
percentage of all PTCLs in North America (3.0%) and 241
Europe (2.3%) but is rare in Asia (0.2%). Median age at
 Chapter 14: T-cell non-Hodgkin lymphoma
time of diagnosis is 34 years, and 68% of cases are in
male patients. Patients present with B-symptoms,
prominent hepatosplenomegaly, anemia, neutropenia,
thrombocytopenia (commonly severe), peripheral
blood lymphocytosis, and lymphadenopathy. It is
often associated with an aggressive clinical course
(median survival 16 months).
Pathology
The enlarged spleen shows involvement of the cords
and sinuses of the red pulp with atrophy of the white
pulp. The liver and bone marrow show a predomi-
nantly intrasinusoidal pattern of infiltration. In the
marrow the intrasinusoidal infiltrate may not be
appreciated without immunohistochemistry. The
neoplastic cells are of medium size and are usually
of γδ T-cell receptor type, but may less commonly
be of αβ type. They are CD3+, but are negative for
CD5. They may express CD56 and CD8, but are
negative for CD4. The pattern of expression of cyto-
toxic granules shows a non-activated cytotoxic T-cell
phenotype as TIA-1 and granzyme M are expressed,
but perforin and granzyme B are usually negative
(Figure 14.5).
Molecular pathology and cytogenetics
Hepatosplenic γδ T-cell lymphoma is associated with a
recurring chromosomal abnormality, i(7)(q10), that is
A B
242 C D
seen in concert with trisomy 8 in a subset of cases.
Interestingly, rare TCR αβ variants of the disease have
been described, with demonstration of identical chro-
mosome aberrations.
Treatment
The clinical course of HSTCL is commonly aggressive
despite multiagent chemotherapy, and median survival
is approximately 1 year. Case reports have described
significant clinical activity with the purine analog, pen-
tostatin, in relapsed patients. Alemtuzumab, both as a
single agent as well as in combination with cladribine
and fludarabine, has also been reported anecdotally to
have responses. A phase II study by Revandi et al.
studied 24 patients with a variety of T-cell leukemias
and lymphomas (three of whom had HSTCL) with a
combination of alemtuzumab 30 mg intravenously
three times weekly for up to 3 months and pentostatin
4 mg/m2 IV weekly for 4 weeks followed by alternate
weekly administration for up to 6 months. Prophylactic
antibiotics, including antiviral, antifungal, and antibac-
terial agents, were administered during the treatment
and for 2 months after its completion. Two of the three
patients with HSTCL achieved a CR with this regimen,
and one of these patients was subsequently able to
receive an allogeneic SCT. The main toxicity was
grade 3/4 infections, most frequently cytomegalovirus
(CMV) reactivation.
Figure 14.5 Hepatosplenic T-cell
lymphoma involving the bone marrow. (A)
The tumor cells in the bone marrow
aspirate resemble blasts. They are
medium-sized with dispersed chromatin
and scant cytoplasm. (B) Two lymphoma
cells display erythrophagocytosis. (C)
Hypercellular bone marrow with almost all
the normal hematopoietic cells replaced
by lymphoma cells growing in sheets. (D)
Lymphoma cells located within the lumen
of a sinus. This image was originally
published in Rizvi MA, Evens AM, Tallman
MS, Nelson BP, Rosen ST. T-cell non-
Hodgkin’s lymphoma. Blood, 2006;107:
1255–1264. © The American Society of
Hematology.

Extranodal NK-/T-cell lymphoma,
nasal and nasal-type
Extranodal NK-/T-cell lymphoma, nasal and nasal-
type, formerly known as angiocentric lymphoma, is
rare in western countries and is more prevalent in
Asia and South and Central America. The disease
commonly presents in men at a median age of 43.
This entity is associated with EBV and is typically
characterized by extranodal presentation and local-
ized stage I/II disease, but with angiodestructive pro-
liferation and an aggressive clinical course. These
tumors have a predilection for the nasal cavity and
paranasal sinuses (“nasal”), although the “nasal-
type” designation encompasses other extranodal
sites of NK-/T-cell lymphomatous disease (skin, gas-
trointestinal, testis, kidney, upper respiratory tract,
and, rarely, orbit/eye).
As part of the International Peripheral T-cell
Lymphoma Project, 136 cases were retrospectively
examined for difference between nasal and extranasal
NK-/T-cell lymphomas. In this international trial,
it was observed that, when compared to nasal
cases, extranasal cases had worse outcomes. The
median OS was better in the nasal compared with
the extranasal cases in early- (2.96 versus 0.36 year;
P < 0.001) and late-stage disease (0.8 versus 0.28
years; P = 0.31).
Pathology
When there is surface mucosa there is usually exten-
sive ulceration. In solid organs such as the testis there
is necrosis. The lymphoma cells show a striking angio-
centric, angioinvasive, and angiodestructive pattern of
infiltration. Cellular morphology is highly variable,
with some cases showing monotonous small cells
while other tumors are composed of cells that are
medium, large, or anaplastic. The cytoplasm is usually
pale or clear and the nuclear chromatin is granular.
Nucleoli are inconspicuous.
In many cases there is an associated mixed
inflammatory cell infiltrate that may mask the
neoplastic population. In the nose/nasal cavity
there may be striking hyperplasia of the surface squa-
mous epithelium (pseudoepitheliomatous hyperpla-
sia), mimicking well-differentiated squamous cell
carcinoma.
Immunophenotypically the cells are positive for
CD2 and CD56. There is cytoplasmic staining for
CD3ε but no surface staining for CD3. There is
Chapter 14: T-cell non-Hodgkin lymphoma
expression of antigens associated with cytotoxic
granules (TIA-1, granzyme, and perforin).
Occasional cases are positive for CD7, and CD30
can be seen in a proportion of cases. There is no
staining for CD5, CD4, CD8, CD43, or CD45RO.
Staining for CD16 and CD57 is also negative. EBV
infection, a defining feature, is best demonstrated by
in situ hybridization for EBV-encoded RNA (EBER).
Cases negative for CD56 are only diagnosed if there is
evidence of EBV infection together with presence of
cytotoxic granules.
Molecular pathology and cytogenetics
Nasal NK-/T-cell lymphoma and NK-/T-cell lympho-
mas of nasal type are mostly derived from cells of NK-
lineage, and therefore no clonal rearrangements for
TCR genes are usually found. However, in some
cases they may be present, pointing to a derivation of
the tumor cells from T-lymphocytes (hence the term
NK-/T-cell lymphoma). In the overwhelming majority
of cases, the tumor cells are infected with the EBV,
implying EBER in situ hybridization as an important
diagnostic tool in this disease.
Treatment
Combined modality therapy incorporating adriamycin-
based chemotherapy (minimum six cycles for patients
with stage III or IV disease), involved-field radiation
(IF-RT; median dose 50 Gy, range 30–67 Gy), and
intrathecal prophylaxis is recommended for extranodal
NK-/T-cell lymphoma, nasal patients, although the
benefit of the addition of chemotherapy to radiation
has not been confirmed for limited-stage disease.
Response rates for NK-/T-cell lymphoma, nasal, have
been reported to be near 85% (two-thirds CR) following
radiation alone, although 50% of patients will experi-
ence local relapse and 25% systemic relapse with a
predilection to extranodal sites such as testis, orbit,
skin, gastrointestinal tract, and central nervous system.
In limited disease, there is a high systemic failure rate
after treatment with radiotherapy alone, and therefore
chemoradiation has been widely used. Kim and col-
leagues showed a 5-year DFS for patients with stage I/II
disease of 34–38% and OS of 57–65%, although systemic
disease progression was often fatal. Li and colleagues in
Taiwan showed that combined chemotherapy/radiation
or radiation alone resulted in better survival compared to
chemotherapy alone (5-year survival rates, 59%, 50%,
and 15%, respectively; P = 0.01) in patients with NK-/ 243
T-cell sinonasal locoregional disease. Li and colleagues
 Chapter 14: T-cell non-Hodgkin lymphoma
retrospectively compared stage IE patients who received
radiation alone versus combined chemotherapy with
radiotherapy. Of note, this group divided IE patients
into limited (confined to nasal cavity) and extensive
(presenting with extension beyond the nasal cavity)
stage IE disease. Limited-stage IE patients survived lon-
ger than extensive IE patients overall (5-year OS 90%
versus 57%, respectively; P < 0.001). Moreover, compar-
ing radiation alone to combined modality therapy, the 5-
year OS was not significantly different for limited-stage
IE, at 89% and 92%, respectively, as well as extensive IE
disease at 54–58%, respectively.
Because of the risk of systemic failure in localized
disease, Kim et al. reported a phase II trial of 30
patients with newly diagnosed stage IE to IIE, nasal
extranodal NK-/T-cell lymphoma who received con-
current chemoradiotherapy with radiation 40–52.8 Gy
and cisplatin 30 mg/m2 weekly. This was followed by
three cycles of VIPD (etoposide 100 mg/m2 day 1–3,
ifosfamide 1200 mg/m2 days 1–3, cisplatin 33 mg/m2
days 1–3, and dexamethasone 40 mg days 1–4). All of
the patients completed chemoradiation and had a
100% response rate, which included 22 CRs and 8
PRs. Twenty-six of 30 completed the subsequent
three cycles of VIPD, and ORR and CR rate were
83% and 80%, respectively. The estimated 3-year PFS
and OS rates were 85% and 86%, respectively. The
results of this study support that nasal extranodal
NK-/T-cell lymphomas have favorable outcomes
with frontline chemoradiation.
Patients with systemic disease have poor long-term
survival (5-year OS 20–25%) with high locoregional
(over 50%) and systemic failure rates (over 70%). The
traditional approach for stage III and IV extranodal
NK-/T-cell lymphoma, nasal is combined modality
therapy with adriamycin-based chemotherapy and radi-
ation therapy. L-asparaginase has promising outcomes
in advanced, relapsed, and refractory disease, and has
been combined with chemotherapy. A retrospective
study of an L-asparaginase-based regimen demonstra-
ted an ORR of 82% (CR 56%; 5-year OS 67%). In
patients with advanced disease, there is evidence that
salvage treatment significantly prolonged the 5-year OS
compared with best supportive care alone (38% versus
0%, respectively; P < 0.001).
Yamaguchi et al. reported a phase I study of a
new chemotherapeutic regimen, SMILE, which con-
sists of dexamethasone, methotrexate, ifosfamide,
L-asparaginase, and etoposide. The rationale for this
244 regimen is based on both in vivo and in vitro data.
Etoposide has shown in vitro and in vivo efficacy for
NK-cell neoplasms, pediatric EBV-related HPS, and
pediatric EBV-associated lymphoproliferative dis-
ease. L-Asparaginase induces the selective apoptosis
of NK lymphoma cells in vitro. Successful results
in NK-cell lymphoma have been reported for
L-asparaginase, either alone or in combination with
other chemotherapy. In a 2012 Blood report, this
group published a follow-up phase II study confirm-
ing the efficacy & SMILE in advanced-stage NK-/T-
cell lymphoma patients. Toxicities were significant,
especially hematologic and real, while the estimated
5-year OS was 50% and 4-year DFS was 64%.
Lee et al. reported on 26 patients with extranodal
NK-/T-cell lymphoma who received ifosfamide,
methotrexate, etoposide, and prednisolone (IMEP)
chemotherapy as first line treatment [ifosfamide 1.5 g/
m2 (days 1–3), methotrexate 30 mg/m2 (days 3 and 10),
etoposide 100 mg/m2 (days 1–3), and prednisolone
120 mg (days 1–5)]. Radiotherapy was administered
only to patients with Ann Arbor stage I/II that had
not achieved CR or for those who developed local fail-
ure after completing chemotherapy. Sixteen patients
(group A) had nasal or upper aerodigestive tract local-
ization (stage I/II) and 10 (group B) had extranasal or
disseminated disease. Of the 14 evaluable patients in
group A, 11 (79%) achieved CR after IMEP alone and
13 (93%) after chemotherapy plus additional radiother-
apy. Of the 11 patients who achieved CR with chemo-
therapy alone, seven developed recurrence. All of the
recurrences were local failure and were successfully
treated by additional curative radiotherapy. However,
patients in group B responded poorly (CR 13%).
Enteropathy-type intestinal T-cell
lymphoma (EITCL)
EITCL (also known as intestinal T-cell lymphoma) is a
rare T-cell lymphoma of intraepithelial lymphocytes
that commonly presents with multiple circumferential
jejunal ulcers in adults. Patients usually have a prior
brief history of gluten-sensitive enteropathy. EITCL
accounts for less than 1% of NHLs and has been recog-
nized to have a poor prognosis, with reported 5-year OS
and DFS rates of 20% and 3%, respectively. This is in
part related to many patients presenting with malnu-
trition and poor PS.
EITCL may present without antecedent celiac dis-
ease history, but most patients have abdominal pain and
weight loss. Evidence of celiac disease serologic markers

such as positive antigliadin antibodies and/or HLA
types (DQA1*0501/DQB1*0201/DRB1*0304) may be
present at diagnosis. Moreover, these genotypes may
represent celiac disease patients at higher risk for devel-
opment of EITCL. Small-bowel perforation or obstruc-
tion, gastrointestinal bleeding, and enterocolic fistulae
are recognized complications of this disease.
Pathology
The ulcerating tumor contains medium to large atyp-
ical lymphoid cells that are associated with an inflam-
matory response that often includes many histiocytes
and eosinophils. If pronounced, the inflammation
may obscure the neoplastic cells. The adjacent
mucosa shows features of enteropathy with an intra-
epithelial lymphocytosis. In the less common type II
enteropathy-type T-cell lymphoma (EATCL), the
neoplastic T-cells are monomorphic and of medium
size. The tumor is less necrotic than in classical
EATCL. The adjacent mucosa shows a prominent
intraepithelial lymphocytosis. In classic EATCL, the
neoplastic T-cells express CD3, CD7, and CD103,
show loss of CD5 and often loss of CD8. They are
CD4-negative. CD30 expression is variable. The
intraepithelial lymphocytosis adjacent to the tumor
may show loss of CD8 and the presence of an identi-
cal T-cell clone to that of the overt lymphoma. In
EATCL type II, the monomorphic neoplastic cells are
CD8-positive and express CD56, which is a similar
immunoprofile to that of the adjacent intraepithelial
lymphocytes.
Molecular pathology and cytogenetics
EATCL has been shown to harbor distinct genetic
alterations as detected by CGH and fluorescent
in situ hybridization (FISH). Recurring genomic losses
were observed in 8p, 9p, and 13q, and gains of chro-
mosomal material were found in 1q, 5q, and 7q. Of
major importance, EATCL harbors a characteristic
chromosomal alteration, a gain in 9q with a minimally
overlapping region at 9q33–34 in approximately 60%
of cases. Interestingly, this chromosomal gain was also
observed in other types of extranodal aggressive PTCL,
namely in four primary gastric and one colonic PTCL,
leading to speculation about a possible association of
9q gains with gastrointestinal PTCL as a whole. In
addition, 9q gains are an infrequent finding in nodal
PTCL-NOS, suggesting profound differences in the
overall genetic profile of EATCL as compared to pri-
mary nodal T-cell lymphomas.
Chapter 14: T-cell non-Hodgkin lymphoma
Treatment
Following diagnosis of EITCL, adriamycin-based
combination chemotherapy should be considered for
each patient and aggressive nutritional support with
parenteral or enteral feeding is critical in the care of
these patients. Patients with known celiac disease
should adhere to a gluten-free diet.
Autologous stem cell transplantation
Because of the poor prognosis with conventional ther-
apy in T-cell lymphoma, there has been interest in the
role of more aggressive treatment approaches, such as
high-dose therapy with autologous stem cell trans-
plantation (autoSCT), which has been considered
standard therapy for many patients with relapsed
aggressive B-cell lymphomas, such as DLBCL.
Frontline consolidative therapy
There have been many retrospective studies with PTCL
patients that have examined upfront (consolidative)
high-dose therapy with autoSCT (Table 14.6). OS
rates in these small studies ranged from 73% at 3 years
to 60% at 5 years. Rodriguez et al. reported on 74
patients with a 5-year OS rate of 68% after 67 months’
follow-up. Feyler et al. reported on 64 patients, with a
3-year OS of 53% after 37 months’ follow-up. However,
in both of these reports ALK+ ALCL patients were
included, likely biasing the results because of their
known superior outcomes compared to other subtypes
of T-cell lymphoma. The largest retrospective study was
by the European Bone Marrow Transplantation Group,
who examined 146 patients with AITL. The OS rate was
67% at 24 months and 59% at 48 months. Nickelson
et al. attempted to use an intensified regimen of
MegaCHOEP (cyclophosphamide, adriamycin, vincris-
tine, etoposide, and prednisone) followed by autoSCT
in patients <60 years old as that strategy had been
successfully used in younger patients with aggressive
B-cell lymphoma. Compared with 84% of B-cell NHL
patients, only 22/33 (67%) patients were able to receive
all treatments, including autoSCT. Sixteen of 33
patients (49%) achieved a CR or CRu, two patients
(6%) achieved a PR, three patients (9%) had stable
disease (SD), nine patients (27%) progressed within 2
months after final restaging or during therapy, one
patient had an unknown response at the end of treat-
ment, and two therapy-related deaths were observed.
The 3-year EFS and OS were only 25.9% and 44.5%, 245
respectively. The group compared outcomes to a
Table 14.6 Studies of autologous stem cell transplantation for peripheral T-cell lymphoma.

Study Study No. Regimen Disease status at time of Response Overall Comments
transplant survival
Consolidative (first remission)
Gisselbrecht et al., 2002 Prospective 76 BEAM NA No data 32% at 5 years Included ALCL and precur-
(n = 43), (for BEAM) sor T-cell lymphoma
diverse
Corradini et al., 2006 Prospective 62 Mito/Mel CR 56%, PR 16%, PD 24% CR 66%, PR 34% at 12 years Included ALK+ ALCL; 74%
or BEAM 18% received autoSCT
D’Amore et al., 2006 Prospective 77 BEAM CR 50%, PR 35% 71% CR/PR 30 of 39 No ALK+ patients; 75%
patients in CR 1 received autoSCT
year post tx
Rodriguez et al., 2007 Retrospective 19 BEAM/ CR1 42%, CR2 15%, PR1 26%, PR2 79% CR, 5% 60% at 5 years Only AITL
BEAC/CVB/ 5%, PD 10% PR
CY + TBI
Rodriguez et al., 2007 Retrospective 74 BEAM/ CR1 100% NA 68% at 5 years Included ALCL
BEAC/CVB/
CY + TBI
Feyler et al., 2007 Retrospective 64 Diverse CR1 48%, CR2 6%, PR 23%, PD 22% NA 53% at 3 years Included ALK+ ALCL
Kyriakou et al., 2008 Retrospective 146 Diverse CR1 34%, CR2 14%, 70% CR, 7% 59% at 4 years Only AITL
chemotherapy-sensitive 38%, PR
chemotherapy-refractory 14%
Reimer et al., 2008 Prospective 83 Cy/TBI CR 39%, PR 40%, 21% PD 58% CR, 8% 48% at 3 years No ALK+ patients, 66%
PR received autoSCT
Relapsed/refractory disease
Kim et al., 2007 Retrospective 29 Diverse CR2 27%, PR 58%, refractory 14% NA 7.4 months CR prior to autoSCT prog-
median nostic: EFS (P ≤ 0.025) and
OS (P ≤ 0.027)
Rodriguez et al., 2007 Retrospective 123 Diverse CR2 36%, PR 16%, RD 9%, non- CR 73%, PR Pre-autoSCT No difference in OS
treated relapse 3% 11%, SD or status: between CR1 or CR2.
PD 16% CR: 5-year Poorest outcomes for PR2
OS 57%, or RD
PR: 5-year
OS 33%
Smith et al., 2007 Retrospective 25 Bu/Cy/ CR2/PR2 68%, RD 32% No data At 5-years: 34%, No difference in OS for CR1
VP16 RFS 18% or CR2/PR/PD
Yang et al., 2009 Retrospective 30 Diverse CR2 17%, PR2 83% No data At 5-years: CR2 CR prior to autoSCT prog-
40%, PR2 12% nostic for OS
NA, not available; autoSCT, autologous stem cell transplantation; OS, overall survival; CR, complete response; AITL, angioimmunoblastic T-cell lymphoma; BEAM, carmustine, etoposide,
cytarabine, melphalan; BEAC, carmustine, etoposide, cytarabine, cyclophosphamide; CVB, cyclophosphamide, etoposide, methotrexate; ALCL, anaplastic large-cell lymphoma; ALK, anaplastic
lymphoma kinase; Mito, mitoxantrone; Mel, melphalan; Bu, busulfan; Cy, cyclophosphamide; VP16, etoposide; TBI, total-body irradiation; PR, partial response; CR1, first complete response;
CR2, second complete response; PD, progressive disease; PR1, first partial response; PR2, second partial response; RD, refractory disease.
 Chapter 14: T-cell non-Hodgkin lymphoma

retrospective group of aggressive B-NHL that showed Relapsed disease

an improved EFS and OS of 60.1% and 63.4%, respec-
In the relapsed refractory setting, there is no prospec-
tively. Conclusions from this study indicated that
tive data; however, there are retrospective studies
MegaCHOP followed by autoSCT does not offer
to guide management. Smith et al. retrospectively
improved outcomes for younger patients with T-cell
analyzed 32 patients with either ALCL (n = 21) or
lymphoma as compared to other high-dose regimens.
PTCL-NOS (n = 11), and 25 of these had relapsed or
There have been only a handful of prospective
refractory disease. All patients received the same prep-
studies of consolidative autoSCT in PTCL
arative regimen consisting of busulfan, etoposide, and
(Table 14.6). Survival in these studies ranges from
cyclophosphamide. The OS and relapsed-free survival
3-year OS rates of 48–86% and 5-year OS rates of
(RFS) at 5 years was 34% and 18%, respectively. There
32–39%. Corradini et al. reported pooled results of
was no significant difference in OS or RFS based on
two prospective phase II studies. With median follow-
histology. There was also no difference in outcome
up of 76 months, they estimated 12-year OS, DFS, and
based on whether they were in first remission or had
EFS rates were 34%, 55%, and 30%, respectively. ALK+
relapsed/refractory disease (P = 0.42 OS; P = 0.24
ALCL was included in this study, which accounted for
RFS). Immunohistochemical staining was available
a population where a stable long-term CR was
for 11 of the ALCL patients, and four of five that
observed. Further, the Nordic Lymphoma Group pre-
were positive for ALK are alive at last follow-up com-
sented data on PTCL-NOS and ALK– ALCL and, with
pared to two of six ALK-negative patients.
77 patients available for analysis, 75% underwent
Rodriguez et al. retrospectively studied 123 relapsed
transplantation with a CR rate of 77%.
or refractory patients with histologic subtypes, includ-
One of the largest prospective, multicenter studies
ing PTCL-NOS (n = 70), anaplastic large T-cell lym-
on high-dose chemotherapy and total-body irradiation
phoma (n = 31), lymphoepithelioid T-cell lymphoma
(TBI) followed by autoSCT in newly diagnosed PTCL
(n = 10), hepatosplenic γδ T-cell lymphoma (n = 2),
was recently reported by Reimer et al. In this study, 83
subcutaneous panniculitis-like T-cell lymphoma
patients were enrolled, with the patient population con-
(n = 2), and intestinal T-cell lymphoma (n = 1). The
sisting of PTCL-NOS (n = 32) and angioimmunoblastic
pre-transplant regimens were varied, but mainly were
T-cell lymphoma (n = 27). The treatment regimen con-
anthracycline-containing regimens. Of assessable
sisted of four to six cycles of cyclophosphamide, doxo-
patients, 87 of 119 (73%) of patients had a CR and
rubicin, vincristine, and prednisone followed by
11% had a PR, and 16% had either SD or progressive
mobilizing therapy with either dexamethasone, carmus-
disease (PD). After a median follow-up of 61 months,
tine, melphalan, etoposide, and cytarabine or etoposide,
47% were still alive. The OS at 5 years was 45% while the
methylprednisolone, cytarabine, and cisplatin with stem
PFS was 34%. On multivariate analysis, the IPI and β2-
cell collection. Patients who achieved a CR or PR under-
microglobulin level predicted outcomes after autoSCT
went myeloablative chemoradiation (fractionated TBI
in this relapsed, refractory population.
and high-dose cyclophosphamide) and autoSCT. Fifty-
Kim et al. retrospectively analyzed 40 patients with
five of 86 patients received transplantation; the main
PTCL, 29 of whom had either relapsed or refractory
reason for not receiving it was progressive disease. Pre-
disease (8 patients in second CR, 17 in second PR, 4
transplant CR rate was 74% (n = 40) and PR 27%
with refractory disease). Univariate analysis showed
(n = 15). Of the 55 who underwent transplantation, 48
that CR status at autoSCT correlated with both EFS
of 55 patients achieved a CR, and 7 achieved a PR, with
(P = 0.023) and OS (P = 0.032). Those in CR at the time
an ORR of 66% post-transplant. With a median follow-
of transplant had a median EFS of 49.6 months, with
up of 33 months, the estimated 3-year OS and DFS rates
OS not reached. This is compared to those not in CR at
for patients in CR were 48% and 53%, respectively. The
the time of autologous transplant, who had an EFS of
3-year PFS rate was 36%. Using a univariate analysis,
3.3 months and OS of 7.4 months. This study did not
Reimer et al. found a significant impact for PIT on OS
distinguish between those in first or second CR. The
and a non-significant trend for improved outcome for
patients that received autoSCT in the frontline setting
low/intermediate–low IPI and for CR status prior to
had an EFS and OS of 49.6 months and 80 months,
transplant. Other authors have reported significantly
respectively. The patients that received autoSCT in the
superior OS and EFS in patients who received their
relapsed or refractory setting had an EFS of 3.6 months
autograft in CR compared with no CR.
and OS of 7.4 months.
247
 Chapter 14: T-cell non-Hodgkin lymphoma

Yang et al. retrospectively evaluated 64 patients who
underwent autoSCT, 30 of whom were in the relapsed
or refractory setting. The 5-year PFS and OS rates for
patients in CR1/PR1 groups were 51% and 76%, respec-
tively, decreasing to 12% and 40%, respectively, for
patients in CR2/PR2. The 3-year OS rate for patients
who achieved CR (n = 21), regardless of transplant
timing, was 71.8%. They found, however, that 3-year
OS rates differed significantly in patients undergoing
transplantation in CR1/PR1 (60%) and those under-
going transplantation in a salvage setting (37.7%).
However, the 3-year OS rate for patients in CR2 was
70.9%, compared with 50% for those in PR1.
Consequently, the achievement of CR at the time of
transplantation was a more significant factor for pre-
dicting survival than was transplant timing. Although
this data is retrospective, it may indicate that there may
be a role for autoSCT even in the relapse/refractory
setting, especially if a CR2 can be achieved.
Allogeneic stem cell transplantation
The relapse rate among relapsed/refractory T-cell lym-
phoma patients is high following standard chemother-
apy as well as autoSCT, therefore there has been interest
in evaluating the use of allogeneic SCT (alloSCT)
(Table 14.7). The Societe Francaise de Greffe de
Moelle et de Therapie Cellulaire conducted a retrospec-
tive analysis of 77 patients with T-cell lymphoma; his-
tologies included ALCL (n = 27), PTCL-NOS (n = 27),
AITL (n = 11), hepatosplenic γδ lymphoma (n = 3),
T-cell granular lymphocytic leukemia (n = 1), enter-
opathy T-cell lymphoma (n = 1), and ATLL (n = 2). All
patients had received at least one line of therapy (mean
2, range 1–5) before alloSCT, with 25% receiving a
previous autoSCT. Fifty-seven patients (74%) received
a myeloablative conditioning regimen. Only 31 of 77
patients were in CR at the time of alloSCT, whereas 26
of 77 were in PR. With mean follow-up of 43 months,
the 5-year OS and EFS rates were 57% and 53%, respec-
tively. In multivariate analysis, chemoresistant disease
at the time of alloSCT and the occurrence of severe
grade 3–4 acute graft-versus-host disease were the
strongest adverse prognostic factors for OS (P = 0.03).
Kyriakou et al. retrospectively analyzed 45 patients
with AITL who underwent alloSCT. The majority of
patients had received over two lines of previous che-
motherapy, and 11 of the 45 patients had treatment
failure to a prior autoSCT. Twenty-five patients under-
went a myeloablative alloSCT and 20 underwent
reduced-intensity alloSCT. PFS and OS rates were
62% and 53% and 66% and 64% at 1 and 3 years,
respectively. Those with chemotherapy-sensitive dis-
ease had significantly improved outcomes. In this
small series, the results of the reduced-intensity
alloSCT were similar to those who received a myelo-
ablative alloSCT.
Shustov et al. retrospectively analyzed 17 patients
with relapsed/refractory (n = 14) or poor-risk newly
diagnosed (n = 3) T-cell and NK-cell lymphomas who

Table 14.7 Studies of allogeneic stem cell transplantation for T-cell lymphoma (non-cutaneous).

Study Type of No. of Regimen Non- Overall Comments

study patients relapse survival
mortality
Corradini Retrospective 17 Non-myeloablative 6% at 2 80% at 2 years Younger popula-
et al., years tion, median
2004 age 47
Gouil Retrospective 77 74% Myeloablative, 33% at 5 57% at 5 years Mean 2 prior
et al., 26% Non-myeloablative years therapies (range
2008 1–5), 25% prior
autoSCT
Kyriakou Retrospective 45 55% Myeloablative, 25% at 1 66% at 1 year, Only AITL, 24%
et al., 44% Non-myeloablative year 64% at 3 years prior autoSCT
2007
Shustov Retrospective 17 Non-myeloablative 19% at 3 59% at 3 years Median 3 prior
et al., years therapies (range
2010 1–7), 35% with
prior autoSCT
autoSCT, Autologous stem cell transplantation; AITL, angioimmunoblastic T-cell lymphoma; No. number.
248

underwent non-myeloablative alloSCT. These
patients were highly pre-treated with median number
of prior treatment three (range 1–7), and seven
patients (41%) received prior autoSCT. Eight (47%)
patients were in CR and four (24%) patients were in
PR at the time of alloSCT. Five (29%) patients had
refractory disease prior to alloSCT. With a median
follow-up of 3.3 years, 9 of 17 patients were still alive.
Estimated 3-year OS and PFS were 59% and 53%,
respectively. Interestingly, five out of nine patients
who had evidence of disease prior to stem cell infu-
sion achieved CR after transplantation, suggesting
the role for graft-versus-lymphoma effect. Similar
conversions from PR to CR after non-myeloablative
conditioning and alloSCT were observed in other
studies in patients with PTCL.
Smith et al. recently published data comparing
post-transplant outcomes among 241 recipients (≤60
years) of autoSCT (n = 115) and alloSCT (n = 126, 76
matched siblings) for T-cell lymphoma between 1996
and 2006, reported to the CIBMTR (Center for
International Blood and Marrow Transplant
Research). The median age was 43 years for autoSCT
and 38 years for alloSCT. In the alloSCT cohort, 59%
received myeloablative conditioning and 60% had
matched sibling donors. AutoSCT, compared to
alloSCT, was used more frequently in ALCL (53%
versus 40%; P = 0.04) and in those with less advanced
and more sensitive disease, including those in first CR
(35% versus 14%; P = 0.001), chemosensitive disease
(86% versus 60%; P < 0.0001), and ≤2 lines prior
therapy (65% versus 44%; P < 0.001). Median follow-
up was 71 months for autoSCT and 49 months for
alloSCT. Treatment-related mortality at 100 days in
the autoSCT and alloSCT cohorts was 2% and 17%,
respectively. For those beyond CR1 in the autoSCT
and alloSCT cohorts, OS at 1 year was 62% versus 52%
and at 3 years 53% versus 41%, respectively. Relapsed
T-cell NHL was the cause of death in 73% of autoSCT
and 44% of alloSCT patients.
Radiation therapy
Because of the paucity of clinical trials, the integration
of radiation therapy into the treatment plan of most
T-cell NHL patients often models that of B-cell NHL.
Compared to DLBCL, T-cell NHL more commonly
presents with stage III/IV disease (48% and 74%, respec-
tively) and T-cell patients frequently present with com-
bined nodal and extranodal disease at diagnosis
(especially PTCL-NOS, ALCL, and AILT), thereby
Chapter 14: T-cell non-Hodgkin lymphoma
often obviating the need for the addition of
radiation. Patients with extranodal NK-/T-cell lym-
phoma can frequently present with localized (stage I
or II) disease that responds well to treatment with
radiation with or without chemotherapy. Studies spe-
cifically addressing the benefits of radiation therapy for
stage I/II non-extranodal NK-/T-cell NHL or bulky sites
with advanced disease have not been done. Despite
lacking data, common practice has been to apply radio-
therapy similar to aggressive B-cell lymphoma.
Emerging and novel therapies
Identification of relevant proto-oncogenes and tumor
suppressor genes involved in the pathogenesis of T-cell
NHL represents potential candidates for molecular-
based therapy. There are many drugs being examined
in patients with T-cell NHL, some of which have been
proven effective in other malignancies (Table 14.8).
Proteasome inhibitors, including bortezomib, rely
on inhibition of the ubiquitin-proteasome pathway,
and reports of phase II trials with single agent borte-
zomib as well as in combination with CHOP show
promising ORRs of approximately 60%. Histone
deacetylase (HDAC) inhibitors trigger both caspase-
dependent and caspase-independent apoptosis; romi-
depsin, a pan-HDAC inhibitor, has shown activity in
both CTCL and PTCL-NOS with ORR 41%. These
responses were durable, with a median of 14.9 (range
1–66) months, and many of these patients were heavily
treated prior to receiving romidepsin. Romidepsin
received accelerated FDA approval in June 2011 for
the treatment of PTCL patients who have received at
least one prior therapy.
Denileukin diftitox (DAB389IL-2, Ontak) is a novel
recombinant fusion protein consisting of peptide
sequences for the enzymatically active and membrane
active and membrane translocation domains of diph-
theria toxin with recombinant interleukin-2. It has
been FDA-approved for cutaneous T-cell NHL,
although clinical benefit has been reported in other
T-cell NHL patients. As a single agent, denileukin
diftitox has a 48% ORR in relapsed/refractory PTCL.
Recently presented phase II data combining denileu-
kin diftitox and CHOP in relapsed/refractory PTCL
showed a promising 67% ORR and 58% CR rate, and
there is a larger multicenter trial planned.
Alemtuzumab is a monoclonal anti-CD52 anti-
body that has shown activity in PTCL. An Italian 249
group combined CHOP with alemtuzumab in 24
 Chapter 14: T-cell non-Hodgkin lymphoma

Table 14.8 Summary of novel therapeutic agents for T-cell lymphoma.

Mechanism/target Examples of agents Current status

Anti-CD30 antibody and SGN-35 Pivotal phase II ALCL trial completed (SGN-35); studies
antibody–drug incorporating SGN-35 into frontline ALCL as well as non-
conjugate ALCL PTCL are ongoing
Antifolate Pralatrexate FDA approved for relapsed PTCL
Antitubulin Vinblastine Significant single agent clinical activity in pediatric ALCL
Proteasome inhibition Bortezomib Ongoing phase II studies combined with chemotherapy
or novel agents for relapsed T-cell NHL
Anti-CD25 drug Denileukin diftitox Modest activity as single agent in PTCL; studies combined
conjugate with CHOP for newly diagnosed T-cell NHL
HDAC inhibitor Romidepsin and vorinostat Both FDA-approved for CTCL; romidepsin approved for
relapsed/refractory PTCL
IMIDs® Thalidomide and lenalidomide Preliminary activity in relapsed/refractory T-cell NHL
Anti-VEGF Bevacizumab Combined with CHOP for newly diagnosed T-cell NHL;
vascular toxicities apparent
131
Radioimmunoconjugates Iodine-anti-CD45 radioantibody, Preclinical and early clinical development
131
Iodine-anti-CD25, 90Yttrium-anti-CD25,
and 90Ytrrium-anti-CD5
ALK inhibition Diaminopyrimidines (NVP-TAE684), dia- Preclinical development and early clinical development
lkoxyquinolines, staurosporine-like
molecules
Signaling pathways Nutlin-3a, flavopiridol, 17-allylamino-17- Preclinical development and early clinical development
downstream of ALK demethoxygeldanamycin (17-AAG), heat
shock protein 90
PTCL, Peripheral T-cell lymphoma; ALK, anaplastic lymphoma kinase; NHL, non-Hodgkin lymphoma; ALCL, anaplastic cutaneous lymphoma;
CTCL, cutaneous T-cell lymphoma; HDAC, histone deacetylase; CHOP, cyclophosphamide, adriamycin, oncovin, and prednisone; IMIDs,
immunomodulatory drugs; VEGF, vascular endothelial growth factor.

patients, 14 of whom were classified as unspecified
(PTCL-u). Of these patients, nine had either a CR or
PR, with five having disease progression. At a mean
follow-up of 495 days, 5/14 patients with PTCL-u were
alive and all were in CR. Toxicity due to infections can
be prohibitive with alemtuzumab and more trials of
this combination are ongoing. In addition, given that
about 30–40% of PTCL-u are CD52+, determining
antigen status is important. Huang and colleagues
studied 17 patients with relapsed T-cell NHL using
13-cis-retinoic acid with IFN-α. A response rate of
31% was documented in four of six ALCL patients
and one of seven PTCL-u, although median survival
for the entire group of patients was 3.6 months.
Pralatrexate (10-propargyl-10-deazaaminopterin)
is an analog of methotrexate and, when compared to
methotrexate, has improved cellular transport via
RFC-1, leading to a greater intracellular concentration
250 of drug and more effective inhibition of dihydrofolate
reductase (DHFR). An international, phase II,
non-randomized, open-label study, PROPEL, treated
117 heavily pre-treated patients with relapsed or
refractory PTCL (n = 109 evaluable for efficacy). The
ORR was 27%, with eight patients having CR and
seven with PR. Of note, only one (5%) of 20 patients
with B-cell lymphoma responded, while 54% of
patients with T-cell lymphoma obtained
remission. Among the T-cell lymphoma patients who
responded, eight (31%) had CR, and six (23%) had PR.
Monitoring for survival is ongoing, but responses lon-
ger than 1 year have been observed. Of note, in this
study 70% of responses were achieved after a single
dose, and five patients were able to undergo hemato-
poietic stem cell transplant (HSCT). Pralatrexate
was overall well tolerated, with 77 patients (69%)
receiving full-dose therapy, and 85% of all scheduled
doses were administered. The most frequent grade
3 and 4 adverse effects were mucositis (17% and
4%, respectively) and thrombocytopenia (14% and
19%, respectively). Nausea and fatigue were also

experienced by a significant number of patients (133).
Based on the results, in 2009 the US Food and
Drug Administration granted accelerated approval to
pralatrexate injection (Folotyn™, made by Allos
Therapeutics, Inc.) for the treatment of patients with
relapsed or refractory PTCL.
Bevacizumab, a humanized monoclonal antibody
directed against vascular endothelial growth factor
(VEGF), has been explored because of the prominent
vascular properties of AITL. There have been two
reports of patients receiving single agent bevacizumab
(one patient 5 mg/kg and one patient 10 mg/kg every 2
weeks) who both attained a CR; however, the durabil-
ity of their responses was short. The phase II data
investigating the combination of bevacizumab with
CHOP as frontline therapy for PTCL was hampered
in part by significant toxicity seen in particular cardiac
events. In 2008, a report of a CR induced by thalido-
mide in a patient with AITL who had failed autologous
transplant sparked interest in exploring its activity
against T-cell lymphomas. Dueck et al. reported lena-
lidomide as monotherapy in 24 previously treated
patients with relapsed and refractory T-cell lympho-
mas (mycosis fungoides excluded) in a phase II trial.
Histologically, patients had PTCL-NOS (n = 10), AITL
(n = 7), ALCL (n = 5), enteropathic T-cell lymphoma
(n = 1), and hepatosplenic γδ T-cell lymphoma (n = 1).
Of the 23 evaluable patients, the ORR was 30% (7 of 23
patients, all with PR), with a median PFS of 96 days
(range 8–696 days). The subtypes that responded
included PTCL-NOS, AITL, and ALCL histologies.
Of the patients that achieved stable disease or better
(n = 9), median PFS was 168 days and median OS had
not been reached with 241–696 days follow-up. The
most common grade 4 adverse event was thrombocy-
topenia, which occurred in one-third of patients.
Neutropenia (20.8%), febrile neutropenia (16.7%),
and unspecified pain (16.7%) comprised the most
common grade 3 adverse effects. The conclusions
from this study were that, in this cohort, lenalidomide
achieved expected results in concordance with the
experience seen in other disease states.
Continuing translational research is needed both
to allow more accurate prediction of course of dis-
ease and to elucidate the molecular oncogenesis of
T-cell NHL further. Prospective, multi-institution
clinical trials remain critically important to deter-
mine the most effective treatment regimens that
will continue to improve cure rates in these often
aggressive diseases.
Chapter 14: T-cell non-Hodgkin lymphoma
Further reading
Advani RH, Hong F, Horning SJ, et al. cardiac toxicity
associated with bevacizumab (Avastin) in combination
with CHOP chemotherapy for peripheral T cell
lymphoma in ECOG 2404 trial. Leuk Lymphoma, 2012;
53:718–720.
Arrowsmith ER, Macon WR, Kinney MC, et al. Peripheral
T-cell lymphomas: clinical features and prognostic
factors of 92 cases defined by the revised European
American lymphoma classification. Leuk Lymphoma,
2003;44:241–249.
Ashton-Key M, Diss TC, Pan L, Du MQ, Isaacson PG.
Molecular analysis of T-cell clonality in ulcerative
jejunitis and enteropathy-associated T-cell lymphoma.
Am J Pathol, 1997;151:493–498.
Attygalle AD, Kyriakou C, Dupuis J, et al. Histologic
evolution of angioimmunoblastic T-cell lymphoma in
consecutive biopsies: clinical correlation and insights into
natural history and disease progression. Am J Surg Pathol,
2007;31(7):1077–1088.
Au WY, Weisenburger DD, Intragumtornchai T, et al.
Clinical differences between nasal and extranasal natural
killer/T-cell lymphoma: a study of 136 cases from the
International Peripheral T-Cell Lymphoma Project.
Blood, 2009;113:3931–3937.
Belhadj K, Reyes F, Farcet JP, et al. Hepatosplenic
gammadelta T-cell lymphoma is a rare clinicopathologic
entity with poor outcome: report on a series of 21
patients. Blood, 2003;102:4261–4269.
Benharroch D, Meguerian-Bedoyan Z, Lamant L, et al. ALK-
positive lymphoma: a single disease with a broad
spectrum of morphology. Blood, 1998;91(6):2076–2084.
Cooke CB, Krenacs L, Stetler-Stevenson M, et al.
Hepatosplenic T-cell lymphoma: a distinct
clinicopathologic entity of cytotoxic gamma delta T-cell
origin. Blood, 1996;88(11):4265–4274.
Croce CM, Isobe M, Palumbo A, et al. Gene for alpha-chain
of human T-cell receptor: location on chromosome 14
region involved in T-cell neoplasms. Science,
1985;227:1044–1047.
Dahmoush L, Hijazi Y, Barnes E, Stetler-Stevenson M, Abati
A. Adult T-cell leukemia/lymphoma: a cytopathologic,
immunocytochemical, and flow cytometric study.
Cancer, 2002;96(2):110–116.
Daum S, Weiss D, Hummel M, et al. Frequency of clonal
intraepithelial T lymphocyte proliferations in
enteropathy-type intestinal T cell lymphoma, coeliac
disease, and refractory sprue. Gut, 2001;49:804–812.
Escalon MP, Liu NS, Yang Y, et al. Prognostic factors and
treatment of patients with T-cell non-Hodgkin
lymphoma: the M. D. Anderson Cancer Center
experience. Cancer, 2005;103:2091–2098.
251
 Chapter 14: T-cell non-Hodgkin lymphoma
Evens AM, Rosen ST, Helenowski I et al. A phase I/II trial of
Bortezomib combined concurrently with Gemcitabine
for relapsed or refractory DLBCL and peripheral T-cell
lymphomas. Br J Haematol, 2013;163:55–56.
Foroni L, Foldi J, Matutes E, et al. Alpha, beta and gamma
T-cell receptor genes: rearrangements correlate with
haematological phenotype in T cell leukaemias. Br J
Haematol, 1987;67:307–318.
Gale J, Simmonds PD, Mead GM, Sweetenham JW, Wright
DH. Enteropathy-type intestinal T-cell lymphoma:
clinical features and treatment of 31 patients in a single
center. J Clin Oncol, 2000;18:795–803.
Gallamini A, Stelitano C, Calvi R, et al. Peripheral T-cell
lymphoma unspecified (PTCL-U): a new prognostic
model from a retrospective multicentric clinical study.
Blood, 2004;103:2474–2479.
Gallamini A, Zaja F, Patti C, et al. Alemtuzumab (Campath-
1H) and CHOP chemotherapy as first-line treatment of
peripheral T-cell lymphoma: results of a GITIL (Gruppo
Italiano Terapie Innovative nei Linfomi) prospective
multicenter trial. Blood, 2007;110:2316–2323.
Gascoyne RD, Aoun P, Wu D, et al. Prognostic significance
of anaplastic lymphoma kinase (ALK) protein expression
in adults with anaplastic large cell lymphoma. Blood,
1999;93:3913–3921.
Gisselbrecht C, Gaulard P, Lepage E, et al. Prognostic
significance of T-cell phenotype in aggressive non-
Hodgkin’s lymphomas. Groupe d’Etudes des
Lymphomes de l’Adulte (GELA). Blood, 1998;92:76–82.
Glass B, Kloess M, Bentz M, et al. Dose-escalated CHOP plus
etoposide (MegaCHOEP) followed by repeated stem cell
transplantation for primary treatment of aggressive high-
risk non-Hodgkin lymphoma. Blood, 2006;107:3058–3064.
Grogg KL, Attygalle AD, Macon WR, et al. Expression of
CXCL13, a chemokine highly upregulated in germinal
center T-helper cells, distinguishes angioimmunoblastic
T-cell lymphoma from peripheral T-cell lymphoma,
unspecified. Mod Pathol, 2006;19(8):1101–1107.
Gutierrez-Garcia G, Garcia-Herrera A, Cardesa T, et al.
Comparison of four prognostic scores in peripheral T-cell
lymphoma. Ann Oncol,
Hasserjian RP, Harris NL. NK-cell lymphomas and
leukemias: a spectrum of tumors with variable
manifestations and immunophenotype. Am J Clin Pathol,
2007;127(6):860–868.
Huang Y, de Reynies A, de Leval L, et al. Gene expression
profiling identifies emerging oncogenic pathways
operating in extranodal NK/T-cell lymphoma, nasal type.
Blood, 115:1226–1237.
Iqbal J, Weisenburger DD, Greiner TC, et al. Molecular
signatures to improve diagnosis in peripheral T-cell
lymphoma and prognostication in angioimmunoblastic
252 T-cell lymphoma. Blood, 115:1026–1036.
Jaffe ES, Chan JK, Su IJ, et al. Report of the workshop on
nasal and related extranodal angiocentric T/natural
killer cell lymphomas. Definitions, differential
diagnosis, and epidemiology. Am J Surg Pathol,
1996;20:103–111.
Jaffe ES, Gaulard P, Ralfkiaer, Cerroni L, Meijer CJLM.
Subcutaneous panniculitis-like T-cell lymphoma. In:
Swerdlow SH, Campo E, Harris NL, et al. (eds). WHO
Classification of Tumours of Haematopoietic and
Lymphoid Tissues, 4th ed. Lyon: IARC; 2008: 294.
Kanavaros P, Lescs MC, Briere J, et al. Nasal T-cell
lymphoma: a clinicopathologic entity associated with
peculiar phenotype and with Epstein–Barr virus. Blood,
1993;81:2688–2695.
Kim SJ, Kim K, Kim BS, et al. Phase II trial of concurrent
radiation and weekly cisplatin followed by VIPD
chemotherapy in newly diagnosed, stage IE to IIE, nasal,
extranodal NK/T-cell lymphoma: Consortium for
Improving Survival of Lymphoma study. J Clin Oncol,
2009;27:6027–6032.
Knudson AG. Antioncogenes and human cancer. Proc Natl
Acad Sci U S A, 1993;90:10914–10921.
Kyriakou C, Canals C, Finke J, et al. Allogeneic stem cell
transplantation is able to induce long-term remissions in
angioimmunoblastic T-cell lymphoma: a retrospective
study from the Lymphoma Working Party of the
European Group for Blood and Marrow Transplantation.
J Clin Oncol, 2009;27:3951–3958.
Marafioti T, Paterson JC, Ballabio E, et al. The inducible T-
cell co-stimulator molecule is expressed on subsets of T
cells and is a new marker of lymphomas of T follicular
helper cell-derivation. Haematologica, 2010;95
(3):432–439.
Melnyk A, Rodriguez A, Pugh WC, Cabannillas F.
Evaluation of the Revised European–American
Lymphoma classification confirms the clinical
relevance of immunophenotype in 560 cases of
aggressive non-Hodgkin’s lymphoma. Blood,
1997;89:4514–4520.
Mourad N, Mounier N, Briere J, et al. Clinical, biologic, and
pathologic features in 157 patients with
angioimmunoblastic T-cell lymphoma treated within the
Groupe d’Etude des Lymphomes de l’Adulte (GELA)
trials. Blood, 2008;111:4463–4470.
O’Connor OA, Horwitz S, Hamlin P, et al. Phase II-I-II study
of two different doses and schedules of pralatrexate, a
high-affinity substrate for the reduced folate carrier, in
patients with relapsed or refractory lymphoma reveals
marked activity in T-cell malignancies. J Clin Oncol,
2009;27:4357–4364.
Pellatt J, Sweetenham J, Pickering RM, Brown L, Wilkins B.
A single-centre study of treatment outcomes and survival
in 120 patients with peripheral T-cell non-Hodgkin’s
lymphoma. Ann Hematol, 2002;81:267–272.

Pileri SA, Ralfkiaer E, Weisenberger DD, et al. Peripheral
T-cell lymphoma, not otherwise specified. In: Swerdlow
SH, Campo E, Harris NL, et al. (eds). WHO Classification
of Tumours of Haematopoietic and Lymphoid Tissues, 4th
ed. Lyon: IARC; 2008: 306–308.
Rabbitts TH. Chromosomal translocation master genes,
mouse models and experimental therapeutics. Oncogene,
2001;20:5763–5777.
Rao PH, Houldsworth J, Dyomina K, et al. Chromosomal
and gene amplification in diffuse large B-cell lymphoma.
Blood, 1998;92:234–240.
Reimer P, Rudiger T, Geissinger E, et al. Autologous stem-
cell transplantation as first-line therapy in peripheral
T-cell lymphomas: results of a prospective multicenter
study. J Clin Oncol, 2009;27:106–113.
Rodriguez J, Conde E, Gutierrez A, et al. Prolonged survival
of patients with angioimmunoblastic T-cell lymphoma
after high-dose chemotherapy and autologous stem cell
transplantation: the GELTAMO experience. Eur J
Haematol, 2007;78:290–296.
Rodriguez J, Conde E, Gutierrez A, et al. The results of
consolidation with autologous stem-cell transplantation
in patients with peripheral T-cell lymphoma (PTCL) in
first complete remission: the Spanish Lymphoma and
Autologous Transplantation Group experience. Ann
Oncol, 2007;18:652–657.
Rudiger T, Weisenburger DD, Anderson JR, et al.
Peripheral T-cell lymphoma (excluding anaplastic large-
cell lymphoma): results from the Non-Hodgkin’s
Lymphoma Classification Project. Ann Oncol,
2002;13:140–149.
Schmitz N, Trumper L, Ziepert M, et al. Treatment and
prognosis of mature T-cell and NK-cell lymphoma: an
analysis of patients with T-cell lymphoma treated in
studies of the German High-Grade Non-Hodgkin
Lymphoma Study Group. Blood, 116:3418–3425.
Shimoyama M. Diagnostic criteria and classification of
clinical subtypes of adult T-cell leukaemia-lymphoma.
A report from the Lymphoma Study Group (1984–87).
Br J Haematol, 1991;79:428–437.
Siegert W, Agthe A, Griesser H, et al. Treatment of
angioimmunoblastic lymphadenopathy (AILD)-type
T-cell lymphoma using prednisone with or without the
COPBLAM/IMVP-16 regimen. A multicenter study.
Kiel Lymphoma Study Group. Ann Intern Med,
1992;117:364–370.
Smith SD, Bolwell BJ, Rybicki LA, et al. Autologous
hematopoietic stem cell transplantation in peripheral
T-cell lymphoma using a uniform high-dose regimen.
Bone Marrow Transplant, 2007;40:239–243.
Smith SM, Burns LJ, van Besien K, et al. Hematopoietic cell
transplantation for systemic mature T-cell non-Hodgkin
lymphoma. J Clin Oncol, 2013;31:3100–3109.
Chapter 14: T-cell non-Hodgkin lymphoma
Sonnen R, Schmidt WP, Muller-Hermelink HK, Schmitz N.
The International Prognostic Index determines
the outcome of patients with nodal mature T-cell
lymphomas. Br J Haematol, 2005;129:366–372.
Stein H, Foss HD, Dürkop H, et al. CD30(+) anaplastic
large cell lymphoma: a review of its histopathologic,
genetic, and clinical features. Blood,
2000;96(12):3681–3695.
Swerdlow SH, Harris CE. WHO Classification of Tumours of
Haematopoietic and Lymphoid Tissues, 4th ed. Lyon:
IARC; 2008.
The International Non-Hodgkin’s Lymphoma Prognostic
Factors Project. A predictive model for aggressive
non-Hodgkin’s lymphoma. N Engl J Med,
1993;329:987–994.
The Non-Hodgkin’s Lymphoma Classification Project. A
clinical evaluation of the International Lymphoma Study
Group classification of non-Hodgkin’s lymphoma. Blood,
1997;89:3909–3918.
Theodorou I, Bigorgne C, Delfau MH, et al. VJ
rearrangements of the TCR gamma locus in peripheral
T-cell lymphomas: analysis by polymerase chain reaction
and denaturing gradient gel electrophoresis. J Pathol,
1996;178:303–310.
Toro JR, Liewehr DJ, Pabby N, et al. Gamma-delta T-cell
phenotype is associated with significantly decreased
survival in cutaneous T-cell lymphoma. Blood,
2003;101(9):3407–3412.
Tsukasaki K, Krebs J, Nagai K, et al. Comparative genomic
hybridization analysis in adult T-cell leukemia/
lymphoma: correlation with clinical course. Blood,
2001;97:3875–3881.
Vose J, Armitage J, Weisenburger D. International
peripheral T-cell and natural killer/T-cell lymphoma
study: pathology findings and clinical outcomes. J Clin
Oncol, 2008;26:4124–4130.
Went P, Agostinelli C, Gallamini A, et al. Marker expression
in peripheral T-cell lymphoma: a proposed clinical-
pathologic prognostic score. J Clin Oncol,
2006;24:2472–2479.
Willemze R, Jansen PM, Cerroni L, et al. Subcutaneous
panniculitis-like T-cell lymphoma: definition,
classification, and prognostic factors: an EORTC
Cutaneous Lymphoma Group Study of 83 cases. Blood,
2008;111:838–845.
Yong W, Zheng W, Zhang Y, et al. L-asparaginase-based
regimen in the treatment of refractory midline nasal/
nasal-type T/NK-cell lymphoma. Int J Hematol,
2003;78:163–167.
Yong W, Zheng W, Zhu J, et al. L-asparaginase in the
treatment of refractory and relapsed extranodal NK/
T-cell lymphoma, nasal type. Ann Hematol,
2009;88:647–652. 253
 Chapter
15
Primary cutaneous lymphoma
Sean Whittaker
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
The skin is the second most frequent extranodal site,
after the gastrointestinal tract, for lymphoma, with an
annual incidence of 0.5–1 per 100 000, although recent
Scandinavian studies have suggested an incidence of 4
per 100 000, possibly because of improved diagnosis
and registration.
Primary cutaneous T-cell lymphoma (CTCL) com-
prises a heterogeneous group of non-Hodgkin’s lym-
phoma (NHL), of which mycosis fungoides (MF) is the
most common clinicopathologic subtype. MF typically
has an indolent course but disease progression may
occur in approximately 35% of patients. Sézary syn-
drome (SS), a leukemic form of CTCL, is very closely
related to MF and has a poor prognosis, with a median
survival of less than 3 years.
Primary cutaneous B-cell lymphomas (CBCL) are
less common, comprising approximately 20% of all
primary cutaneous lymphomas. They typically present
with cutaneous papules, plaques, or nodules, and can
be broadly divided into follicle center cell lymphoma,
marginal zone lymphoma, and large B-cell lymphoma.
The WHO–EORTC classification system
(Table 15.1) has clarified the classification of primary
cutaneous lymphomas, and the distinction of rare
CTCL variants from MF/SS is critical as the prognosis
is poorer and treatment options are different.
Primary cutaneous T-cell
lymphomas (CTCL)
Mycosis fungoides
MF is generally associated with an indolent clinical
course and is characterized by polymorphic atrophic
erythematous patches and scaly plaques. Some
patients progress from having limited patches and
254 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
plaques to extensive thick plaques or tumors and
even erythroderma, and a minority (approximately
30%) will present with advanced disease. However,
many patients do not develop disease progression. A
number of clinical variants are recognized, including
hypopigmented and folliculotropic variants.
Staging is based on cutaneous surface area involved
and type of skin lesion; patches and plaques involving
less than 10% of the body surface area (stage T1/IA;
Figure 15.1) and more than 10% (stage T2/IB;
Figure 15.2), tumors (stage T3/IIB; Figure 15.3) and
erythrodermic disease (stage T3/III; Figure 15.2). In
2007 a revised staging system was proposed, incorp-
orating stratification of early-stage skin disease into
patches only compared to those with patches and pla-
ques as well as a detailed histologic and molecular
classification of node and peripheral blood involvement
(Table 15.2). In 2010 this proposed staging model was
validated on a large cohort of 1502 patients. The prog-
nosis of those with stage IA and IB disease is excellent,
with 5-year disease-specific survival rates of 98% and
89%, respectively. However, those patients with patches
only (T1a/T2a) have a significantly better prognosis than
those with patches and plaques (T1b/T2b), with disease-
specific survival rates of 100%/96% compared to 90%/
86%, respectively. In contrast, the presence of tumors
(IIB) or erythroderma (III) is associated with a worse
prognosis (5-year disease-specific survival rates of 52%/
48%, respectively) (Table 15.3). Notably, the presence of
a peripheral blood T-cell clone identical to the cutane-
ous T-cell clone (B0b) is associated with a significantly
worse prognosis. Histologic involvement of lymph
nodes (5-year disease-specific survival of 25% and
median survival of 2 years) and visceral disease are
associated with a poor prognosis (5-year disease-specific
survival 18% and median survival 18 months).
 Chapter 15: Primary cutaneous lymphoma

Table 15.1 WHO–EORTC classification of cutaneous lymphomas with primary cutaneous manifestations.

Cutaneous T-cell and NK-cell lymphomas

Mycosis fungoides
MF variants and subtypes
Folliculotropic MF
Pagetoid reticulosis
Granulomatous slack skin
Sézary syndrome
Primary cutaneous CD30+ lymphoproliferative disorders
Primary cutaneous anaplastic large-cell lymphoma
Lymphomatoid papulosis
Subcutaneous panniculitis-like T-cell lymphoma*
Primary cutaneous peripheral T-cell lymphoma, unspecified
Primary cutaneous aggressive epidermotropic CD8+ T-cell lymphoma (provisional)
Cutaneous/T-cell lymphoma (provisional)
Primary cutaneous CD4+ small-/medium-sized pleomorphic T-cell lymphoma (provisional)
Cutaneous B-cell lymphomas
Primary cutaneous marginal zone B-cell lymphoma
Primary cutaneous follicle center lymphoma
Primary cutaneous diffuse large B-cell lymphoma, leg type
Primary cutaneous diffuse large B-cell lymphoma, other intravascular large B-cell lymphoma

nodes should be biopsied. Staging computed tomo-

graphic (CT) scans of the chest, abdomen, and pelvis
are indicated in all those patients with stage IIA–IV
disease. Bone marrow aspirate/trephine biopsies are
recommended only in those with unexplained hema-
tologic abnormalities as early bone marrow involve-
ment is rare and does not influence staging.

Figure 15.1 Mycosis fungoides, patch stage.
Multivariate analysis confirmed that stage, age, male
gender, lactate dehydrogenase (LDH), B0b, and follicu-
lotropic MF are independent prognostic factors.
Common sites of visceral involvement include pulmo-
nary, nasopharynx, and CNS.
All patients should have a full clinical examination
and adequate diagnostic biopsies for histology, immu-
nophenotypic, and molecular studies. Peripheral
blood samples should be taken for routine haematol-
ogy, biochemistry, serum LDH, Sézary cells, lympho-
cyte subsets, HTLVI serology, and T-cell receptor
(TCR) gene analysis. Any palpable bulky peripheral
Sézary syndrome
SS is defined by a clinical triad of erythroderma, var-
iable presence of peripheral lymphadenopathy and
atypical mononuclear cells (Sézary cells) comprising
>20% of total lymphocyte count or a total Sézary count
>1000 × 109/l (B2), (Figures 15.4 and 15.5), although
the distinction from erythrodermic MF patients with
early blood involvement (5% or more of peripheral
blood lymphocytes (B1)) is questionable, particularly
as the prognosis for B1 and B2 patients is similar. SS
patients usually have a modest CD4 lymphocytosis
with a blood T-cell clone as indicated either by aber-
rant expression of pan-T-cell antigens, cytogenetics, or
TCR gene analysis helping to distinguish SS from an
inflammatory erythroderma.
Patients present with a generalized exfoliative
erythroderma, ectropion, scalp alopecia, palmoplantar
hyperkeratoses, and nail dystrophy associated with
severe intractable pruritus. The prognosis is poor,
with a median survival of 3 years and 5-year disease- 255
specific survival of 30%. The majority die of
 Chapter 15: Primary cutaneous lymphoma
Figure 15.2 Mycosis fungoides, stage IB.
opportunistic infection. The prognosis is worse for
those with lymph node involvement and a high
blood tumor burden (Sézary count >10 000/μL).
Pathology of MF and Sézary syndrome
In the early stages the diagnosis of MF is difficult to
distinguish from dermatitis. Typical features, such as
the presence of intraepidermal lymphocytes that are
larger than normal lymphocytes with convoluted, cer-
ebriform nuclei, and a clear rim of cytoplasm (halo
cells) or a linear pattern of infiltration along the der-
moepidermal junction, are not specific. Pautrier
microabscesses (clusters of atypical lymphoid cells in
the epidermis) are only found in 10% of cases
(Figure 15.6). There is usually little spongiosis.
Plasma cells and eosinophils are usually absent in
early patch stages of MF.
In thin plaques the features are more established,
256 with a dense infiltrate that includes cells with linear
distribution in the basal layer of the epidermis and
Figure 15.3 Mycosis fungoides, tumor stage.
single cell epidermotropism. The cells are often small
with round or slightly irregular nuclei. In some cases
halo cells predominate. There is a background of small
lymphocytes, eosinophils, plasma cells, and histiocytes
(Figure 15.7).
In thick plaques there is a dense band-like subepi-
thelial infiltrate that includes a high number of atypical
lymphoid cells with cerebriform nuclei. Small clusters
of intraepidermal lymphoid cells are frequent but
Pautrier microabscesses are present in only 10%.
In tumors the infiltrate is more diffuse and epider-
motropism may be absent. The cellular infiltrate is
more polymorphic, with small, medium, and large
cells with cerebriform nuclei and occasional large
transformed, blastic cells that have prominent nucle-
oli. Large-cell transformation may occur in all stages
but is more common in tumors and is defined by
cohesive nodules of large pleomorphic or blast-like
cells.
In the folliculotropic subtype there is a dense infil-
trate of small- to medium-sized cells that infiltrate
around and within the hair follicles with sparing of
the intervening areas. The follicles show cystic dilata-
tion and plugging and there may be extracellular dep-
osition of mucinous material.
 Chapter 15: Primary cutaneous lymphoma

Table 15.2 TNMB stages.

Skin
T1 Limited patches, *papules, and/or plaques covering <10% of the skin surface
May further stratify into T1a (patch-only) versus T1b. (plaque ± patch)
T2 Patches, papules, or plaques covering ≥10% of the skin surface. May further stratify intoT2a
(patch-only) versus T2b, (plaque ± patch)
T3 One or more tumors‡ (≥1 cm diameter)
T4 Confluence of erythema covering ≥80% body surface area
Node
N0 No clinically abnormal peripheral lymph nodes§; biopsy not required
N1 Clinically abnormal peripheral lymph nodes; histopathology Dutch grade 1 or NCI LN02
N1a Clone-negative#
N1b Clone-positive#
N2 Clinically abnormal peripheral lymph nodes; histopathology Dutch grades or NCI LN3
N2a Clone-negative#
N2b Clone-positive#
N3 Clinically abnormal peripheral lymph nodes; histopathology Dutch grades 3–4 or NCI LN4;
clone-positive or -negative
Nx Clinically abnormal peripheral lymph nodes; no histologic confirmation
Visceral
Mo No visceral organ involvement
M1 Visceral involvement (must have pathology confirmation, and organ involved
should be specified)
Bleed
B0 Absence of significant blood involvement ±5% of peripheral blood lymphocytes
are atypical (Sézary) cells
B0a Clone-negative#
B0b Clone-positive#
B1 Low blood tumor burden: >5% of peripheral blood lymphocytes are atypical (Sézary)
cells but does not meet the criteria of B2
B1a Clone-negative#
B1b Clone-positive#
B1 High blood tumor burden: ≥1000/μL Sézary cells with positive clone#

In pagetoid reticulosis there is an acanthotic epi-
dermis that is infiltrated by halo cells, either individu-
ally or in small clusters.
Granulomatous slack skin shows a band-like infil-
trate in early stages, but as the disease progresses there is
infiltration of the dermis by small cells. In all cases there
is an associated population of multinucleate giant his-
tiocytic cells that often contains elastic fibers (elastoly-
sis) or lymphocytes within their cytoplasm.
In SS the features can be similar to classical MF,
with epidermotropism and Pautrier microabscesses
present, but the pathology may also be non-diagnostic
in almost one-third of patients with proven SS.
Lymph node involvement by MF is characterized by
dermatopathic type changes (NCI LN1), with atypical
lymphocytes that are either scattered individually (NCI
LN2) or present in small clusters (NCI LN3). With
more advanced infiltration of the nodes there is partial
effacement of the node by neoplastic cells with a mainly
paracortical distribution or diffuse infiltration and total
effacement of the node (NCI LN4).
The typical immunophenotype is that of a mature
T-cell and shows staining for CD2, CD3, and CD5. The
cells are usually CD4-positive, with lack of CD8
although CD8-positive cases are described, particularly
in childhood MF and hypopigmented variants. There is
usually no expression of CD30 in early stages of disease
but TCRαβ is usually expressed. There is frequent loss
of CD7 and as the disease progresses there may also be 257
loss of CD2 and CD5, particularly in the intraepidermal
Table 15.3 Prognosis in CTCL and CBCL.

Stage IA IB IIA IIB IIIA/ IVA1/ IVB LYP pcALCL SPTCL pcPTL, Pc Pc small/
MF/SS or IIIB IVA2 unspecified epidermotropic medium
disease CD8+ CTCL pleomorphic
entity CTCL
Disease- 98 89 89 56 54/48 41/23 18 100 90–95 82 16 18 60–80
specific
survival at 5
years (%)
Disease- 95 77 67 42 45 20 –
specific
survival at
10 years (%)
Median 35.5 21.5 15.8 4.7 4.7/3.4 3.8/2.1 1.4
survival
(years)
Risk of 8% 21% 17% 48% 53/82% 62/77% 82%
disease
progression
at 5 years

Prognosis in CBCL.
Disease entity pcMZL pcFCL pcLCL
Disease-specific survival 5 years (%) 99 95 50–55

Figure 15.4 Sézary syndrome.
component. Expression of cutaneous lymphoma-
associated antigen (CLA; associated with homing of
lymphocytes to the skin) is present in the majority of
cases. In large-cell transformation the tumor cells may
express CD30.
Molecular pathology of cutaneous
lymphomas
Primary cutaneous lymphomas include mature B-cell
and T-/NK-cell neoplasms that show entirely
Chapter 15: Primary cutaneous lymphoma
Figure 15.5 Sézary cell morphology.
Figure 15.6 Mycosis fungoides (skin). A dense infiltrate of atypical
lymphoid cells with convoluted nuclei focally extending into the
epidermis.
different molecular genetic characteristics and clini-
cal features. Their prognosis may be fundamentally
different from their histologically similar systemic
counterparts.
MF, the prototype of cutaneous T-cell lymphoma,
initially presents in the skin and shows a characteristic
stepwise clinical progression that probably reflects
cumulative alterations of various genetic events
involved in its pathogenesis. Monoclonal rearrange-
ment of TCR gamma genes is a common finding in
the tumor stage of MF, but is found in only half of early-
stage cases. Molecular cytogenetic analysis has identi-
fied common genetic alterations in MF and SS that
represent the systemic form of the disease. The most
frequent chromosomal losses involve chromosomes 1p,
17p, 10/10q, and 19, and chromosomal gains involve 4/
4q, 18, and 17/17q. Numerical aberrations of chromo- 259
somes 6, 13, 15, and 17, and structural aberrations of
 Chapter 15: Primary cutaneous lymphoma
Figure 15.7 Mycosis fungoides pathology.
chromosomes 3, 9, and 13 have been reported in MF,
including early lesions. A pseudodicentric translocation
(17;8)(p11.2;p11.2), identified by spectral karyotyping
(SKY) analysis, may represent a recurrent structural
aberration in these entities.
Recently, deletions or translocations affecting the
human homolog of unc-53, the NAV3 gene at chromo-
some12q, were reported in MF and SS patients. The
function of NAV3 in human lymphoid cells is not
known; however, preliminary studies aiming at the
silencing of NAV3 with small interfering RNAs showed
increased interleukin (IL)-2, but not CD25 expression,
suggesting that NAV3 may contribute to the growth,
differentiation, and apoptosis of the tumor cells as well
as in skewing from a Th1 to Th2 phenotype during
disease progression. In addition, one tumor sample was
identified that harbored a NAV3 deletion in one allele
and a missense mutation of the other allele, supporting
the hypothesis that NAV3 may act as a classical tumor
suppressor gene. Taken together, NAV3 alterations
were found in four of eight (50%) patients with early
MF (stages IA –IIA) and 11 of 13 (85%) patients with
advanced MF and SS, suggesting that it may be the most
frequent genetic lesion in CTCL. All patients with
260
NAV3 deletion or translocation suffered from frequent
relapses or died from disease.
In addition, the inactivation via deletions or pro-
moter hypermethylation of several tumor suppressor
genes, including SHP-1, p15, p16, and hMLH1 has been
reported. SHP-1 is an important negative regulator
involved in signaling through receptors for cytokines
and growth factors, whose expression is often lost in MF
and SS samples. Inactivation of p15 and p16 is reported
in both early- and advanced-stage MF and SS by allelic
loss or aberrant promoter methylation. Recently, a
genome-wide scale differential methylation hybridiza-
tion analysis revealed a hypermethylated status of sev-
eral additional putative tumor suppressor genes such as
BCL-7a, PTPRG, and thrombospondin. Whether these
events are causative or an epiphenomenon remains to
be elucidated; nevertheless, these findings suggest the
possible experimental use of demethylation agents in
the therapy of these tumors.
Clonal rearrangement of TCR genes can be detected
in most cases of granulomatous slack skin (GSS) dis-
ease, referred to as a subtype of MF. Further molecular
genetic data are not available because of the rarity of the
disease; however, trisomy 8 was reported in two cases.
Characteristic cytogenetic abnormalities in SS
include genomic losses at 1p, 10q, 14q, and 15q that
are obviously associated with clonal evolution of the
tumor cells during disease progression. More recently,
oligonucleotide array analysis provided interesting data
based on the comparison between SS cells and CD4+ T-
cells isolated from the peripheral blood of patients with
atopic or chronic dermatitis and healthy volunteers. The
SS samples displayed a relatively homogeneous gene
expression pattern with consistent upregulation of the
two genes Twist and EphrinA4 (EphA4). The Twist gene
encodes a basic helix–loop–helix family transcriptional
factor that is normally not expressed in lymphoid tissue
and may inhibit apoptosis via antagonizing p53. EphA4
belongs to the Eph receptor subfamily of transmem-
brane protein tyrosine kinases and probably acts
through an activation of the JAK/STAT pathway. A
filter array analysis compared partially purified SS cells
with in vitro Th1- and Th1-skewed peripheral blood
cells and found that the loss of STAT4 expression,
together with the increased expression of RhoB and
other genes, can discriminate SS patients from patients
with inflammatory diseases. The discrepant data from
both studies are probably because of the differences in
study designs, patient sampling, and data analysis.

Treatment of MF and SS
The choice of initial treatment for patients with MF
and SS is dependent on the stage of the disease and the
patient’s performance status. The rationale for therapy
in CTCL has been shaped by a randomized controlled
trial in CTCL which compared palliative skin-directed
therapy (topical mechlorethamine, superficial radio-
therapy, and phototherapy) with combined total skin
electron beam and multiagent chemotherapy in 103
patients. Response rates were higher in the chemo-
therapy group but morbidity was greater and there
was no difference in disease-free survival (DFS) or
overall survival (OS).
Skin-directed therapy includes topical treatments,
phototherapy, and radiotherapy. It is the first line
treatment for most patients with early-stage IA–IIA
disease. Systemic treatments for advanced (IIB–IVB)
or treatment-resistant early-stage disease should be
given under the supervision of a multidisciplinary
team (MDT), preferably as part of a randomized
trial. The therapies for CTCL are listed below and
treatment options in the different stages of disease
are shown in Table 15.4.
Skin-directed therapies
Topical corticosteroids
Emollients and topical corticosteroids are appropriate
for patients with stage IA/B disease to relieve symp-
toms of pruritus. Moderately potent topical cortico-
steroids may be used regularly for patches and thin
plaques. Potent topical corticosteroids can produce a
clinical response, although this is usually short-lived
and prolonged use can cause cutaneous atrophy.
Topical chemotherapy
Topical mechlorethamine (nitrogen mustard), 0.01%
or 0.02% daily, either as an aqueous solution or oint-
ment base, is effective for early-stage disease, with
response rates of 51–80% for IA, 26–68% for IB, and
61% for IIA disease. Nitrogen mustard can cause an
irritant or allergic contact dermatitis (35–58%). It
must not be used in pregnancy and is carcinogenic,
with rare reports of non-melanoma skin cancer.
Topical carmustine (BCNU) is an alternative top-
ical chemotherapeutic agent in CTCL with similar
efficacy to nitrogen mustard. Alternate or daily treat-
ment with 10 mg of BCNU dissolved in 60 mL of 95%
alcohol, or 20–40% BCNU ointment may be used.
Hypersensitivity reactions are less frequent than with
Chapter 15: Primary cutaneous lymphoma
nitrogen mustard, occurring in 5–10% of patients.
Unlike nitrogen mustard all patients should have
regular monitoring of full blood counts, and treat-
ment is normally given for limited periods to avoid
myelosuppression.
Other topical treatments
The rexinoid RXR (retinoid X-receptor) agonist, 1%
targretin (bexarotene) gel, is FDA-approved for top-
ical therapy in stage I MF patients who are resistant or
intolerant of other topical therapies. Response rates of
63% in IA/B disease with complete responses (CR) in
21% and median duration of 2 years have been
reported.
Other topical treatments used in small numbers of
patients with some efficacy include imiquimod, but
formal studies are lacking at present.
Phototherapy and photochemotherapy
PUVA (psoralen with UVA), broad band UVB, nar-
row band UVB, and high-dose UVA-1 phototherapy
have all been used in MF with considerable success,
but there have been no adequate comparative studies
of different phototherapy regimes in CTCL. UVB may
be associated with a lower risk of cutaneous carcino-
genesis than PUVA, but is less effective in patients
with thick plaques. A retrospective study of 56 patients
with early-stage MF (stage 1A and 1B) suggested that
narrow band UVB is at least as effective as PUVA in
terms of both response and relapse-free interval, but
PUVA therapy remains the standard of care for early-
stage disease.
The clinical benefit of PUVA (photochemotherapy)
in early-stage disease was noted three decades ago with
complete response rates (CRR) of 79–88% in stage IA
and 52–59% in stage IB disease reported. Treatment
schedules have varied but normally involve treatment
2–3× per week with oral 8-methoxypsoralen and UVA
in regular dosage increments until maximum tolerated
dose is reached. Treatment is continued until complete
or best partial response has been achieved. Flexural sites
(“sanctuary sites”) often fail to respond completely and
the duration of response varies. UVA will not penetrate
deeply enough to ensure effective treatment of follicu-
lotropic disease or tumors and is rarely tolerated by
erythrodermic patients.
One study has shown that 56% of stage IA and 39%
of stage IB complete PUVA responders had no recur-
rence of disease after 44 months of follow-up without 261
maintenance therapy. However, relapse following
 Chapter 15: Primary cutaneous lymphoma

Table 15.4 Therapeutic options in primary CTCL and primary CBCL.

Stage/disease First line Second line Novel/

entity developing
Therapies
MF IA No therapy or SDT No therapy or SDT
MF IB SDT PUVA plus IFN-2α and/or bexarotene; TSEB Denileukin
diftitox, HDACi
MF IIA SDT PUVA plus IFN-2α and/or bexarotene; TSEB Denileukin
diftitox, HDACi
MF IIB Radiotherapy/TSEB; single agent IFN-2α Denileukin diftitox; bexarotene; HDACi RIC allo-SCT
chemotherapy (liposomal doxorubicin/
gemcitabine)
MF/SS III PUVA ± IFN-2α; ECP ± IFN-2α and/or Denileukin diftitox; TSEB; HDACi; RIC allo-SCT
bexarotene; methotrexate single agent chemotherapy (chlorambucil
or deoxycoformycin); antibody therapy
(alemtuzumab)
MF/SS IVA Radiotherapy/TSEB; single agent IFN-2α; HDACi; Denileukin diftitox, multiagent RIC allo-SCT
chemotherapy (gemcitabine/ liposomal chemotherapy; antibody therapy
doxorubicin) (alemtuzumab)
MF/SS IVB Palliative radiotherapy/ chemotherapy RIC allo-SCT
LYP Expectant approach UVB or PUVA, SGN-35
radiotherapy, methotrexate antibody
pcALCL Surgical excision, radiotherapy Methotrexate; chemotherapy for SGN-35
extracutaneous disease antibody
SPTCL Radiotherapy; corticosteroids Single agent chemotherapy (liposomal Autologous
doxorubicin); multiagent chemotherapy stem cell trans-
(CHOP) plant; RISTs
pc Liposomal doxorubicin; TSEB Multiagent chemotherapy Autologous
epidermotropic stem cell
transplant,
CD8+ CTCL (CHOP) RISTs
pc small/ Primary excision; radiotherapy Chemotherapy (cyclophosphamide); IFN-2α
medium
pleomorphic
CTCL
pcMZL Expectant policy, primary excision, Chlorambucil; IFN-2α (sc or intralesional) Intralesional
radiotherapy rituximab
pcFCL Radiotherapy Liposomal doxorubicin; multiagent Intralesional
chemotherapy with rituximab (CHOP-R) rituximab
pcLBCL Radiotherapy; multiagent
chemotherapy with rituximab (CHOP-R)
ECP, Extra Corporeal Photopheresis; HDACi, Histone Deacetylase Inhibitors; RIST, Reduced Intensity Stem cell Transplant; SDT, Skin Directed
Therapy; TSEB, Total Skin Electron Beam.

PUVA therapy is commonly reported and patients
often require repeated courses over many years. Many
patients will inevitably have a high total cumulative
UVA dose and the risks of non-melanoma and mela-
noma skin cancer are consequently increased for these
262
patients. Maintenance therapy is only rarely effective at
preventing relapse and therefore should, if possible, be
avoided, particularly as this group of patients already
have an increased risk of secondary malignancies.
Effort should be made to restrict the total PUVA
dose to less than 200 treatment sessions or a total
cumulative dose of 1200 J/cm2. However, in

exceptional circumstances patients may receive a
greater total dose if clinically justified and with the
consent of the patient.
Radiotherapy and total skin electron beam therapy
MF and other CTCL variants are extremely radiosen-
sitive. Individual thick plaques, eroded plaques, or
tumors can be treated successfully with localized and
relatively low-dose superficial orthovoltage radiother-
apy (2 or 3 fractions of 400 cGy at 80–120 kV). Large
tumors may require a higher energy. Closely adjacent
and overlapping fields can often be retreated because
of the low doses used.
Electron beam therapy is also used to treat local
disease and the energy used is dependent on the depth
of disease. Whole-body total skin electron beam
(TSEB) therapy has also been extensively used to
treat resistant widespread cutaneous disease.
A meta-analysis of open uncontrolled and mostly
retrospective studies of TSEB as monotherapy in 952
patients with CTCL established that responses are
stage-dependent, with CR of 96% in stage IA–IIA
disease, but relapse rates are high, indicating that this
approach is not curative in early-stage disease. In stage
IIB disease CRs are less common (36%), but erythro-
dermic (stage III) disease shows CR of 60%. Greater
skin surface dose fractionation (32–36 Gy) and higher
energy (4–6 MeV electrons) are associated with a
higher rate of CR, and 5-year relapse-free survivals of
10–23% have been reported. A comparative retrospec-
tive study of TSEB versus topical mechlorethamine in
early stage MF showed similar response rates and
duration of response, suggesting that TSEB therapy
should be reserved for those who fail first and second
line therapies.
Adverse effects of TSEB include temporary alope-
cia, telangiectasia, and skin malignancies. Although
TSEB is usually only given once in a lifetime, several
reports have documented patients who have received
two or three courses, although the total dose tolerated
and duration of response have been lower with sub-
sequent courses.
Systemic therapies
Extracorporeal photopheresis
Extracorporeal photopheresis (ECP) involves admin-
istration of oral psoralen, followed by ex vivo collec-
tion of an enriched buffy coat preparation using a cell
separator. These leukocytes are then exposed to UVA
Chapter 15: Primary cutaneous lymphoma
before being returned to the patient. This regimen is
repeated on two successive days and the 2-day cycle
repeated monthly or fortnightly. In vitro evidence
suggests a proportion of the UVA-exposed leukocytes,
including some tumor lymphocytes, undergo apopto-
sis and that dendritic cells are activated during the ex
vivo circulation with induction of a host antitumor
immune response after the treated cells are returned to
the patient. Recent evidence has also shown activation
of dendritic cells during an expanded period of ex vivo
incubation overnight (transimmunization).
ECP is FDA-approved for the treatment of CTCL,
but there are no randomized studies to clarify
whether ECP has an impact on OS. The original
open study of ECP in 29 patients with erythrodermic
CTCL (stage III/IVA) reported a response rate of
73%, but response rates in patients with earlier stages
were much lower (38%). Subsequently, a median
survival of 62 months was reported in the original
cohort of 29 erythrodermic patients, which compares
favorably with historical controls (30 months),
although a later study of 33 patients with SS treated
with ECP reported a median survival of 39 months. A
systematic review of response rates in erythrodermic
disease (stage III/IVA) with ECP has shown overall
responses of 35–71%, with CRs of 14–26%. ECP has
not been found to be of benefit in stage IB patients
who have molecular evidence of a peripheral blood
clone. There have been claims that the CD8 count is
critical in predicting whether patients will respond to
ECP, although others have provided evidence that the
total baseline Sézary count is the best predictor of
response.
Preliminary non-randomized cohort studies suggest
that the combination of interferon-α and ECP is more
effective than ECP alone. There are also isolated case
reports of combined ECP and interferon-γ which have
induced complete clinical and molecular responses.
Immunotherapy
Immunotherapy may enhance the antitumor host
immune response by promoting the generation of
cytotoxic T-cells and TH1 cytokine responses.
Interferon-α has overall response rates (ORRs) of
45–74% and CRRs of 10–27%.Various dosage sched-
ules have been employed (3MU × 3/week–36 MU/
day), and it appears that response rates are higher for
larger doses (overall responses of 78% compared to
37% for the lower dose schedule), but dose-limiting 263
toxicities include fever, chills, malaise, leukopenia,
 Chapter 15: Primary cutaneous lymphoma
thrombocytopenia, and liver dysfunction. ORRs are
also higher in early (IB–IIA, 88%) compared to late
(III–IV, 63%) stages of disease.
Combined interferon-α and retinoic acid recep-
tor (RAR) retinoids produce similar response
rates to interferon alone and are not recommended.
A randomized controlled study comparing PUVA
and interferon-α with interferon-α and acitretin
in early-stage disease showed CRRS of 70% and
38%, respectively, but there are no data on duration
of response. Uncontrolled studies of combined
PUVA and interferon-α (maximum tolerated dose
12 MU/m2 × 3/week) have shown ORRs of 100%,
with CRRs of 62%. This combination may also be
useful in patients with resistant early-stage disease
such as those with thick plaques and folliculotropic
disease.
Small pilot studies have shown that interferon-γ
and IL-12 can produce clinical responses in CTCL, but
the therapeutic value remains to be established.
Monoclonal antibody therapy
CAMPATH (anti-CD52/alemtuzumab) has shown
high response rates in advanced disease, especially
erythrodermic MF and SS, but is associated with a
high rate of CMV reactivation and prolonged
immunosuppression.
Retinoids/rexinoids
The retinoids acitretin and isotretinoin show similar
but low rates of response in MF/SS. Side effects
include teratogenicity, dryness of the mucous mem-
branes, and hyperlipidemia. Bexarotene (Targretin)
is an FDA-approved rexinoid which selectively binds
and activates the RXR receptor and has been shown
to promote apoptosis and inhibit cell proliferation. In
phase II and III studies, response rates from 20% to
67% have been reported.The most effective tolerated
oral dose is 300 mg/m2/day although responses
improve with higher doses. It is teratogenic and
most patients develop hyperlipidemia, requiring
prior treatment with lipid-lowering agents and cen-
tral hypothyroidism.
Open studies comparing PUVA with combined
PUVA and acitretin have shown a similar CRR (73/
72%), although the cumulative dose to best response
was lower in patients receiving the combination
therapy. Results of randomized studies comparing
264 PUVA with PUVA and bexarotene are awaited
(2011).
Systemic chemotherapy
MF and SS are relatively chemoresistant malignancies.
CRs are seen in about 33% but are frequently short-
lived. Chemotherapy is not suitable or clinically effec-
tive for early-stage disease (IA–IIA). In addition, indi-
vidual tumors (IIB) and effaced lymph nodes (IVA)
will respond to local radiotherapy, but chemotherapy
should be considered for those with good performance
status who have extensive nodal or visceral disease.
CHOP may produce CRs of 38% in stage IIB–IVB
but the duration of response is often very short.
Single agent chemotherapy which has been shown
to produce a clinical response in late-stage disease
includes oral chlorambucil, methotrexate, and etopo-
side, and intravenous 2-deoxycoformycin, liposomal
doxorubicin, gemcitabine, 2-chlorodeoxyadenosine,
and fludarabine. Open studies of 2-deoxycoformycin
in MF and SS have reported response rates of 35–
71%, with CRRs of 10–33%. Methotrexate (single
weekly oral doses of 5–125 mg) has been reported to
produce a CRR of 41% in erythrodermic disease, with
a median survival of 8.4 years. Deoxycoformycin and
methotrexate are now the preferred second line
therapies for stage III erythrodermic disease.
Liposomal doxorubicin has shown response rates of
88% and, although responses are often short-lived, is
now the preferred chemotherapy regime for patients
with extensive cutaneous tumors. Several studies
have also shown similar high response rates for gem-
citabine, which is also now a preferred choice for late-
stage (IIB–IV) disease.
Toxin therapies
Denileukin diftitox, a DAB389-IL-2 fusion toxin
(ONTAK), is FDA-approved for the treatment of
resistant or recurrent CTCL. It is a recombinant fusion
toxin which inhibits protein synthesis in tumor cells
expressing high levels of the IL-2 receptor (CD25+
cells), resulting in apoptosis. Phase III studies of 71
heavily pre-treated patients with stage IB–IVA showed
an ORR of 30%, including 10% with CRs. The median
duration of response was 6.9 (range 2.7–46.1) months.
A subsequent randomized placebo-controlled study of
144 patients confirmed these results, with a signifi-
cantly better and durable ORR of 44%, with 10%
CRR and improved PFS compared to placebo.
Adverse effects include fever, chills, myalgia, nau-
sea and vomiting, and a mild increase in transaminase
levels. A vascular leak syndrome characterized by
hypotension, hypoalbuminemia, and edema occurs

in 25% of patients but can be prevented by concurrent
steroid therapy. Myelosuppression is rare.
Histone deacetylase (HDAC) inhibitors
Both the HDAC inhibitors oral suberoylanilide
hydroxamic acid (SAHA; Vorinostat) and intrave-
nous depsipeptide (Romidepsin) are FDA-approved
for advanced MF/SS. In a phase II open label study of
oral SAHA in 74 patients, ORRs of 30% have been
reported.
A phase II open study of Romidepsin in 96 patients
reported an ORR of 34%, with 6% CRs and a median
duration of response of 15 months. The most common
adverse events were fatigue, diarrhea, nausea, non-
specific and reversible electrocardiogram changes,
and thrombocytopenia.
Bone marrow transplantation
Small cohort studies assessing the use of TSEB and or
total body irradiation combined with high-dose con-
ditioning chemotherapy prior to autologous stem cell
transplantation (ASCT) in patients with stage IIB/
IVA disease have shown good clinical responses, but
high relapse rates, and autologous transplants are not
currently considered for MF/SS patients.
Myeloablative allogeneic stem cell or bone marrow
transplantation has only been used in a few patients,
with encouraging results, but the associated mortality
suggests that this approach is difficult to justify.
However, a graft-versus-lymphoma effect may be
therapeutically important and results of non-
myeloablative allogeneic transplantation for selected
patients with advanced stages of MF/SS have shown
very encouraging results. A review of the European
Group for Blood and Marrow Transplantation
(EBMT) experience of allogeneic SCT in 60 patients
with MF/SS has confirmed that RIC allo-SCT has
lower transplant-related mortality than myeloabla-
tive allo-SCT and that conditioning, including
T-cell depletion, increases the risk of early relapse
significantly. Reported non-relapse mortality is
approximately 25% at year 1, with OS of 50% at
year 3.
Emerging novel therapies
A humanized chimeric anti-CD4 monoclonal antibody
(HuMax-CD4/zanolimumab) has shown response rates
of 56%. Patients with a clinical response develop an
eczema-like reaction and CD4 depletion may be pro-
longed for up to 1 year. A radiolabeled anti-CD5
Chapter 15: Primary cutaneous lymphoma
antibody has also shown objective responses in small
cohorts. A variety of other agents have shown efficacy in
small phase II studies, including pralatrexate (folate
antagonist) and bortezomib (proteasome inhibitor),
but pivotal studies of these agents are awaited.
MF clinicopathologic variants
Pagetoid reticulosis
Pagetoid reticulosis is a rare localized solitary variant
of CTCL, characterized clinically by an isolated, per-
sistent scaly plaque, commonly involving an acral site
and histologically by intense epidermotropism and an
aberrant T-cell phenotype. A more generalized variant
with multiple plaques at other sites has also been
described, but it is likely that this represents the
aggressive CD8+ epidermotropic CTCL variant. The
natural history of pagetoid reticulosis is of very slow
local extension with an excellent prognosis.
Treatment: successful remission and cure has been
reported with both surgical excision and low-dose
superficial radiotherapy.
Folliculotropic and syringotropic MF
Folliculotropic and syringotropic clinicopathologic var-
iants of MF are common. Clinically these variants con-
sist of multiple small (1–3 mm) red macules and papules
clustered around hair follicles and frequently associated
with alopecia. Histologically the atypical lymphocytes
show tropism for the hair follicle or sweat gland epithe-
lium but may demonstrate epidermotropism.
Follicular mucinosis may occur in association with
MF or as a separate entity. Clinically there is follicular
plugging with hair loss, and boggy erythematous pla-
ques may be present. Histologically, mucinous degen-
eration can be seen deposited throughout the follicular
epithelium. If the hair follicles have been destroyed,
scarring alopecia may occur. Studies using multivari-
ate analysis suggest that the prognosis of folliculo-
tropic variants of MF are significantly worse.
Treatment: because of the perifollicular localization
of the dermal infiltrates, folliculotropic MF is often less
responsive to skin-targeted therapies, such as PUVA
and topical nitrogen mustard, than classical plaque-
stage MF, but localized disease may respond well to
radiotherapy. In patients with extensive disease, TSEB
irradiation is an effective treatment, but sustained com-
plete remissions are rarely achieved. PUVA combined 265
with retinoids or interferon-α should also be considered.
 Chapter 15: Primary cutaneous lymphoma
Granulomatous variants of MF
GSS presents with redundant erythematous folds typ-
ically in flexural sites. The histology is granulomatous,
with extensive areas of elastolysis, and the atypical
lymphocytic infiltrate may be sparse and may not
show epidermotropism. The condition must be distin-
guished from other forms of cutis laxa. While GSS is
considered to be a variant of MF, patients rarely show
typical MF patches and plaques. Granulomatous infil-
tration has also been described in plaques of MF with-
out the typical features of slack skin.
Treatment: GSS responds to radiotherapy. Rapid
recurrences after surgical excision have been reported.
CD30+ T-cell lymphoproliferative
disorders of the skin
Primary cutaneous CD30+ lymphoproliferative disor-
ders represent a spectrum of disorders consisting of
lymphomatoid papulosis and primary cutaneous
CD30+ large-cell anaplastic lymphoma. Critically, pri-
mary cutaneous CD30+ lymphoproliferative disorders
are associated with a good prognosis, in contrast to
transformation of mycosis fungoides in which tumor
cells express CD30, and systemic nodal CD30-positive
lymphomas.
Lymphomatoid papulosis
The term lymphomatoid papulosis (LYP) was first
coined in 1968 by Macaulay to describe a “self-healing
rhythmical paradoxical papular eruption, histologically
malignant but clinically benign.” Recurrent crops of
papular/nodular lesions predominantly affect the
limbs, although any body site may be involved and
regional localized variants may occur. These lesions
may grow rapidly over a few days and develop ulcerated
necrotic centers. Healing occurs slowly, with fine atro-
phic or varioliform scars. Recurrent lesions are com-
mon but the prognosis is excellent (Table 15.3). Staging
investigations are not indicated in the absence of
peripheral lymphadenopathy. Patients are very rarely
at risk of developing CD30-positive large-cell anaplastic
T-cell lymphoma and Hodgkin’s disease. Patients with
MF may also develop LYP, which is a favorable prog-
nostic feature in MF. Overall, the incidence of such
lymphomas in LYP is estimated to be <5%.
Pathology
266 Three histological patterns of LYP have been
described. Type A lesions contain scattered large
Figure 15.8 Lymphomatoid papulosis.
cells that resemble Reed–Sternberg cells within a
dense infiltrate that includes neutrophils, eosino-
phils, histiocytes, and a few plasma cells
(Figure 15.8). Type B lesions are more reminiscent of
MF, with a band-like infiltrate of cerebriform-type
cells that show epidermotropism. Type C lesions con-
tain prominent clusters/cohesive sheets of atypical
large anaplastic cells.
Immunophenotypically the cells are mature
T-cells with expression of CD3. In most cases there
is expression of CD2 and/or CD5, but both may be
lost. Staining for CD7 is often absent. The cells are
usually CD4-positive with lack of CD8 and there is
expression of CD25 and HLA-DR. In type A and C
lesions the large cells are positive for CD30 and
negative for CD15. The small cerebriform cells of
type B lesions are CD30-negative. There is expres-
sion of TIA-1 and granzyme in the majority of cases.
Rare cases of LYP are reported in which the pheno-
type of the atypical cells is CD8+, and this has been
classified as type D.
Treatment: There is no current treatment that
alters the clinical course of LYP. Lesions are
extremely radiosensitive but, as they resolve sponta-
neously, this is only necessary for larger necrotic
lesions. Topical or intralesional steroids and topical
nitrogen mustard applied to developing lesions
may accelerate clearance, but have little effect on
well-developed lesions. Narrow band UVB therapy
or PUVA may be preventative whilst patients are on
treatment. Low-dose oral methotrexate is highly
effective and is the treatment of choice. There is
no evidence that intensive combination chemother-
apy is effective for LYP, and chemotherapy is
contraindicated. Therapeutic options are shown in
Table 15.4.

Primary cutaneous (anaplastic) CD30-
positive large-cell lymphoma
Primary cutaneous CD30-positive anaplastic large-cell
lymphoma (pcALCL) is usually seen in adults and
presents as large solitary nodules or plaques which
may ulcerate. Multiple lesions may occur but, crucially,
there are no associated typical atrophic patches or pla-
ques characteristic of MF. Individual lesions may par-
tially regress spontaneously, and disease may remain
localized to a limb. Progression to extracutaneous sites
is rare but has been reported in 10%. The prognosis for
patients with primary cutaneous disease is excellent
(5-year survival rates of 90–95%; Table 15.3).
However, staging consisting of bone marrow and CT
scans is required to exclude secondary cutaneous
involvement in primary nodal CD30+ anaplastic lym-
phoma kinase(ALK)+ lymphomas (Figure 15.9).
Pathology
Morphologically these lesions are characterized by a
dense dermal infiltrate of large cells with immunoblas-
tic, pleomorphic, or anaplastic morphology.
Epidermotropism may be seen. Large bizarre cells
with irregular nuclei and multiple nucleoli and abun-
dant pale cytoplasm are present, and some of these
Chapter 15: Primary cutaneous lymphoma
Figure 15.9 Primary cutaneous CD30-
positive ALCL.
may have multiple nucleoli. There may be an associ-
ated infiltrate of small lymphocytes but eosinophils
and plasma cells are not prominent.
Immunophenotypically CD30 is present in over
75% of large cells. These cells also express CD2, CD3,
CD5, and CD25, although variable loss of CD2, CD3,
and CD5 is common. The cells are positive for cyto-
toxic granules (TIA-1 and granzyme) and MUM-1 but
do not express epithelial membrane antigen (EMA)
and are negative for ALK.
Treatment: both excision and localized radiother-
apy are appropriate for isolated lesions. Nodal involve-
ment may also respond to radiotherapy. Oral
methotrexate is also effective. Multiagent chemother-
apy, including CHOP, can be effective, but is not rec-
ommended unless extracutaneous sites are involved. A
phase II study of anti-CD30 antibodies in pcCD30+
CTCL variants has shown responses in 16 of 23 patients,
with 10 patients achieving a complete remission, includ-
ing patients with pcALCL. Therapeutic options are
shown in Table 15.4.
Molecular pathology of CD30+ T-cell
lymphoproliferative disorders of the skin (CD30+ LPD)
TCR genes are clonally rearranged in the vast majority of 267
cases. Nucleophosmin (NPM)–ALK fusion transcripts
 Chapter 15: Primary cutaneous lymphoma
generated from the translocation t(2;5)(p23;q35) that are
characteristic of systemic nodal ALCL are invariably not
detected in cutaneous ALCL and LYP samples. Rare
cases reported may represent a cutaneous manifestation
of primary systemic disease. Comparative genomic
hybridization (CGH) studies revealed genomic gains at
chromosomes 1/1p, 5, 6, 7, 8/8p, and 19. Recently,
microarray-based CGH analysis was performed in a
series of cutaneous ALCL that demonstrated copy num-
ber gains of FGFR1 (8p11), NRAS (1p13.2), MYCN
(2p24.1), RAF1 (3p35), CTSB (8p22), FES (15q26.1),
and CBFA2 (21q22.3) oncogenes. In the same series,
amplifications of CTSB, RAF1, REL (2p13), JUNB
(19p13.2), MYCN, and YES1 (18p11.3) were found.
Subcutaneous panniculitis-like T-cell
lymphoma
Subcutaneous panniculitis-like T-cell lymphoma
(SPTCL) is a lymphoproliferative disease originating
and presenting in the subcutaneous tissue and predom-
inantly affecting adults. It is characterized clinically by
solitary or multiple subcutaneous nodules and plaques,
which mainly involve the legs. SPTCL is an indolent
lymphoma derived from alpha/beta T-cells that are
CD8-positive and show histologic involvement con-
fined to the subcutis. This CTCL variant should be
distinguished from a primary cutaneous gamma/delta
T-lymphoma which has a poor prognosis and is usually
CD56-positive with histologic evidence of both epider-
motropic and subcutaneous involvement.
A hemophagocytic syndrome is a recognized com-
plication of SPTCL and cutaneous γδ T-cell lympho-
mas in which patients present with fever,
pancytopenia, and hepatosplenomegaly, and this may
herald a fulminant clinical course. Dissemination to
lymph nodes and other organs is uncommon.
Pathology
The infiltrate is characteristically subcutaneous, with
involvement of the lobules of the subcutaneous fat
resembling lobular panniculitis. There is prominent
apoptosis and fat necrosis is frequently present.
Characteristically, there is rimming of the lobular adi-
pocytes by atypical lymphoid cells. The neoplastic cell
morphology is variable, ranging from small cells with
round nuclei and inconspicuous nucleoli to larger cells
with dense chromatin. Plasma cells and reactive lym-
268 phoid follicles are usually absent (Figure 15.10).
Figure 15.10 Subcutaneous panniculitis-like T-cell lymphoma
(skin). The subcutaneous fat is infiltrated by atypical lymphoid cells
that surround adipocytes. Apoptotic debris is prominent.
Immunophenotypically the cells express a mature
T-cell phenotype and express CD2, CD3, and CD5. The
cells are characteristically positive for CD8, lack CD4,
and express cytotoxic granule-related proteins TIA-1,
granzyme, and perforin. CD30 is usually negative. The
cells express TCRαβ and are EBV-negative. In contrast,
primary cutaneous γδ T-cell lymphomas are usually
negative for CD4 and CD8 and may express CD56.
Treatment: recent studies suggest that most
patients have an indolent disease that can be success-
fully treated with systemic corticosteroids and/or local
radiation therapy. However, in patients with extensive
refractory disease, doxorubicin-based chemotherapy
may be appropriate. Specifically in those with primary
cutaneous γδ T-cell lymphoma, preconditioning che-
motherapy followed by an autologous/allogeneic bone
marrow transplant may be the only therapeutic option
for improving survival. Therapeutic options are
shown in Table 15.4.
Primary cutaneous peripheral T-cell
lymphoma (PTL), unspecified
This category encompasses all primary CTCLs that are
not currently well defined on the basis of clinicopatho-
logic criteria outlined for other CTCL variants above.
It includes primary cutaneous aggressive epidermo-
tropic CD8+ T-cell lymphoma, cutaneous gamma/
delta-positive T-cell lymphoma, which has been dis-
cussed above in the context of SPTCL, and primary
cutaneous CD4+ small-/medium-sized pleomorphic
T-cell lymphoma.

Primary cutaneous aggressive
epidermotropic CD8+ T-cell lymphoma
Primary cutaneous aggressive CD8+ CTCL typically
presents with widespread papules, plaques, nodules,
and/or tumors that may show erosions, ulceration,
and necrosis. Mucosal involvement may occur. There
are no polymorphic scaly or atrophic patches and/or
plaques characteristic of MF. These lymphomas express
CD8 and cytotoxic proteins with a striking epidermo-
tropic infiltrate. Although there are currently few
reports of this entity, the distinctive pathological and
immunophenotypic features and poor prognosis (5-
year survival 18%; Table 15.3) suggest that it represents
a distinct subtype of CTCL.CNS involvement can occur
and nodal dissemination is uncommon.
Treatment: there is a poor response to radiother-
apy and chemotherapy. Patients are generally treated
with doxorubicin-based multiagent chemotherapy
and younger patients may be eligible for autologous/
allogeneic bone marrow transplant. Therapeutic
options are shown in Table 15.4.
Primary cutaneous CD4+ small-/medium-
sized pleomorphic T-cell lymphoma
This CTCL variant presents with non-specific eryth-
ematous nodules on the face, the neck, or the upper
trunk. Multiple lesions may occur but with no clinical
evidence of MF. Histologically there are nodular and
diffuse infiltrates of small- to medium-sized pleomor-
phic T-cells that may infiltrate into the subcutis.
Epidermotropism may be present. Recent reports sug-
gest that a CD8+ variant exists which often has a
similar clinical presentation. In most cases there is a
favorable clinical course and systemic involvement is
unusual (5-year survival 75%; Table 15.3).
Treatment: in view of the excellent prognosis,
patients with solitary localized skin lesions may be
treated with surgical excision or radiotherapy.
Cyclophosphamide as single agent therapy and
interferon-α are effective in patients with more gener-
alized skin disease.Therapeutic options are shown in
Table 15.4.
Primary cutaneous B-cell lymphomas
Primary CBCLs constitute approximately one-quarter
of all primary cutaneous lymphomas. Full staging
investigations are essential to exclude a primary
nodal lymphoma with secondary cutaneous
Chapter 15: Primary cutaneous lymphoma
involvement. A staging system for primary cutaneous
lymphomas other than MF/SS has been proposed as
the current American Joint Committee on Cancer
(AJCC) staging systems for nodal lymphomas are
inappropriate for primary cutaneous lymphomas.
The prognosis of the various diagnostic entities is
shown in Table 15.3.
Molecular analysis of primary cutaneous B-cell
lymphoma demonstrates clonally rearranged immu-
noglobulin receptor genes in the majority of cases. Few
data are available concerning the immunoglobulin
receptor configuration; however, preferential usage
of the IgVH family member 2–70, ongoing mutations,
and low rates of replacement mutations in framework
regions have been reported. These findings suggest a
negative antigen selection pressure and indicate that
local antigens could modulate the growth of primary
cutaneous B-cell lymphoma.
Primary cutaneous marginal zone B-cell
lymphoma/immunocytoma
Primary cutaneous marginal zone B-cell lymphomas
(pcMZLs) are indolent lymphomas that present as
asymptomatic solitary or multiple dermal papules,
plaques, or nodules on any body site, although the
trunk is most often involved. Benign monoclonal
paraproteinemia may be present. There is a slight
male predominance. The estimated 5-year survival is
excellent at 98–100%; Table 15.3.
Pathology
pcMZLs are derived from post-germinal-center cells
and are characterized by a proliferation of small lym-
phocytes, lymphoplasmacytoid cells and plasma cells
with monotypic cytoplasmic immunoglobulin.
Molecular pathology
The translocation t(11;18)(q21;q21), representing
the most common translocation in extranodal mar-
ginal zone B-cell lymphoma, has not been demon-
strated in pcMZL. The translocation t(14;18)(q32;
q21), involving the IgH locus and the MALT1 gene,
is reported frequently and, according to some
studies, can be demonstrated in one-third of cases.
The recently described t(3;14)(p13;q32) that results
in the deregulation of the FOXP1 gene has also
been detected in pcMZL; however, its overall fre-
quency is not established. Similar to the API2–
MALT1 fusion, aberrations involving the BCL-10 269
gene have not been identified in pcMZL. Recent
 Chapter 15: Primary cutaneous lymphoma
studies suggest that rare cases of primary cutaneous
B-cell lymphoma may be associated with Borrelia
burgdorferi infection.
Treatment: patients with solitary or few cutaneous
lesions can be treated with radiotherapy or surgical
excision or an expectant approach may be followed. In
rare patients with associated B. burgdorferi infection,
systemic antibiotics should be considered. For patients
presenting with multifocal skin lesions, radiotherapy
can still be used if the extent of disease is limited, but
chlorambucil or intralesional/subcutaneous adminis-
tration of interferon-α may produce CRs in approx-
imately 50% of patients. Excellent results have also
been obtained with systemic and intralesional use of
anti-CD20 antibody (rituximab). Patients with secon-
dary nodal involvement may require more chemother-
apy treatment. Therapeutic options are shown in
Table 15.4.
Primary follicle center-cell lymphoma
Primary cutaneous follicle center-cell lymphoma
(pcFCL), previously known as Crosti’s lymphoma, is
an indolent low-grade B-cell lymphoma. Patients
present with clinically solitary or grouped papules,
plaques, or nodules, most commonly on the head,
neck, or trunk, although any body site may be involved
(Figure 15.11). Lesions may extend slowly without
treatment over a period of years. The estimated
5-year survival of pcFCL is 94–97%; Table 15.3.
In contrast to systemic follicular lymphoma, the
t(14;18) is not detected in primary follicle center-cell
lymphoma.Those tumors with predominant diffuse
populations of centrocytes must be distinguished
from large B-cell lymphoma.
270
Figure 15.11 Primary cutaneous follicle center-like lymphoma.
Pathology
The growth pattern of these lesions may be follicular,
follicular and diffuse, or purely diffuse. The infiltrate
is within the dermis and may extend into the
subcutaneous tissue. In most cases there is a Grenz
zone between the infiltrate and the overlying epider-
mis. There is a variable mixture of centrocyte and
centroblast cells which, in the follicular type, gives a
monotonous appearance to the follicle centers
without the characteristic zoning of reactive germi-
nal centers. Macrophages with tangible bodies are
scanty. In contrast to nodal follicular lymphoma,
grading of these lymphomas is not performed
as this has not been shown to be prognostically
useful.
Immunophenotypically the cells are positive for
CD20 and CD79a and lack CD5. Expression of CD23
may be seen in some cases. Staining for CD10 is
variable but most cases are positive for BCL-6. The
cells lack BCL-2 protein expression in most cases.
There is no expression for MUM1/IRF4. Where a
follicular growth pattern is present, underlying follic-
ular dendritic cell meshworks can be highlighted by
CD21, CD23, or CD35.
Molecular cytogenetics
In contrast to nodal follicular lymphoma, pcFCL gen-
erally lack the translocation t(14;18)(q32;q21) that
results in the deregulated expression of BCL-2, but
small proportions of pcFCL do show a t(14;18)
although this does not appear to influence outcome.
In pcFCL, inactivation of p15INK4b and p16INK4a
tumor suppressor genes is frequently reported, most
often as a result of promoter hypermethylation. High-
level amplifications of c-REL at 2p16 and deletions at
chromosomal band 14q32 are reported. pcFCL are
characterized by a germinal-center B-cell (GCB)-like
expression pattern and overexpression of the SPINK2
gene.
Treatment: in patients with localized or few scat-
tered skin lesions, radiotherapy is the preferred mode
of treatment. Cutaneous relapses can be treated with
further radiotherapy. Anthracycline-based chemo-
therapy (such as doxorubicin) is required only in
patients with very extensive cutaneous disease and
patients developing extracutaneous disease. Systemic
or intralesional anti-CD20 antibody (rituximab)
therapy has been effective in small numbers of
patients. Therapeutic options are shown in
Table 15.4.

Primary cutaneous large B-cell lymphoma
(leg type and cutaneous large B-cell
lymphoma at other sites)
Primary cutaneous large B-cell lymphomas
(pcLBCL) typically present as large dermal nodules
or tumors that are either solitary or multifocal and
characterized by a diffuse proliferation of large
B-cells consisting of centroblasts and immunoblasts.
These lymphomas tend to develop on the lower limbs
(Figure 15.12). This has led to the classification of
these lymphomas as leg type although pcLBCL can
also occur less commonly at other sites. This condi-
tion affects an elderly population with a female pre-
dominance. The prognosis of pcLBCL is poor, with a
5-year survival of 52–58% (Table 15.3).
Figure 15.12 Large B-cell lymphoma involving the leg.
Chapter 15: Primary cutaneous lymphoma
Pathology
There is a diffuse dermal infiltrate of large lymphoid
cells with destruction of the adnexal structures.
Extension into the subcutaneous tissue may be
present. There is usually a subepidermal Grenz zone.
The cells resemble centroblasts with vesicular nuclei
that contain multiple nucleoli that are often attached
to the nuclear membrane. Some cases have significant
numbers of immunoblast-like cells.
Immunophenotypically the cells express CD20 and
CD79a and lack CD5 and CD10. Expression of BCL-6
is variable but is usually positive. There is strong
expression of BCL-2 and MUM1/IRF4 but CD138 is
not expressed.
Molecular pathology
Cases of pcLBCL lack translocations commonly
encountered in nodal DLBCL. Recent array-based
CGH studies demonstrated different genomic imbal-
ances in pcFCL and pcLBCL-leg type. The most fre-
quent alterations in pcLBCL are amplifications at
18q21, including both BCL-2 and MALT1 genes,
and deletions of a small region at 9p21 containing
the CDKN2A, CDKN2B, and NSG-x genes. Gene
expression profiling of cutaneous B-cell lymphomas
demonstrated increased expression of several genes
associated with proliferation (Pim1, Pim2, MYC,
Mum1/IRF4, and Oct2) in pcLBCL leg-type and an
activated B-cell (ABC)-like signature.
Treatment: in elderly patients with solitary
tumors radiotherapy may be appropriate, but multi-
agent chemotherapy is usually required for multi-
focal disease. The role of rituximab in primary
cutaneous disease has yet to be determined but
this has proved effective in relapsed nodal
DLBCL in combination with CHOP chemotherapy.
Intralesional rituximab may be effective. Therapeutic
options are shown in Table 15.4.
Further reading
Agar N, Wedgeworth E, Crichton S, et al. Survival
outcomes and prognostic factors in mycosis
fungoides and Sézary syndrome: validation of the
revised International Society for Cutaneous
Lymphomas and the European Organisation for
Treatment and Research staging proposal. J Clin Oncol,
2010;28:4730–4739.
Bekkenk MW, Geelen FAMJ, van Voorst Vader PC, et al.
Primary and secondary cutaneous CD30 positive
lymphoproliferative disorders. Blood, 271
2000:95;3653–3661.
 Chapter 15: Primary cutaneous lymphoma
Duarte R, Canals C, Onida F, et al. Allogeneic hematopoietic
cell transplantation for patients with mycosis fungoides
and Sézary syndrome: a retrospective analysis of the
Lymphoma Working Party of the European group for
Blood and Bone Marrow transplantation. J Clin Oncol,
2010;28:4492–4499.
Duvic M, Hymes K, Heald P, et al. Bexarotene is effective and
safe for treatment of refractory advanced-stage cutaneous
T-cell lymphoma: multinational phase II–III trial results.
J Clin Oncol, 2001;19:2456–2471.
Jones GW, Kacinski BM, Wilson LD, et al. Total skin electron
radiation in the management of mycosis fungoides:
consensus of the European Organization for Research and
Treatment of Cancer (EORTC) Cutaneous Lymphoma
Project Group. J Am Acad Dermatol, 2002;47(3):364–370.
Kaye FJ, Bunn PA Jr, Steinberg SM, et al. A randomized trial
comparing combination electron beam radiation and
chemotherapy with topical therapy in the initial
treatment of mycosis fungoides. N Engl J Med,
1989;321:1784–1790.
Kim Y, Willemze R, Pimpinelli N, et al. TNM
classification system for primary cutaneous lymphomas
other than mycosis fungoides and Sézary syndrome: a
proposal of the International Society for Cutaneous
Lymphoma (ISCL) and the Cutaneous Lymphoma Task
Force of the European Organisation for Research and
Treatment of Cancer (EORTC). Blood,
2007;110:479–484.
Lundin J, Hagberg H, Repp R, et al. Phase 2 study of
alemtuzumab (anti-CD52 monoclonal antibody) in
patients with advanced mycosis fungoides/Sézary
syndrome. Blood, 2003;101:4267–4272.
Olsen E, Kim Y, Kuzel T, et al. Phase IIB multicenter trial of
vorinostat in patients with persistent, progressive or
treatment refractory cutaneous T-cell lymphoma. J Clin
Oncol, 2007;25:1–9.
Olsen E, Vonderheid E, Pimpinelli N, et al. Revisions to the
staging and classification of mycosis fungoides and
Sézary syndrome: a proposal of the International Society
for Cutaneous Lymphomas (ISCL) and the cutaneous
lymphoma task force of the European Organisation of
Research and Treatment of Cancer (EORTC). Blood,
2007;110:1713–1722.
Pimipinelli N, Olsen E, Santucci M, et al. Defining early
mycosis fungoides. J Am Acad Dermatol,
2005;53:1053–1063.
Prince M, Whittaker S, Hoppe R. How we treat
mycosis fungoides and Sézary syndrome. Blood,
2009;114:4337–4353.
Prince M, Duvic M, Martin A, et al. Phase III placebo
controlled trial of denileukin diftitox for patients with
272
cutaneous T-cell lymphoma. J Clin Oncol,
2010;28:1870–1877.
Quereux G, Marques S, Nguyen J-M, et al. Prospective
multicenter study of pegylated liposomal doxorubicin
treatment in patients with advanced or refractory mycosis
fungoides or Sézary syndrome. Arch Dermatol,
2008;144:727–733.
Senff N, Hoefnagel J, Jansen P, et al. Reclassification of 300
primary cutaneous B-cell lymphomas according to the
new WHO-EORTC classification for cutaneous
lymphomas: comparison with previous classifications
and identification of prognostic markers. J Clin Oncol,
2007;25:1581–1587.
Senff N, Hoefnagel J, Nelis K, et al. Results of radiotherapy in
153 primary cutaneous B-cell lymphomas classified
according to the WHO-EORTC classification. Arch
Dermatol, 2007;143:1520–1526.
Senff N, Noordijk E, Kim Y, et al. European Organization
for Research and Treatment of Cancer and
International Society for Cutaneous Lymphoma
consensus recommendations for the management of
cutaneous B-cell lymphomas. Blood,
2008;112:1600–1609.
Trautinger F, Knobler R, Willemze R, et al. EORTC
consensus recommendations for the treatment of
mycosis fungoides/Sézary syndrome. Eur J Cancer,
2006;42:1014–1030.
Vonderheid EC, Bernengo MG, Burg G, et al. Update on
erythrodermic cutaneous T-cell lymphoma: report of the
International Society for Cutaneous Lymphomas. J Am
Acad Dermatol, 2002;46(1):95–106.
Whittaker S, Demierre MF, Kim E, et al. Final results from a
multicentre international pivotal study of romidepsin in
refractory cutaneous T-cell lymphoma. J Clin Oncol,
2010;28:4485–4491.
Willemze R, Jaffe ES, Burg G. WHO-EORTC
classification for cutaneous lymphomas. Blood,
2005;105(10):3768–3785.
Willemze R, Jansen P, Cerroni L, et al. Subcutaneous
panniculitis-like T-cell lymphoma: definition,
classification and prognostic factors: an EORTC
cutaneous lymphoma group study of 83 cases. Blood,
2008;111:838–845.
Yu J, McNiff J, Lund M, Wilson L. Treatment of primary
cutaneous CD30+ anaplastic large cell lymphoma with
radiation therapy. Int J Radiat Oncol Biol Phys,
2008;70:1542–1545.
Zinzani P, Baliva G, Magagnoli M, et al.
Gemcitabine treatment in pretreated cutaneous
T-cell lymphoma: experience in 44 patients. J Clin
Oncol, 2000;18:2603–2606.
 Chapter
16
Lymphoma in the immunosuppressed
Michele Spina, Robert Marcus, Shireen Kassam, and Umberto Tirelli
Pathology: Andrew Wotherspoon Molecular cytogenetics: Andreas Rosenwald and German Ott
Immunodeficient patients are at higher risk for devel-
oping lymphoproliferative disorders (LPDs), above all
non-Hodgkin lymphomas (NHLs). Even though the
association between primary immunodeficiency dis-
eases (i.e. X-linked lymphoproliferative syndrome,
common variable immunodeficiency, ataxia telangiec-
tasia, and Wiskott–Aldrich syndrome) and LPDs, on
the one hand, and between LPDs and autoimmune
diseases, on the other, is well known, the leading causes
of immunosuppression are, at present, organ trans-
plantation and human immunodeficiency virus (HIV)
infection. Lymphomas in the immunocompromised
host have been traditionally regarded as having a
more aggressive course, including extranodal involve-
ment, a more rapid clinical course, poorer response to
conventional therapies, and poorer outcome, but this
has been called into question both in post-transplant
lymphoproliferative disorders (PTLD) following solid
organ transplant (SOT) and HIV-related lymphoma
by more recent publications.
Tumors following solid organ
transplantation (PTLD)
PTLDs are well recognized and potentially life-
threatening complications after SOT. PTLD is seen
in up to 10% of all SOT recipients. It is the most
common form of post-transplant malignancy in chil-
dren and in adults it is the second most common
malignancy after skin cancer. In both children and
adults it is the most common cause of cancer-related
mortality after SOT, with the reported overall mortal-
ity for PTLD exceeding 50%. Up to 85% of PTLDs
are of B-cell lineage and most of these (over 80%)
are associated with Epstein–Barr virus (EBV)
infection. Around 10–15% of PTLDs are of T-cell
lineage, around 30% of which are associated with EBV.
Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
According to the WHO classification, B-cell
PLTDs can be divided into three main groups:
1. Diffuse B-cell hyperplasia, characterized by
differentiated plasma cells and preservation of the
normal lymphoid architecture.
2. Polymorphic PTLDs characterized by nuclear
atypia, tumor necrosis, and destruction of the
underlying lymphoid architecture.
3. Monomorphic PTLDs, which includes high-grade
invasive lymphoma of B- or T-lymphocyte
centroblasts.
Monomorphic B-cell PTLDs can be further divided
into diffuse large cell lymphomas and Burkitt or
Burkitt-like lymphomas that display their character-
istic chromosomal translocations. Monomorphic
T-cell PTLDs can be further divided into large cell,
anaplastic, or unspecified.
Hodgkin lymphoma (HL) is rare but well described
in this setting. EBV is closely involved in the patho-
genesis and the majority of cases arise in response to
primary infection with EBV or to reactivation of pre-
viously acquired EBV.
Other low-grade B-cell lymphomas of B-cell ori-
gin; marginal zone lymphoma and lymphoplasma-
cytoid lymphoma also have an increased incidence
after SOT and are associated with prior exposure to
Helicobacter pylori and hepatitis C, respectively.
The incidence of PTLD after SOT varies between
children and adults and also according to the organ
transplanted. In adult recipients, PTLD occurs in
1–2.3% of kidney transplants, 1–2.8% of liver trans-
plants, 1–6.3% of heart transplants, 2.4–5.8% of heart–
lung transplants, 4.2–10% of lung transplants, and up
to 20% of small bowel transplants. The incidence of
PTLD is significantly higher in pediatric recipients,
where it has been reported in 1.2–10.1% of kidney
273
 Chapter 16: Lymphoma in the immunosuppressed
transplants, 4–15% of liver transplants, and 6.4–19.5%
of lung, heart, and heart–lung transplants.
Although PTLD may occur at any time after trans-
plantation, the risk of developing PTLD is greatest
within the first year and declines over time. A report
by the Transplant Collaborative Study showed the
incidence of PTLD to be 224/100 000 in the first
year, 54/100 000 in the second year, and 31/100 000
in the sixth year following transplantation. A more
recent study from the French Transplant Registry sug-
gests an overall incidence in renal transplant recipients
of between 0.2% and 0.4%, with a recent decline in
incidence perhaps related to lower intensity immuno-
suppression (IS) being used in the post-transplant
setting and a confirmation of an increased association
with both antilymphocyte globulin (ATG) usage and
prior EBV exposure. A biphasic distribution was seen
in renal transplant recipients, with the incidence rising
in the first year, declining in years 2–4, and rising again
thereafter. These findings were also observed in a large
US registry series.
PTLD arises, in large part, as a consequence of the
potent immunosuppressive agents necessary to pre-
vent allograft rejection. The risk following administra-
tion of specific immunosuppressive agents is still a
matter of controversy. The incidence of lymphoma
was found to increase after the introduction of cyclo-
sporine but may be associated with other cofactors. In
other large series, treatment with cyclosporine did not
enhance the risk for lymphoma compared to azathio-
prine. However, a number of small retrospective stud-
ies have suggested that the routine introduction of
cyclosporine as an immunosuppressive agent in the
1980s was associated with an increased incidence of
PTLD. Experience with tacrolimus is more limited,
although an increased incidence of lymphomas has
been reported both in adults and pediatric patients. It
has been speculated that the overall intensity of immu-
nosuppression, rather than an individual agent, plays a
key role in the development of lymphoma, which is
supported by the observation of a higher incidence of
lymphoma in non-renal transplant recipients receiv-
ing higher doses of immunosuppression. Patients
receiving monoclonal antibodies, such as OKT3 or
ATG, demonstrate a significantly higher incidence of
PTLD. There is also some evidence that the high early
incidence of aggressive PTLD in heart–lung recipients
relates to the large volume of potentially EBV-infected
lymphoid tissue. EBV plays a major role in the patho-
274 genesis of PTLD in the pediatric context and in early
polymorphic PTLD seen in adults, with a close associ-
ation between rising EBV copy number and the devel-
opment of PTLD. There is also some preliminary
evidence of an association of specific HLA subtypes
in the recipient that may be protective against PTLD
development.
Patients who have compromised immune systems
prior to transplantation, such as the elderly or those
with pre-existing CMV infection or hepatitis C, may be
particularly vulnerable to over-immunosuppression
and hence PTLD. Recently, prognostic indices have
been developed that differ subtly but definitely from
those devised for diffuse large B-cell lymphoma
(DLBCL) in the non-immunosuppressed context. The
simplest of these was devised by Choquet and col-
leagues, and consists of Eastern Cooperative Oncology
Group (ECOG) performance status 0–1 versus 2–4,
elevated lactate dehydrogenase (LDH), and age >60 or
<60 years. Patients with no risk factors had a 2-year
survival of 88%, those with a single risk factor 50%, and
in those with two or three risk factors there were no
long-term survivors.
A later refinement of this approach in a larger
series recently published by Caillard and colleagues
confirms that age, LDH, and stage of disease are
major prognostic factors, together with extranodal
sites, including CNS, also having a major impact on
survival, with an overall survival (OS) in this 500-
patient series of just 50% at 5 years. Superior results
have been obtained since the introduction of
rituximab.
Diagnosis can only be made on biopsy of affected
tissues, and suspicion may also be raised by the classi-
cal systemic symptoms of lymphoma: fever, sweats,
weight loss, or pruritus, or, specifically in the PTLD
context, a rise in EBV copy number or deterioration of
organ function. Staging should include CT and PET
where available, bone marrow biopsy and MRI of CNS
where indicated. Assessment of comorbidities (specif-
ically cardiac, lung, and renal function) will need to be
assessed prior to commencement of any therapeutic
strategy.
Surgical resection and/or local radiotherapy
should be considered in the rare patient with localized
disease.
The initial treatment in all patients with PTLD is
to reduce immunosuppression in the hope that rising
T-cell numbers will induce antitumor activity. The
approach to reducing immunosuppressive therapy
needs to be carefully individualized and will depend

on the nature and extent of disease and the type of
transplant recipient, i.e. whether they have a life-
supporting graft (e.g. heart or heart–lung, small
bowel, or liver) or a non-life-supporting graft (e.g.
kidney). Cyclosporine or tacrolimus dosage are typi-
cally reduced over 4–6 weeks to give whole blood
trough levels of around 25–50% that of normal ther-
apeutic levels, depending on the extent of disease and
type of graft, and transplanted organ function carefully
monitored. Myelosuppressive drugs such as azathio-
prine or mycophenolate mofetil should be reduced or
stopped if at all possible. A response to reduction in
immunosuppression is usually seen within 2–4 weeks,
with long-term remissions in around 50% of patients.
The traditional approach has been to regard poly-
clonal PTLD developing during the first year after
transplantation (early disease) as likely to be respon-
sive to reduced immunosuppression, whereas mono-
clonal tumors are generally regarded as being
unresponsive to a reduction in immunosuppression
(RIS) and require more intensive approaches.
The first systematic study to test this was published
by Tsai and colleagues. In their series, patients treated
with RIS, patients with organ dysfunction, a raised
LDH, and multisite disease fared poorly, whereas
those patients with none of these factors had a 90%
response rate. RIS then is a prerequisite for subsequent
therapy but may be insufficient in itself.
Another approach has been to treat PTLD identi-
cally to NHL arising in the non-transplant context. An
early series of patients with B-cell PTLD treated with
the conventional CHOP (cyclophosphamide, doxoru-
bicin, vincristine, prednisolone) regimen showed a
5-year progression-free survival (PFS) of 43%, with
the majority of failures due to recurrence.
In the rituximab era a number of approaches have
been used to assess the value of rituximab alone or
with chemotherapy. Such results are, of course, only
applicable to the majority of patients with B-cell
PTLD. The recommended therapies for T-cell and
Hodgkin PTLD are discussed below.
The first of these was published by Choquet and
colleagues: 46 patients were treated with rituximab
alone, with a response rate of 44% and a 5-year survival
of 67%. Once again, raised LDH, age, and multisite
disease were associated with a poor outcome.
In a small series of 18 PTLD patients, Taylor et al.
demonstrated a high response with R-CHOP, with a
62% PFS at 5 years. Such intensive regimens are not
without toxicity and may not be necessary in a
Chapter 16: Lymphoma in the immunosuppressed
proportion of patients; they may pose particular prob-
lems for patients with compromised cardiac function.
These observations led to a prospective European
study where all patients with PTLD were treated first
with rituximab as a single agent; those responding
underwent short-course rituximab maintenance ther-
apy, while those that did not received four cycles of
R-CHOP chemotherapy. A total of 91 patients have
now been reported; the trial included patients with
both early and late PTLD, with the majority having
undergone renal or liver transplantation. Interestingly,
and contrary to early reports, almost 50% of the
so-called “late” patients had evidence of persistent EBV
positivity. The overall response rate (ORR) to single
agent rituximab was 54%, with 32% of patients entering
complete remission (CR). Those patients not entering
CR then underwent four cycles of R-CHOP chemother-
apy. The ORR was >90%, with similar numbers disease-
free at 1 year. Long-term follow-up data are awaited.
A retrospective study from Evens and colleagues
also provides useful guidance. Eighty patients with
PTLD (with a median time from transplant to PTLD
of 48 months) were reported. All received RIS as initial
strategy; 59 then received rituximab ± chemotherapy,
with a 3-year PFS of 57% and OS of 73%. Once again,
22/30 of the “early” patients were EBV-positive and 17/
50 of the “late” patients were EBV-positive in this series.
Current UK and National Comprehensive Cancer
Network (NCCN) guidelines follow these results: after
staging, all patients with multifocal disease are recom-
mended to receive RIS as initial therapy; those enter-
ing CR receive no further therapy; patients with
residual or progressive disease then receive single
agent rituximab, and, for those not entering remission,
R-CHOP is recommended. Special provision will need
to be made for those patients with heart transplants
where anthracyclines may be contraindicated.
T-cell post-transplant
lymphoproliferative disorders
This is a rare complication, with fewer than 15% of
patients having T-cell phenotype in most series. The
outlook here is poorer than in B-cell PTLD, perhaps
reflecting the increased chemotherapy resistance of
T-cell tumors generally and the lack of an effective
monoclonal antibody strategy in this group. Published
series are rare but the general consensus is that patients
should be treated with RIS followed by CHOP or
CHOP-like therapy. 275
 Chapter 16: Lymphoma in the immunosuppressed
Hodgkin lymphoma
Despite the well-established association between EBV
and HL, case series of HL as PTLD are rare. Once
again, the general consensus is that patients should
undergo standard therapy with ABVD (doxorubicin,
bleomycin, vinblastine, dacarbazine) subject to
adequate cardiac function with an expectation of sim-
ilar results to those observed in the non-PTLD setting.
Burkitt lymphoma
These are rare in the PTLD setting after SOT and often
are disseminated at the time of presentation. Intensive
regimens such as CODOX-M/IVAC (cyclophospha-
mide, vincristine, doxorubicin, methotrexate/ifosfa-
mide, etoposide, cytarabine) and DA-EPOCH
(dose-adjusted etoposide, prednisone, vincristine,
cyclophosphamide, doxorubicin) have been used in
this setting, with special precautions required in
those with compromised organ function.
Lymphomas in patients with HIV
infection
The widespread use of highly active antiretroviral
therapy (HAART) has completely changed the
approach to patients with HIV infection and non-
Hodgkin lymphoma (HIV–NHL).
Clinical features and natural history of HIV–NHL
differ greatly from those observed in the general
population. At onset, HIV–NHL fall within stages III
and IV in over 70% of cases, with extranodal involve-
ment in 70–98% and B-symptoms in 75% of patients.
Typically HIV–NHL are often bulky, have a rapid
growth rate, and serum LDH levels above normal
range. In recent years a huge amount of data has dem-
onstrated the beneficial effect of HAART on the clinical
presentation of HIV–NHL. A comparison between the
clinical features of the disease in HAART-treated and
HAART-naïve patients has actually shown that the
patients who develop NHL while taking HAART are
older and present with a minor degree of extranodal
involvement, especially of bone marrow and meninges.
Unfortunately, a comparison with aggressive NHLs in
the general population still produces a negative effect.
In fact, in HIV patients, Burkitt lymphoma (BL) is
more frequent than in the general population and
NHL presents with more advanced disease, more exten-
sive extranodal involvement, higher International
276 Prognostic Index (IPI), and, frequently, higher LDH
levels. Interestingly, HAART has a positive impact on
the outcome of patients, with significantly improved
survival and disease-free survival (DFS) rates. In two
of the most common lymphoma subtypes seen in
HIV-positive patients, DLBCL and BL, the CR and OS
rates are now similar to those in HIV-negative patients.
Diffuse large B-cell lymphoma
Before the introduction of HAART the treatment of
DLBCL represented a challenge, with disappointing
results in terms of CR rate and OS. Low-dose M-
BACOD (methotrexate, bleomycin, doxorubicin,
cyclophosphamide, vincristine, dexamethasone) was
associated with CR rates (46–56%) and median survival
(6.5–8 months) that were comparable with those
achieved with standard doses (CR >50%; median sur-
vival 7–8 months), but severe toxicity was significantly
lower with the administration of lower doses (51%
versus 70%; P <0.008). The French–Italian cooperative
group conducted a large randomized study (HIV–
NHL-93) of risk-adapted, intensive chemotherapy in
patients with HIV-related NHL, the majority of
whom had DLBCL. The study showed that, in patients
without unfavorable prognostic factors, both ACVBP
(doxorubicin, cyclophosphamide, vindesine, bleomy-
cin, prednisolone) and CHOP arms reached similar
CR (66% versus 60%) and 5-year survival rates (51%
versus 47%). In patients with only one unfavorable
prognostic factor, CR rate was significantly higher
after the standard CHOP arm than with low-dose
CHOP (49% versus 32%; P <0.05), even if 5-year OS
rates are similar (28% versus 24%); whereas, in patients
with more than one unfavorable prognostic factor the
CR rate was much higher in the low-dose CHOP arm
than in the group that had received vincristine plus
prednisolone (20% versus 5%). The 5-year OS rates
were similar (11% for low-dose CHOP versus 3% for
vincristine plus prednisolone). The only factors influ-
encing survival rates were the administration of
HAART, HIV score, and the IPI score. The study,
which was terminated before monoclonal antibodies
became available, concluded that CHOP was the stand-
ard chemotherapy regimen for HIV–NHL, as for NHL
in the general population, and that increasing the treat-
ment dose-intensity in patients with good prognosis
had no effect on outcome.
In recent years, the introduction of rituximab (R)
has significantly improved the survival of NHL of the
general population as against patients receiving CHOP
alone. Based on these data, several authors have
explored the feasibility and effectiveness of rituximab

in combination with chemotherapy in patients with
HIV–NHL. The Italian Cooperative Group on AIDS
and Tumors (GICAT) performed a study on the admin-
istration of R-CDE with concomitant HAART in 74
patients, and 70% achieved a CR. Treatment-related
toxicity was acceptable for this patient setting except
for infectious events. After a median follow-up of 16
months (range 1–57 months), median survival has not
been reached, yet, and 2-year OS rate is 62%, DFS is
89%, and PFS is 86%. Other studies have evaluated the
effectiveness of rituximab in combination with chemo-
therapy in the treatment of HIV–NHL. The first trial
took place in France and used R-CHOP to treat 61
patients, with a 77% CR rate and a 2-year OS of 75%
and PFS of 69%. A phase II study was performed in
Spain on the same treatment regimen. Out of 81
patients the following rates were achieved: CR 69%,
3-year OS 56%. The only randomized phase III study
of CHOP with and without rituximab, conducted by the
AIDS-Malignancies Consortium (AMC) in the USA,
led to controversy as to the benefit of rituximab. In all,
150 patients were enrolled and no differences were
observed in CR (58% in R-CHOP versus 47% in
CHOP group) and OS rates. However, there was a
significantly increased risk of death from infectious
complications in the R-CHOP group compared to the
CHOP alone group (14% versus 2%; P = 0.035).
However, death from infection occurred mainly in
those with a CD4 count below 50 cells/mm3, and
many patients received maintenance rituximab follow-
ing completion of chemotherapy, an approach not cur-
rently used in the HIV-negative setting. Subsequently,
the AMC conducted a randomized phase II study of
DA-EPOCH with either concurrent or sequential ritux-
imab. It is worth mentioning that many centers in the
USA favor the infusional DA-EPOCH regimen over
CHOP in HIV-related DLBCL as some phase II studies
demonstrated a better response rate (RR) and OS. The
addition of rituximab to this regimen did not result in
increased risk of death from infection, and concurrent
administration of rituximab was superior to sequential
administration. A recent meta-analysis of prospective
studies has confirmed the benefit in RR and OS of the
addition of rituximab to chemotherapy. Therefore, the
addition of rituximab to chemotherapy is considered
the gold standard for the treatment of patients with HIV
infection and DLBCL.
Whether infusional chemotherapy could be con-
sidered more effective than classical chemotherapy is
still a matter of debate. A pooled analysis of both the
Chapter 16: Lymphoma in the immunosuppressed
AMC studies mentioned above suggests that the addi-
tion of rituximab to the infusional regimen, DA-
EPOCH, results in a better event-free survival and
OS than R-CHOP. However, this is a post-hoc analysis
of data from two studies in which patients were treated
over different time periods and thus other factors,
including differences in supportive care, may have
confounded the results. On the basis of these results,
however, DA-EPOCH plus rituximab is considered
the chemotherapy regimen of choice in the USA,
whereas in Europe R-CHOP remains the standard of
care. A randomized trial of these treatment regimens is
required to truly answer the question of superiority.
An additional controversy, and one that has not
been explored in a randomized study, is whether
HAART should be continued during chemotherapy
or not. In the USA, centers that use the DA-EPOCH
regimen suspend cART because of potential adverse
pharmacokinetics and pharmacodynamics interac-
tions with chemotherapy and the potential for
increased toxicity. This strategy does not appear to
affect lymphoma outcomes adversely or increase
infectious complications. In Europe, it is usual to
continue HAART during chemotherapy, avoiding
ritonavir-boosted regimens wherever possible, as
they are associated with greater toxicity.
Burkitt lymphoma (BL)
Even in the HAART era, treatment of BL remains
challenging since cure requires the use of short-course,
dose-intense chemotherapy regimens such as hyper-
CVAD (cyclophosphamide, vincristine, doxorubicin,
dexamethasone alternating with methotrexate and
cytarabine) and CODOX-M/IVAC. CHOP-based reg-
imens, as used for DLBCL, have little efficacy in BL,
producing survival rates of only a few months. There
has been reluctance to use more intensive chemother-
apy regimens for HIV-associated BL because of con-
cerns about toxicity, in particular the increased risk of
opportunistic infections. However, evidence is emerg-
ing to suggest that these regimens can be safely deliv-
ered, achieving similar response rates and OS as
HIV-negative patients with BL. American and
Spanish investigators report a 63–71% CR rate, a
46–60% failure-free survival rate at 2 years, and the
same toxicity as in the general population. A retro-
spective study addresses concerns regarding pro-
longed immunosuppression and loss of viral control
following intensive chemotherapy. Excellent immune 277
 Chapter 16: Lymphoma in the immunosuppressed
recovery was demonstrated in 30 HIV-positive
patients with BL treated with CODOX-M/IVAC. The
viral load was undetectable in 88% and 87% of patients
at 6 and 12 months, respectively, following chemo-
therapy. The CD4 count was greater than 200/mL in
58% and 80% of patients at 6 and 12 months,
respectively.
Given the benefits of rituximab in the treatment of
DLBCL and the strong expression of CD20 in Burkitt
cells, its use in HIV-negative BL is becoming more
widespread. However, data in the HIV-positive setting
is limited. A prospective study of 36 patients with BL
treated with intensive chemotherapy and rituximab
included 19 HIV-positive patients. Although HIV-
positive patients experienced more severe mucositis
and a higher incidence of infection, their outcome
was not significantly different to HIV-negative
patients, with a CR rate of 84% and a 2-year OS of
73%. A recent retrospective analysis of 80 patients with
BL treated with CODOX-M/IVAC with or without
rituximab included 14 patients who were HIV-
positive, 10 of whom received rituximab. The results
demonstrated that there were fewer relapses in
patients treated with rituximab, but only a non-
significant trend to improved survival. Importantly,
the outcome for those with HIV infection was com-
parable to the HIV-negative patients.
In the USA, to overcome concerns of treatment
toxicity in HIV-associated BL, the DA-EPOCH regi-
men plus rituximab is favored. Although numbers
treated are small, CR rates of 63–100% have been
reported.
Primary effusion lymphoma (PEL)
PEL is a rare lymphoma that has a dismal prognosis,
with a median survival of about 6 months. Given its
rarity, however, there are very few longitudinal observa-
tional series of patients with PEL, and no large prospec-
tive trials have ever defined optimal treatment strategies.
Furthermore, the vast majority of the observed and
published series on HIV–NHL have all been comprised
of more common histological types, such as BL or
DLBCL, and it is not clear that their findings and obser-
vations apply to PEL, given its unique clinical presenta-
tion and pathogenesis. In cohorts of HIV-positive
patients, prior to the introduction of HAART, the ther-
apeutic results were unsatisfactory despite the use of
aggressive polychemotherapy regimens, including
278 anthracyclines. The significant improvement of AIDS–
NHL prognosis observed in the HAART era also applies
to the PEL setting. However, the improvement in
patients with PEL is below what has been reported in
patients with HIV infection and DLBCL. As mentioned
before, clinical series report the prognostic impact of
HAART, in association with chemotherapy, on PEL
outcome. Complete remissions of PEL have been
reported after implementation of HAART alone, with-
out chemotherapeutic agents. Although this experience
is based on single case reports only, a role for immune
reconstitution in control of this aggressive lymphoma
has been suggested. Thus, initiating or continuing
HAART as part of supportive therapy is recommended
when commencing treatment for HIV-positive patients
with PEL. High-dose chemotherapy with autologous
stem cell transplant has been reported in only two indi-
vidual case reports of PEL and currently there is no
evidence to support using high-dose chemotherapy as
consolidation therapy in first remission. Novel
approaches for PEL therapy outside traditional chemo-
therapy have also been suggested. These include the
addition of antiviral therapy as well as inhibition of
specific cellular targets. Antitumor activity of antiviral
therapy directed against Kaposi’s sarcoma-associated
herpesvirus (KSHV)/human herpesvirus 8 (HHV8)
infection has been reported. However, this experience
is based on single case reports. Patients with a diagnosis
of PEL, related or not to HIV infection, experienced
prolonged complete remission with cidofovir, an anti-
viral agent with a broad activity against multiple DNA
viruses and one of the most effective agents to inhibit
KSHV/HHV8 replication. The inhibition of viral repli-
cation and a direct proapoptotic effect on lymphoma
cells by the high concentration of cidofovir achieved
through the intracavitary administration probably
explain this relevant clinical effect. Another approach
may be to target NF-κB through the use of proteasome
inhibition with drugs such as bortezomib, which induces
apoptosis of PEL cell lines in vitro alone or in combina-
tion with chemotherapy. Other approaches that have
been proposed include targeting latency phase genes
such as LANA-1, using small RNA transcripts to silence
essential viral genes, and augmenting host immunity
against HHV8.
Plasmablastic lymphoma (PBL)
PBL accounts for 2.6% of all HIV–NHL, suggesting an
increase in the past years. It is generally associated with
severe immunodeficiency and predominantly occurs
in males. The most notable feature of PBL is its pred-
ilection for the oral cavity even though one-third of

patients have extraoral presentation. Other patients
present with typical B-symptoms of lymphoma (i.e.
fever, night sweats, and unintentional weight loss) but
no symptoms are particularly sensitive or specific for
PBL. Despite the introduction of HAART, the prog-
nosis of PBL remains very poor. Three reports have
noted initial regression of PBL after administration of
HAART only; two of these patients subsequently
relapsed and control of the tumor was achieved with
both HAART and chemotherapy; the third patient was
also treated with chemotherapy with good response,
but follow-up was only available at 1 month. The most
commonly reported lymphoma-specific chemother-
apy is CHOP, and two retrospective case series report
a CR rate of around 66%. Despite this encouraging
response rate, PFS is only a few months, with a median
survival of less than 1 year. More intensive chemo-
therapy regimens do not appear to improve prognosis,
and a significantly worse outcome has been associated
with an ECOG performance score ≥2, advanced clin-
ical stage, and the presence of myc gene rearrange-
ments, the latter being found in 40% of patients in
one series. Since conventional chemotherapy is largely
ineffective, novel agents are being investigated and
drugs such as bortezomib and lenalidomide may
prove to be useful.
Primary central nervous system lymphoma
Primary central nervous system lymphoma (PCNSL)
occurs in the setting of severe immunosuppression,
thus in the HAART era the incidence has reduced
markedly. Historically, standard treatment has been
whole brain irradiation, but this is merely palliative.
Survival has improved with the use of HAART and the
resultant immune recovery. This has encouraged the
use of CNS penetrating, systemic chemotherapy regi-
mens as in the HIV-negative setting. A small study in
15 HIV-positive patients treated with high-dose
methotrexate, 3 g/m2, reported a CR rate of 47% and
a median survival of 19 months. Other strategies being
explored are combinations with rituximab and
temozolamide.
Salvage treatment
Lymphoma progression is the leading cause of death
in 35–55% of patients with HIV–NHL receiving che-
motherapy, of whom around half need second line
chemotherapy following progression or relapse of the
Chapter 16: Lymphoma in the immunosuppressed
disease. To date, the results achieved by salvage thera-
pies that do not include a standard high-dose chemo-
therapy regimen with peripheral blood stem cell
transplant have been very frustrating (median survival
2–4 months). With the introduction of HAART into
clinical practice, more aggressive treatment protocols,
whose effectiveness has already been documented in
HIV-negative lymphoma patients, can be taken into
consideration. Several studies support the feasibility of
high-dose chemotherapy in combination with autolo-
gous stem cell transplant in patients with HIV–NHL,
chemosensitive recurrence, or partial remission (PR)
after first line chemotherapy, proving that peripheral
blood stem cell collections are adequate, anchoring
rates are similar to those recorded in HIV-negative
patients, and high-dose chemotherapy is well tolerated
with no increase in the incidence of opportunistic
infections, at least in the short term. Recently, within
the GICAT, a study has been performed on a group of
patients with NHL or refractory or recurred Hodgkin’s
disease: the results support the feasibility of an
adequate peripheral blood stem cell collection, with
no transplant-related mortality and a very good out-
come, similar to that observed in HIV-negative
patients.
Allogeneic transplantation remains an experimen-
tal procedure in patients with HIV infection, with no
prospective trials. However, it does appear to be fea-
sible in HIV-positive patients with hematological
malignancies. A retrospective study from the Center
for International Blood Marrow Transplant Research
in the USA has reported the outcome for 23 HIV-
positive patients who received allogeneic transplants
between 1987 and 2003, 10 of whom had a diagnosis of
NHL. In 19 patients (83%) the donor was an HLA-
identical sibling, and 20 patients (87%) received mye-
loablative conditioning. The incidence of acute and
chronic graft-versus-host disease was 30% and 28%,
respectively, and the 2-year OS was 30%. A signifi-
cantly better outcome was apparent for those patients
treated after the introduction of HAART.
Hodgkin lymphoma
Hodgkin lymphoma is a non-AIDS defining malig-
nancy with an increased incidence in HIV-positive
patients. Unexpectedly, some studies have reported a
further increase in incidence in the HAART era with
possibly a higher risk shortly after commencing
HAART. With the introduction of HAART, there 279
has been a shift in histologies, with the mixed
 Chapter 16: Lymphoma in the immunosuppressed
cellularity subtype of HL occurring most commonly in
the context of severe immunosuppression and the
nodular sclerosis subtype occurring most commonly
in those with a higher CD4 count. The prognosis for
patients with HIV-related HL has certainly improved
in the HAART era and it is becoming the norm to treat
patients with chemotherapy regimens used in the
HIV-negative setting; however, no randomized trials
but only prospective phase II studies have been per-
formed in this patient group.
A GICAT prospective trial carried out between
March 1989 and March 1992 enrolled 17 patients to
test a combined chemotherapy regimen based on epi-
rubicin, bleomycin, and vinblastine (EBV protocol).
Overall, the CR rate was 53%, with an 11-month
median survival and a 2-year DFS rate of 55%. In an
attempt to improve the above findings, another trial
was performed between 1993 and 1997 on the feasi-
bility of the EBVP (epirubicin, bleomycin, vinblastine,
and prednisone) combination. Thirty-five patients
were admitted. The results showed a CR rate of 74%,
but a recurrence was observed after 2 years in 38% of
the cases. Survival and 3-year DFS rates were 32% and
53%, respectively. ACTG reported the findings of a
phase II study on 21 patients receiving six cycles of
ABVD regimen. Antiretroviral therapy was not
administered concomitantly but at the end of chemo-
therapy. A CR rate of 43% was observed, with an
18-month median survival.
The GICAT has reported the final results of a
prospective phase II study to determine the feasibility
of short-term (12 weeks) chemotherapy with classical
Stanford V regimen and adjuvant radiotherapy. In the
time period from May 1997 to October 2001, 59
patients were treated and 69% completed the treat-
ment plan. No dose reduction or delayed administra-
tion was required. The main dose-limiting side effects
were myelosuppression and peripheral neurotoxicity.
Overall, 89% of the patients had an objective response,
with CR and PR rates of 81% and 8%, respectively. The
estimated 3-year OS, DFS, and time-to-progression
rates are 51%, 68%, and 60%, respectively.
Multivariate analysis showed that the IPS was the
only significant survival-related variable (p = 0.001)
with higher survival in patients with an IPS <2 com-
pared to those with an IPS >2.
The results of the combined BEACOPP (bleomy-
cin, etoposide, doxorubicin, cyclophosphamide, vin-
cristine, procarbazine, prednisone) regimen on a small
280 group of patients with Hodgkin disease and HIV
infection have been reported. All patients have
achieved CR, and nine are still alive and disease-free
after a median follow-up of 49 months. The results of a
large prospective phase II study with ABVD have been
published. Within a cooperative network in Spain, 62
patients with HIV–HL received the standard ABVD
plus HAART. The scheduled six to eight ABVD cycles
were completed in 82% of cases. Six patients died
during induction, 54 (87%) achieved a CR, and two
were resistant. The 5-year OS and event-free survival
probabilities were 76% and 71%, respectively. The
immunological response to HAART had a positive
impact on OS (P = 0.002) and EFS (P = 0.001). The
GICAT have also reported a prospective phase II study
aiming to evaluate feasibility and activity of a novel
regimen including epirubicin, bleomycin, vinorelbine,
cyclophosphamide, and prednisone (VEBEP regi-
men). Seventy percent of patients had advanced stages
of disease, and 45% had an IPS >2. The CR was 67%,
and 2-year OS, DFS, time to failure (TTF), and EFS
were 69%, 86%, 59%, and 52%, respectively. Recently, a
retrospective study with long follow-up has demon-
strated that, in patients treated with ABVD chemo-
therapy, the outcome is similar to that of HIV-negative
HL patients. In this study, which included 93 HIV-
positive patients with HL, the complete remission/
complete remission – unconfirmed (CR/CRu) rate
was 77%, with a 5-year DFS and OS of 87% and 81%,
respectively. Similarly to HIV-negative patients,
[18F]-fluoro-2-deoxy-D-glucose positron emission
tomography [FDG–PET] after 2–3 cycles of ABVD
chemotherapy appears to be predictive of outcome.
In a retrospective study of 23 HIV-positive patients
with advanced HL, treatment failure was not seen in
those with a negative interim FDG–PET
scan. Table 16.1 summarizes all these studies.
In general, continuation of HAART during che-
motherapy for HL is recommended. However, care is
required in the choice of antiviral combination as
interactions with chemotherapy have been reported
to increase toxicity. To this end, ritonavir-boosted
regimens are avoided because of reports of severe
neurotoxicity.
It can be drawn from the above that prognosis for
patients with lymphomas and concomitant HIV infec-
tion could be improved by using better combined
chemotherapy protocols, incorporating anticancer
treatments and antiretroviral drugs. The administra-
tion of the newly available effective antiretroviral
drugs (protease inhibitors) during chemotherapy
 Chapter 16: Lymphoma in the immunosuppressed

Table 16.1 Results of prospective studies in HIV–HL.

Regimen Number of Stage Response Complete Overall

patients III–IV rate remission rate survival
EBV 17 88% 82% 53% 11 months
EBVP 35 83% 91% 74% 16 months
ABVD 21 81% 62% 43% 18 months
Stanford V 59 71% 89% 81% 59% at 5 years
BEACOPP 12 92% 100% 100% 75% at 3 years
ABVD 62 100% 87% 87% 76% at 5 years
VEBEP 71 70% 78% 67% 69% at 2 years
ABVD 93 80% 91% 74% 81% at 5 years

together with nucleoside analogs can improve control
of the underlying HIV infection. Actually, a reduction
in the viral load to indeterminable levels and an
increased CD4+ count may lower the risk for oppor-
tunistic infections during chemotherapy. The inclu-
sion of hematopoietic growth factors in the treatment
of this patient group makes it possible to increase
chemotherapy doses and prolong the administration
of antiretroviral drugs with an intent to improve
survival.
Further reading
An J, Sun Y, Fisher M, et al. Antitumor effects of bortezomib
(PS-341) on primary effusion lymphomas. Leukemia,
2004;18(10):1699–1704.
Armstrong R, Bradrick J, Liu YC. Spontaneous regression of
an HIV-associated plasmablastic lymphoma in the oral
cavity: a case report. J Oral Maxillofac Surg,
2007;65:1361–1364.
Balsalobre P, Díez-Martín JL, Re A, et al. Autologous stem-
cell transplantation in patients with HIV-related
lymphoma. J Clin Oncol, 2009;27(13):2192–2198.
Barnes JA, Lacasce AS, Feng Y, et al. Evaluation of the
addition of rituximab to CODOX-M/IVAC for Burkitt's
lymphoma: a retrospective analysis. Ann Oncol,
2011;22(8):1859–1864.
Barta SK, Lee JY, Kaplan LD, Noy A, Sparano JA. Pooled
analysis of AIDS malignancy consortium trials evaluating
rituximab plus CHOP or infusional EPOCH
chemotherapy in HIV-associated non-Hodgkin
lymphoma. Cancer, 2011.
Baumgartner JE, Rachlin JR, Beckstead JH, et al. Primary
central nervous system lymphomas: natural history and
response to radiation therapy in 55 patients with acquired
immunodeficiency syndrome. J Neurosurg,
1990;73(2):206–211.
Bibas M, Grisetti S, Alba L, Picchi G, Del Nonno F, Antinori
A. Patient with HIV-associated plasmablastic lymphoma
responding to bortezomib alone and in combination with
dexamethasone, gemcitabine, oxaliplatin, cytarabine, and
pegfilgrastim chemotherapy and lenalidomide alone.
J Clin Oncol, 2010; 28(34):e704–708.
Biggar RJ, Jaffe ES, Goedert JJ, Chaturvedi A, Pfeiffer R,
Engels EA. Hodgkin lymphoma and immunodeficiency
in persons with HIV/AIDS. Blood, 2006;108
(12):3786–3791.
Bose P, Thompson C, Gandhi D, Ghabach B, Ozer H.
AIDS-related plasmablastic lymphoma with dramatic,
early response to bortezomib. Eur J Haematol,
2009;82(6):490–492.
Bohlius J, Schmidlin K, Boue F, et al. HIV-1-related
Hodgkin lymphoma in the era of combination
antiretroviral therapy: incidence and evolution of
CD4(+) T-cell lymphocytes. Blood, 2011;117
(23):6100–6108.
Boue F, Gabarre J, Gisselbrecht C, et al. Phase II trial of
CHOP plus rituximab in patients with HIV-associated
non-Hodgkin’s lymphoma. J Clin Oncol,
2006;24:4123–4128.
Boulanger E, Daniel MT, Agbalika F, et al. Combined
chemotherapy including high-dose methotrexate in
KSHV/HHV8-associated primary effusion lymphoma.
Am J Hematol, 2003;73(3):143–148.
Boulanger E, Gérard L, Gabarre J, et al. Prognostic factors
and outcome of human herpesvirus 8-associated primary
effusion lymphoma in patients with AIDS. J Clin Oncol,
2005;23(19):4372–4380.
Bower M, McCall-Peat N, Ryan N, et al. Protease inhibitors
potentiate chemotherapy-induced neutropenia. Blood,
2004;104(9): 2943–2946.
Caillard S, Porcher R, Provot F, et al. Post-transplantation
lymphoproliferative disorder after kidney 281
transplantation: report of a nationwide French registry
 Chapter 16: Lymphoma in the immunosuppressed
and the development of a new prognostic score. J Clin
Oncol, 2013;31(10):1302–1309.
Castillo JJ, Echenique IA. Rituximab in combination with
chemotherapy versus chemotherapy alone in HIV-associated
non-Hodgkin lymphoma: A pooled analysis of 15
prospective studies. Am J Hematol, 2012;87(3):330–333.
Castillo J, Pantanowitz L, Dezube BJ. HIV-associated
plasmablastic lymphoma: lessons learned from 112
published cases. Am J Hematol, 2008;83(10):804–809.
Chen YB, Rahemtullah A, Hochberg E. Primary effusion
lymphoma. Oncologist, 2007;12(5):569–576.
Cheung MC, Hicks LK, Leitch HA. Excessive neurotoxicity
with ABVD when combined with protease inhibitor-
based antiretroviral therapy in the treatment of AIDS-
related Hodgkin lymphoma. Clin Lymph, myel leuk,
2010;10(2):E22–25.
Choquet S, Levy V, Leblond V. Identification of prognostic
factors in post-transplantation lymphoproliferative
disorders. Bull Cancer, 2004;91(4):373–378.
Choquet S, Leblond V, Herbrecht R, et al. Efficacy and safety of
rituximab in B-cell post-transplantation
lymphoproliferative disorders: results of a prospective
multicenter phase 2 study. Blood, 2006;107(8):3053–3057.
Choquet S, Trappe R, Leblond V, et al. CHOP-21 for the
treatment of post-transplant lymphoproliferative
disorders (PTLD) following solid organ transplantation.
Haematologica, 2007;92(2):273–274.
Coiffier B, Lepage E, Briere J, et al. CHOP chemotherapy
plus rituximab compared with CHOP alone in elderly
patients with diffuse large-B-cell lymphoma. N Engl J
Med, 2002;346:235–242.
Dharnidharka VR, Lamb KE, Gregg JA, et al. Associations
between EBV serostatus and organ transplant type in
PTLD risk: an analysis of the SRTR National Registry Data
in the United States. Am J Transplant, 2012;12:976–983.
Diamond C, Taylor TH, Im T, Anton-Culver H. Presentation
and outcomes of systemic non-Hodgkin’s lymphoma: a
comparison between patients with acquired
immunodeficiency syndrome (AIDS) treated with highly
active antiretroviral therapy and patients without AIDS.
Leuk Lymphoma, 2006;47(9):1822–1829.
Díez-Martín JL, Balsalobre P, Re A, et al. Comparable
survival between HIV+ and HIV– non-Hodgkin and
Hodgkin lymphoma patients undergoing autologous
peripheral blood stem cell transplantation. Blood,
2009;113(23):6011–6014.
Dunleavy K, Wilson WH. How I treat HIV-associated
lymphoma. Blood, 2012;119(14): 3245–3255.
Dunleavy K, Little RF, Pittaluga S, et al. The role of tumor
histogenesis, FDG-PET, and short-course EPOCH
with dose-dense rituximab (SC-EPOCH-RR) in
282 HIV-associated diffuse large B-cell lymphoma. Blood,
2010;115(15):3017–3024.
Engels EA, Pfeiffer RM, Fraumeni JF Jr., et al. Spectrum of
cancer risk among US solid organ transplant recipients.
JAMA, 2011;306:1891–1901.
Errante D, Tirelli U, Gastaldi R, et al. Combined antineoplastic
and antiretroviral therapy for patients with Hodgkin’s
disease and human immunodeficiency virus infection. A
prospective study of 17 patients. Cancer, 1994;73:437–444.
Errante D, Gabarre J, Ridolfo AL, et al. Hodgkin’s disease in 35
patients with HIV infection: an experience with epirubicin,
bleomycin, vinblastine and prednisone chemotherapy in
combination with antiretroviral therapy and primary use of
G-CSF. Ann Oncol, 1999;10:189–195.
Evens AM, David KA, Helenowski I, et al. Multicenter
analysis of 80 solid organ transplantation recipients with
post-transplant lymphoproliferative disease: outcomes
and prognostic factors in the modern era. J Clin Oncol,
2010;28(6):1038–1046.
Gabarre J, Marcelin AG, Azar N. High-dose therapy plus
autologous hematopoietic stem cell transplantation for
human immunodeficiency virus (HIV)-related
lymphoma: results and impact on HIV disease.
Haematologica, 2004;89(9):1100–1108.
Gupta V, Tomblyn M, Pedersen TL, et al. Allogeneic
hematopoietic cell transplantation in human
immunodeficiency virus-positive patients with
hematologic disorders: a report from the center for
international blood and marrow transplant research. Biol
Blood and Marrow transpl, 2009;15(7):864–871.
Habermann TM, Weller EA, Morrison VA, et al.
Rituximab-CHOP versus CHOP alone or with
maintenance rituximab in older patients with diffuse large
B-cell lymphoma. J Clin Oncol, 2006;24(19):3121–3127.
Halfdanarson TR, Markovic SN, Kalokhe U, et al. A non-
chemotherapy treatment of a primary effusion
lymphoma: durable remission after intracavitary
cidofovir in HIV negative PEL refractory to
chemotherapy. Ann Oncol, 2006;17(12):1849–1850.
Hartmann P, Rehwald U, Salzberger B, et al. BEACOPP
therapeutic regimen for patients with Hodgkin’s disease
and HIV infection. Ann Oncol, 2003;14:1562–1569.
Hentrich M, Maretta L, Chow KU, et al. Highly active
antiretroviral therapy (HAART) improves survival in
HIV-associated Hodgkin's disease: results of a
multicenter study. Ann Oncol, 2006;17(6):914–919.
Hocqueloux L, Agbalika F, Oksenhendler E, Molina JM.
Long-term remission of an AIDS-related primary
effusion lymphoma with antiviral therapy. AIDS,
2001;15(2):280–282.
Hoffmann C, Tabrizian S, Wolf E, et al. Survival of AIDS
patients with primary central nervous system lymphoma
is dramatically improved by HAART-induced immune
recovery. AIDS, 2001;15(16):2119–2127.
Jacomet C, Girard PM, Lebrette MG, Farese VL, Monfort L,
Rozenbaum W. Intravenous methotrexate for primary

central nervous system non-Hodgkin's lymphoma in
AIDS. AIDS, 1997;11(14):1725–1730.
Jiang Y, Villeneuve PJ, Wielgosz A, et al. The incidence of cancer
in a population-based cohort of Canadian heart transplant
recipients. Am J Transplant, 2010;10:637–645.
Kaplan LD, Straus DJ, Testa MA, et al. Low-dose compared
with standard-dose m-BACOD chemotherapy for non-
Hodgkin’s lymphoma associated with human
immunodeficiency virus infection. National Institute of
Allergy and Infectious Diseases AIDS Clinical Trials
Group. N Engl J Med, 1997;336:1641–1648.
Kaplan LD, Lee JY, Ambinder RF, et al. Rituximab does not
improve clinical outcome in a randomized phase 3 trial of
CHOP with or without rituximab in patients with HIV-
associated non-Hodgkin lymphoma: AIDS-Malignancies
Consortium Trial 010. Blood, 2005;106:1538–1543.
Klass CM, Krug LT, Pozharskaya VP, et al. The targeting of
primary effusion lymphoma cells for apoptosis by inducing
lytic replication of human herpesvirus 8 while blocking
virus production. Blood, 2005;105(10):4028–4034.
Krishnan AY, Molina A, Zaia J, et al. Durable remissions with
autologous stem cell transplantation for high-risk HIV-
associated lymphomas. Blood, 2005;105(2):874–878.
LaCasce AS. Post-transplant lymphoproliferative disorders.
Oncologist, 2006;11:674–680.
Lanoy E, Rosenberg PS, Fily F, et al. HIV-associated Hodgkin
lymphoma during the first months on combination
antiretroviral therapy. Blood, 2011;118(1):44–49.
Lester R, Li C, Phillips P, et al. Improved outcome of human
immunodeficiency virus-associated plasmablastic
lymphoma of the oral cavity in the era of highly active
antiretroviral therapy: a report of two cases. Leuk
Lymphoma, 2004;45:1881–1885.
Levine AM, Li P, Cheung T, et al. Chemotherapy consisting
of doxorubicin, bleomycin, vinblastine and dacarbazine
with granulocyte-colony-stimulating factor CSF in HIV
infected patients with newly diagnosed Hodgkin’s
disease: a prospective, multi-institutional AIDS Clinical
Trials Group study (ACTG 149). J Acquir Immune Defic
Syndr, 2000;24:444–450.
Lim ST, Karim R, Nathwani BN, et al. AIDS-related Burkitt’s
lymphoma versus diffuse large-cell lymphoma in the
pre-highly active antiretroviral therapy (HAART) and
HAART eras: significant differences in survival with
standard chemotherapy. J Clin Oncol, 2005;23:4430–4438.
Little RF, Pittaluga S, Grant N, et al. (2003). Highly effective
treatment of acquired immunodeficiency syndrome-
related lymphoma with dose-adjusted EPOCH: impact of
antiretroviral therapy suspension and tumor biology.
Blood, 2003;101(12):4653–4659.
Luppi M, Trovato R, Barozzi P, et al. Treatment of herpesvirus
associated primary effusion lymphoma with intracavity
cidofovir. Leukemia, 2005;19(3):473–476.
Chapter 16: Lymphoma in the immunosuppressed
Montoto S, Wilson J, Shaw K, et al. Excellent immunological
recovery following CODOX-M/IVAC, an effective
intensive chemotherapy for HIV-associated Burkitt's
lymphoma. AIDS, 2010;24(6): 851–856.
Montoto S, Shaw K, Okosun J, et al. HIV status does not
influence outcome in patients with classical Hodgkin
lymphoma treated with chemotherapy using
doxorubicin, bleomycin, vinblastine, and dacarbazine in
the highly active antiretroviral therapy era. J Clin Oncol,
2012;30(33):4111–4116.
Mounier N, Spina M, Gabarre J, et al. AIDS-related non-
Hodgkin lymphoma: final analysis of 485 patients treated
with risk-adapted intensive chemotherapy. Blood,
2006;107:3832–3840.
Nasta SD, Carrum GM, Shahab I, et al. Regression of a
plasmablastic lymphoma in a patient with HIV on highly
active antiretroviral therapy. Leuk Lymphoma,
2002;43:423–426.
Navarro JT, Lloveras N, Ribera JM, Oriol A, Mate JL. Feliu E
(2005). The prognosis of HIV-infected patients with
diffuse large B-cell lymphoma treated with
chemotheraphy and highly active antiretroviral therapy is
similar to that of HIV-negative patients receiving
chemotherapy. Haematologica, 2005;90(5): 704–706.
Okosun J, Warbey V, Shaw K, et al. Interim fluoro-2-deoxy-
D-glucose-PET predicts response and progression-free
survival in patients with Hodgkin lymphoma and HIV
infection. AIDS, 2012;26(7):861–865.
Opelz G, Daniel V, Naujokat C, Dohler B. Epidemiology of
pretransplant EBV and CMV serostatus in relation to
posttransplant non-Hodgkin lymphoma.
Transplantation, 2009;88:962–967.
Oriol A, Ribera JM, Esteve J, et al. Lack of influence of
human immunodeficiency virus infection status in the
response to therapy and survival of adult patients with
mature B-cell lymphoma or leukemia. Results of the
PETHEMA-LAL3/97 study. Haematologica,
2003;88(4):445–453.
Oriol A, Ribera JM, Bergua J, et al. High-dose chemotherapy
and immunotherapy in adult Burkitt lymphoma:
comparison of results in human immunodeficiency
virus-infected and noninfected patients. Cancer,
2008;13(1):117–125.
Panos G, Karveli EA, Nikolatou O, et al. Prolonged survival
of an HIV-infected patient with plasmablastic lymphoma
of the oral cavity. Am J Hematol, 2007;82:761–765.
Parker A, Bowles K, Bradley JA, et al. Management of post-
transplant lymphoproliferative disorder in adult solid
organ transplant recipients–BCSH and BTS guidelines.
Br J Haematol, 2010;149(5):693–705.
Pfreundschuh M, Trumper L, Osterborg A, et al. CHOP-like
chemotherapy plus rituximab versus CHOP-like
chemotherapy alone in young patients with good- 283
prognosis diffuse large-B-cell lymphoma: a randomised
 Chapter 16: Lymphoma in the immunosuppressed
controlled trial by the MabThera International Trial
(MInT) Group. Lancet Oncol, 2006;7(5):379–391.
Polesel J, Clifford GM, Rickenbach M, et al. Non-Hodgkin
lymphoma incidence in the Swiss HIV Cohort Study
before and after highly active antiretroviral therapy.
AIDS, 2008;22(2):301–306.
Quinlan SC, Pfeiffer RM, Morton LM, Engels EA. Risk
factors for early-onset and late-onset post-transplant
lymphoproliferative disorder in kidney recipients in the
United States. Am J Hematol, 2011;86:206–209.
Raez LE, Patel P, Feun L, Restrepo A, Raub WA, Jr., Cassileth
PA. Natural history and prognostic factors for survival in
patients with acquired immune deficiency syndrome
(AIDS)-related primary central nervous system
lymphoma (PCNSL). Crit Rev oncogen,
1998;9(3–4):199–208.
Ratner L, Lee J, Tang S, et al. Chemotherapy for human
immunodeficiency virus-associated non-Hodgkin's
lymphoma in combination with highly active
antiretroviral therapy. J Clin Oncol,
2001;19(8):2171–2178.
Re A, Cattaneo C, Michieli M, et al. High-dose therapy and
autologous peripheral-blood stem-cell transplantation as
salvage treatment for HIV-associated lymphoma in
patients receiving highly active antiretroviral therapy.
J Clin Oncol, 2003;21(23):4423–4427.
Re A, Michieli M, Casari S, et al. High-dose therapy and
autologous peripheral blood stem cell transplantation as
salvage treatment for AIDS-related lymphoma: long-term
results of the Italian Cooperative Group on AIDS and
Tumors (GICAT) study with analysis of prognostic
factors. Blood, 2009;114(7):1306–1313.
Ribera JM, Oriol A, Morgades M, et al. Safety and efficacy
of cyclophosphamide, adriamycin, vincristine,
prednisone and rituximab in patients with human
immunodeficiency virus-associated diffuse large B-cell
lymphoma: results of a phase II trial. Br J Haematol,
2008;140(4):411–419.
Riedel DJ, Gonzalez-Cuyar F, Zhao XF, et al. Plasmablastic
lymphoma of the oral cavity: a rapidly progressive
lymphoma associated with HIV infection. Lancet Infect
Dis, 2008;8:261–267.
Schichman SA, McClure R, Schaefer RF, Mehta P. HIV and
plasmablastic lymphoma manifesting in sinus, testicles,
and bones: a further expansion of the disease spectrum.
Am J Hematol, 2004;77:291–295.
Siddiqi T, Joyce RM. A case of HIV-negative primary
effusion lymphoma treated with bortezomib, pegylated
liposomal doxorubicin, and rituximab. Clin Lymphoma
Myeloma, 2008;8(5):300–304.
Simonelli C, Spina M, Cinelli R, et al. Clinical features and
outcome of primary effusion lymphoma in HIV-infected
284 patients: a single-institution study. J Clin Oncol,
2003;21(21):3948–3954.
Sparano JA. Clinical aspects and management of AIDS-
related lymphoma. Eur J Cancer, 2001;37:1296–1305.
Sparano JA, Lee JY, Kaplan LD, et al. Rituximab plus
concurrent infusional EPOCH chemotherapy is highly
effective in HIV-associated B-cell non-Hodgkin’s
lymphoma. Blood, 2010;115(15):3008–3016.
Spina M, Gaidano G, Carbone A, Capello D, Tirelli U.
Highly active antiretroviral therapy in human
herpesvirus-8-related body-cavity-based lymphoma.
AIDS, 1998;12 (8):955–956.
Spina M, Vaccher E, Juzbasic S, et al. Human
immunodeficiency virus-related non-Hodgkin
lymphoma: activity of infusional cyclophosphamide,
doxorubicin and etoposide as second-line chemotherapy
in 40 patients. Cancer, 2001;1(92):200–206.
Spina M, Gabarre J, Rossi G, et al. Stanford V and
concomitant HAART in 59 patients with Hodgkin’s
disease and HIV infection. Blood, 2002;100:1984–1988.
Spina M, Carbone A, Vaccher E, et al. Outcome in patients
with non-Hodgkin lymphoma and with or without
human immunodeficiency virus infection. Clin Infect Dis,
2004;38(1):142–144.
Spina M, Jaeger U, Sparano JA, et al. Rituximab plus
infusional cyclophosphamide, doxorubicin, and
etoposide (R-CDE) in HIV-associated non-Hodgkin’s
lymphoma: pooled results from three phase II trials.
Blood, 2005;105(5):1891–1897.
Spina M, Simonelli C, Talamini R, Tirelli U. Patients with
HIV with Burkitt’s lymphoma have a worse outcome
than those with diffuse large-cell lymphoma also in the
highly active antiretroviral therapy era. J Clin Oncol,
2005;23(31):8132–8133.
Spina M, Rossi G, Antinori A, et al. VEBEP regimen and
highly active antiretroviral therapy (HAART) in patients
(pts) with HD and HIV infection (HD-HIV). Ann Oncol,
2008;19:iv152, abs 227.
Taylor AL, Bowles KM, Callaghan CJ, et al. Anthracycline-
based chemotherapy as first-line treatment in adults with
malignant post-transplant lymphoproliferative disorder
after solid organ transplantation. Transplantation,
2006;82(3):375–381.
Tirelli U, Spina M, Vaccher E, et al. Clinical evaluation of 451
patients with HIV related non-Hodgkin’s lymphoma:
experience of the Italian Cooperative Group on AIDS and
Tumors (GICAT). Leukemia Lymphoma, 1995;20:91–96.
Trappe R, Oertel S, Leblond V, et al. Sequential treatment
with rituximab followed by CHOP chemotherapy in
adult B-cell post-transplant lymphoproliferative disorder
(PTLD): the prospective international multicentre phase
2 PTLD-1 trial. Lancet Oncol, 2012;13(2):196–206.
Tsai De, Hardy CL, Tomaszewski JE, et al. Reduction in
immunosuppression as initial therapy for posttransplant
lymphoproliferative disorder: analysis of prognostic

variables and long-term follow-up of 42 adult patients.
Transplantation, 2001;71(8):1076–1088.
Vaccher E, Spina M, Talamini R, et al. Improvement of
systemic human immunodeficiency virus-related
non-Hodgkin’s lymphoma outcome in the era of highly
active antiretroviral therapy. Clin Infect Dis,
2003;37:1556–1564.
Vaccher E, Tirelli U, Spina M, et al. Age and serum lactate
dehydrogenase level are independent prognostic factors
in HIV-related non-Hodgkin’s lymphoma: a single
institution study of 96 patients. J Clin Oncol,
1996;14:2217–2233.
Waddington TW, Aboulafia DM. Failure to eradicate AIDS-
associated primary effusion lymphoma with high-dose
chemotherapy and autologous stem cell reinfusion: case
report and literature review. AIDS Patient Care STDS,
2004;18(2):67–73.
Wang ES, Straus DJ, Teruya-Feldstein J, et al. Intensive
chemotherapy with cyclophosphamide, doxorubicin,
high-dose methotrexate/ifosfamide, etoposide, and high-
dose cytarabine (CODOX-M/IVAC) for human
Chapter 16: Lymphoma in the immunosuppressed
immunodeficiency virus-associated Burkitt lymphoma.
Cancer, 2003;98:1196–1205.
Weiss R, Mitrou P, Arasteh K, et al. Acquired
immunodeficiency syndrome-related lymphoma:
simultaneous treatment with combined
cyclophosphamide, doxorubicin, vincristine, and
prednisone chemotherapy and highly active antiretroviral
therapy is safe and improves survival–results of the
German Multicenter Trial. Cancer,
2006;106(7):1560–1568.
Won JH, Han SH, Bae SB, et al. Successful eradication
of relapsed primary effusion lymphoma with
high-dose chemotherapy and autologous stem cell
transplantation in a patient seronegative for human
immunodeficiency virus. Int J Hematol, 2006;83
(4):328–330.
Xicoy B, Ribera JM, Miralles P, et al. Results of treatment with
doxorubicin, bleomycin, vinblastine and dacarbazine and
highly active antiretroviral therapy in advanced stage
human immunodeficiency virus-related Hodgkin’s
lymphoma. Haematologica, 2007;92:191–198.
285
 Chapter
17
Atypical lymphoproliferative, histiocytic,
and dendritic cell disorders
Matthew S. Davids and Jennifer R. Brown
Introduction
Atypical lymphoproliferative, histiocytic, and dendritic
cell disorders often mimic their more common benign
or malignant counterparts, and thus can easily be mis-
diagnosed. Although it may be challenging to identify
these unusual disease entities precisely, doing so may
significantly alter further evaluation, prognosis, and
management of the patient. Multidisciplinary collabor-
ation between clinicians and pathologists is often
required to arrive at a definitive diagnosis. Critical to
this process is obtaining adequate tissue, ideally through
excisional lymph node biopsy. Although core-needle
biopsy may sometimes yield a diagnosis, for many of
these entities it will be inadequate. Fine needle aspira-
tion has been demonstrated to be of little value in the
diagnosis of lymphoproliferative disorders, as it does
not provide an adequate assessment of tissue architec-
ture. In cases where a biopsy is equivocal, additional
biopsies may be necessary to establish a firm diagnosis.
In this chapter, we will summarize the pathophysi-
ology, presentation, diagnosis, prognosis, and treatment
of atypical lymphoproliferative, histiocytic, and dendritic
cell disorders. Three main categories of these disorders
have been defined (Table 17.1). Benign conditions, such
as Kikuchi’s disease, often behave indolently and can
usually be managed with observation. Histiocytic disor-
ders exhibit a wide range of behavior, and may require
more aggressive therapies. Atypical lymphoproliferative
disorders can be polyclonal, such as Castleman’s disease,
or malignant, such as lymphomatoid granulomatosis,
and are treated in a variety of different ways.
While benign lymphoproliferative disorders are
generally self-limiting, histiocytic disorders and atyp-
ical lymphoproliferative disorders have the potential
to evolve into more aggressive and eventually malig-
nant phenotypes. As the understanding of the biology
286 Lymphoma, Second Edition, ed. Robert Marcus, John W. Sweetenham, and Michael E. Williams. Published by
Cambridge University Press. © Cambridge University Press 2014.
underlying these disorders has grown, it is increasingly
apparent that each disease category encompasses a
whole spectrum of distinct pathologic and clinical
entities. For example, Castleman’s disease can be sub-
divided into unicentric or multicentric variants, and
by the presence or absence of human herpes virus 8
(HHV8). Appreciating these biological differences will
be important as we strive to improve our treatments
for these rare disorders.
Benign lymphoproliferative disorders
Kikuchi’s disease
Presentation/diagnosis
Kikuchi’s disease, also known as Kikuchi–Fujimoto
disease or Kikuchi’s histiocytic necrotizing lymphade-
nitis, is a rare, benign syndrome characterized by
cervical lymphadenopathy and fever. Although the
underlying etiology is largely unknown, some evidence
points to a possible autoimmune process triggered by
an inciting infectious agent. The disease was first
described in young, otherwise healthy, women, but is
now thought to present in men in nearly equal num-
bers. The median age at diagnosis is 30, with reports
published showing presentation at ages ranging from 6
to 80 years. Kikuchi’s disease was first described in
Japan, and is likely more common in Asia, but cases
have now been documented worldwide. Low-grade
fever is often an initial presenting symptom, and typ-
ically lasts for 1–4 weeks. Cervical lymphadenopathy,
which generally develops during this time period, is
usually unilateral, less than 4 cm, and may be painful.
Subsequently, additional symptoms such as night
sweats, weight loss, and, more rarely, gastrointestinal
symptoms can develop. Dermatologic manifestations

Table 17.1 Classification of rare lymphoproliferative
disorders.
Benign lymphoproliferative disorders
* Kikuchi’s disease
*
Progressive transformation of germinal centers
* Vascular transformation of sinuses
* Inflammatory pseudotumor of lymph nodes
Histiocytic disorders
*
Rosai–Dorfman disease
* Histiocytic sarcoma
* Langerhans cell histiocytosis
* Interdigitating dendritic cell sarcoma
* Follicular dendritic cell sarcoma
* Erdheim–Chester disease
* Hemophagocytic lymphohistiocytosis
Atypical lymphoproliferative disorders
* Polyclonal
* Unicentric Castleman’s disease
* Multicentric Castleman’s disease
* Malignant
* Lymphomatoid papulosis
* Lymphomatoid granulomatosis
may be present in more advanced cases. Most com-
monly, these include a transient skin rash similar to a
drug eruption or viral exanthem, but a variety of other
skin lesions have been described.
Excisional biopsy of an involved lymph node is the
preferred diagnostic procedure for Kikuchi’s disease, as
an evaluation of the tissue architecture is crucial to dis-
tinguish it from malignant lymphoma. Inflammatory
markers, such as the erythrocyte sedimentation rate,
are usually elevated, and mild leukopenia and anemia
are also commonly seen. Although serologic testing with
antinuclear antibodies (ANA) or rheumatoid factor (RF)
is usually negative, an ANA should be checked at base-
line because patients with Kikuchi’s disease are at
increased risk for the development of systemic lupus
erythematosus (SLE), with at least 35 such cases reported
in the literature. No clear predictive markers for the
development of SLE in Kikuchi’s patients have been
identified.
Prognosis/treatment
Kikuchi’s disease is generally self-limited, with most
patients having resolution of their symptoms by 3
months. In a series of 102 patients followed in Korea
from 2001 to 2006, eight patients relapsed early and
13 relapsed late. When patients did relapse, their
symptoms tended to last longer than at initial
Chapter 17: Atypical disorders
presentation. In cases where symptoms are more
severe, a trial of glucocorticoids may be effective, and
anecdotal success has been reported with hydroxy-
chloroquine. In general, patients should be followed
for several years after the resolution of symptoms,
given the risk of late recurrence and the increased
likelihood of the development of SLE. Patients with
recurrent disease often are symptomatic for longer
than their initial episode, though there is no evidence
that early intervention can prevent recurrence or
decrease its severity.
Progressive transformation
of germinal centers
Progressive transformation of germinal centers
(PTGC) occurs primarily in young, asymptomatic
adults, who develop otherwise unexplained lymph-
adenopathy. There is approximately a 3:1 male pre-
dominance. Cervical lymphadenopathy is the most
common presentation, although extranodal presenta-
tions of PTGC in oral and gastrointestinal mucosa,
most frequently in the colon, have also been reported.
In contrast to nodal PTGC, extranodal cases have a
female predominance and tend to occur at an older
age. Both nodal and extranodal PTGC can occur at
multiple sites. Enlarged lymph nodes may wax and
wane over a period of months or years, and sometimes
regress spontaneously. No specific treatment is
generally required. Of note, PTGC can be associated
with both classical Hodgkin lymphoma and nodular
lymphocyte-predominant Hodgkin lymphoma
(NLPHD). Small prospective studies have shown that
about 2.5% of patients with PTGC will later develop
NLPHD, but causal factors underlying this association
have not been identified. In rare cases, florid presenta-
tions of PTGC have been reported, and in such sit-
uations they can be mistaken for lymphoma. It is
important to distinguish these cases pathologically
from lymphoma, as they will often be self-limited
and may not require treatment. These more aggressive
cases tend to occur in young men, and present most
often as a rapidly enlarging, solitary lymph node.
Vascular transformation of sinuses
Although vascular transformation of the sinuses
(VTS) can present as lymphadenopathy, more com-
monly it is an incidental finding noted on pathologic
examination of lymph nodes. The condition usually 287
arises in abdominal lymph nodes, but cervical lymph
 Chapter 17: Atypical disorders
node presentations have also been reported. Since VTS
may be caused by lymph node obstruction, the diag-
nosis should prompt an evaluation for causes of nodal
obstruction. A variety of factors can contribute to
lymph node obstruction, including vascular thrombo-
sis, heart failure, previous surgery, or radiotherapy.
Occult malignancy leading to obstruction is a theoret-
ical concern, but little has been published describing
this phenomenon. A case of a plexiform variant of
VTS associated with myelodysplastic syndrome was
reported, but there are not enough data to conclude
whether a causal relationship exists between these two
entities.
Inflammatory pseudotumor
of lymph nodes
Inflammatory pseudotumor of lymph nodes (IPLN),
also known as plasma cell granuloma, is a benign
condition that usually occurs in young adults who
present with unexplained lymphadenopathy. Patients
with a single site of disease may be asymptomatic, but
those with multiple sites often also have systemic
symptoms and elevated inflammatory markers in the
blood. Occasionally, IPLN will be discovered as part of
a fever of unknown etiology work-up. As the disease
develops, marked enlargement of both peripheral and
central lymph nodes can occur, and it is not uncom-
mon for nodes to enlarge to >3 cm. These large nodes
are usually self-contained, but focal infiltration of the
liver and spleen by nodes growing in, as well as paren-
chymal involvement, have also been described.
IPLN is thought to be a consequence of a resolving
infection. Although the inciting pathogens are not well
defined, an association with resolved toxoplasmosis
infection has been proposed based on a similar patho-
logic appearance. Treatment is usually not required,
although surgery has been used either with curative
intent in single site disease, or for palliation of symp-
toms in multifocal disease.
Histiocytic disorders
Histiocytic disorders comprise a wide range of disease
entities that vary greatly in their pathological features
and clinical behavior. These extremely rare conditions
in aggregate comprise less than 1% of all lymphopro-
liferative disorders. They may represent benign pro-
liferations of mononuclear phagocytes, such as
macrophages or dendritic cells, or true malignant dis-
288 eases of histiocytic origin. Immunohistochemistry is
often the key to confirming these diagnoses. For exam-
ple, expression of CD1a is helpful in confirming the
diagnosis of Langerhans cell histiocytosis. Clarifying
the diagnosis often may require multiple biopsies, but
is essential, as approaches to management vary widely.
Rosai–Dorfman disease
Presentation/diagnosis
Rosai–Dorfman disease (RDD), also known as sinus
histiocytosis with massive lymphadenopathy, is a rare
disorder characterized by histiocytic proliferation in
lymph node sinuses. The disease is more common in
males and in those of African descent. RDD has been
described following allogeneic stem cell transplanta-
tion for hematologic malignancies and concurrently
or after treatment for both Hodgkin and non-Hodgkin
lymphoma (NHL).
Patients generally present in early adulthood with
massive painless lymphadenopathy, predominantly in
the cervical area. Although the majority of cases occur
within lymph nodes, extranodal presentations are
common, occurring in about 40% of cases, usually in
the cervical region. Isolated RDD in the CNS has been
reported, generally in an extra-axial or dura-based
location. These patients can present with headache,
seizure, or other neurologic manifestations. CNS pre-
sentations occur more often in older patients and can
be mistaken radiographically for meningioma.
Extranodal presentations have also been described in
a variety of other locations; for example, in the thyroid,
kidney, and bone, the latter of which can manifest as
lytic lesions. A cutaneous form of the disease, with
isolated papular or firm nodular lesions, has also
been described. Whether cutaneous RDD is truly a
distinct entity is not certain. Systemic symptoms,
such as fever or night sweats, may be present, as may
inflammatory markers, such as the erythrocyte sedi-
mentation rate and polyclonal elevation in immuno-
globulin levels.
Prognosis/treatment
RDD patients are generally observed, since about half of
cases resolve spontaneously over the course of weeks to
months, and most other cases remain stable or wax and
wane for months or even years without causing signifi-
cant sequelae. If particular lesions grow to the point
of impairing nearby organs or threatening the
airway, surgical resection can sometimes be helpful.
Chemotherapy and radiation therapy are generally

ineffective; however, steroids may temporarily help to
decrease the size of the masses and improve systemic
symptoms in many cases. Often, low-dose prednisone is
adequate, but the optimal dosing and duration of treat-
ment is not known. A recent case report showed a
complete response to the tyrosine kinase inhibitor ima-
tinib, suggesting that targeted therapy has the potential
to play a role in this disease. The authors propose that
direct activity of the drug against histiocytes, modula-
tion of cytokine expression within the RDD lesions, or
both, probably through the inhibition of platelet-derived
growth factor receptor-beta (PDGFRB) and KIT,
account for its efficacy in this disease.
Serious complications of RDD are rare, but death
from visceral involvement or immune dysregulation can
occur, in one series at a rate of up to 7%. Patients with
immunologic abnormalities at presentation, more wide-
spread nodal disease, or extranodal presentations tend to
have a poorer prognosis. Those patients with life-
threatening disease may be candidates for chemotherapy,
with agents such as 2-chlorodeoxyadenosine (2-CDA),
low-dose methotrexate (MTX), or 6-mercaptopurine
(6-MP) showing efficacy in case reports. Optimally,
these agents should be given as part of a clinical trial.
Histiocytic sarcoma
Presentation/diagnosis
Only small series of patients with histiocytic sarcoma
have been reported in the literature. Patients tend to
present most commonly in middle age, with a median
age at diagnosis of 52, although cases have been
reported in both the elderly and pediatric populations.
Some studies have suggested a slight male predomi-
nance. Most patients present with extranodal disease,
typically in the skin, which can manifest as rash
or frank tumors, and the gastrointestinal tract
which can present as obstruction. Muscle, or other
soft tissue presentations, can occur, but are rare.
Lymphadenopathy has been described, but is also
unusual. Occasionally, patients have a single site of
disease at diagnosis, although more florid presenta-
tions with multiple sites of involvement are more
common, a condition sometimes referred to as “malig-
nant histiocytosis.” Using the NHL staging system,
about 70% of patients present with stage III/IV disease.
Even in patients with unifocal disease, systemic symp-
toms at presentation are common. As the disease pro-
gresses, patients may eventually become pancytopenic
and develop hepatosplenomegaly.
Chapter 17: Atypical disorders
Prognosis/treatment
Histiocytic sarcoma generally carries a poor prognosis,
with about 60–80% disease-related mortality. Those
with unifocal disease and no systemic symptoms tend
to have a somewhat better prognosis. Combination
chemotherapy regimens, such as cyclophosphamide,
doxorubicin, vincristine, and prednisone (CHOP), are
often employed, with variable efficacy. If patients are
able to achieve remission, consolidation with autolo-
gous stem cell transplantation and thalidomide is a
reasonable consideration.
Langerhans cell histiocytosis
Langerhans cell histiocytosis (LCH), also called “his-
tiocytosis X,” was historically known as Hand–
Schüller–Christian disease, Letterer–Siwe disease,
eosinophilic granuloma, or diffuse reticuloendothe-
liosis. These latter terms have fallen out of favor and,
for clarity, the disease should now be referred to only
as LCH.
Presentation/diagnosis
Pediatric LCH is a rare disorder, estimated to occur in
about 3–5 individuals per million children. The etiol-
ogy is unknown, although scattered reports of familial
cases have been published, and an epidemiologic study
suggested a possible association with perinatal infec-
tion and parental solvent exposure. There is a male
predominance of 3.7:1, and the disease occurs more
commonly in Caucasians of northern European
descent.
LCH often presents at a single site. For example,
skin presentations can occur in infants who have
purple/brown papules. Older children may develop
a red papular rash, ulcerative lesions, and/or sebor-
rhea of the scalp. Oral manifestations can include
ulcerations or gingival hypertrophy. Lymph node
involvement is seen frequently, with cervical node
enlargement often accompanied by lymphedema.
The most frequently involved site in pediatric LCH
is an isolated lytic lesion of the skull. These lesions
may or may not be painful, and involvement of other
bones, though less common, does occur. Bony lesions
of the skull may lead to mass effect, which, if imping-
ing on the dura, can lead to intracranial tumor exten-
sion and CNS sequelae. It should be noted that in
several reports, upon further evaluation, systemic
involvement is uncovered in nearly 60% of these
patients who appear initially to only have a single 289
site of disease.
 Chapter 17: Atypical disorders
Multisystem disease presentations are more com-
mon in infants than in older children, with involve-
ment of the lungs, bone marrow, spleen, liver, CNS,
gastrointestinal, and endocrine systems being most
common. Pulmonary manifestations can include
progressive fibrosis and pneumothorax. Bone mar-
row involvement is typically seen in young children
with diffuse disease, and can lead to pancytopenia.
Hemophagocytic syndrome may also be associated
with LCH, and can further suppress hematopoiesis.
Splenomegaly and hypersplenism may further
exacerbate cytopenias, and liver involvement can
lead to hepatomegaly and liver dysfunction, includ-
ing coagulopathy. CNS disease may lead to blockage
of cerebrospinal fluid (CSF) circulation as well as a
neurodegenerative disorder characterized by bilat-
eral symmetric cerebellar or basal ganglia lesions on
MRI. Diabetes insipidus (DI) is the most common
endocrine abnormality in LCH, and is thought to be
due to direct pituitary infiltration. Deficiencies in
growth hormone and other factors have also been
described.
Adult LCH is even rarer than the pediatric disease,
with an estimated incidence of only 1–2 per million
adults. The median age at diagnosis is 32 years, but a
range of 21–69 years has been reported. As with chil-
dren, dermatologic presentations are most common, but
pulmonary symptoms, bone pain, polydipsia and poly-
uria, lymphadenopathy, B-symptoms, and ataxia have all
been reported. Given the rarity of the disease, it can often
take several years before a diagnosis of LCH is made.
Skin presentations in adults typically involve a
reddish/brown papular rash, sometimes with oral
mucosal involvement. DI that arises without another
explanation has been found in two studies to be due to
LCH in about 15% of cases. Unlike children, where the
skull is frequently involved, adults with bony manifes-
tations typically have jaw, vertebral, and extremity
disease. These lesions are generally painful, but rarely
can be asymptomatic. Pulmonary involvement is more
common in smokers, and a reduced carbon monoxide
diffusing capacity (DLCO) is seen on pulmonary func-
tion testing. Chest CT usually reveals a characteristic
pattern of cysts and nodules, but lung biopsy is still
needed to confirm the diagnosis.
When LCH is suspected, a detailed diagnostic eval-
uation is warranted, including CT scan to determine
the extent of disease, and biopsy of suspicious lesions
with staining for CD1a and anti-langerin (CD207).
290
Prognosis/treatment
The overall prognosis for children with single site
disease without major organ involvement is excellent,
with an estimated survival of >90% in most series. In
contrast, multisystem disease carries a much more dire
prognosis, with a reported 66% mortality for young
children who do not respond promptly to therapy.
Bone marrow, liver, and lung involvement are consid-
ered to be the most significant risk factors. Progression
from single site to multisite involvement is rare, but
has been reported, particularly in infants, who overall
carry a poorer prognosis than older children or adults.
Treatment of pediatric LCH depends on whether
the patient has low-risk, single site disease, or high-risk,
multisystem disease. Low-risk patients generally receive
12 months of prednisone alone, vinblastine and predni-
sone in combination, or curettage of bony lesions. Bony
lesions at risk for collapse or CNS impingement can be
amenable to radiation. Close monitoring for evidence
of systemic disease is important. An ongoing study is
evaluating whether low-risk patients can be treated
effectively with 6 months of therapy rather than 12.
Patients with multisystem disease require intensi-
fied therapy, which can include combinations of cladri-
bine, cytarabine, and vincristine. Most of these patients
will initially achieve a complete remission, but nearly
half of these patients will relapse within 5 years. Salvage
regimens include high-dose cladribine and cytarabine.
A second complete remission can be achieved in 85% of
these cases, and there are promising but limited data
supporting the use of allogeneic bone marrow trans-
plantation as consolidation in patients achieving a com-
plete remission. Despite the potential for an aggressive
disease course, multisystem pediatric LCH still has
about an 80% chance of cure with proper therapy.
In adult patients with lung-only disease, smoking
cessation can lead to improvement in lung LCH
involvement. Those with the typical nodular pattern
often receive steroids, which may slow the progression
of pulmonary LCH. Unlike in pediatric LCH,
treatment algorithms for multisystem involvement
are less well defined. A commonly used regimen con-
sists of 6 weeks of daily prednisone and weekly vinblas-
tine followed by 28 weeks of daily 6-mercaptopurine
with vinblastine/prednisone pulses and etoposide.
Remissions have been reported with cladribine,
and a recent paper on the efficacy of methotrexate,
doxorubicin, cyclophosphamide, vincristine, predni-
sone, bleomycin (MACOP-B) in a series of seven

patients suggested that there may be a role for this
combination. There has been growing interest in
using targeted therapy with imatinib to inhibit differ-
entiation of dendritic cells from peripheral blood pro-
genitor cells. A recent report also described recurrent
BRAF mutations in LCH, suggesting that a subset of
patients may respond to RAF pathway inhibitors.
Interdigitating dendritic cell sarcoma
Only single case reports or very small series have been
published on the extremely rare interdigitating den-
dritic cell sarcoma (IDC). In the largest series of four
cases, the disease occurred predominantly in older
adults, and may be associated with NHL. The median
age at diagnosis of reported cases is 51, with a range of
2–86 years. Most commonly, patients are asympto-
matic, and present with a single enlarged lymph
node. Extranodal presentations may occur in the
lung, bone marrow, bladder, testis, oral cavity, or
breast. More extensive disease is frequently accompa-
nied by systemic symptoms, such as weight loss, night
sweats, fever, or fatigue. Hepatosplenomegaly or more
diffuse lymphadenopathy are unusual.
The disease tends to behave aggressively, with pro-
gressive growth of lesions that can then invade into
organs, eventually affecting the liver, lungs, or kidneys.
About half of patients eventually die of the disease, and
effective treatment options have been limited. For local-
ized disease, radiation has been reported to be useful.
With extensive disease, attempts have been made to
utilize NHL-like therapy, with limited success. The
Hodgkin chemotherapy regimen, doxorubicin, vinblas-
tine, bleomycin, dacarbazine (ABVD), was also utilized
successfully in a patient with extensive IDC.
Follicular dendritic cell sarcoma
Follicular dendritic cell (FDC) sarcoma is a rare neo-
plasm of unknown etiology. The hyaline-vascular type
of Castleman’s disease may be associated with FDC,
usually developing either prior to or simultaneous
with FDC. The disease presents at a median age at
diagnosis of 44 years. FDC is confined to the lymph
nodes in up to two-thirds of cases, most often in the
cervical region. Intra-abdominal lymph nodes may be
larger, as they often present later in the course of dis-
ease. When FDC sarcoma spreads, the lungs and liver
are the most common sites involved. Extranodal pre-
sentations have also been described in up to one-third
of patients, including in the gastrointestinal tract, oral
Chapter 17: Atypical disorders
cavity, skin, soft tissue, liver, and spleen. The most
common presentation is that of a painless mass that
slowly enlarges, but can grow to large sizes (median size
5 cm). Systemic symptoms are uncommon, but there
are reports of a possible association with pemphigus.
The disease course is generally indolent, and most
patients can be successfully treated with complete sur-
gical excision. Local recurrence occurs in more than
50% of cases, and may be delayed for several years.
Larger tumors (>6 cm), high-grade histology, or an
intra-abdominal location are all poor prognostic fac-
tors. Adjuvant radiotherapy and/or chemotherapy
have been utilized in selected cases, but are generally
reserved for recurrent disease. More widespread dis-
ease may eventually develop in up to 25% of patients,
and at least 10% of patients ultimately die of the
disease.
Erdheim–Chester disease
Erdheim–Chester disease (ECD) is a rare disorder
similar in its distribution of disease sites to LCH but
pathologically distinct, in that ECD is typically S-100
negative, and lacks Birbeck granules. In contrast to
LCH, ECD tends to present at an older age, with a
median age at diagnosis of 53 years. ECD also involves
symmetric, sclerotic lesions in the long bones, as
opposed to the lytic, axial lesions in LCH. The most
common symptom at presentation in ECD is bone
pain, with systemic symptoms also frequently present
at diagnosis. Cough and dyspnea are frequent, as 20–
30% of patients have lung involvement, often seen as
interstitial infiltrates or pleural disease on X-ray. X-
rays may also reveal sclerosis of the diaphyses and
metaphyses, which typically spares the epiphyses.
Radiologic circumferential sheathing, or coating, of
the aorta, is considered to be pathognomonic of
ECD. Extraosseous disease can include pituitary
involvement, leading to DI, cerebral infiltration, pul-
monary fibrosis, pericardial infiltration, renal infiltra-
tion, retroperitoneal involvement, and cutaneous
xanthomas. In general, the greater the burden of extra-
osseous disease the more aggressive the clinical course.
Most patients die within 3 years of diagnosis, often
due to progressive renal dysfunction, pulmonary fib-
rosis, and/or congestive heart failure. Various combi-
nations of corticosteroids, chemotherapy, and
radiation have been attempted, but none have shown
a consistent effect in reducing the mortality of the
disease. Improvement of clinical manifestations, 291
including regression of bony lesions, has been
 Chapter 17: Atypical disorders
reported with interferon alfa-2a, which is now used
commonly as a first line treatment for ECD. As under-
standing of the molecular underpinnings of this dis-
order improves, there is hope that more targeted
agents will be even more effective. A recent report
found PDGFRB expression in 32/37 patients (86.5%)
and utilized imatinib to achieve stable disease in 2/6
(33%) patients with advanced, treatment-refractory
disease. Another group recently reported using imati-
nib as frontline therapy in one patient with ECD, and
was able to achieve both symptomatic and radio-
graphic improvement.
Hemophagocytic lymphohistiocytosis
Presentation/diagnosis
Primary hemophagocytic lymphohistiocytosis
(HLH) is a rare, aggressive, and often life-threatening
disease that most commonly occurs in infants under
the age of 18 months, but may also present in older
children and adults. The incidence has been reported
to be of the order of 1.2 cases per million, but the true
incidence is likely higher given the difficulty of mak-
ing the diagnosis. In contrast, secondary HLH is a
more common phenomenon, occurring secondary to
a variety of infections, autoimmune conditions,
immune deficiencies, drug-induced hypersensitivity
syndromes, leukemias, and lymphomas. Symptoms
are similar, with fever the most common presenting
symptom, and hepatosplenomegaly is present in
nearly all cases. In fact, the signs and symptoms of
HLH can mimic common infections, and the disease
is part of the differential diagnosis of fever of
unknown origin. Neurologic symptoms (including
encephalitis), rash, and lymphadenopathy are also
commonly present at diagnosis. In severe cases,
acute respiratory distress syndrome attributable to
HLH has been described. Familial cases comprise
about 25% of all primary HLH, and are frequently
seen in association with consanguinity, consistent
with their autosomal recessive inheritance.
Mutations have been described in Munc 13–4, per-
forin, and syntaxin, and testing for these mutations is
available to confirm the diagnosis of primary HLH
and to screen family members.
The diagnosis of HLH can be difficult because of the
large number of mimicking conditions, so a high index
of suspicion is required. The major diagnostic criteria
include tissue demonstration of hemophagocytosis in
292
the setting of cytopenias (of at least two lines), fever,
splenomegaly, and hypertriglyceridemia or hypofibri-
nogenemia. Minor criteria include serum ferritin con-
centration >500 ug/L, and specialized testing, such as
the soluble CD25 (sIL-2 receptor) >2400 U/mL, and low
or absent NK-cell activity. If the clinical suspicion is
high, but hemophagocytosis cannot be demonstrated
pathologically, a repeat biopsy should be performed, as
this finding is frequently missed on marrow
examination. Notably, serum ferritin concentrations
may be highly elevated (in the tens of thousands
range) and values in this range should prompt consid-
eration of HLH in the differential diagnosis.
Prognosis/treatment
The prognosis and treatment of patients with secondary
HLH depends on the underlying inciting etiology.
Prompt treatment of this underlying disorder often
results in resolution of the manifestations of HLH. In
contrast, the prognosis for young children diagnosed
with primary HLH is grim, with over 90% mortality
despite treatment. Therapy for primary HLH should be
initiated immediately upon diagnosis, to avoid the com-
plications of multiorgan system failure seen with more
advanced disease. In the HLH-94 protocol, children
were given induction with dexamethasone, etoposide,
cyclosporine, and intrathecal methotrexate, followed by
pulses of dexamethasone and etoposide for up to 1 year.
This study showed an overall survival (OS) of 55%, with
a median follow-up of 3.1 years. For patients respond-
ing poorly to induction therapy, those with familial
HLH, and those with CNS disease, allogeneic stem cell
transplantation provides the best hope of cure, with
64% of patients on the HLH-94 protocol who went on
to have transplantation alive, with a median follow-up
of 4.1 years. Refinements to this regimen and to the
diagnostic criteria for HLH have been made following
the HLH-2004 study. A recent report also demonstrated
significantly improved survival in patients undergoing
reduced intensity stem cell transplantation compared to
myeloablative conditioning.
Of note, the related macrophage activation syn-
drome (MAS) can be considered an acquired (or sec-
ondary) form of HLH that is most commonly seen in
patients with systemic lupus erythematosus or juvenile
rheumatoid arthritis. Treatment is generally more
straightforward and effective than treatment of pri-
mary HLH. Immunosuppression, usually including a
combination of cyclosporine and steroids, leads to
resolution of fever and cytopenias within a few days
of starting treatment in most patients.

Atypical lymphoproliferative
disorders
Polyclonal disorders
Unicentric Castleman’s disease
Presentation/diagnosis
In 1956, Castleman and colleagues described a series
of patients with mediastinal masses with hyaline vas-
cularization that were similar, but pathologically dis-
tinct from thymomas. As is typical in unicentric
Castleman’s disease (UCD), these patients tended to
be young adults who presented with large solitary
mediastinal masses or enlarged single peripheral
lymph nodes in the absence of constitutional symp-
toms. Since the majority of patients are asympto-
matic, today the disease is most commonly detected
incidentally on imaging. Most of the lesions discov-
ered are nodal, though extranodal presentations have
also been described. About 70% of the time, the
lesion is located in the mediastinum or hilum, with
the abdomen being the next most common site of
disease.
Although UCD was initially believed to be
one disease, a second, less common variant was sub-
sequently described in 10–20% of patients, and
demonstrated extensive plasmacytosis in the inter-
follicular region. This plasma cell variant occurs in a
similar age group, and also often presents with a
mediastinal mass; however, there are important dif-
ferences with the hyaline vascular type of UCD. First,
patients with the plasma cell variant usually present
with an aggregate of enlarged lymph nodes, as
opposed to a single node. Second, about 50% of
patients will have constitutional symptoms, includ-
ing fever, weight loss, night sweats, and fatigue.
Many patients will also have laboratory abnormal-
ities, including leukocytosis, anemia, hypoalbumin-
emia, and polyclonal hypergammaglobulinemia.
These systemic manifestations of the disease are
thought possibly to be mediated by interleukin-6
(IL-6), which has been shown to have elevated levels
in patients with the plasma cell, but not the hyaline
vascular, form of UCD.
Prognosis/treatment
It is not clear whether early treatment of UCD leads to
improved outcomes. Nonetheless, most patients do
opt for surgical resection of their lesion(s). Complete
Chapter 17: Atypical disorders
resection is generally curative, with no recurrences
reported in published series. When it is not possible
to achieve complete surgical resection, the prognosis
remains excellent. Indeed, partially resected or unre-
sected masses may remain stable and not cause symp-
toms for many years. In UCD patients with the plasma
cell variant, anti-IL-6 therapy could theoretically help
ameliorate constitutional symptoms, but thus far the
use of this therapy has mainly been studied in patients
with multicentric disease.
Multicentric Castleman’s disease
Presentation/diagnosis
Although pathologically similar to the plasma cell
variant of UCD, multicentric Castleman’s disease
(MCD) is distinct with regard to its presentation,
clinical behavior, and treatment. First described in
1978, MCD typically presents in older individuals,
with a median age at diagnosis between 50 and 60. It
occurs at multiple anatomic sites, most commonly
lymph nodes, but also frequently with hepatic and
splenic involvement. Patients commonly present with
fever, night sweats, fatigue, and weight loss, and also
often have palpable hepatosplenomegaly. Laboratory
abnormalities similar to those seen in the plasma cell
variant of UCD are also frequent.
An important observation was that the disease is
more common in patients with immune dysregulation,
in particular patients with human immunodeficiency
virus (HIV). HIV-negative patients who get MCD tend
to be older and may also have altered immunoregula-
tion, such as immune senescence, that makes them
more susceptible to developing the disease. The subse-
quent discovery of the association of MCD with
Kaposi’s sarcoma (KS) linked MCD to HHV8, which
is known to be the cause of KS as well as primary
effusion lymphomas in HIV patients. Several studies
have now demonstrated a clear link between HHV8 and
MCD. The virus is universally detectable in HIV-
positive MCD patients, and is present in up to 50% of
HIV-negative MCD patients. It is unusual for HHV8
to be found in patients with UCD, although it has also
been described.
Prognosis/treatment
The clinical course of MCD is highly variable. At one
extreme is the chronic, persistent form of the disease,
which can behave quite indolently. This indolent var-
iant can often be observed for years without the need
for intervention. The other extreme is a highly 293
 Chapter 17: Atypical disorders
aggressive, rapidly progressive form of the disease,
which can lead to death within weeks of diagnosis.
This more aggressive variant appears to be more com-
mon in HIV-positive patients. In between these two
extremes, the disease exhibits a large variety of behav-
ior. For example, many patients have disease that
behaves in an aggressive manner for a limited period
of time, only to have spontaneous regression for a
period of time, and then subsequent relapse.
When viewed in the aggregate, however, MCD
carries a poor prognosis. HIV-positive patients with
MCD fare the worst, with a median survival time
of only 8–14 months, and an overall mortality rate of
70–85%. HIV-negative MCD patients do slightly bet-
ter, but still only have median survival times of 26–30
months. The most common causes of death are the
development of overwhelming infection or of second
malignancies, such as KS and/or NHL. In particular,
growing evidence suggests that MCD patients are at
risk for developing HHV8-associated, plasmablastic
microlymphomas within affected nodes, which are
difficult to treat and confer a poor prognosis.
A wide variety of treatment options have been tried
in MCD, but no clear standard of care has been
defined. Unlike in UCD, surgery generally is not fea-
sible or effective in MCD, because of the multifocal
nature of the disease. In HIV-negative MCD patients,
steroids may help alleviate systemic symptoms, and in
some cases induce a transient remission. The response
rate is of the order of 65%, with about 15% achieving a
complete response. HIV-positive patients generally
benefit more from chemotherapy-based approaches.
Both etoposide and vinblastine have been studied as
single agents, with most patients at least achieving a
partial response and relief of symptoms; however, the
disease tends to relapse upon discontinuation of ther-
apy, leading some to advocate for intermittent main-
tenance therapy. Patients with more aggressive MCD,
regardless of HIV status, should be considered for
combination chemotherapy approaches. NHL-type
regimens such as CHOP have demonstrated about a
50% durable complete response rate.
Recently, there has been promising activity of the
anti-CD20 monoclonal antibody rituximab in MCD.
In HIV-positive patients recently treated with single
agent chemotherapy, four weekly infusions of rituxi-
mab were able to maintain a sustained remission in
over 70% of patients at 1 year. Another study utilized
weekly rituximab for 4 weeks, and achieved a partial
294 response in two-thirds of patients, with 2-year relapse-
free survival of nearly 80%, though reactivation of KS
was observed in several patients. There has been inter-
est in blocking the cytokine-mediated symptoms in
MCD that are primarily related to elevated IL-6 levels.
Neutralizing IL-6 monoclonal antibodies have been
studied, and appear to be efficacious in providing
symptom relief, but the results are usually transient,
with a return of symptoms upon discontinuation of
therapy. Antiviral agents, such as ganciclovir, are able
to reduce HHV8 viral load, but have shown only
modest clinical benefit in the small number of patients
studied to date.
Malignant disorders
Lymphomatoid papulosis
Lymphomatoid papulosis (LP) is a histologically
malignant cutaneous T-cell lymphoproliferative dis-
order that clinically usually behaves indolently. The
disease typically presents with multiple, small
(<3 cm) skin papules that can wax and wane in size
over a period of several months to years. The disease
tends to be a chronic, self-healing condition, but does
carry a substantial risk of transformation into frank
lymphoma, in particular cutaneous T-cell lym-
phoma, Hodgkin lymphoma, or anaplastic large cell
lymphoma. Estimates of the overall risk for progres-
sion to lymphoma vary widely in the literature, from
20% to 80%. It does appear that the risk of progres-
sion is cumulative, as summarized in a review of
four major studies which each showed substantial
cumulative lymphoma risk over time. Some have
proposed that, because of this substantial risk of
transformation, early treatment with agents such as
methotrexate or interferon-alpha may slow or
even prevent this progression. However, it should
be noted that, to date, no currently available treat-
ment has been convincingly shown to lower the
risk of transformation. Furthermore, the survival of
patients who develop lymphoma after LP has not
been systematically studied. Until more data are
available, treatment of LP should therefore generally
be reserved for symptomatic patients.
Lymphomatoid granulomatosis
Presentation/diagnosis
Lymphomatoid granulomatosis (LYG) was first
described in the early 1970s, primarily in patients with

multiple nodular pulmonary lesions that pathologically
resembled a pulmonary angitis similar to Wegener’s
granulomatosis. Most patients with LYG are middle-
aged, with a median age at diagnosis of 48 years. A 2:1
male predominance has been described, and the disease
is more common in western countries. Patients most
commonly present with respiratory symptoms, includ-
ing cough and shortness of breath, and fever is also
often present at initial diagnosis. Other systemic symp-
toms, such as malaise and weight loss, are present in
about one-third of patients. On chest imaging, a bilat-
eral, nodular pattern of infiltrates is often observed,
classically with small lesions (1 cm or less) in the mid
to lower lungs fields. Extrapulmonary presentations are
less common, with the skin (25–50%), kidney (32%),
and neurologic system (20–30%) involved most com-
monly. Skin lesions can vary from an erythematous
rash to subcutaneous nodules or ulceration. Lymph
node involvement is less common at diagnosis, but
may develop as the disease progresses.
The lung is almost always involved at some point
in the disease even if the primary site of presentation
is extrapulmonary. The neoplastic cell component is
mainly angiocentric and angioinvasive but there is
often a florid background population of reactive
small lymphocytes, immunoblastic cells, plasma
cells, and histiocytes. Lesions are graded according
to the mix of neoplastic cellular appearances. Grade 1
lesions contain a lymphoid infiltrate with absent large
cells and little pleomorphism, grade 2 lesions have
scattered large cells, while grade 3 lesions have more
abundant large cells that are easily apparent and may
include cells with morphology reminiscent of Reed–
Sternberg cells.
Immunophenotypically the neoplastic cells of this
disorder are CD20-positive B-cells. CD79a may be
absent in some cases. Staining for CD30 is variable.
There is always a population of Epstein–Barr virus
(EBV)-positive cells that usually express LMP-1, and
the proportion of these cells correlates with the histo-
logical grading, being scanty in grade 1 lesions and
numerous in grade 3 lesions. The background popu-
lation is mainly T-cells, most of which express CD4.
Prognosis/treatment
The prognosis for LYG patients is quite variable; while
spontaneous resolution occurs in about 20% of
patients, the majority of patients develop progressive
disease. In a large study of 152 patients, the overall
mortality rate due to LYG was 67%, with a median OS
Chapter 17: Atypical disorders
of 14 months. Patients most commonly succumb to
pulmonary complications, but development of CNS
disease, infection, or lymphoma can also be potentially
fatal. Treatment strategies are guided by the histopa-
thologic grade, severity of symptoms, and extent of
extrapulmonary disease. Asymptomatic patients with
low-grade (grade 1 or 2) disease in the lungs only can
be followed closely with observation, including serial
imaging. Many of these patients will spontaneously
remit. Those who develop progressive disease, those
with grade 3 disease, or those with CNS involvement
require more aggressive systemic therapy. A small
prospective study of 15 patients suggested that combi-
nation cyclophosphamide and steroids could lead to
prolonged complete remissions in the majority of
patients. Low-risk patients have also been treated
with some success with the antiproliferative cytokine
interferon alfa-2b. A recent NIH study of 31 grade I/II
LYG patients treated with interferon alfa-2b showed a
61% complete response (CR) rate, with a 90% CR rate
in patients with CNS disease. They also reported a
progression-free survival of 56%, with a median
follow-up time of 5 years. High-risk (grade 3) patients
may benefit from combination chemotherapy regi-
mens such as CHOP in combination with rituximab,
which was highly efficacious in a recent case report.
Autologous stem cell transplantation may also play a
role in the treatment of LYG, although more data are
needed.
Further reading
Andriko JA, Morrison A, Colegial CH, Davis BJ, Jones RV.
Rosai–Dorfman disease isolated to the central nervous
system: a report of 11 cases. Mod Pathol,
2001;14(3):172–178.
Appel S, Boehmler AM, Grunebach F, et al. Imatinib
mesylate affects the development and function of
dendritic cells generated from CD34+ peripheral blood
progenitor cells. Blood, 2004;103(2):538–544.
Badalian-Very G, Vergilio JA, Degar BA, et al. Recurrent
BRAF mutations in Langerhans cell histiocytosis. Blood,
16(11):1919–1923.
Beck JT, Hsu SM, Wijdenes J, et al. Brief report: alleviation of
systemic manifestations of Castleman’s disease by
monoclonal anti-interleukin-6 antibody. N Engl J Med,
1994;330(9):602–605.
Bower M, Powles T, Williams S, et al. Brief communication:
rituximab in HIV-associated multicentric Castleman
disease. Ann Intern Med, 2007;147(12):836–839.
Cabanillas F, Armitage J, Pugh WC, Weisenburger D, Duvic 295
M. Lymphomatoid papulosis: a T-cell dyscrasia with a
 Chapter 17: Atypical disorders
propensity to transform into malignant lymphoma. Ann
Intern Med, 1995;122(3):210–217.
Casper C, Nichols WG, Huang ML, Corey L, Wald A.
Remission of HHV-8 and HIV-associated multicentric
Castleman disease with ganciclovir treatment. Blood,
2004;103(5):1632–1634.
Castleman B, Iverson L, Menendez VP. Localized
mediastinal lymph node hyperplasia resembling
thymoma. Cancer, 1956;9(4):822–830.
Chadburn A, Cesarman E, Nador RG, Liu YF, Knowles DM.
Kaposi’s sarcoma-associated herpesvirus sequences in
benign lymphoid proliferations not associated with
human immunodeficiency virus. Cancer,
1997;80(4):788–797.
Chang Y, Cesarman E, Pessin MS, et al. Identification
of herpesvirus-like DNA sequences in AIDS-
associated Kaposi’s sarcoma. Science, 1994;266
(5192):1865–1869.
Davis RE, Warnke RA, Dorfman RF. Inflammatory
pseudotumor of lymph nodes. Additional observations
and evidence for an inflammatory etiology. Am J Surg
Pathol, 1991;15(8):744–756.
Derenzini E, Fina MP, Stefoni V, et al. MACOP-B regimen
in the treatment of adult Langerhans cell histiocytosis:
experience on seven patients. Ann Oncol,
2010;21(6):1173–1178.
Dunleavy K, Chattopadhyay P, Kawada J, et al. Immune
Characteristics Associated with Lymphomatoid
Granulomatosis and Outcome Following Treatment with
Interferon-Alpha. ASH Annual Meeting; 2010.
Fauci AS, Haynes BF, Costa J, Katz P, Wolff SM.
Lymphomatoid granulomatosis. Prospective clinical and
therapeutic experience over 10 years. N Engl J Med,
1982;306(2):68–74.
Foucar E, Rosai J, Dorfman RF. Sinus histiocytosis with
massive lymphadenopathy. An analysis of 14 deaths
occurring in a patient registry. Cancer,
1984;54(9):1834–1840.
Foucar E, Rosai J, Dorfman R. Sinus histiocytosis with
massive lymphadenopathy (Rosai–Dorfman disease):
review of the entity. Semin Diagn Pathol,
1990;7(1):19–73.
Frizzera G. Castleman’s disease and related disorders. Semin
Diagn Pathol, 1988;5(4):346–364.
Frizzera G, Peterson BA, Bayrd ED, Goldman A. A systemic
lymphoproliferative disorder with morphologic features
of Castleman’s disease: clinical findings and
clinicopathologic correlations in 15 patients. J Clin Oncol,
1985;3(9):1202–1216.
Gaba AR, Stein RS, Sweet DL, Variakojis D. Multicentric
giant lymph node hyperplasia. Am J Clin Pathol,
1978;69(1):86–90.
296
Gadner H, Grois N, Arico M, et al. A randomized trial of
treatment for multisystem Langerhans’ cell histiocytosis.
J Pediatr, 2001;138(5):728–734.
Gadner H, Grois N, Potschger U, et al. Improved outcome
in multisystem Langerhans cell histiocytosis is
associated with therapy intensification. Blood,
2008;111(5):2556–2562.
Gerard L, Berezne A, Galicier L, et al. Prospective study of
rituximab in chemotherapy-dependent human
immunodeficiency virus associated multicentric
Castleman’s disease: ANRS 117 Castleman Trial. J Clin
Oncol, 2007;25(22):3350–3356.
Gergis U, Dax H, Ritchie E, et al. Autologous hematopoietic
stem-cell transplantation in combination with thalidomide
as treatment for histiocytic sarcoma: a case report and
review of the literature. J Clin Oncol, 2011 Jan 10.
Gessain A, Briere J, Angelin-Duclos C, et al. Human herpes
virus 8 (Kaposi’s sarcoma herpes virus) and malignant
lymphoproliferations in France: a molecular study of 250
cases including two AIDS-associated body cavity based
lymphomas. Leukemia, 1997;11(2):266–272.
Gruber R, Sepp NT, Fritsch PO, Schmuth M. Prognosis of
lymphomatoid papulosis. Oncologist,
2006;11(8):955–957; author reply 7.
Haroche J, Amoura Z, Trad SG, et al. Variability in the
efficacy of interferon-alpha in Erdheim–Chester disease
by patient and site of involvement: results in eight
patients. Arthritis Rheum, 2006;54(10):3330–3336.
Haroche J, Amoura Z, Charlotte F, et al. Imatinib
mesylate for platelet-derived growth factor receptor-
beta-positive Erdheim–Chester histiocytosis. Blood,
2008;111(11):5413–5415.
Hehn ST, Grogan TM, Miller TP. Utility of fine-needle
aspiration as a diagnostic technique in lymphoma.
J Clin Oncol, 2004;22(15):3046–3052.
Henter JI, Horne A, Arico M, et al. HLH-2004: diagnostic
and therapeutic guidelines for hemophagocytic
lymphohistiocytosis. Pediatr Blood Cancer,
2007;48(2):124–131.
Herrada J, Cabanillas F, Rice L, Manning J, Pugh W. The
clinical behavior of localized and multicentric Castleman
disease. Ann Intern Med, 1998;128(8):657–662.
Horne A, Janka G, Maarten Egeler R, et al. Haematopoietic
stem cell transplantation in haemophagocytic
lymphohistiocytosis. Br J Haematol,
2005;129(5):622–630.
Horneff G, Jurgens H, Hort W, Karitzky D, Gobel U. Sinus
histiocytosis with massive lymphadenopathy
(Rosai–Dorfman disease): response to methotrexate and
mercaptopurine. Med Pediatr Oncol, 1996;27(3):187–192.
Hornick JL, Jaffe ES, Fletcher CD. Extranodal histiocytic
sarcoma: clinicopathologic analysis of 14 cases of a rare

epithelioid malignancy. Am J Surg Pathol,
2004;28(9):1133–1144.
Jang YJ, Park KH, Seok HJ. Management of Kikuchi’s
disease using glucocorticoid. J Laryngol Otol,
2000;114(9):709–711.
Janku F, Amin HM, Yang D, et al. Response of histiocytoses
to imatinib mesylate: fire to ashes. J Clin Oncol,
28(31):e633–636.
Jindal B, Vashishta RK, Bhasin DK. Vascular transformation
of sinuses in lymph nodes associated with
myelodysplastic syndrome–a case report. Indian J Pathol
Microbiol, 2001;44(4):453–455.
Johnston A, Coyle L, Nevell D. Prolonged remission of
refractory lymphomatoid granulomatosis after
autologous hemopoietic stem cell transplantation with
post-transplantation maintenance interferon. Leuk
Lymphoma, 2006;47(2):323–328.
Jung KH, Sung HJ, Lee JH, et al. A case of pulmonary
lymphomatoid granulomatosis successfully treated by
combination chemotherapy with rituximab.
Chemotherapy, 2009;55(5):386–390.
Katzenstein AL, Carrington CB, Liebow AA. Lymphomatoid
granulomatosis: a clinicopathologic study of 152 cases.
Cancer, 1979;43(1):360–373.
Keller AR, Hochholzer L, Castleman B. Hyaline-vascular
and plasma-cell types of giant lymph node hyperplasia of
the mediastinum and other locations. Cancer,
1972;29(3):670–683.
Kojima M, Nakamura N, Sakamoto K, et al. Progressive
transformation of the germinal center of extranodal
organs: a clinicopathological, immunohistochemical,
and genotypic study of 14 cases. Pathol Res Pract,
2010;206(4):235–240.
Kudo K, Ohga S, Morimoto A, et al. Improved outcome of
refractory Langerhans cell histiocytosis in children with
hematopoietic stem cell transplantation in Japan. Bone
Marrow Transplant, 2010;45(5):901–906.
Leger-Ravet MB, Peuchmaur M, Devergne O, et al.
Interleukin-6 gene expression in Castleman’s disease.
Blood, 1991;78(11):2923–2930.
Marsh RA, Vaughn G, Kim MO, et al. Reduced-intensity
conditioning significantly improves survival of patients
with hemophagocytic lymphohistiocytosis
undergoing allogeneic hematopoietic cell
transplantation. Blood, 2010;116(26):5824–5831.
McCarty MJ, Vukelja SJ, Banks PM, Weiss RB.
Angiofollicular lymph node hyperplasia (Castleman’s
disease). Cancer Treat Rev, 1995;21(4):291–310.
Minkov M, Grois N, Heitger A, et al. Response to initial
treatment of multisystem Langerhans cell histiocytosis:
an important prognostic indicator. Med Pediatr Oncol,
2002;39(6):581–585.
Chapter 17: Atypical disorders
Minkov M, Prosch H, Steiner M, et al. Langerhans cell
histiocytosis in neonates. Pediatr Blood Cancer,
2005;45(6):802–807.
Minkov M, Steiner M, Potschger U, et al. Reactivations in
multisystem Langerhans cell histiocytosis: data of the
international LCH registry. J Pediatr,
2008;153(5):700–705.
Miras-Parra FJ, Parra-Ruiz J, Gomez-Morales M,
Gomez-Jimenez FJ, de la Higuera-Torres-Puchol J.
Inflammatory pseudotumor of lymph nodes with focal
infiltration in liver and spleen. Dig Dis Sci, 2003; 48(10):
2003–2004.
Nakamura S, Koshikawa T, Kitoh K, et al.
Interdigitating cell sarcoma: a morphologic and
immunologic study of lymph node lesions in four cases.
Pathol Int, 1994;44(5):374–386.
Oksenhendler E, Duarte M, Soulier J, et al.
Multicentric Castleman’s disease in HIV infection: a
clinical and pathological study of 20 patients. AIDS,
1996;10(1):61–67.
Olnes MJ, Nicol T, Duncan M, Bohlman M, Erlich R.
Interdigitating dendritic cell sarcoma: a rare malignancy
responsive to ABVD chemotherapy. Leuk Lymphoma,
2002;43(4):817–821.
Osborne BM, Butler JJ, Gresik MV. Progressive
transformation of germinal centers: comparison of 23
pediatric patients to the adult population. Mod Pathol,
1992;5(2):135–140.
Peterson BA, Frizzera G. Multicentric Castleman’s disease.
Semin Oncol, 1993;20(6):636–647.
Pulsoni A, Anghel G, Falcucci P, et al. Treatment of sinus
histiocytosis with massive lymphadenopathy (Rosai–
Dorfman disease): report of a case and literature review.
Am J Hematol, 2002;69(1):67–71.
Rezai K, Kuchipudi S, Chundi V, et al. Kikuchi–Fujimoto
disease: hydroxychloroquine as a treatment. Clin Infect
Dis, 2004;39(12):e124–126.
Rosai J, Dorfman RF. Sinus histiocytosis with
massive lymphadenopathy. A newly recognized
benign clinicopathological entity. Arch Pathol,
1969;87(1):63–70.
Samet A, Gilbey P, Talmon Y, Cohen H. Vascular
transformation of lymph node sinuses. J Laryngol Otol,
2001;115(9):760–762.
Scott D, Cabral L, Harrington WJ, Jr. Treatment of
HIV-associated multicentric Castleman’s
disease with oral etoposide. Am J Hematol,
2001;66(2):148–150.
Song JY, Lee J, Park DW, et al. Clinical outcome
and predictive factors of recurrence among patients
with Kikuchi’s disease. Int J Infect Dis,
2009;13(3):322–326. 297
 Chapter 17: Atypical disorders
Soulier J, Grollet L, Oksenhendler E, et al. Kaposi’s
sarcoma-associated herpesvirus-like DNA sequences
in multicentric Castleman’s disease. Blood,
1995;86(4):1276–1280.
Spannuth WA, Leath CA, 3rd, Conner MG, Alvarez RD.
Inflammatory pseudotumor of pelvic
lymph nodes. Obstet Gynecol, 2006;108(3 Pt 2):
779–782.
Utikal J, Ugurel S, Kurzen H, et al. Imatinib as a
treatment option for systemic non-
Langerhans cell histiocytoses. Arch Dermatol,
2007;143(6):736–740.
Verma A, Stock W, Norohna S, et al. Progressive
transformation of germinal centers. Report of 2 cases
and review of the literature. Acta Haematol,
2002;108(1):33–38.
298
Veyssier-Belot C, Cacoub P, Caparros-Lefebvre D, et al.
Erdheim–Chester disease. Clinical and radiologic
characteristics of 59 cases. Medicine (Baltimore),
1996;75(3):157–169.
Wang KH, Chen WY, Liu HN, et al. Cutaneous Rosai–
Dorfman disease: clinicopathological profiles, spectrum
and evolution of 21 lesions in six patients. Br J Dermatol,
2006;154(2):277–286.
Wilson WH, Kingma DW, Raffeld M, Wittes RE, Jaffe ES.
Association of lymphomatoid granulomatosis
with Epstein–Barr viral infection of B
lymphocytes and response to interferon-alpha
2b. Blood, 1996;87(11):4531–4537.
Zietz C, Bogner JR, Goebel FD, Lohrs U. An unusual cluster
of cases of Castleman’s disease during highly active
antiretroviral therapy for AIDS. N Engl J Med,
1999;340(24):1923–1924.
Index
ABVD regimen, 71, 73, 74, 280
ABVE-PC regimen, 80
accelerated titration method, 52
acitretin, 264
acrocyanosis, 141, 142
activated B cell (ABC)-like DLBCL, 16
acute lymphoblastic leukemia
(ALL), 172
See also lymphoblastic lymphoma.
acute myelogenous leukemia, 145, 163
acute respiratory distress syndrome
(ARDS)
complication of therapy, 41–42
ACVB regimen, 234
ACVBP regimen, 197
adnexal ocular MALT lymphoma
association with Chlamydophila
psittaci, 24, 114–115
adolescents
treatment of Hodgkin lymphoma,
80–81
Adriamycin, 55, 69, 74, 79, 80, 234, 243,
244, 245
adult T-cell leukemia/lymphoma
(ATLL), 226, 230–232
association with HTLV-1, 8, 230
cytogenetics, 230
molecular pathology, 230
pathology, 230
subtypes, 230
treatment, 230–232
aetiology of lymphoma, 7–10
autoimmune disease risk factor, 9
challenges for aetiological studies, 7
effects of immunosuppression, 8–9
environmental risk factors, 9
InterLymph consortium studies, 9–10
organ transplantation risk factor, 9
role of Helicobacter pylori, 8
role of immune system
dysregulation, 7–10
role of infection, 7–8
sex-specific associations with
lymphomas, 9
age
influence on lymphoma rates, 2–6
prognostic factor for Burkitt
lymphoma, 24–25
AIDS. See HIV/AIDS; HIV-related
lymphomas
alcohol
potential risk factor for lymphoma, 9
alemtuzumab, 129–130, 144, 240, 242,
249–250, 264
ALK protein
expression in PTCLs, 23
alkeran, 74
alkylator based therapy, 93
allele-specific oligonucleotide
polymerase chain reaction
(ASO-PCR), 125
allogenic stem cell transplantation, 77–78
in chronic lymphocytic leukemia, 131
in DLBCL, 204
in follicular lymphoma, 99–100
in mantle cell lymphoma, 165
in secondary CNS lymphomas, 222
in T-cell non-Hodgkin lymphoma,
248–249
alpha-heavy-chain disease, 115
American Society of Hematology, 48
amoxicillin, 111
amphotericin B, 41
ampicillin, 116
anaplastic large-cell lymphoma
(ALCL), 23
anaplastic lymphoma kinase
negative (ALK-) subtype, 226
anaplastic lymphoma kinase positive
(ALK+) subtype, 226
anaplastic large-cell lymphoma
(ALCL), systemic type, 236–240
cytogenetics, 239
distinction from primary cutaneous
ALCL, 236
division into ALK-positive and ALK-
negative types, 236
incidence, 236
molecular pathology, 239
pathology (ALK-negative), 239
pathology (ALK-positive), 236–239
treatment, 239–240
angiocentric lymphoma, 243
angioimmunoblastic
lymphadenopathy with
dysproteinemia, 234
angioimmunoblastic T-cell lymphoma
(AITL), 23, 226, 234–236
cytogenetics, 236
disease features, 234
incidence, 234
molecular pathology, 236
pathology, 234–236
prognostic factors, 234
treatment, 236
angioinvasive aspergillosis, 42–43
Ann Arbor staging classification, 38, 177
DLBCL, 12–14
Ann Arbor staging classification,
Cotswolds revision, 27, 38
Hodgkin lymphoma, 26–27, 68
ansoprazole, 111
anthracycline, 94, 191, 236
antibody-dependent cell-mediated
cytotoxicity (ADCC), 147
anti-CD20 monoclonal antibodies, 18
anti-CD30 monoclonal antibodies/
immunotoxins, 78–79
antilymphocyte globulin (ATG), 274
ara-C. See cytarabine
ASHAP regimen, 76, 77
L-asparaginase, 244
ataxia telangiectasia, 8, 158, 273
atypical lymphoproliferative
disorders
classification, 286
diagnostic challenges, 286
lymphomatoid granulomatosis,
294–295
lymphomatoid papulosis, 294
malignant disorders, 294–295
multicentric Castleman’s disease,
293–294
polyclonal disorders, 293–294
unicentric Castleman’s disease, 293
autoimmune disease
association with lymphoproliferative
disorders, 273
associations with lymphomas, 9
autologous stem cell transplantation
(ASCT), 75–77
in Burkitt lymphoma, 181
in CNS lymphoma, 214
in DLBCL, 199–200
in follicular lymphoma first
remission, 96
in mantle cell lymphoma, 162, 163–164
in relapsed DLBCL, 203–204
in secondary CNS lymphoma, 299
221–222
 Index
autologous stem cell transplantation
(ASCT) (cont.)
in T-cell non-Hodgkin lymphoma,
245–247
relapse following, 77–78
Avastin, 240
azathioprine, 274, 275
bacterial infection
role in lymphoma aetiology, 8
B-cell acute lymphoblastic leukemia, 172
See also lymphoblastic lymphoma.
B-cell lymphoma, unclassifiable, with
features intermediate between
DLBCL and Burkitt lymphoma
cytogenetics, 175
molecular pathology, 175
pathology, 173
B-cell lymphoma, unclassifiable, with
features intermediate between
DLBCL and cHL, 66–67
B-cell prolymphocytic leukemia, 122
B-cell receptor (BCR) pathway
inhibitors
mantle cell lymphoma studies, 168
bcl-2 inhibitors, 168
BCNU, 181, 214, 261
BDR regimen, 149
BEACOPP regimen, 71–75, 84, 280
BEAM regimen, 76, 181, 214
Bence Jones proteinuria, 143
bendamustine, 97, 98, 144, 162
mantle cell lymphoma studies,
165–166
use in Waldenström’s
macroglobulinemia, 148
benign lymphoproliferative disorders,
286–288
classification, 286
inflammatory pseudotumor of
lymph nodes, 288
Kikuchi’s disease, 286–287
progressive transformation of
germinal centers, 287
vascular transformation of sinuses,
287–288
bevacizumab, 240, 251
bexarotene, 261, 264
Bexxar, 205
BFM 90 regimen, 179, 186
BFM 95 regimen, 186
BFM regimen, 183
BH3 mimetics, 168
bleomycin, 55, 69, 70, 74, 80, 92, 94,
197, 234, 276, 280
lung toxicity, 41–42
blood-brain barrier (BBB)
effects in CNS lymphoma
chemotherapy, 209–210
300 reversible disruption, 214
blood hyperviscosity syndrome,
139–141
bone marrow transplant, 241
Borrelia burgdorferi
association with cutaneous MALT
lymphoma, 24, 114–115
bortezomib, 98, 144, 149, 168, 201, 240,
249, 265, 278, 279
mantle cell lymphoma studies, 166
use in Waldenström’s
macroglobulinemia, 147
brentuximab vedotin (SGN-35), 240
Bruton’s tyrosine kinase (Btk)
inhibitors, 206
bryostatin, 206
B-symptoms
definition, 27
prognostic significance in Hodgkin
lymphoma, 27
Burkitt lymphoma
as post-transplant
lymphoproliferative disorder,
276
association with EBV, 8
B-cell lymphoma, unclassifiable,
with features intermediate
between DLBCL and Burkitt
lymphoma, 173, 175
classification, 173
clinical presentation, 172
clinical variants, 172
CNS prophylaxis and treatment, 221
cytogenetics, 174–175
endemic Burkitt lymphoma, 172
HIV-associated Burkitt lymphoma,
182
immunodeficiency associated
Burkitt lymphoma, 172
immunophenotypes, 173
in HIV patients, 277–278
management of relapse, 181
molecular pathology, 174–175
natural history, 172
pathology, 172–173
prognostic factors, 24–25
age, 24–25
at diagnosis, 24–25
clinical factors at diagnosis,
24–25
refractory disease, 25
relapsed disease, 25
treatment-related, 25
sporadic Burkitt lymphoma, 172
St Jude classification, 38
staging, 177
supportive care, 187–188
treatment, 178–181
treatment principles, 177–178
treatment recommendations, 182
tumor lysis syndrome, 187
CALGB regimen, 179, 182
Campath, 129, 240, 264
Campylobacter jejuni
association with IPSID, 116
association with Miller Fisher
syndrome, 142
association with small intestine
MALT lymphoma, 114
cancer registration
impact of classification issues, 2
candida infection
imaging, 43
carboplatin, 76, 77, 203
cardiac toxicity
treatment related, 82
carmustine, 76, 203, 261
lung toxicity, 41–42
CD30 + T-cell lymphoproliferative
disorders, 266–268
CD30 expression
in HRS cells, 30
prognostic significance in Hodgkin
lymphoma, 30
cell cycle inhibitors, 168
centrocytic lymphoma, 155
chemotherapy
therapeutic strategies, 11
Chlamydophila psittaci
association with adnexal ocular
MALT lymphoma, 24, 114–115
chlorambucil, 41, 93, 94, 111, 127, 128,
144, 264
use in Waldenström’s
macroglobulinemia, 145–146
2-chlorodeoxyadenosine, 264
CHOEP regimen, 199, 234
chonic lymphocytic leukemia
cell immunophenotypes, 122
CHOP regimen, 14, 55, 94, 161, 196,
198, 234, 236, 239, 241, 264,
277, 279, 289, 294
addition of rituximab, 17
variations, 17
CHOP-bleomycin regimen, 94
CHOP-DHAP regimen, 163
CHOP-rituximab regimen, 199
ChP regimen, 94
chromosome translocations
MALT lymphomas, 24
chronic lymphocytic leukemia 121
advances in understanding and
treatment, 121
allogenic stem cell transplantation,
131
B-cell prolymphocytic leukemia, 122
chemotherapeutic approach, 126
choice of first line therapy, 127–129
choice of therapy in relapse,
130–131
differential diagnosis, 160

factors defining choice of therapy,
126–127
future treatment challenges, 134–135
lymph-node structure, 121–122
minimal residual disease, 124–125
monoclonal antibody therapy,
129–130
novel therapeutic agents, 131–134
recommended therapeutic
approaches, 134
treatment by risk stratification,
123–124
use of alkylating agents, 127
CHVP/interferon regimen, 95
cisplatin, 76, 77, 163, 203, 234, 244
cladribine, 144, 242, 290
use in Waldenström’s
macroglobulinemia, 146
clarithromycin, 111
classical Hodgkin lymphoma (cHL)
associations with chronic
inflammatory rheumatic
disease, 9
characteristic cell types, 61
Hodgkin Reed-Sternberg (HRS)
cells, 61, 68
pathology, 63–64
Reed–Sternberg cells, 63–64, 66
classification of haematological
malignancies
challenges, 1
WHO classification, 1
clinical features
use in classification of lymphomas, 1
clinical trials in lymphoma
accelerated titration method, 52
confidence intervals for results, 48
continual reassessment method
(CRM), 51–52
contribution to therapy
development, 58
Cox regression model, 50
definition of p-value, 50
definition of randomization, 50
definition of sample size, 50
definition of significance level, 50
definition of significance test, 50
definitions of endpoints, 47
design complexity and challenges, 45
determining endpoints, 46–47
development of new therapeutic
approaches, 45
dose-limiting toxicity (DLT), 50–51
error control, 49–50
evaluating treatment effect,
47–48
event-free survival (EFS) endpoint,
47, 50
future developments, 58
hazard function, 50
historical background, 45
informed consent, 46
intent-to-treat analysis, 50, 57
interim analysis, 50, 53
Kaplan-Meier estimator, 50
lymphoma specific survival
endpoint, 47
maximum tolerated dose (MTD),
50–52
Modified Cheson criteria, 47
outcome data, 47
overall survival (OS) endpoint, 47
patient eligibility criteria, 48–49
phase I trials, 50–52
phase II trials, 52–53, 54
phase III trials, 54–57
power of a study, 50
precision, 48–50
progression-free survival (PFS)
endpoint, 47
qualitative outcome measures, 47
quantitative outcome measures, 47
randomization methods, 57
randomized phase II trials, 54
response criteria, 47
study population, 48–49
time to event data, 47
time to progression (TTP)
endpoint, 47
type I and type II errors, 49–50
use of placebo, 54–55
CNS lymphomas
aggressive lymphomas, 222–223
allogenic stem cell transplantation,
222
autologous stem cell transplantation
(ASCT), 221–222
chemotherapy, 209–212
combined chemo-radiotherapy,
213–214
conventional therapeutic strategies
for primary CNS lymphoma,
208–209
corticosteroid therapy, 212
cytarabine therapy, 210
dural based MALT lymphoma, 208
EBV positive CNS lymphomas, 208
effects of the blood-brain barrier
(BBB), 209–210
experimental therapeutic strategies,
214
high dose chemotherapy with ASCT,
214
histopathology, 208
HIV-related lymphomas, 217–220
immunocompromised patients, 208,
217–220
immunocytoma, 222
indolent lymphomas, 222
intraocular lymphoma, 216–217
Index
intravascular large B-cell lymphoma,
222
leptomeningeal lymphoma,
214–216
low-grade marginal zone B-cell
lymphoma (MALT type), 222
methotrexate therapy, 210
plasmacytoma, 222–223
primary CNS Hodgkin lymphoma,
223
primary DLBCL of the CNS, 208
prophylaxis against CNS relapse,
220–221
radioimmunotherapy, 212
radiotherapy, 212–213
rare forms, 222–223
relapsed disease, 217
reversible BBB disruption, 214
rituximab therapy, 212
small lymphocytic lymphoma, 208
spinal cord lymphoma, 217
T/NK cell lymphoma, 208
T-cell phenotype of primary CNS
lymphoma, 222
temozolomide therapy, 212
topotecan therapy, 212
treatment of secondary CNS
lymphomas, 221–222
types of, 208
vanishing tumor effect of
corticosteroids, 212
whole-brain radiotherapy (WBRT),
212–213
CODOX-M regimen, 180, 182
CODOX-M/IVAC regimen, 277
coeliac disease
association with lymphomas, 9
cold agglutinin hemolytic anemia, 142
common variable immunodeficiency,
273
COMP regimen, 183
Competence Network Malignant
Lymphoma, 48
complete response (CR)
IWC definition, 39
complete response unconfirmed (CRu)
IWC definition, 39
complications of therapy
acute respiratory distress syndrome
(ARDS), 41–42
angioinvasive aspergillosis, 42–43
bleomycin-related lung injury, 41–42
carmustine-related lung injury, 41–42
cyclophosphamide-related lung
injury, 41–42
diffuse alveolar damage, 41–42
frank radiation pneumonitis, 42
imaging, 41–43
immunocompromised patients,
42–43 301
 Index
complications of therapy (cont.)
non-specific interstitial pneumonia
(NSIP), 42
organizing pneumonia, 42
paramediastinal fibrosis, 42
pulmonary haemorrhage, 42
radiotherapy, 42
computed tomography (CT), 32
PET-CT, 37–38, 48
staging of lymphomas, 32–34
treatment response assessment,
38–40
computerized axial tomography
(CT scan), 48
confidence interval
definition, 50
for study results, 48
continual reassessment method
(CRM), 51–52
COPBLAM regimen, 234
COPDAC regimen, 80
COPP regimen, 80
COPP/ABVD regimen, 72
core biopsy, 32
corticosteroids, 94, 241, 261
use in CNS lymphomas, 212
vanishing tumor effect, 212
Cotswolds modification. See Ann
Arbor classification
Cox regression model, 50, 57
Crohn’s disease
association with lymphomas, 9
cryoglobulinemia, 141
Cumulative Illness Rating Scale
(CIRS), 126
cutaneous large B-cell lymphoma, 271
cutaneous MALT lymphoma
association with Borrelia burgdorferi,
24, 114–115
cutaneous γδ T-cell lymphoma, 226,
241
pathology, 241
treatment, 241
CVP regimen, 94, 234
CVP-bleo regimen, 92
cyclophosphamide, 17, 25, 55, 69, 72,
77, 78, 80, 92, 93, 94, 95, 127,
144, 162, 163, 180, 181, 183,
196, 197, 199, 234, 245, 247,
276, 280, 295
lung toxicity, 41–42
cyclosporine, 236, 274, 275, 292
cytarabine, 41, 55, 76, 77, 163, 178, 180,
181, 197, 203, 214, 221, 234, 290
use in CNS lymphomas, 210
cytomegalovirus (CMV), 274
imaging of infection, 43
cytophagic histiocytic panniculitis
(CHP), 240
302 cytosine arabinoside, 76
dacarbazine, 70, 79, 80, 84, 280
DA-EPOCH regimen, 277
DA-EPOCH-R regimen, 181, 182, 199
data reporting
challenges related to cancer
registration, 2
decadron, 234
dendritic cell disorders
classification, 286
diagnostic challenges, 286
follicular dendritic cell sarcoma, 291
interdigitating dendritic cell
sarcoma, 291
denileukin diftitox, 236, 240, 249, 264
2-deoxycoformycin, 264
Dexa-BEAM regimen, 76
dexamethasone, 55, 76, 77, 129, 149,
163, 181, 203, 234, 244, 276, 292
DHAP regimen, 76, 77, 203, 234
diagnosis of haematological neoplasms
multiple parameters used, 1
diagnosis of lymphoma
changing diagnostic practice, 7
methods used, 1
diffuse alveolar damage
complication of therapy, 41–42
imaging, 41–42
diffuse large B-cell lymphoma,
(DLBCL)
activated B cell (ABC)-like DLBCL, 16
addition of new agents to
chemotherapy, 201
addition of rituximab to
chemotherapy, 200–201
allogenic stem cell transplantation, 204
Ann Arbor classification, 12–14
anti-Helicobacter therapy, 111
associated with inflammation, 193
association with autoimmune
diseases, 9
B-cell lymphoma, unclassifiable,
with features intermediate
between DLBCL and Burkitt
lymphoma, 173, 175
B-cell lymphoma, unclassifiable,
with features intermediate
between DLBCL and cHL,
66–67
centroblastic lymphoma, 191
classification of subtypes, 15
clinical follow-up, 113–114
clinical presentation, 191
cytogenetics, 194–195
diffuse large B-cell lymphoma NOS,
191–192
dose dense therapy, 198–199
dose intensified therapy, 198–199
EBV-positive DLBCL of the elderly,
193
extranodal disease, 191
factors predicting response to H.
pylori eradication, 112–113
germinal center-like (GC) DLBCL, 16
high dose therapy and ASCT,
199–200
high dose therapy and ASCT in
relapsed disease, 203–204
histone deacetylase (HDAC)
inhibitors, 205–206
immunoblastic variant, 191–192
immunophenotypes, 192
in HIV patients, 276–277
intravascular large B-cell lymphoma,
193
IPI and R-IPI classification systems,
12–14
management of antibiotic-resistant
cases, 114
management of relapsed DLBCL,
203–206
mediastinal (thymic) large B-cell
lymphoma, 193–194
molecular follow-up, 113–114
molecular pathology, 194–195
morphological variants, 191–192
mTOR inhibitors, 206
new therapeutic targets, 205–206
pathology, 191–194
post-treatment histologic evaluation,
112
primary DLBCL of the CNS, 208
prognostic factors, 12–18, 196
at diagnosis, 12–16
at relapse, 17–18
CHOP regimen, 17
clinical factors at diagnosis,
12–14
clinical factors at relapse, 17–18
DLBCL subtypes, 15
dose intensity and schedule, 17
gene expression profiling, 16
gene expression profiling at
relapse, 18
genetic abnormalities, 16
histological factors, 15–16
immunophenotypes, 15–16
PET-18F-FDG, 17
relapsed DLBCL, 204–205
time to complete remission (CR),
16–17
treatment-related factors, 16–17
treatment-related factors at
relapse, 17–18
protein kinase C β (PKCβ)
inhibitors, 206
radiolabeled monoclonal antibodies
in relapsed disease, 205
second line therapy prior to ASCT,
203
staging investigations, 195

T-cell/histiocyte rich B-cell
lymphoma (TCHRBCL), 192
treatment approaches, 191
treatment of advanced stage DLBCL,
198–202
treatment of early stage DLBCL,
196–198
treatment of primary extranodal
disease, 198
treatment of primary mediastinal
large B-cell lymphoma,
202–203
use of IF-RT in advanced DLBCL,
201–202
with expression of full length ALK, 193
diffuse large B-cell lymphoma
associated with inflammation,
193
diffuse large B-cell lymphoma NOS,
191–192
diffuse large B-cell lymphoma with
expression of full length ALK,
193
diffuse reticuloendotheliosis, 289
dose-limiting toxicity (DLT), 50–51
doxorubicin, 17, 70, 76, 77, 80, 93, 94,
95, 163, 180, 181, 196, 197, 199,
234, 241, 264, 276, 280
dural based MALT lymphoma
CNS lymphomas, 208
Eastern Cooperative Oncology Group
(ECOG), 45
Eastern Solid Tumor Group, 45
EBV regimen, 280
EBVP regimen, 70, 280
endemic Burkitt lymphoma, 172
endpoints
definitions of endpoints, 47
determining endpoints in clinical
trials, 46–47
enteropathy-associated T-cell
lymphoma, 226
enteropathy-type intestinal T-cell
lymphoma (EITCL), 244–245
cytogenetics, 245
molecular pathology, 245
pathology, 245
treatment, 245
environment
risk factors for lymphoma, 9
enzastaurin, 168, 206
eosinophilic granuloma, 289
epidemiology of lymphomas
aetiology, 7–10
age-related effects, 2–6
changes in rates over time, 6–7, 9
classification challenges, 1
data reporting and collection
challenges, 2
Index
descriptive epidemiology, 7 extranodal NK/T-cell lymphoma, nasal
gender-related effects, 2–6 and nasal-type, 243–244
geographic variation, 2 cytogenetics, 243
Hodgkin lymphoma (HL), 61 molecular pathology, 243
incidence of lymphoma, 1 pathology, 243
information on disease burdens and treatment, 243–244
patterns, 2
WHO classification, 1 family history of haematological
epidoxorubicin, 74 cancers
epirubicin, 70, 280 risk factor for lymphoma, 9
EPOCH-R regimen, 199 FCM regimen, 162
epratuzumab, 201 FCR regimen, 127, 128
Epstein–Barr virus (EBV), 67, 217, 236 fertility
association with lymphomas, 8 effects of treatment on, 82
association with post-transplant fine needle aspiration, 32
lymphoproliferative disorders, flavopiridol, 133, 168
273 Fleming multi-stage trial design, 53
pediatric EBV associated flow cytometry, 7
lymphoproliferative disease, 244 fludarabine, 78, 94, 97, 127, 144, 162,
pediatric EBV-related 165, 242, 264
hemophagocytic syndrome, 244 use in Waldenström’s
role in T-cell non-Hodgkin macroglobulinemia, 146
lymphoma, 226 fluorescent in situ hybridization
Epstein–Barr virus (EBV)-positive CNS (FISH), 127
lymphomas, 208 fluorodeoxyglucose (FDG). See
Epstein–Barr virus (EBV)-positive PET-18F-FDG
DLBCL of the elderly, 15, 193 follicular dendritic cell sarcoma, 291
ER-CHOP regimen, 201 follicular lymphoma
Erdheim–Chester disease, 291–292 association with inflammatory
error control in clinical trials, 49–50 disease, 9
ESHAP regimen, 77, 234 chromosome aberrations, 89–90
esomeprazole, 111 classification, 89
etiology of lymphomas. See aetiology of clinical presentation, 87
lymphomas clinicopathological variants, 89
etoposide, 55, 69, 74, 76, 77, 80, 93, 95, cytogenetics, 89–90
178, 180, 181, 197, 199, 203, disease course, 87
214, 234, 244, 245, 264, 280, gene expression profiling, 90
290, 292, 294 grading, 87
European Bone Marrow Transplant histology, 87–89
(EBMT) registry, 150 immunophenotypes, 89
European Mantle Cell Network, 21 incidence, 18, 87
European Organization for Research molecular pathology, 89–90
and Treatment of Cancer pediatric, 89
(EORTC), 27 prognostic factors, 18–21, 91
European Union at diagnosis, 18–20
Clinical Trials Directive legislation, 46 at relapse, 21
event-free survival (EFS) clinical factors at diagnosis, 18–19
definition, 47, 50 gene expression profiling, 20
everolimus, 77, 79, 144, 206 genetic abnormalities, 19–20
mantle cell lymphoma studies, 167 histological factors, 19
use in Waldenström’s treatment-related factors, 20–21
macroglobulinemia, 148 use of prognostic indexes, 18–19
extended-field radiotherapy (EF-RT), staging investigations, 91
70, 91–92 treatment
extracorporeal photopheresis (ECP), 263 advanced disease in asymptomatic
extranodal marginal zone B-cell patients, 92–93
lymphoma. See MALT advanced disease in symptomatic
lymphoma patients, 93–97
extranodal marginal zone lymphoma allogenic stem cell
(MZL), 24 transplantation, 99–100 303
 Index
follicular lymphoma (cont.)
autologous stem cell
transplantation (ASCT) in first
remission, 96
combined immunochemotherapy
(first line), 94–95
conventional chemotherapy, 94
early stage disease, 91–92
first line treatment in advanced
disease, 92–97
high dose therapy and ASCT, 99
high grade transformation,
100–101
immunochemotherapy (relapsed
disease), 97–98
immunotherapy, 94
management of relapsed follicular
lymphoma, 97–101
monoclonal antibody therapy, 94
novel agents, 98–99
radioimmunotherapy, 96
radioimmunotherapy (relapsed
disease), 100
rituximab, 94
rituximab maintenance therapy
(first line), 95–96
rituximab maintenance therapy
(relapsed disease), 98
treatment selection considerations, 18
Follicular Lymphoma International
Prognostic Index (FLIPI), 12,
18–19
Food and Drug Administration (US), 46
fostamatinib, 168, 206
frank radiation pneumonitis, 42
French–American–British (FAB)
classification, 172
fulminant hemophagocytosis, 241
functional MRI
staging of lymphomas, 36
GA-101 monoclonal antibody, 98,
133–134
Galen, 45
gastric MALT lymphoma. See MALT
lymphoma
G-CSF, 74
GELA grading system, 112
gemcitabine, 76, 77, 203, 234, 264
gender
differences in lymphoma rates, 2–6
sex-specific associations with
lymphomas, 9
gene amplification, 227–228
gene expression profiling, 11
genetic abnormalities
DLBCL, 16
follicular lymphoma, 19–20
mantle cell lymphoma, 21–23
304 use in classification of lymphomas, 1
genetic polymorphisms
potential risk factors for lymphoma,
9–10
geographic variation in lymphoma
rates, 2
German Hodgkin Lymphoma Study
Group (GHSG), 28
German Low Grade Lymphoma Study
Group, 21
germinal center-like (GC) DLBCL, 16
graft versus host disease, 279
granulomatous slack skin disease, 257,
260, 266
Groupe d’Etudes des Lymphomes de
l’Adulte (GELA), 69
Guillain–Barré syndrome, 142
GVD regimen, 76, 77
Haematological Malignancy Research
Network (UK), 3–6
Hand–Schuller–Christian disease, 289
Hashimoto thyroiditis, 114
association with thyroid lymphoma, 9
hazard function, 50
definition, 50
hazard ratio, 57
Helicobacter pylori
anti-Helicobacter therapy in DLBCL,
111
association with MALT lymphomas,
8, 24
clinical follow-up after treatment,
113–114
eradication in gastric MALT
lymphoma, 110–111
involvement in gastric MALT
lymphoma, 109–110
management of antibiotic-resistant
cases, 114
molecular follow-up after treatment,
113–114
role in lymphoma etiology, 8
role in T-cell non-Hodgkin
lymphoma, 226
hemophagocytic lymphohistiocytosis,
292
hemophagocytic syndrome
(HPS), 241
hepatitis B virus (HBV), 8
hepatitis C virus (HCV), 8, 114, 116,
118, 138, 274
role in T-cell non-Hodgkin
lymphoma, 226
hepatosplenic T-cell lymphoma
(HSTCL), 226, 241–242
cytogenetics, 242
molecular pathology, 242
pathology, 242
treatment, 242
hepatosplenic γδ T-cell lymphoma, 23
herpes virus
imaging of infection, 43
herpes zoster, 149
highly active antiretroviral therapy
(HAART), 182, 217–219, 276,
277, 278, 279
histiocytic disorders, 288–292
classification, 286
diagnostic challenges, 286
Erdheim–Chester disease, 291–292
follicular dendritic cell sarcoma, 291
hemophagocytic
lymphohistiocytosis, 292
histiocytic sarcoma, 289
interdigitating dendritic cell
sarcoma, 291
Langerhans cell histiocytosis,
289–291
Rosai–Dorfman disease, 288–289
histiocytic sarcoma, 289
histiocytosis X, 289
histone acetyltransferases (HATs), 79
histone deacetylase (HDAC) inhibitors,
79, 249, 265
mantle cell lymphoma studies, 168
use in DLBCL, 205–206
HIV/AIDS
association with B-cell lymphomas,
7–8
potential influence on lymphoma
rates, 7
HIV-related lymphomas, 276–281
AIDS primary CNS lymphomas,
217–220
Burkitt lymphoma, 172, 182, 277–278
CNS involvement, 217–220
diffuse large B-cell lymphoma
(DLBCL), 276–277
Hodgkin lymphoma, 279–280
improving treatment protocols,
280–281
non-Hodgkin lymphoma, 276
plasmablastic lymphoma, 278–279
primary CNS lymphoma, 279
primary effusion lymphoma (PEL),
278
salvage treatment, 279
See also highly active antiretroviral
therapy (HAART).
HLH-94 regimen 292
Hodgkin lymphoma (HL)
Ann Arbor classification, 26–27
approaches to treatment, 61
as post-transplant
lymphoproliferative disorder,
276
B-cell lymphoma, unclassifiable, with
features intermediate between
DLBCL and cHL, 66–67
characteristic cell types, 61

clinical presentation, 61
cytogenetics, 67–68
diagnosis, 68
epidemiology, 61
Hodgkin Reed-Sternberg (HRS)
cells, 61
in HIV patients, 279–280
lymphocytic and histiocytic (L&H)
cells, 61
molecular pathology, 67–68
pathogenesis, 82–83
pathology
classical HL, 63–64
lymphocyte-depleted HL, 66
lymphocyte-rich classical
HL, 64
mixed cellularity HL, 65
nodular lymphocyte predominant
HL, 61–63
nodular sclerosis HL, 65–66
primary CNS Hodgkin lymphoma,
223
prognostic factors, 26–30
at diagnosis, 26
at relapse, 30
B-symptoms, 27
CD30 expression, 30
clinical factors at diagnosis, 26
dose intensity and time to
complete remission (CR), 30
histological factors, 29–30
International Prognosis Score
(IPS), 28–29
new biological parameters, 30
PET-18F-FDG, 30
prognostic models, 28–30
related to interaction of tumor and
host, 27
related to patient characteristics, 27
related to tumor spread and tumor
burden, 26–27
treatment-related factors, 30
rare patterns of nodal infiltration, 66
response assessment, 69
risk stratification, 68–69
staging, 68
treatment
advanced HL, 73–75
allogenic stem cell
transplantation, 77–78
anti-CD30 monoclonal antibodies/
immunotoxins, 78–79
cardiac toxicity, 82
early-stage favorable disease,
70–71
early-stage unfavorable disease,
71–73
fertility impairment, 82
first line treatment, 70–75
HDCT/ASCT regimen, 75–77
histone deacetylase (HDAC)
inhibitors, 79
HL in children and adolescents,
80–81
lenalidomide, 79
mTOR inhibitors, 79
nodular lymphocyte predominant
HL, 80
primary refractory HL, 75–77
protein-specific “small
molecules”, 79
relapse after HDCT and ASCT,
77–78
relapse after initial chemotherapy,
75–77
relapse after initial radiotherapy, 75
relapsed HL, 75–77
rituximab, 79
salvage radiotherapy, 77
summary of treatment strategies,
83–84
targeted therapies, 78–79
targeting the microenvironment,
79
treatment-related adverse effects,
81–82
treatment-related secondary
malignancies, 81–82
treatment-related toxicity (long
term), 81
treatment-related toxicity (short
term), 81
use as a reporting category, 2
Hodgkin Reed-Sternberg (HRS) cells,
61, 62, 66
CD30 expression, 30
cytogenetics, 67–68
molecular pathology, 67–68
See also Reed–Sternberg cells.
human herpesvirus 8 (HHV-8) 278, 293
association with lymphomas, 8
role in T-cell non-Hodgkin
lymphoma, 226
human T-cell leukemia virus 1 (HTLV-1)
association with ATLL, 8, 230
modes of transmission, 230
role in T-cell non-Hodgkin
lymphoma, 226
hydrocortisone, 221
hyper-CVAD regimen, 163, 180, 181,
182, 183, 277
hyper-CVAD/MA regimen, 234
90
Y-ibritumomab tiuxetan (Zevalin),
205, 212
ibrutinib, 129, 132, 168
ICE regimen, 76, 77, 203, 234
idelalisib, 129, 132, 167–168
ifosfamide, 76, 77, 180, 197, 203, 221,
234, 244
Index
IGEV regimen, 76, 77
IgM gammopathy, 143
IgM monoclonal gammopathy of
unknown significance (MGUS),
138, 139
IgM related neuropathy, 141–142
IL-12, 264
imaging
angioinvasive aspergillosis, 42–43
complications of therapy, 41–43
diffuse alveolar damage, 41–42
infection in immunocompromised
patients, 42–43
Pneumocystis carinii pneumonia
(PCP), 43
pneumonia, 42
radiation therapy complications, 42
role of the radiologist in lymphoma
management, 32
staging classifications, 38
staging techniques, 32–38
treatment response assessment, 38–41
imatinib, 291, 292
IMEP regimen, 244
immune system
dysregulation in lymphoma, 7–10
immunocompromised patients
CNS lymphomas, 208, 217–220
complications of therapy, 42–43
infection risk, 42–43
immunocytoma
CNS involvement, 222
immunodeficiency
HIV association with B-cell
lymphomas, 7–8
risk factor for primary CNS
lymphoma, 25
immunodeficiency associated Burkitt
lymphoma, 172
immunophenotype
use in classification of lymphomas, 1
immunoproliferative small intestinal
disease (IPSID), 115–116
immunosuppressed patients
HIV-related lymphomas, 276–281
post-transplant lymphoproliferative
disorders, 9, 273–276
risk factors for lymphoma, 273
immunosuppression
lymphoma risk factor, 8–9
immunotherapy
for primary cutaneous T-cell
lymphoma, 263–264
incidence of lymphoma, 1
infection
history of atopic infections, 9
potential risk factor for lymphoma, 9
role in lymphoma etiology, 7–8
infection risk
immunocompromised patients, 42–43 305
 Index
infertility, 80
treatment related, 82
inflammatory conditions
associations with lymphomas, 9
inflammatory pseudotumor of lymph
nodes, 288
informed consent, 46
inotuzumab ozogamicin
(CMC-544), 98
in-situ mantle cell lymphoma, 157
intent-to-treat analysis, 57
definition, 50
interdigitating dendritic cell sarcoma,
291
interferon alpha, 18, 95, 118, 241, 264,
294
interferon alpha-2b, 295
interferon gamma, 241, 264
interim analysis
definition, 50
in clinical trials, 53
InterLymph consortium
lymphoma etiology studies, 9–10
intermediate lymphocytic lymphoma,
155
International Agency for Research on
Cancer (IARC), 2
International Prognosis Score (IPS)
for Hodgkin lymphoma, 28–29
International Prognostic Index (IPI), 12
IPI and R-IPI classification of
DLBCL, 12–14
limitations in mantle cell lymphoma,
160
prognostic value in peripheral T-cell
lymphoma, 26–23
use in follicular lymphomas, 18
International Prognostic Score (IPS)
for Hodgkin lymphoma, 69
International Workship Criteria (IWC)
treatment response assessment,
38–41
intestinal T-cell lymphoma, 244
intraocular lymphoma, 216–217
intravascular large B-cell lymphoma, 193
CNS involvement, 222
involved-field radiotherapy (IF-RT),
70, 73, 91–92, 196
involved-node radiotherapy (IN-RT),
70–71, 73
isotretinoin, 264
IVAC regimen, 180, 182
juvenile rheumatoid arthritis, 292
Kaplan–Meier estimator, 50, 57
Kaposi’s sarcoma (KS), 293
Kaposi’s sarcoma herpesvirus
(KSHV), 278
306 association with lymphomas, 8
Kiel classification, 155, 156
Kikuchi’s disease, 286–287
Knudson two-hit model, 227
Langerhans cell histiocytosis, 289–291
lenalidomide, 51, 94, 98, 129, 131, 149,
206, 251, 279
Hodgkin lymphoma treatment, 79
mantle cell lymphoma studies, 167
use in Waldenström’s
macroglobulinemia, 147–148
leptomeningeal lymphoma,
214–216
Letterer-Siwe disease, 289
leucovorin, 180
livedo reticularis, 142
LMB 89 protocol, 178
LMB 96 regimen, 181
logrank statistics, 57
lomustine, 74
LSA2L2 regimen, 183
lymphoblastic lymphoma, 176
ALL-type therapy, 183–186
CNS treatment, 186–187
cytogenetics, 175–176
management of relapse, 187
molecular pathology, 175–176
natural history, 172
pathology, 174
precursor B lymphoblastic
lymphoma/leukemia, 174,
175–176
precursor T lymphoblastic
lymphoma/leukemia, 174, 176
staging, 177
supportive care, 187–188
treatment, 183–187
treatment principles, 182–183
treatment recommendations, 187
tumor lysis syndrome, 187
lymphocyte-depleted HL
pathology, 66
lymphocyte-rich classical HL
pathology, 64
lymphocytic and histiocytic (L&H)
cells, 61–63
lymphoma specific survival
definition, 47
lymphomatoid granulomatosis,
294–295
lymphomatoid papulosis, 266, 294
lymphoplasmacytic lymphoma, 138
See also Waldenström’s
macroglobulinemia.
MACOP-B regimen, 55, 203, 290
macrophage activation syndrome
(MAS), 292
magnetic resonance imaging (MRI)
staging of lymphomas, 35–36
MALT lymphoma, 24
anti-Helicobacter therapy in DLBCL,
111
association with autoimmune
disease, 9
association with Helicobacter pylori,
8, 24
chromosome translocations, 24
clinical features, 104–105
clinical follow-up, 113–114
cytogenetics, 106–107
diagnosis of gastric MALT
lymphoma, 109–110
differential diagnosis, 106
disease course, 104–105
disseminated disease, 104
factors predicting response to
H. pylori eradication, 112–113
gastric MALT lymphoma and
H. pylori, 24
GELA grading system, 112
H. pylori eradication in gastric
MALT lymphoma, 110–111
immunophenotypes, 106
immunoproliferative small intestinal
disease (IPSID), 115–116
incidence, 104
localization, 104
low-grade marginal zone B-cell
lymphoma of the CNS, 222
management, 110
management of antibiotic-resistant
cases, 114
management of Borrelia burgdorferi-
associated cutaneous
lymphomas, 114–115
management of Chlamydophila
psittaci-associated lymphoma,
114–115
management of H.pylori negative
cases, 114
management of non-gastric
locations, 114–115
molecular biology, 106–107
molecular follow-up, 113–114
nodal marginal zone lymphoma,
118–119
pathology, 105–106
post-treatment histologic evaluation,
112
primary splenic marginal zone
lymphoma, 116–118
prognostic factors, 24, 104–105
staging, 108–110
staging of gastric MALT lymphoma,
109–110
Wotherspoon score, 112
MALT lymphoma in transformation,
24
mantle cell lymphoma

allogenic stem cell transplantation,
165
autologous stem cell transplantation
(ASCT), 162
B-cell receptor (BCR) pathway
inhibitors, 168
bcl-2 inhibitors/BH3 mimetics, 168
bendamustine studies, 165–166
bortezomib studies, 166
cell cycle inhibitors, 168
chromosomal aberrations, 157–158
classification, 155, 156
clinical presentation, 155–156
clinical prognostic markers, 160–161
CNS relapse, 155
conventional chemotherapy, 161
cytogenetics, 157–158
differential diagnosis, 160
diffuse type, 21
disease course and timescale, 155–156
dose-intensified regimens, 163
everolimus studies, 167
familial aspect, 155
future directions, 168
HDAC inhibitors, 168
immunomodulatory drugs, 167
immunophenotypes, 157
incidence, 155
initial therapy, 161
lymph node architecture, 156–157
management of relapsed disease,
165–166
minimal residual disease, 161
molecular markers, 159–160
molecular pathogenesis, 157–158
monoclonal antibody and combined
immunochemotherapy,
161–162
morphologic subtypes, 159
mTOR inhibitor studies, 167
myeloablative therapy followed by
ASCT, 163–164
nodular type, 21
novel therapeutic approaches,
167–168
pathology, 156–157
phenotypic markers, 159–160
PI3K/AKT pathway therapeutic
targets, 167–168
prognostic factors, 159–161
at diagnosis, 11–12
clinical factors at diagnosis, 21
genetic abnormalities, 21–23
histological factors, 21
prognostic indexes, 21
radioimmunotherapy, 162
radioimmunotherapy in relapsed/
refractory disease, 166
rituximab maintenance therapy,
164–165
sequential dose intensification and
autologous transplantation,
163–164
staging, 159
temsirolimus studies, 167
therapeutic challenges, 155
Mantle Cell Lymphoma International
Prognostic Index (MIPI), 21, 161
mantle zone lymphoma, 155
marginal zone lymphomas (MZL)
association with autoimmune
diseases, 9
normal counterparts, 24
prognostic factors, 24
subtypes, 24
maximum tolerated dose (MTD),
50–52
M-BACOD regimen, 55, 234, 276
MCP regimen, 94
MDX-060 (Medarex), 78, 240
MDX-1401, 240
MEC regimen, 73
mechlorethamine, 70, 74, 93, 261
mediastinal (thymic) large B-cell
lymphoma, 193–194
Mediterranean lymphoma, 115
MegaCHOEP regimen, 245
melphalan, 76, 78, 181, 203, 214
6-mercaptopurine, 290
methotrexate, 26, 55, 77, 93, 163, 178,
180, 181, 183, 186, 197, 213,
214, 219, 221, 234, 236, 241,
244, 264, 276, 292, 294
use in CNS lymphomas, 210
methyl-GAG, 234
methylprednisolone, 77
methylprednisone, 234
metronidazole, 111, 116
Miller Fisher syndrome
association with Campylobacter
jejuni, 142
mini-BEAM regimen, 203
mitoxantrone, 94, 162, 234
mixed cellularity HL
pathology, 65
Modified Cheson criteria, 47
monoclonal antibodies, 11, 18, 94
novel agents, 98–99, 133–134
use in chronic lymphocytic
leukemia, 129–130
use in mantle cell lymphoma,
161–162
use in Waldenström’s
macroglobulinemia, 130,
146–147 See also specific agents.
MOPP regimen, 84
MOPP/ABV regimen, 71
MOPP/ABVD regimen, 70
motor neuron disease, 142
mTOR inhibitors, 77, 79
Index
mantle cell lymphoma, 167
use in DLBCL, 206
mucosa-associated lymphoid tissue. See
MALT lymphoma
multicentric Castleman’s disease,
293–294
multi-parameter, four-color flow
cytometry (MRD flow), 125
mycophenolate mofetil, 275
mycosis fungoides, 254
clinical features, 255–255
clinicopathologic variants, 265–266
folliculotropic variant, 265
granulomatous variants, 266
molecular pathology, 259–260
pagetoid reticulosis, 265
pathology, 256–259
staging, 255–255
syringotropic variant, 265
treatment, 261–265
myelodysplasia, 145, 163
National Cancer Institute (NCI), 45, 46
navitoclax (ABT-263), 132–133, 168
neuropathy
IgM related, 141–142
nitrogen mustard, 261
NK/T-cell lymphoma (NK/TCL), nasal
type, 226
NK-cell neoplasms, 244
nodal marginal zone lymphoma, 24
clinical features, 118
genetics, 118
management, 118–119
pathology, 118
nodular lymphocyte predominant HL
characteristic cell types, 61
lymphocytic and histiocytic (L&H)
cells, 61–63
pathology, 61–63
popcorn cells, 61–63
treatment, 80
nodular sclerosis HL
pathology, 65–66
non-Hodgkin lymphoma (NHL)
association with HIV/AIDS, 7–8
change in rates over time, 9
changes in rates over time, 6–7
HIV-related non-Hodgkin
lymphoma, 276
International Prognostic Index (IPI),
12, 13
use as a reporting category, 2
See also specific subtypes.
non-Hodgkin’s Lymphoma Pathologic
Classification Project, 49
non-specific interstitial pneumonia
(NSIP)
complication of therapy, 42
nuclear factor kappa B (NFκB), 24 307
 Index
obatoclax (GX 15–070), 168
obesity
potential risk factor for
lymphoma, 9
OEPA regimen, 80
ofatumumab, 98, 133, 144
use in Waldenström’s
macroglobulinemia, 147
OKT3, 274
omeprazole, 111
Oncovin, 72, 74, 77
Ontak, 240, 249, 264
OPPA regimen, 80
organ transplantation
lymphoma risk, 9
potential influence on lymphoma
rates, 7
risk factor for primary CNS
lymphoma, 25
See also post-transplant
lymphoproliferative disorders.
organizing pneumonia
complication of therapy, 42
outcome measures, 47
overall survival (OS)
definition, 47, 50
endpoint, 12
pagetoid reticulosis, 265
panobinostat, 77
pantoprazole, 111
paramediastinal fibrosis, 42
partial response (PR)
IWC definition, 39
PCI-37265, 206
PD 0332991, 168
pediatric conditions
EBV associated lymphoproliferative
disease, 244
EBV-related hemophagocytic
syndrome, 244
follicular lymphoma, 89
hemophagocytic
lymphohistiocytosis, 292
Langerhans cell histiocytosis,
289–291
St Jude classification, 38
treatment of Hodgkin lymphoma in
children, 80–81
See also Burkitt lymphoma.
pegylated liposomal doxorubicin, 77
pentostatin, 242
peripheral T-cell lymphoma
prognostic factors, 23
ALK protein expression, 23
at diagnosis, 23
clinical factors at diagnosis, 23
histological factors, 23
treatment-related factors, 23
308 prognostic value of IPI, 26–23
peripheral T-cell lymphoma, not
otherwise specified (PTCL-
NOS), 23, 226, 232–234
clinical presentation, 232
cytogenetics, 233
incidence, 232
molecular pathology, 233
pathology, 232–233
treatment, 233–234
PET scan
role in response evaluation, 47–48
PET-18F-FDG, 11, 70
in HIV positive HL patients, 280
predictive value in DLBCL, 17
prognostic value in Hodgkin
lymphoma, 30
staging of lymphomas, 36–38
treatment response assessment, 40–41
PET-18FLT
staging of lymphomas, 38
PET-CT, 48
staging of lymphomas, 37–38
photochemotherapy, 261–263
phototherapy, 261–263
PI3K/AKT pathway
therapeutic targets, 167–168
plasmablastic lymphoma, 278–279
plasmacytoma
CNS involvement, 222–223
plerixafor, 132
Pneumocystis carinii pneumonia
(PCP)
imaging, 43
pneumonia
imaging, 42
POEMS (polyneuropathy,
organomegaly, endocrinopathy,
M protein, and skin changes)
syndrome, 142
polymerase chain reaction (PCR), 7
popcorn cells, 61–63
positron emission tomography. See
PET
post-transplant lymphoproliferative
disorders, 9, 273–276
B-cell lymphomas, 273
Burkitt lymphoma, 276
diagnosis, 274
etiology, 274
Hodgkin lymphoma, 276
incidence, 273–274
prognostic factors, 274
T-cell lymphomas, 273, 275
treatment, 274–275
power of a study, 50
pralatrexate, 250–251, 265
prednisolone, 55, 92, 94, 95, 244
prednisone, 17, 25, 55, 69, 70, 72, 74, 80,
93, 181, 183, 196, 197, 199, 234,
236, 245, 280, 290
pregabelin, 149
primary CNS lymphoma, 208
HIV related, 279
incidence, 25
prognostic factors, 25–26
at diagnosis, 25–26
clinical factors at diagnosis, 25–26
histological factors, 26
immunodeficiency, 25
organ transplantation, 25
treatment-related factors, 26
See also CNS lymphomas.
primary cutaneous (anaplastic) CD30
positive large cell lymphoma,
268
primary cutaneous aggressive
epidermotropic CD8+ T-cell
lymphoma, 269
primary cutaneous anaplastic large-cell
lymphoma (ALCL), 236
primary cutaneous B-cell lymphoma,
254, 269
classification system, 254
cutaneous large B-cell lymphoma, 271
primary cutaneous large B-cell
lymphoma, leg type, 271
primary cutaneous marginal zone
B-cell lymphoma/
immunocytoma, 269–270
primary follicle centre-cell
lymphoma, 270
primary cutaneous CD4-small/
medium-sized pleomorphic
T-cell lymphoma, 269
primary cutaneous DLBCL, leg type, 15
primary cutaneous follicle centre
lymphoma, 89
primary cutaneous large B-cell
lymphoma, leg type, 271
primary cutaneous lymphoma
classification, 254
incidence of skin lymphoma, 254
primary cutaneous B-cell
lymphoma, 254
See also primary cutaneous B-cell
lymphoma; primary cutaneous
T-cell lymphoma.
primary cutaneous marginal zone
B-cell lymphoma/
immunocytoma, 269–270
primary cutaneous peripheral T-cell
lymphoma, unspecified,
268–269
primary cutaneous T-cell lymphoma,
254
CD30 + T-cell lymphoproliferative
disorders, 266–268
classification system, 254
granulomatous slack skin disease,
257, 260, 266

lymphomatoid papulosis, 266
molecular pathology, 259–260
mycosis fungoides, 255–266
pathology, 256–259
primary cutaneous (anaplastic)
CD30 positive large cell
lymphoma, 267–268
primary cutaneous aggressive
epidermotropic CD8+ T-cell
lymphoma, 269
primary cutaneous CD4+ small/
medium-sized pleomorphic
T-cell lymphoma, 269
primary cutaneous peripheral T-cell
lymphoma, unspecified,
268–269
Sézary syndrome, 255–265
subcutaneous panniculitis-like T-cell
lymphoma, 268
treatment of mycosis fungoides,
261–265
treatment of Sézary syndrome,
261–265
primary DLBCL of the CNS, 15
primary effusion lymphoma (PEL), 278
primary endpoint
definition, 50
primary follicle centre-cell lymphoma,
270
primary mediastinal large B-cell
lymphoma, 191
treatment, 202–203
procarbazine, 69, 70, 72, 74, 80, 93, 234,
280
prognostic factors
approach to identification, 11
building prognostic indices, 11–12
Burkitt lymphoma, 24–25
diffuse large B-cell lymphoma
(DLBCL), 12–18, 196
follicular lymphoma, 18–21, 91
Hodgkin lymphoma, 26–30
MALT lymphoma, 24, 104–105
mantle cell lymphoma, 11–12,
159–161
marginal zone lymphomas (MZL), 24
peripheral T-cell lymphoma, 23
post-transplant lymphoproliferative
disorders, 274
primary CNS lymphoma, 25–26
relapsed DLBCL, 204–205
risk stratification for Hodgkin
lymphoma, 68–69
role of prognostic indices, 11
T-cell non-Hodgkin lymphoma,
228–229
Waldenström’s macroglobulinemia,
138, 144
prognostic indices
accuracy, 11
dealing with continuous variables, 12
descriminatory ability, 12
endpoint definition, 12
for follicular lymphoma, 18–19
for mantle cell lymphoma, 21
goals of, 11
methodology for building, 11–12
progression-free survival (PFS)
endpoint, 12
role of, 11
simplicity of usage, 11–12
use of OS as endpoint, 12
validation (internal and external), 12
prognostic models
for Hodgkin lymphoma, 28–30
progression-free survival (PFS)
definition, 47
endpoint, 12
progressive disease (PD)
IWC definition, 39
progressive transformation of germinal
centers, 287
prolymphocytic leukemia, 160
ProMACE-CytaBOM regimen, 55
ProMACE-CytoBOM regimen, 51
ProMACE-MOPP regimen, 92
protein kinase C β (PKCβ) inhibitors,
206
proteosome inhibitors, 249
psoralen, 241
pulmonary haemorrhage
complication of therapy, 42
purine analogues, 236
PUVA, 241, 261–263, 264
p-value
definition, 50
qualitative outcome measures, 47
quantitative outcome measures, 47
R2-CHOP regimen, 51
radioimmunotherapy
follicular lymphoma treatment, 96
relapsed follicular lymphoma, 100
use in CNS lymphomas, 212
use in mantle cell lymphoma, 162
radiologist
role in lymphoma diagnosis and
management, 32
radiotherapy
complications of therapy, 42
extended-field radiotherapy
(EF-RT), 70, 91–92
for cutaneous γδ T-cell lymphoma,
241
for primary cutaneous T-cell
lymphoma, 263
imaging of complications, 42
involved-field radiotherapy (IF-RT),
70, 71–73, 91–92, 243
Index
involved-node radiotherapy
(IN-RT), 70–71, 73
risk of second malignancy, 42
use of IF-RT in advanced DLBCL,
201–202
randomization
definition, 50
methods for clinical trials, 57
rapamycin (mTOR), 167
RAR retinoids, 264
Raynaud’s phenomenon, 141
Raynaud’s syndrome, 142
R-CHOP regimen, 162
R-DHAP regimen, 203
Reed–Sternberg cells, 61, 63–64, 65–66
See also Hodgkin Reed–Sternberg
(HRS) cells.
relapsed disease
IWC definition, 39
treatment strategies, 11
response evaluation, 47–48
retinoids, 264
Revised European–American
Lymphoma (REAL)
classification (1994), 7, 49
Revised International Prognostic Index
(R-IPI), 14
rexinoids, 264
R-FC regimen, 162
R-FCM regimen, 162
R-GDP regimen, 203
rheumatoid arthritis
association with lymphomas, 9
ribavirine, 118
R-ICE regimen, 203
risk stratification for Hodgkin
lymphoma, 68–69
ritonavir, 277
rituximab, 14, 15, 24, 51, 92, 93, 94, 98,
127, 133, 144, 149, 181, 199,
203, 236
addition to CHOP regimen, 17
addition to DLBCL chemotherapy,
200–201
Hodgkin lymphoma treatment, 79
immunochemotherapy, 94–95
in relapsed follicular lymphoma, 97
maintenance therapy in follicular
lymphoma, 95–96
maintenance therapy in mantle cell
lymphoma, 164–165
maintenance therapy in relapsed
follicular lymphoma, 98
maintenance therapy in
Waldenström’s
macroglobulinemia, 150
use in Burkitt lymphoma, 181
use in CNS lymphomas, 212
use in HIV-related lymphoma,
276–277, 278 309
 Index
rituximab (cont.)
use in mantle cell lymphoma, 161–162
use in multicentric Castleman’s
disease, 294
use in post-transplant
lymphoproliferative disorders,
275
use in Waldenström’s
macroglobulinemia, 146–147
romidepsin, 249, 265
Rosai-Dorfman disease, 288–289
salivary gland lymphoma
association with Sjögren’s
syndrome, 9
salvage radiotherapy, 77
sample size
definition, 50
Schnitzler syndrome, 143
secondary CNS lymphoma, 208
See also CNS lymphomas.
secondary endpoints
definition, 50
secondary leukaemia, 80
secondary malignancies
treatment related, 81–82
Sézary syndrome, 254
clinical features, 255–256
molecular pathology, 259–260
pathology, 256–259
prognostic factors, 255–256
treatment, 261–265
SGN-30, 78, 240
SGN-35, 78–79, 240
significance level
definition, 50
significance test
definition, 50
Simon min-max trial design, 53
Simon optimal trial design, 53
Sjögren’s syndrome, 114
association with salivary gland
lymphoma, 9
small intestine MALT lymphoma
association with Campylobacter
jejuni, 114
small lymphocytic lymphoma
advances in understanding and
treatment, 121
CNS lymphomas, 208
See also chronic lymphocytic
leukemia.
SMILE regimen, 244
smoking
potential risk factor for lymphoma, 9
solumedrol, 76
Southwest Oncology Group (SWOG), 70
spinal cord lymphoma, 217
splenic lymphoma with villous
310 lymphocytes (SLVL), 160
splenic mantle zone lymphoma, 24
splenic marginal zone lymphoma, 24
clinical course, 117–118
clinical features, 116
differential diagnosis, 117
immunophenotypes, 116–117
management, 117–118
molecular pathogenesis, 117
pathology, 116–117
sporadic Burkitt lymphoma, 172
St Jude staging classification, 38, 177, 178
stable disease (SD)
IWC definition, 39
staging classifications
St Jude classification, 38, 177, 178
See also Ann Arbor staging
classification.
staging techniques
computed tomography (CT), 32–34
functional MRI, 36
imaging techniques, 32–38
magnetic resonance imaging (MRI),
35–36
PET-18F-FDG, 36–38
PET-18FLT, 38
PET-CT, 37–38
Stanford V regimen, 85, 280
STLI regimen, 71
subcutaneous panniculitis-like T-cell
lymphoma, 226, 240–241, 268
cytophagic histiocytic panniculitis
(CHP), 240
pathology, 241
phenotypes, 240–241
treatment, 241
sun exposure
potential risk factor for lymphoma, 9
Surveillance, Epidemiology and End
Results (SEER) Program (US), 7
systemic anaplastic large-cell
lymphoma (ALCL), 23
systemic lupus erythematosus (SLE),
287, 292
association with lymphomas, 9
T/NK cell lymphoma
CNS lymphomas, 208
tacrolimus, 274, 275
Targretin, 261, 264
T-cell acute lymphoblastic leukemia,
172
See also lymphoblastic lymphoma.
T-cell lymphoma
association with autoimmune
diseases, 9
T-cell non-Hodgkin lymphoma
adult T-cell leukemia/lymphoma
(ATLL), 230–232
allogenic stem cell transplantation,
248–249
anaplastic large-cell lymphoma
(ALCL), systemic type,
236–240
angioimmunoblastic T-cell
lymphoma (AITL), 234–236
autologous stem cell transplantation
(ASCT), 245
chromosome rearrangements,
226–227
classification of non-cutaneous
subtypes, 226
cutaneous γδ T-cell lymphoma, 241
emerging and novel therapies,
249–251
enteropathy-type intestinal T-cell
lymphoma (EITCL), 244–245
etiology, 226–228
extranodal NK/T-cell lymphoma,
nasal and nasal-type, 243–244
front line therapy, 245–247
gene amplification, 227–228
gene expression profiling, 228
genetic alterations, 226–228
hepatosplenic T-cell lymphoma
(HSTCL), 241–242
high dose therapy with ASCT,
245–247
inactivation of tumor suppressor
genes, 227
incidence, 226
peripheral T-cell lymphoma, not
otherwise specified
(PTCL-NOS), 232–234
prognosis, 228–229
proteosome inhibitors, 249
radiation therapy, 249
relapsed disease, 247–248
role of infection in lymphogenesis,
226
subcutaneous panniculitis-like T-cell
lymphoma, 240–241
T-cell/histiocyte rich B-cell lymphoma
(TCHRBCL), 192
T-cell/histiocyte-rich large B-cell
lymphoma, 15
T-cell/histiocyte-rich primary DLBCL
of the CNS, 15
temozolomide
use in CNS lymphomas, 212
temsirolimus, 206
mantle cell lymphoma studies, 167
teniposide, 234
tetracycline, 116
thalidomide, 144, 149
mantle cell lymphoma studies, 167
use in Waldenström’s
macroglobulinemia,
147–148
therapeutic strategies
use of prognostic indices, 11

thyroid lymphoma
association with Hashimoto
thyroiditis, 9
time to event data, 47
time to progression (TTP)
definition, 47
endpoint, 12
time to treatment failure (TTF)
endpoint, 12
topical steroids, 241
topotecan
use in CNS lymphomas, 212
131
I tositumomab, 205
total skin electron beam (TSEB)
therapy, 263
toxin therapies, 264–265
treatment response assessment
International Workship Criteria
(IWC), 38–41
PET-18F-FDG, 40–41
use of CT, 38–40
use of imaging, 38–41
tumor lysis syndrome, 178, 187
tumor necrosis factor alpha (TNF-α)
inhibitors, 226
tumor suppressor genes
inactivation, 227
ultrasound (US) examination, 32
unicentric Castleman’s disease, 293
urticaria, 143
vanishing tumor effect of
corticosteroids, 212
vascular transformation of sinuses,
287–288
VEBEP regimen, 280
Velcade, 240
VIM-3 regimen, 234
vinblastine, 70, 74, 79, 84, 280,
290, 294
vincristine, 17, 25, 55, 69, 70, 72, 74, 77,
80, 92, 93, 94, 163, 180, 181, 183,
196, 199, 234, 245, 276, 280, 290
vindesine, 74, 197, 234
vinorelbine, 76, 77, 280
VIPD regimen, 244
viral infections
association with marginal zone
lymphomas, 114
role in lymphoma aetiology, 7–8
vorinostat, 168, 265
Wald statistics, 57
Waldenström’s macroglobulinemia
auto-antibody activity, 141–142
bendamustine therapy, 148
biochemical investigations, 143
blood hyperviscosity syndrome,
139–141
bone marrow findings, 143–144
bone marrow involvement, 139
bortezomib therapy, 147
chlorambucil therapy, 145–146
classification, 138
clinical features, 139
cold agglutinin hemolytic anemia, 142
combination therapies, 148–150
cryoglobulinemia, 141
cytogenetics, 138–139
effects of tissue infiltration by
neoplastic cells, 142–143
epidemiology, 138
etiology, 138
everolimus therapy, 148
familial aspect, 138
hematological abnormalities, 143
high-dose therapy and stem cell
transplantation, 150
IgM related neuropathy, 141–142
imaging studies, 144
immunomodulatory agents, 147–148
Index
incidence, 138
laboratory investigations and
findings, 143–144
lenalidomide therapy, 147–148
lymph node biopsy findings, 144
maintenance rituximab
therapy, 150
monoclonal antibody therapies,
146–147
morbidity mediated by effects of
IgM, 139–143
nature of the clonal cell, 139
nucleoside analog therapies, 146
ofatumumab therapy, 147
prognosis, 144
response criteria, 150–151
rituximab therapy, 146–147
serum viscosity investigation, 143
thalidomide therapy, 147–148
tissue deposition, 142
treatment indications, 147–144
treatment options, 144–150
whole brain radiotherapy (WBRT),
212–213
Wiskott–Aldrich syndrome, 8, 273
World Health Organization (WHO)
classification of hematological
malignancies, 1
classification, fourth edition, 49
International Agency for Research
on Cancer (IARC), 2
PTLD classification, 9
Wotherspoon score, 112
X-linked lymphoproliferative
syndrome, 273
zanolimumab, 265
Zevalin, 205, 212
zidovudine, 220
Zubrod, Gordon, 45
311