Hematopathology / ORIGINAL ARTICLE
Flow Cytometric and Immunohistochemical Analysis
of Small Lymphocytic Lymphoma, Mantle Cell Lymphoma,
and Plasmacytoid Small Lymphocytic Lymphoma
Joseph A. Tworek, M D , Timothy P. Singleton,
and Charles W. Ross, M D
Key Words: Small lymphocytic lymphoma; Mantle cell lymphoma; CD23; FMC7
Small lymphocytic lymphoma (SLL), mantle cell
lymphoma (MCL), and SLL plasmacytoid (SLLP) are
malignant neoplasms of small B cells that may have
overlapping cytologic features. Entities such as SLL with
irregular nuclear contours may pose additional
diagnostic difficulties. We investigated the utility of flow
cytometric analysis and immunohistochemistry studies in
distinguishing these disorders from each other. We
reviewed 29 lymphomas and classified them as SLL (13
cases), MCL (8 cases), and SLLP (8 cases) based on
histology and expression of cytoplasmic immunoglobulin
light chain. Paraffin section immunohistochemistry was
performed for CD5, CD20, CD23, CD43, CD45RA,
CD45RO, and K and A, light chains. Flow cytometric
analysis was carried out by 2-color direct immuno-
fluorescence for CD5, CDllc, CD19, CD20, CD22,
CD23, FMC7, and K and X light chains. By immuno-
histochemistry, we found that the expression ofCD23 in
SLL discriminates between SLL and MCL and that the
expression ofCD23 and CD43 in SLL discriminates
between SLL and SLLP. By flow cytometric analysis, we
found that CDllc* and dim fluorescence intensity ofslg
and dim fluorescence intensity ofFMC7 in SLL
distinguish SLL from MCL, and the expression ofCD5+,
CD23+ and dim fluorescence intensity ofFMC7 in SLL
distinguishes SLL from SLLP. We found no immuno-
phenotypic difference between SLL and SLL with
irregular nuclear contours.
582 Am J Clin Pathol 1998; 110:582-589
Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
M D , Bertram Schnitzer, M D , Eric D. Hsi, M D ,
Histologically, small lymphocytic lymphoma (SLL),
mantle cell lymphoma (MCL), and SLL plasmacytoid (SLLP)
are lymphomas composed of small lymphocytes. Distin-
guishing these lymphomas from each other on histologic
grounds can be difficult. Particularly when SLL is composed
of lymphocytes with irregular nuclear contours, it may be
difficult to distinguish SLL from MCL. Yet this distinction is
important clinically inasmuch as MCL has a shorter median
survival than SLL or SLLP, and it is treated more aggressively
than SLL or SLLP.1 Furthermore, it is important to distinguish
SLL from SLLP because SLLP is often associated with
macroglobulinemia and its hyperviscosity complications.2
Immunophenotypic characterization of these disorders by
flow cytometry and immunohistochemistry has been shown to
be a useful adjunct to their histologic classification.3-16
CD23 is a B-cell activation antigen that functions in IgE-
mediated presentation to T cells17 and prevents apoptosis of
B cells in germinal centers.18 Recently, assessment of CD23
expression by flow cytometry has been found to be useful in
distinguishing SLL from MCL 56 and discriminating SLL
from SLLP.6 By both flow cytometry and immunohistochem-
istry, CD23 has tended to be present in SLL, absent in
MCL)4,7,10,1U3,19 a n d a b s e n t i n SLLP.
478 13
' However, there
have been reports of CD23 expression in a majority of
SLLP" or MCL cases.20 The use of anti-CD23 by immuno-
histochemistry has been successful with frozen section mate-
rial. Recently, two published studies reported success with
anti-CD23 in paraffin-embedded material.7,21
Our objectives in this study were 3-fold: investigate the
utility of immunohistochemical detection of CD23 expres-
sion in paraffin-embedded material to aid in the distinction of
SLL from MCL and SLL from SLLP; investigate whether a
battery of other immunophenotypic markers can help in
making these distinctions by immunohistochemical and flow
© American Society of Clinical Pathologists

cytometric analysis; and determine whether SLL with irreg-
ular nuclear contours (SLLIR) differs immunophenotypically
from SLL without irregular nuclear contours.
Materials and Methods
Case Selection
The 29 cases selected for this study had cryopreserved
cells available for flow cytometric immunophenotyping (22
cases) or had recently been characterized by flow cytometric
analysis (7 cases). In addition, all cases had paraffin-
embedded tissue (fixed with formalin, B5, or both) available
for immunohistochemistry.
Morphology
Diagnosis and classification of the lymphomas were
based on histologic review. The histologic reviewers were
blinded to the flow cytometric results. With the exception of
K and A, light chain immunoglobulin results to confirm SLLP,
the histologic reviewers were also blinded to the other
immunohistochemical results. The SLL and MCL cases
were classified based on the results of lymph node biopsies.
The SLLP cases were classified on both bone marrow (5
cases) and lymph node biopsies (3 cases).
Small Lymphocytic Lymphoma
Cases defined as SLL consisted predominantly of small
lymphocytes with round nuclear contours, but larger
lymphoid cells (prolymphocytes and paraimmunoblasts) were
Hmage I I Small lymphocytic lymphoma. A, Low-power image showing proliferation center (lower left, H&E, x85). B, High-
power image showing numerous small lymphocytes with round nuclear contours (H&E, x1,320).
© American Society of Clinical Pathologists
Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
Hematopathology / ORIGINAL ARTICLE
always present. The large cells were found either in discrete
proliferation centers or scattered more evenly throughout the
lymph node Umage II. Cases designated as SLLIR had
features of SLL, except that the cells had irregular nuclear
contours in both formalin- and B5-fixed material Hmage 21
The nuclear morphology on paraffin-embedded material was
compared to that of peripheral blood or bone marrow smear
material when the data were available (9 of 13 cases). All
SLL cases were analyzed with CD45RO to determine
whether a greater number of T cells were present in SLLIR
than in SLL without irregular nuclear contours.
Mantle Cell Lymphoma
Cases classified as MCL were composed of small- to
medium-sized lymphocytes arranged in a diffuse or vaguely
nodular pattern. Large lymphocytes were rare or absent.
Three cases of MCL had features of the blastoid variant.
Small Lymphocytic Lymphoma Plasmacytoid
Cases classified as SLLP (lymphoplasmacytoid lymphoma/
immunocytoma in the Revised European-American Lymphoma
classification) consisted of small lymphocytes, plasmacytoid
lymphocytes, and monoclonal plasma cells. Cytoplasmic
immunoglobulin light chain restriction was confirmed by
immunohistochemistry on paraffin-embedded tissue.
Immunohistochemistry
Formalin-fixed, paraffin-embedded tissues were deparaf-
finized in xylene and alcohol and then stained for the
following: CD5 (CD5/54/B4, titer 1:20, manufactured by
Novocastra Laboratories, Newcastle upon Tyne, United
Am J Clin Pathol 1998:110:582-589 583
 Tworek et al / IMMUNOPHENOTYPIC ANALYSIS OF CHRONIC LYMPHOPROLIFERATIVE DISORDERS

Ilmage 21 Small lymphocytic lymphoma with irregular contou rs. A, Low- power image showing proliferation center (left of
center, H&E, x165). B, High-power image showing numerous small lymphocytes with irregular nuclear contours and
occasional paraimmunoblasts (H&E, x1,320).

Kingdom, and distributed by Vector Laboratories, The immunohistochemistry studies were performed
Burlingame, Calif); CD20 (L26, 1:500, DAKO Corporation, with an automated stainer at 36°C (Ventana 320, Ventana
Carpinteria, Calif); CD23 (BU38, 1:20, Binding Site, San Medical Systems) with an avidin-biotin-peroxidase complex.
Diego, Calif); CD43 (Leu-22, 1:200, Becton Dickinson, San For CD5, the primary antibody was incubated with the tissue
Jose, Calif); CD45RA (4KB5, 1:160, DAKO), CD45RO (A6, overnight at room temperature after which it was subjected
1:50, Zymed, San Francisco, Calif); and K (peroxidase-conju- to detection with the avidin-biotin-peroxidase complex on an
gated rabbit polyclonal, 1:80, DAKO) and X (peroxidase- automated stainer (Biotek Techmate, Ventana Medical
conjugated rabbit polyclonal, 1:40, DAKO) light chains. Systems). Staining for the peroxidase-conjugated antibodies
For CD5, CD20, CD45RA, CD45RO, and CD43, was done manually.
antigen retrieval was performed by microwaving tissue
sections in 10 mM citrate buffer (pH 6.0) inside a pressure Flow Cytometric Analysis
cooker (Tender Cooker from Nordic Ware, Minneapolis, Specimens for flow cytometric analysis included cells
Minn). Sections for CD23 antigen retrieval were pretreated from the following: bone marrow (8 cases), peripheral blood
with protease (with a Ventana detection kit, Ventana Medical (17 cases), lymph nodes (3 cases), and spleen (1 case). Cells
Systems Inc, Tucson, Ariz). from lymph nodes and spleen were obtained by mechanical
disaggregation. Mononuclear cells were isolated by Ficoll-
Hypaque density gradient centrifugation. Surface membrane
•Table II immunophenotyping was performed by 2-color direct
immunofluorescence with the antibodies listed in I Table II.
Antibodies Used in Flow Cytometric Analysis The criterion for immunophenotypic marker positivity was
Antibody Source Fluorochrome labeling of at least 20% of the cells by the marker in question.
Although the authors recognize that this is an arbitrary crite-
CD5(Leu-1) B-D PE
rion, there is no consensus on the method to reliably distinguish
CD11c(Leu-M5) B-D PE
CD19(Leu-12) B-D PE dim positive and negative populations by flow cytometry.
CD20(Leu-16) B-D FITC For CD20, FMC7, and surface immunoglobulin, inten-
CD22(Leu-14) B-D FITC
CD23 (Leu-20) B-D PE sity of fluorescence was scored as either dim or bright on a
FMC7 AMAC FITC logarithmic scale. Dim expression was characterized by peak
K light chains B-D FITC
X light chains B-D PE channels within the first logarithmic decade, whereas bright
B-D = Becton Dickinson, San Jose, Calif; PE = phycoerythrin; FITC = fluorescein
expression was characterized by peak channels within the
isothiocyanate.
*AMAC, Westbrook, Me. second or third logarithmic decade.

584 Am J Clin Pathol 1998:110:582-589 © American Society of Clinical Pathologists

Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
 Hematopathology / ORIGINAL ARTICLE

Statistical Analysis immunohistochemical staining with CD45RO. Furthermore,

Testing for significance of the relationship between anti-
body marker reactivity and histologic subtypes was conducted
by using Fisher's exact test with 2 x 2 contingency tables
using Sigmastat software (San Rafael, Calif). The histologic
subtypes that were analyzed included SLL vs MCL and SLL
vs SLLR For the purpose of further statistical analysis, the
cases of SLLIR (6 cases) were combined with the SLL cases
(7 cases) inasmuch as there was no significant statistical
difference between SLL and SLLIR in the expression of
immunophenotypic markers by either flow cytometry or
immunohistochemistry. Comparisons of the flow cytometric
markers CD20, FMC7, and sig were made on the basis of
strong expression of these markers vs weak or no expression
of these markers. Comparisons of all other immunohistochem-
ical and flow cytometric markers were made simply by
considering the expression vs lack of expression of a marker.
Results
SLL vs SLLIR
Neither flow cytometry nor immunohistochemistry
revealed any statistically significant difference between the
immunophenotype of SLLIR and the immunophenotype of
SLL without irregular nuclear contours. Likewise, no statisti-
cally significant difference between SLLIR and SLL without
irregular nuclear contours was evident when the proportion
of T cells in paraffin-embedded tissue was determined by
we did not discover a correlation between the degree of
nuclear contour irregularity of SLL in paraffin-embedded
material and the degree of nuclear contour irregularity found
in peripheral blood or bone marrow aspirate smears.
Flow Cytometry
The percentages of marker expression in SLL, MCL,
and SLLP are summarized in ITable 21. In most cases SLL
was CD5 + , CDllc + , CD19+, CD22", and CD23 + and had
dim surface intensity CD20, FMC7, and sig; MCL was
CD5\ CD lie", CD19+, CD22+, and CD23" and had bright
surface intensity CD20, FMC7, and sig; and SLLP was
CD5", CDllc + , CD19+, CD22+, and CD23" and had bright
surface intensity CD20, FMC7, and sig. Six of the SLL cases
differed from the general flow cytometric immunophenotype
by having bright surface fluorescence intensity (FI) expres-
sion of CD20, FMC7, or sig (1 case had only bright FI of
sig, 3 cases had bright FI of CD20 and sig, 1 case had bright
FI of CD20 and FMC7, and 1 case had bright FI of CD20,
FMC7, and sig). Histologic re-review of all 6 of these cases
confirmed their classification as SLL.
The tests for significance of marker expression are
summarized in ITable 31. In SLL, the immunophenotype of
CDllc + , dim FI of FMC7, and dim FI of sig discriminated
with statistical significance between SLL and MCL. Markers
in SLL that were statistically significant in discriminating
from SLLP included CD5+, CD23+, and dim surface inten-
sity expression of FMC7. Although the FI of CD20 was not
significant in discriminating between SLL and MCL or

ITable 21 •Table 31
Immunophenotypes of SLL, MCL, and SLLP Statistical Correlations of Histology and Immunophenotype
SLL, % MCL, % SLLP, % SLL vs MCL SLL vs SLLP
(n = 13) (n = 8) (n = 8)
Flow Cytometry PValue* PValue*
Flow Cytometry CD5 .381 .007
+
CD5 100 88 43 CD11c .001 .305
CD11c 83* 0f 57 f CD20t .067 .067
CD19 100* 100 100 CD22 .067 .070
CD20 100 100 100' CD23 .387 .007
CD22 38 88 86 f FMC7f .002 .018
CD23 6? 38 0 s!gt .018 .367
FMC7 93 100 100 Immunohistochemistry
sig (KA) 100 100 100 CD20 1.000 1.000
Immunohistochemistry CD23 .005 .005
CD5 17 13 0 CD43 133 .001
CD20 100 100 100 CD45R 1.000 1.000
CD23 69 0 0
CD43 100 75 25 * A P value of <.05 indicates indicates a statistically significant difference in
CD45RA 100 100 100 expression of an antigen between the histologic categories.
CD45R0 0 0 0 'Comparisons made between bright surface intensity expression vs dim or no
clg 0 0 100 expression.

SLL = small lymphocytic lymphoma; MCL = mantle cell lymphoma; SLLP =

plasmacytoid small lymphocytic lymphoma.
*n = 12.
*n = 7.

© American Society of Clinical Pathologists Am J Clin Pathol 1998; 110:582-589 585

Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
 Tworek et al / IMMUNOPHENOTYPIC ANALYSIS OF CHRONIC LYMPHOPROLIFERATIVE DISORDERS
SLL B MCL
1 *! .V'..'. i'„^j!S(
IFigure I I Two-color flow cytometric analysis of CD23 and FMC7 in small lymphocytic lymphoma (SLL), mantle cell
lymphoma (MCL), and plasmacytoid small lymphocytic lymphoma (SLLP). Note dim fluorescence intensity of FMC7 in
small lymphocytic lymphoma.
between SLL and SLLP, a strong trend was evident. Specifi-
cally, dim FI of CD20 was seen more often in SLL than in
MCL or in SLLP. The FI of slg was not helpful in discrimi-
nating between SLL and SLLP. CD23 was not helpful in
distinguishing between SLL and MCL but was a significant
discriminator between SLL and SLLP. Although FMC7 was
expressed by almost all cases of SLL, it tended to be dim
IFigure II and provided significant discrimination between
SLL and MCL and between SLL and SLLP.
Immunohistochemistry
Most often SLL was CD20 + , CD23 + , CD43 + , and
CD45RA+; MCL was CD20+, CD23" CD43+, and CD45RA+;
and SLLP was CD20+, CD23", CD45RA+, and clg+ (Table 2).
•Image 31 CD23 immunoperoxidase preferentially staining
paraimmunoblasts in a proliferation center (left of center,
H&E, x330).
586 Am J Clin Pathol 1998;110:582-589
Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
SLLP
Individual markers that allowed a high degree of discrimina-
tion included CD23, which discriminated between SLL and
MCL and between SLL and SLLP, and CD43, which discrimi-
nated between SLL and SLLP (Table 3). CD23 tended to
stain large cells (paraimmunoblasts) preferentially, and it
also stained follicular dendritic cells llmage 31. CD5 detec-
tion in paraffin-embedded material was not successful; only
2 cases of SLL and 1 case of MCL had CD5 detectable by
this method. CD45RA was universally expressed by all
disorders and thus was not a useful discriminator.
Only 6 of 13 cases of SLL expressed CD23 by both flow
cytometry and immunohistochemistry. However, 2 cases of
SLL that expressed CD23 by flow cytometry and lacked
CD23 by immunohistochemistry had poor antigen preserva-
tion in paraffin. Three other cases of SLL that expressed CD23
by immunohistochemistry and were interpreted as CD23" by
flow cytometry had between 12% and 18% of the neoplastic
cells in question expressing CD23 by flow cytometry.
Discussion
Classification of SLL, MCL, and SLLP based on
histology alone can be difficult, particularly when SLL is
composed of lymphocytes with irregular nuclear contours.
These lymphomas express pan-B-cell markers but have
heterogeneous expression of other markers such as CD5 and
CD23. We sought to use flow cytometry and immunohisto-
chemistry to identify immunophenotypic markers that would
aid in discriminating among these lymphomas.
CD23 expression by paraffin-section immunohisto-
chemistry was statistically significant in discriminating
between SLL and MCL and discriminating between SLL
and SLLP; CD23 was preferentially expressed in SLL and
not expressed in MCL or SLLP. With a few exceptions, our
© American Society of Clinical Pathologists

findings are similar to those reported in the literature
concerning both frozen section material4''013 and paraffin-
embedded material.7'21 In the case of frozen section material,
1 study20 reported that 4 of 4 MCL cases expressed CD23.
By using paraffin-embedded material, Murray et al7 detected
CD23 expression in 1 of 3 cases of MCL. Kumar et al,21
who also used paraffin-embedded material, detected CD23
expression in 1 of 39 cases of MCL; the single positive case
was a blastoid variant of MCL. Three of our MCL cases
were blastoid variants, and they did not express CD23.
Examination of CD23 expression in SLLP has only
been reported in the literature for a small number of cases.
With the exception of 1 series" in which 7 of 11 cases of
SLLP had CD23 expression in frozen material, most studies
have reported no expression of CD23 in frozen material4 or
only a small minority of SLLP cases with expression of
CD23 in both frozen13 and paraffin-embedded material.7'21
Another immunohistochemical marker that is helpful in
the differential diagnosis is the aberrant coexpression of CD43.
CD43 expression was seen frequently in both SLL and MCL
but was uncommon in SLLP. This pattern of CD43 expression
allowed for statistically significant discrimination between SLL
and SLLP and is similar to that reported in other studies.13,22
Flow cytometric detection of CD23 did not discriminate
between SLL and MCL. All cases of MCL that expressed
CD23 by flow cytometry had dim FI, whereas those cases of
SLL that expressed CD23 tended to have bright expression
of the antigen; this trend did not achieve statistical signifi-
cance (P < .132). Flow cytometric detection of CD23 did
significantly discriminate between SLL and SLLP.
By flow cytometry, CD20 and sig were universally
expressed by SLL, SLLP, and MCL in our study. The intensity
at which these markers fluoresced, however, tended to
discriminate among these lymphomas. Dim FI of sig was
more frequent in SLL than in MCL or SLLP and allowed for
statistically significant discrimination between SLL and MCL.
The FI of CD20 also tended to be dimmer in SLL than in
MCL or SLLP. A subset of SLL (46%), however, had bright
FI of CD20 or sig or both. The literature on FI of CD20 and
sig in SLL mirrors our findings. Generally, SLL/chronic
lymphocytic leukemia (CLL) has been characterized as having
dim FI of CD20 and dim FI of sig.15 However, several studies
have reported a small percentage of SLL cases having classic
SLL morphology with bright FI of CD20 or sig or
both.516'23'24 The percentage of bright FI cases in these studies
has ranged from 7% to 26% for CD20 and from 4% to 24%
for sig.524 These findings beg the question of whether cases of
SLL with bright FI of CD20 or sig or both might be better
classified as MCL. We were unable to reclassify any of our
SLL cases as MCL upon histologic re-review of biopsy mate-
rial. It is thought that the heterogeneity in FI of surface
markers in SLL may reflect the developmental diversity of
© American Society of Clinical Pathologists
Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
H e m a t o p a t h o l o g y / ORIGINAL ARTICLE
lymphocytes.23 No difference in prognosis has been reported
between cases of SLL with bright FI of CD20 or sig and cases
of SLL with dim FI of these antigens.24
FMC7 detects subgroups of B-cell leukemias that have
arisen in relatively late stages of maturation.3,25 In our study,
FMC7 was nearly universally expressed by SLL, MCL, and
SLLP. However, there was a difference in FI of FMC7
among these disorders. FMC7 tended to have dim FI in SLL
and bright FI in SLLP and MCL, allowing for statistically
significant distinction between SLL and MCL and between
SLL and SLLP.
Review of the literature indicates that FMC7 has often
been reported to be absent in SLL/CLL; FMC7 has often
reported to be present in MCL, SLLP, prolymphocytic
leukemia or prolymphocytic transformation of SLL/CLL.6'14
In those studies in which FMC7 has been detected in
SLL/CLL, morphologic examination has often resulted in
reclassification of these disorders. 312 In one study, most
cases were reclassified as prolymphocytic transformation of
CLL.3 In another study of chronic lymphoproliferative disor-
ders, most FMC7-positive cases were classified as MCL or
prolymphocytic transformation of CLL.12
Upon histologic re-review of our SLL cases that
expressed FMC7, we were not able to reclassify these cases
as another disorder. It is important to note that for a case to be
considered as having FMC7 expression, we arbitrarily
required that at least 20% of the cells express FMC7. By
contrast, others have set cutoffs to define FMC7 positivity as
high as 30%.12 Had we applied the 30% cutoff to our group
of SLL cases, only 4 of them would have surpassed this
threshold for FMC7-positivity. We conclude that dim FI
expression of FMC7 is compatible with the diagnosis of SLL.
Other flow cytometric markers that were statistically
significant discriminators in this study included CD5 and
CD lie. The FAB group has indicated that CD5 is typically
absent in SLLP14 However, some have reported CD5 posi-
tivity in up to 20% of cases of SLLP.6 In our study, CD5 was
expressed in 43% of SLLP cases, which is a statistically
significant difference from the universal expression of CD5
in SLL. We observed a striking discrepancy between flow
cytometry and immunohistochemistry in our ability to detect
CD5 on neoplastic B cells. The anti-CD5 that we used for
immunohistochemistry appears to have very poor sensitivity
in paraffin sections; these results confirm an earlier report.26
CD1 lc belongs to a class of adhesion receptor molecules
called integrins. The CDllc/CD18 complex is found in tissue
macrophages and monocytes, and is necessary for cell-cell and
cell-surface adhesion.27 According to the literature, CDllc
may occasionally be expressed in SLL9 but is typically absent
in MCL.5 In our study, CDllc was frequently expressed in
both SLL and SLLP but was absent in MCL. CDllc was a
significant discriminator between SLL and MCL.
Am J Clin Pathol 1998;110:582-589 587
 Tworek et al / IMMUNOPHENOTYPIC ANALYSIS OF CHRONIC LYMPHOPROLIFERATIVE DISORDERS

SLL with irregular nuclear contours can pose particular References

difficulty in the differential diagnosis of MCL. It was the pres- 1. Weisenburger DD, Duggan M], Perry DA, et al. Non-
ence of proliferation centers in a diffuse lymphoma with irreg-
ular nuclear contours that initially led Perry et al28 to recognize 1991;26:139-158.
this lymphoma as a form of SLL that is distinct from MCL and
that has a better prognosis than MCL. Since its recognition, little Wilkins, 1992:611.
has been written regarding the immunophenotype of SLLIR.
Zukerberg et al13 found no immunophenotypic or clinical differ-
ences after comparing 2 cases of SLL with cleaved lymphocytes
and 23 cases of SLL without cleaved lymphocytes. Similarly,
our study by flow cytometry and immunohistochemistry did not
reveal any statistically significant difference between the
immunophenotype of SLLIR and the immunophenotype of
SLL without irregular nuclear contours. We also investigated
whether T cells, which often have irregular nuclear contours,
were more frequent in cases of SLLIR than in SLL without
irregular nuclear contours and thus may have accounted for the
irregular nuclear contours of SLLIR. When we used CD45RO,
we found no statistically significant difference in the proportion7. Murray PG, Janmohamed RMI, Crocker J. CD23 expression
of T cells in SLLIR when compared to SLL without irregular
nuclear contours. Our recognition of SLLIR was based on
nuclear irregularity detected in paraffin-embedded material. We
were not able to correlate the nuclear features in paraffin-
embedded material with those found in peripheral blood or bone
marrow aspirate smears, and thus the nuclear irregularities may
simply reflect differences in fixation.
From our study, we draw 5 major conclusions:
1. The detection of CD23 by immunohistochemistry on
paraffin-embedded material provides statistically
significant, high-specificity discrimination between SLL
and MCL and between SLL and SLLP.
2. Immunohistochemical expression of CD43 is strongly
associated with SLL and MCL but not with SLLP.
3. By flow cytometry, CD1 lc + , dim FI of slg, and dim FI of
FMC7 in SLL discriminates SLL from MCL.
+ +
4. By flow cytometry, CD5 , CD23 , and dim FI of FMC7
in SLL discriminates SLL from SLLP.
5. There is no immunophenotypic difference between SLL
with irregular nuclear contours and SLL without
irregular nuclear contours.
From the Department of Pathology, University of Michigan
Hospitals, Ann Arbor, Michigan.
Presented in part at the annual meeting of United States and 17. Santamaria LF, Bheekha R, vanReijsen FC, et al. Antigen
Canadian Association of Pathology, Washington DC, March, 1996. focusing by specific monomeric immunoglobulin E bound to
Address reprint requests to Dr Ross: Department of
Pathology, University of Michigan Medical Center, Medical
Science I Building M5242, 1301 Catherine Rd, Ann Arbor, MI
48109-0602.
Acknowledgments: The authors wish to thank Usha Kota,
Laura Glashauser, Judy Tschantz, Mary Gerard, and Karen
Peterson for their assistance with flow cytometric analyses.
Hodgkin's lymphoma of mantle zone origin. Pathol Annu.
2. Ben-Ezra J. Small lymphocytic lymphoma. In: Knowles DM,
ed. Neoplastic Hematopathology. Baltimore, Md: Williams &
3. Catovsky D, Cherchi M, Brooks D, et al. Heterogeneity of
B-cell leukemias demonstrated by the monoclonal antibody
FMC7. Blood. 1981;58:406-408.
4. Dorfman DM, Pinkus GS. Distinction between small
lymphocytic and mantle cell lymphoma by immunoreactivity
forCD23.ModPat/io!. 1994;7:326-331.
5. Kilo MN, Dorfman DM. The utility of flow cytometric
immunophenotypic analysis in the distinction of small
lymphocytic lymphoma/chronic lymphocytic leukemia from
mantle cell lymphoma. Am] Clin Pathol. 1996;105:451-457.
6. Matutes E, Owusu-Ankomah K, Morilla R, et al. The
immunological profile of B-cell disorders and proposal of a
scoring system for the diagnosis of CLL. Leukemia.
1994;8:1640-1645.
in non-Hodgkin lymphoma: immunohistochemical
demonstration using the antibody BU38 on paraffin sections.
] Pathol. 1991;165:125-128.
8. Orazi A, Cattoretti G, Polli N, et al. Distinct morphophenotypic
features of chronic B-cell leukemias identified with CDlc and
CD23 antibodies. Eur J Haematol. 1991;47:28-35.
9. Oertel J, Kastner M, Bai AR, et al. Analysis of chronic
lymphoid leukemias according to FAB. Leuk Res.
1992;16:919-927.
10. Salter DM, Krajewski AS, Cunningham S. Activation and
differentiation antigen expression in B-cell non-Hodgkin's
lymphoma. J Pathol. 1988;154:209-222.
11. Stein H, Lennert K, Feller AC, et al. Immunohistological
analysis of human lymphoma: correlation of histological and
immunological categories. Adv Cancer Res. 1984;42:67-147.
12. Huh YO, Pugh WC, Kantarjian HM, et al. Detection of
subgroups of chronic B-cell leukemias by FMC7 monoclonal
antibody. Am J Clin Pathol 1994;101:283-289.
13. Zukerberg LR, Medeiros JL, Ferry J A, et al. Diffuse low-grade
B-cell lymphomas: four clinically distinct subtypes defined by
a combination of morphologic and immunophenotypic
features. Am J Clin Pathol. 1993;100:373-385.
14. Bennett JM, Catovsky D, Daniel M-T, et al. Proposals for the
classification of chronic (mature) B and T lymphoid
leukemias. J Clin Pathol. 1989;42:567-584.
15. Almasri NM, Duque RE, Itunaspe ], et al. Reduced expression
of CD20 antigen as a characteristic marker for chronic
lymphocytic leukemia. Am J HematoL 1992;40:259-263.
16. Witzig TE, Li C-Y, Tefferi A, et al. Measurement of the
intensity of cell surface antigen expression in B-cell chronic
lymphoid leukemia. Am J Clin Pathol. 1994;101:312-317.
CD23 on Epstein-Barr virus-transformed B cells. Hum lmunol.
1993;37:23-30.
18. Liu YJ, Mason DY, Johnson GD, et al. Germinal center cells
express bcl-2 protein after activation by signals which prevent
their entry into apoptosis. Eur J Immunol. 1991;21:1905-1910.
19. Schuurman H], van Baarlen J, Huppes W, et al. Immuno-
phenotyping of non-Hodgkin's lymphoma. Am J Pathol.
1987;129:140-151.

588 Am J Clin Pathol 1998;110:582-589 © American Society of Clinical Pathologists

Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018

20. Williamson JMS, Grigor I, Smith MEF, et al. Cluster
differentiation antigen expression, proliferative activity and
clinical stage in centroblastic centrocytic lymphomas. J Pathol.
1986;150:51-59.
21. Kumar S, Green GA, Teruya-Feldstein J, et al. Use of
CD23(BU38) on paraffin sections in the diagnosis of small
lymphocytic lymphoma and mantle cell lymphoma. Mod
Pathol. 1996;9:925-929.
22. Said JW, Stall PN, Shintaku P, et al. Leu-22. A preferential
marker for T-lymphocytes in paraffin sections. Staining profile
in T-and B-cell lymphomas, Hodgkin's disease, other
lymphoproliferative disorders, myeloproliferative disorders
and various neoplastic processes. Am ] Clin Pathol.
1989;91:542-549.
23. Maddy AH, Sanderson A, Mackie MJ, et al. The role of cell
maturation in the generation of phenotypic heterogeneity in
B-cell chronic lymphocytic leukemia. Immunol.
1989;68:346-352.
24- Tefferi A, Bartholmai BJ, Witzig TE, et al. Heterogeneity and
clinical relevance of the intensity of CD20 and immunoglobulin
light-chain expression in B-cell chronic lymphocytic leukemia.
AmJCUnPathol. 1996;106:457^61.
© American Society of Clinical Pathologists
Downloaded from https://academic.oup.com/ajcp/article-abstract/110/5/582/1757757
by guest
on 20 February 2018
Hematopathology / ORIGINAL ARTICLE
25. Zola H. Human leukocyte markers defined by antibodies
derived from somatic cell hybrids IV. A monoclonal antibody
reacting specifically with a subpopulation of human B
lymphocytes. J Immunol. 1981 ;26:1373.
26. Ben-Ezra JM, Kornstein MJ. Antibody NCL-CD5 fails to
detect neoplastic CD5+ cells in paraffin sections. Am J Clin
Pathol. 1996;106:370-373.
27. Hanson CA, Gribbin TE, Schnitzer B, et al. CD1 lc (Leu-M5)
expression characterizes a B-cell chronic lymphoproliferative
disorder with features of both chronic lymphocytic leukemia
and hairy cell leukemia. Blood. 1990;76:2360-2367.
28. Perry DA, Bast MA, Armitage JO, et al. Diffuse intermediate
lymphocytic lymphoma: a clinicopathologic study and
comparison with small lymphocytic lymphoma and diffuse
small cleaved cell lymphoma. Cancer. 1990;66:1995-2000.
AmJCUnPathol 1998;110:582-589 589