Nat Rev Drug Dis - 2005

May 2005 Vol 4 No 5
S5 | Editorial
doi:10.1038/nrd1748
S6 | Jean-Pierre Issa, Hagop Kantarjian &

Peter Kirkpatrick
doi:10.1038/nrd1726
S8 | Lee M. Ellis & Peter Kirkpatrick

doi:10.1038/nrd1727
S10 | Richard Goldberg & Peter Kirkpatrick

doi:10.1038/nrd1728
S12 | Ching-Hon Pui, Sima Jeha &

Peter Kirkpatrick
doi:10.1038/nrd1729
S14 | Jonathan Dowell, John D. Minna &

Peter Kirkpatrick
doi:10.1038/nrd1730
S16 | David Gibbs, Ann Jackman &

Peter Kirkpatrick
doi:10.1038/nrd1731
S19 | doi:10.1038/nrd1737
INTRODUCTION
EDITORIAL OFFICE
Nature Publishing Group
The Macmillan Building
4 Crinan Street
London N1 9XW, UK
Tel: +44 (0)20 7833 4000
Fax: +44 (0)20 7843 3629
Supplement Editor: Peter Kirkpatrick
Editorial Production: Susanne Harris
Copy Editor: Daniel Jones
Managing Production Editor: Judith Shadwell
Electronic Production: Alexander Thurrell
Sponsorship: Claire Hines
Design and Layout (Sponsor Profiles):
Suzanne Coleman
Production: Kelly Hopkins
Publisher: Adam Smith
Editor-in-Chief, Nature Publications: Philip Campbell
Publishing Director: Peter Collins
World Wide Web:
www.nature.com/nrd/focus/hotdrugs/2005
NATURE REVIEWS JOURNALS
Nature Reviews Cancer
Chief Editor: Ezzie Hutchinson
www.nature.com/reviews/cancer
Nature Reviews Drug Discovery
Chief Editor: Peter Kirkpatrick
www.nature.com/reviews/drugdisc
Nature Reviews Genetics
Chief Editor: Magdalena Skipper
www.nature.com/reviews/genetics
Nature Reviews Immunology
Chief Editor: Elaine Bell
www.nature.com/reviews/immunol
Nature Reviews Microbiology
Chief Editor: David OConnell
www.nature.com/reviews/micro
Nature Reviews Molecular
Cell Biology
Chief Editor: Arianne Heinrichs
www.nature.com/reviews/molcellbio
Nature Reviews Neuroscience
Chief Editor: Rachel Jones
www.nature.com/reviews/neuro
elcome to Hot Drugs Cancer 2005, a collection of

articles on the new oncology drugs approved by the
US FDA last year.
The articles provide a two-page snapshot of key information
on the featured drugs. The first page discusses the science behind
the drug and how it was developed, including background on
the disease area, the rationale for the development of the drug,
its mechanism of action and major clinical results. And on the
second page, experts in the field consider questions related to
advances in the treatment of the disease in question and the
clinical impact of the new drug.
The drugs featured represent an intriguing mix of old and new
approaches to cancer therapy. Azacitidine, now approved for the
treatment of myelodysplastic syndromes, was originally turned
down for approval as a cytotoxic agent more than twenty years
ago. However, since then, its ability to exert antineoplastic effects
by reversing aberrant DNA methylation has been recognized.
And clofarabine and pemetrexed are both examples of how
optimization of traditional anticancer drug classes nucleoside
analogues and antifolates, respectively can still result in
important treatment advances.
The other three drugs highlighted all come from the new
breed of molecularly targeted agents. Bevacizumab, an antibody
against vascular endothelial growth factor, could be viewed
as a culmination of more than thirty years of research aimed
at targeting tumour angiogenesis as a therapeutic strategy.
Completing the collection, the antibody cetuximab and the
small-molecule kinase inhibitor erlotinib represent different
approaches to targeting another key signalling pathway in
cancer, the epidermal growth factor receptor pathway. Finally,
complementary market data for oncology drug development is
also provided.
We are pleased to acknowledge the financial support of
Genentech, ImClone Systems and Roche in the production
of Hot Drugs Cancer 2005. Thanks to this support, the content
will be freely available online for six months on a dedicated web
focus, which can be found at http://www.nature.com/nrd/focus/
hotdrugs/2005/index.html. The editorial content is the sole
responsibility of the Nature Publishing Group.
M AY 20 05 | S5
2005 Nature Publishing Group
HOT DRUGS | CANCER
Azacitidine
Azacitidine
Azacitidine (Vidaza; Pharmion), an inhibitor

of DNA methylation, was approved by the
US FDA for the treatment of myelodysplastic
syndromes in May 2004. It is the first drug to
be approved by the FDA for treating this rare
family of bone-marrow disorders, and has
been given orphan-drug status. It is also a
pioneering example of an agent that targets
epigenetic gene silencing, a mechanism
that is exploited by cancer cells to inhibit the
expression of genes that counteract the
malignant phenotype.
Myelodysplastic syndromes (MDS) are a heterogeneous group of bone-marrow disorders

that are characterized initially by ineffective
haematopoesis (the formation and development of blood cells) and subsequently by the
frequent development of acute myeloid leukaemia (AML)1,2. Typical symptoms of haematopoietic failure include infection, bleeding,
bruising and fatigue1,2.
Standard treatment for patients with MDS
consists of supportive measures such as blood
transfusions and administration of haematopoetic factors to correct cytopaenias, and
antibiotics to treat opportunistic infections1,2.
At present, the primary potentially curative
treatment is allogeneic stem-cell transplantation, but such therapy is often not appropriate
for MDS patients owing to their advanced age
or accompanying diseases1,2. There is therefore
a great need for novel agents for the management of MDS. Advances in the understanding
of the pathogenesis of MDS have led to the
testing of a number of such agents, including
DNA-methylation inhibitors such as azacitidine
(FIG. 1).
Drug properties
Azacitidine is a nucleoside analogue of cytidine that specifically inhibits DNA methylation by trapping DNA methyltransferases4
(FIG. 1). It was originally developed as a cytotoxic agent, and an application to the FDA
requesting its approval as such was turned
down more than 25 years ago. The discovery
in the early 1980s that it was a hypomethylating agent4, and the elucidation of the role of
DNA hypermethylation in cancer, have
prompted its re-evaluation and eventually
led to its recent approval. Azacitidine is
thought to exert its antineoplastic effects in
part by causing hypomethylation of DNA
and consequent reactivation of previously
silenced genes, including tumour-suppressor
genes4,5. It might also have activity through
incorporation into RNA, as well as direct
cytotoxicity.
Clinical data
A randomized, open-label, controlled trial

compared the safety and efficacy of subcutaneous azacitidine plus supportive care with
supportive care alone (observation group) in
~200 patients with any of the five subtypes of
MDS 5. Azacitidine was administered at a
subcutaneous dose of 75 mg per m2 daily for
seven days every 4 weeks; the dose was
increased to 100 mg per m2 if no beneficial
effect was seen after two treatment cycles5.
Patients in the observation group were allowed
to cross over to the azacitidine group on the
basis of several criteria of worsening disease5.
S6
Indications
Azacitidine is approved by the FDA for the

treatment of patients with the following MDS
subtypes: refractory anaemia or refractory
anaemia with ringed sideroblasts (if accompanied by neutropaenia or thrombocytopaenia
or requiring transfusions), refractory anaemia
with excess blasts, refractory anaemia with
excess blasts in transformation, and chronic
myelomonocytic leukaemia5.
Basis of discovery
DNA methylation is a key epigenetic mechanism that results in the heritable silencing of
genes without a change in their coding
sequence3,4. Epigenetic processes are required
for the normal development of mammalian
cells, but are not used for the dynamic regulation of gene expression1. However, it is now
known that malignant cells can exploit the
process of DNA methylation to silence the
expression of genes that counteract the malignant phenotype, such as tumour-suppressor
genes3. Leukaemias and MDS are characterized by the hypermethylation and consequent silencing of multiple genes1,3, which
has led to interest in inhibiting DNA methylation as a therapeutic strategy for treating
these diseases.
The primary efficacy endpoint was

response rate. After excluding several patients
who were found to have AML at baseline, the
overall response rate (complete response and
partial response) in patients who were
treated with azacitidine was 15.7%, which
was statistically significantly higher than the
response rate of 0% in the observation group
(before crossing over) 5. Response occurred in
all MDS subtypes, as well as in patients with
baseline diagnosis of AML, and the median
duration of clinical response (complete or
partial response) was estimated as 330 days;
75% of the responding patients were still in
partial response or better at completion of
treatment5. Patients who crossed over to
receive azacitidine had a response rate of
12.8%5.
Azacitidine was also studied in a similar
manner in two single-arm, open-label studies
in patients with MDS and AML. Similar proportions of patients had a complete or partial
response to azacitidine as in the randomized
trial5.
NH2
N
HO
N
N
OH OH
Azacitidine
4-amino-1--Dribofuranosyl-striazin-2(1H)-one;
C8H12N4O5; Mr = 244;
CAS number: 320-67-2
DNA
replication
Strand
separation
Z
Z
DNMTs
Figure 1 | Azacitidine and DNA methylation. a | Azacitidine. b | A family of DNA methyltransferases

(DNMTs) catalyse the methylation of the 5 position of the cytosine ring. After intracellular conversion to
5-aza-2-deoxycytidine (decitabine), azacitidine (Z) is incorporated in place of cytidine into DNA, where it acts
as a direct and irreversible inhibitor of DNMTs. Cells then divide in the absence of DNMTs, which results in
progressive DNA hypomethylation and reactivation of previously silenced genes1,3,4. Azacitidine also
incorporates into RNA, but very little is known about the effects of this. Pink circles, methylated CpG;
yellow circles, unmethylated CpG. Adapted from REF. 4.
| MAY 2005
HOT DRUGS | CANCER
EPIGENETIC THERAPIES AND MDS | VIEW FROM THE CLINIC

Analysing clinical issues for myelodysplastic
syndrome (MDS) and epigenetic therapies in
general are Jean-Pierre Issa, M.D., Director of
the Translational Research Laboratories in the
Department of Leukemia, and Hagop
Kantarjian, M.D., Chairman of the Department
of Leukemia at the University of Texas MD
Anderson Cancer Center, USA. Dr Issas interests
are the roles of epigenetic changes in ageing and
cancer, and the rapid translation of basic research
in epigenetics to the treatment of patients with
leukaemias and solid tumours. Dr Kantarjians
interests are novel therapies for leukaemias.
What are the current key issues in the treatment

of MDS?
MDS is a very heterogeneous group of diseases
with vastly different outcomes depending on
clinico-pathological features1: it can range
from a chronic, indolent disease to an aggressive
malignancy with a median survival similar to
that of metastatic lung cancer. A key issue in
the management of MDS is to reduce this heterogeneity using molecular profiles, which will
then be essential to select patients for therapy.
The International Prognostic Scoring System
(IPSS) uses a combination of clinical data
with cytogenetics to achieve a certain degree of
selection6, but it remains imperfect. More
importantly, the IPSS score does not predict
who is going to respond to azacitidine or other
treatment modalities in MDS. A concerted
effort to supplement the IPSS using geneexpression profiles, DNA-methylation profiles
or higher-resolution chromosomal analysis is
essential to making progress in this disease.
Given that the reported major response rate to
azacitidine is less than 20%7, this effort should
then be applied to identifying those patients
most likely to respond to this drug.
How would you expect MDS treatment to
evolve in the next 5 years?
Azacitidine should be the cornerstone of efforts
to move MDS treatment towards real lifeprolonging options. Given that MDS is primarily a disease of older individuals, aggressive
therapies such as combination chemotherapy
and stem-cell transplantation are simply not
realistic for most patients. There is much interest therefore in exploring less toxic agents in
this disease, and learning how to integrate them
into a multi-agent therapeutic approach.
Among the classical therapies, the cytotoxic
cytosine analogue cytarabine has a response rate
of about 15% in MDS when used at low doses,
and the topoisomerase inhibitor topotecan also
has reported activity in MDS8. There is also
reported synergy between hypomethylating

agents and topoisomerase inhibitors9, which
should be explored in MDS. A drawback for
these approaches is toxicity, which has to be
kept at a minimum in an elderly population.
Given that all patients who initially respond
to azacitidine eventually relapse, it might be
attractive to investigate whether agents such
as cytarabine and topotecan can prolong
remissions in this disease.
Newer agents in MDS that are attracting
attention include the nucleoside analogue
clofarabine10, the farnesyl-transferase inhibitor
tipifarnib8 and the thalidomide analogue
lenalidomide11. Although all these agents are
clinically promising, there is no compelling
rationale for combining them with azacitidine.
The deoxyribose form of azacitidine, decitabine,
also has significant clinical activity in MDS. This
compound is a more potent hypomethylating
agent, but does not incorporate into RNA. Its
relative clinical efficacy compared to azacitidine
deserves investigation. It is also not known
whether there is complete overlap between the
mechanism of clinical activity of the two compounds. From a molecular standpoint, the most
rational combinations involving azacitidine or
decitabine are combination epigenetic therapy,
and combinations that exploit gene reactivation. DNA methylation-associated gene silencing in cancer involves a cascade of orchestrated
molecular events12 that include methyl-binding
proteins, histone deacetylases, histone methylases and polycomb-group proteins. Each of
these pathways is attractive as a target for epigenetic therapy, and histone-deacetylase
inhibitors are already in clinical trials. Valproic
acid, a widely used medication for epilepsy, is
also a histone-deacetylase inhibitor. Remarkably, it has single-agent activity (though
modest) in MDS13, and trials of combinations
of valproic acid with hypomethylating agents
are ongoing. Finally, given that azacitidine activates the expression of readily targetable pathways4, combinations that exploit this are also
attractive. For example, hypomethylating agents
sensitize many cells to the action of retinoic
acid, and trials of combinations of azacitidine
with all-trans retinoic acid are ongoing.
How applicable might DNA-methylation
inhibitors be to other leukaemias and cancers?
One of the most important areas of research
is understanding the azacitidineMDS connection. There is nothing extraordinary about
DNA methylation in MDS15, and why that
disease should be particularly responsive is
unknown. There is evidence that azacitidine has
NATURE REVIEWS | DRUG DISCOVERY
activity in AML16, but it could be argued that

AML and MDS form a continuum of diseases
to explain this fact. Possible explanations for the
current clinical state of affairs can be found in
the facts that hypomethylating agents work
better and with less toxicity at low doses17, and
that epigenetic therapy works slowly7. MDS is a
disease of elderly patients with few therapeutic
options and a more indolent course than acute
leukaemias. By necessity, investigators interested
in MDS had to test agents at low, non-toxic
doses, and could wait several months before
determining response or failure. The discovery
of MDS as a disease responsive to hypomethylating agents might owe more to these issues
than to real biology. If this hypothesis is correct,
epigenetic therapy at low doses extended over a
period of several months of treatment might
well prove useful in other malignancies.
Jean-Pierre Issa and Hagop Kantarjian are at the
Department of Leukemia, University of Texas MD
Anderson Cancer Center, Houston, Texas 77030,
USA. Peter Kirkpatrick is at Nature Reviews Drug
Discovery. Correspondence to J.-P. I. & P. K.
e-mails: jpissa@mdanderson.org;
p.kirkpatrick@nature.com
doi:10.1038/nrd1726
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
List, A. F., Vardiman, J., Issa, J.-P. & DeWitte, T. M.

Myelodysplastic syndromes. Hematology (Am. Soc.
Hematol. Educ. Program) 297317 (2004).
Hofmann, W.-K. et al. Myelodysplastic syndromes.
Hematol. J. 5, 18 (2004).
Herman, J. G. & Baylin, S. B. Gene silencing in cancer in
association with promoter hypermethylation. N. Engl. J.
Med. 349, 20422054 (2003).
Egger, G., Liang, G., Aparicio, A. & Jones, P. A. Epigenetics
in human disease and prospects for epigenetic therapy.
Nature 429, 457463 (2004).
FDA labelling information [online], <http://www.fda.gov/
cder/foi/label/2004/050794lbl.pdf> (2004).
Greenberg, P. et al. International scoring system for
evaluating prognosis in myelodysplastic syndromes. Blood
89, 20792088 (1997).
Silverman, L. R. et al. Randomized controlled trial of
azacitidine in patients with the myelodysplastic syndrome:
a study of the cancer and leukemia group B. J. Clin. Oncol.
20, 24292440 (2002).
Faderl, S. & Kantarjian, H. M. Novel therapies for
myelodysplastic syndromes. Cancer 101, 226241 (2004).
Anzai, H., Frost, P., & Abbruzzese, J. L. Synergistic
cytotoxicity with 2-deoxy-5-azacytidine and topotecan in
vitro and in vivo. Cancer Res. 52, 21802185 (1992).
Kantarjian, H. et al. Phase 2 clinical and pharmacologic
study of clofarabine in patients with refractory or relapsed
acute leukemia. Blood 102, 23792386 (2003).
List, A. et al. Efficacy of lenalidomide in myelodysplastic
syndromes. N. Engl. J. Med. 352, 549557 (2005).
Bird, A. DNA methylation patterns and epigenetic memory.
Genes Dev. 16, 621 (2002).
Kuendgen, A. et al. Treatment of myelodysplastic
syndromes with valproic acid alone or in combination with
all-trans retinoic acid. Blood 104, 12661269 (2004).
Wijermans, P. et al. Low-dose 5-aza-2-deoxycytidine, a
DNA hypomethylating agent, for the treatment of high-risk
myelodysplastic syndrome: a multicenter phase II study in
elderly patients. J. Clin. Oncol. 18, 956962 (2000).
Claus, R. & Lubbert, M. Epigenetic targets in hematopoietic
malignancies. Oncogene 22, 64896496 (2003).
Saiki, J. H. et al. 5-azacytidine in acute leukemia. Cancer
42, 21112114 (1978).
Issa, J. P. et al. Phase I study of low-dose prolonged
exposure schedules of the hypomethylating agent 5-aza2-deoxycytidine (Decitabine) in hematopoietic
malignancies. Blood 103, 16351640 (2004).
MAY 2005 | S 7
HOT DRUGS | CANCER
Bevacizumab
Bevacizumab
Bevacizumab (Avastin; Genentech/Roche),

an antibody against vascular endothelial
growth factor, was approved by the US FDA
in February 2004 for the first-line treatment
of metastatic colorectal cancer in
combination with 5-fluorouracil-based
chemotherapy. It is the first approved agent
to target tumour angiogenesis.
Angiogenesis the development of new blood

vessels from pre-existing vasculature has a
key role in normal development, but also in
several diseases, such as cancer. For a tumour to
grow beyond a certain size, it needs a network of
blood vessels that supply nutrients and oxygen,
and remove waste products1. So, understanding
how angiogenesis is regulated in cancer has been
a major area of cancer research, particularly in
the past decade.
tumour angiogenesis was mediated by diffusible factors produced by tumour cells, and
this hypothesis stimulated efforts to identify
these factors1.
In the 1980s, such efforts led to the isolation of vascular endothelial growth factor
(VEGF), a potent stimulator of the growth
of endothelial cells, the main type of cell in
the inside lining of blood vessels1. It is now
known that VEGF, which activates receptor
tyrosine kinases on the surface of endothelial
cells (FIG. 1), is a key regulator of normal and
pathological blood vessel growth 1. In the
early 1990s, the demonstration that inhibition of VEGF-induced angiogenesis using a
monoclonal antibody against VEGF markedly
suppressed tumour growth in vivo3 led to the
development of bevacizumab.
Drug properties
Basis of discovery
The hypothesis that inhibiting angiogenesis

might be an effective anticancer strategy was
put forward more than 30 years ago2. Earlier
experiments had provided evidence that
Bevacizumab is a humanized version of the

monoclonal antibody against VEGF that was
used in early proof-of-principle experiments3,4.
It binds to VEGF, which prevents its interaction with the VEGF receptor tyrosine kinases
VEGF
VEGFR1
Bevacizumab
Decoy effect on VEGF signalling

Induction of uPA, tPA, MMP9
Vascular-bed-specific release of growth factors
VEGFR2
Proliferation
Migration
Survival
Angiogenesis
Permeability
Figure 1 | Simplified view of VEGF signalling and tumour angiogenesis. The receptors for vascular
endothelial growth factor (VEGF, also known as VEGF-A) VEGFR1 (also known as Flt1) and VEGFR2
(also known as Flk1 or KDR) are expressed on the surface of blood endothelial cells. VEGFR2 is
thought to be the major mediator of endothelial cell mitogenesis, survival and microvascular permeability,
whereas VEGFR1 does not seem to mediate an effective mitogenic signal in endothelial cells, but does
have other activities that can be important in tumour growth and metastasis, including the induction of
matrix metalloproteinases (MMPs). tPA, tissue plasminogen activator; uPA, urokinase-type plasminogen
activator. Adapted from REF. 1.
S8
| MAY 2005
VEGFR1 and VEGFR2 (FIG. 1), and inhibits the

growth of human tumour cell lines in mice1.
Following early clinical trials showing that
bevacizumab as a single agent was relatively
non-toxic, and that it could be added to standard cytotoxic chemotherapy regimes, large
clinical trials were initiated in several cancer
types, including colorectal cancer1.
Clinical data
It has been estimated that colorectal cancer

is the fourth largest cause of cancer deaths5.
For many years, chemotherapy of the disease
has been based on traditional cytotoxic
drugs, in particular 5-fluorouracil (5-FU) in
combination with leucovorin (LV). More
recently, the addition of the cytotoxic drugs
irinotecan (Camptosar; Pfizer) or oxaliplatin
(Eloxatin; Sanofi-Aventis) to 5-FU/LV
regimes has been shown to prolong survival
in advanced colorectal cancer.
Bevacizumab, in combination with 5-FUbased chemotherapy, was studied in randomized, controlled trials as a first-line treatment
for patients with metastatic carcinoma of the
colon or rectum6,7. In the pivotal Phase III
study, which involved more than 800 patients,
bevacizumab (5 mg per kg every 2 weeks) or
placebo was given to the patients in addition
to bolus-IFL (irinotecan 125 mg per m2
intravenously, 5-FU 500 mg per m2 intravenously and LV 20 mg per m2 intravenously
once weekly for 4 weeks every 6 weeks).
Median overall survival was significantly
increased from 15.6 months in the bolus-IFL
+ placebo arm to 20.3 months in the bolusIFL + bevacizumab arm6,7. Similar increases
were also seen in progression-free survival
(6.4 versus 10.6 months), overall response
rate (35 % versus 45%) and duration of
response (7.1 months versus 10.4 months).
Bevacizumab was generally well tolerated
in these clinical trials; however, some serious
and unusual toxicities were noted. In particular,
bevacizumab was associated with gastrointestinal perforations and wound healing
complications in ~2% of patients 7. Other
adverse events associated with bevacizumab
use include thromboembolic complications,
hypertension, bleeding and proteinuria7.
Indications
Bevacizumab, used in combination with intravenous 5-FU-based chemotherapy, is approved

by the FDA for the first-line treatment of
patients with metastatic carcinoma of the colon
or rectum7.
HOT DRUGS | CANCER
ANTI-VEGF THERAPIES | VIEW FROM THE CLINIC

Analysing clinical issues for anti-VEGF
therapies is Lee M. Ellis, M.D., Professor of
Surgical Oncology and Professor of Cancer
Biology at the University of Texas MD
Anderson Cancer Center, Texas, USA. His
research interests are defining the role of
angiogenesis in primary and metastatic human
colorectal, pancreatic and gastric cancers, and
evaluating mechanisms of anti-angiogenic
therapies, such as those targeted against VEGF.
What impact have anti-VEGF therapies such as

bevacizumab had on the treatment of cancer?
The results from the original Phase III trials
comparing IFL to IFL and bevacizumab6, and
the subsequent FDA approval of bevacizumab
in combination with intravenous 5-FU-based
regimens, have changed the standard of care
for patients with metastatic colorectal cancer
(CRC). More recently, it has been announced
that in Phase III trials, bevacizumab provided
benefit to patients with advanced non-small-cell
lung cancer (NSCLC). In addition, Phase I/II
studies have suggested that the addition of
bevacizumab to other therapeutic modalities
(chemotherapy and radiation therapy) might
be efficacious in patients with other malignancies, such as pancreatic carcinoma, renal-cell
carcinoma and locally advanced rectal cancer.
Another Phase III trial in metastatic CRC
evaluated patients randomized to FOLFOX
with or without the VEGF receptor tyrosine
kinase inhibitor PTK787/ZD222584. It has been
announced that after ~1,200 patients had been
accrued , an investigator analysis demonstrated
an improvement in progression-free survival,
but a central radiology review did not note a
statistically significant improvement in progression-free survival. Data on overall survival
should be available in 2006.
How well are the mechanisms of benefit of antiVEGF therapies understood?
It is important to differentiate anti-VEGF
therapy from other anti-angiogenic therapies,
many of which are specifically intended to
disrupt endothelial cell signalling or survival
pathways. VEGF is a pluripotent factor that
has multiple effects on the vasculature, including induction of angiogenesis and enhancement of endothelial cell survival, the ability to
induce vascular permeability, and production
of a dilated and disorganized vascular network. This altered tumour vascular network
leads to inefficient blood flow, which can theoretically be normalized with anti-VEGF
therapy, augmenting delivery of chemotherapy
and oxygen8.
It was initially believed that the VEGF receptors (VEGFRs) are present only on endothelial
cells, but recent studies have demonstrated that
VEGFRs are present on tumour cells9,10. So,
another potential mechanism of action of antiVEGF therapy is a direct effect on tumour cells,
where it might inhibit processes involved in
tumour progression and metastasis.
In addition to the above novel mechanisms
for anti-VEGF therapy, it is still believed that
anti-VEGF therapy can inhibit tumour angiogenesis. A detailed analysis has shown that the
addition of bevacizumab to chemotherapy
leads to a remarkable improvement in progression-free survival relative to the incremental
improvement in response rate as seen with
chemotherapy alone (A. Grothey, personal
communication). This observation suggests
that anti-VEGF therapy might indeed be antiangiogenic, although this is indirect evidence.
How do you see anti-VEGF therapy evolving in
the next few years?
Although most studies with bevacizumab have
been carried out in patients with advanced-stage
disease, future studies will include its use in
combination with other therapies in the neoadjuvant and adjuvant settings. Interesting
results have already been obtained with the use
of bevacizumab in addition to chemoradiation
therapy for locally advanced rectal and pancreatic cancers11,12, and ongoing trials in CRC
should provide some insight into the appropriate use of this agent in the adjuvant setting.
However, one must consider the long-term
effects of bevacizumab. In a meta-analysis of
randomized Phase III trials, a 2.3-fold increase
in arterial thrombotic events, including stroke,
myocardial infarction, angina and transient
ischaemic attacks, was noted. This is particularly
important in patients who will be receiving
bevacizumab in the adjuvant setting, and longterm follow up of cardiovascular events is crucial in such studies, especially as many patients
would remain disease-free without therapy or
with chemotherapy alone as adjuvant therapy.
Although anti-angiogenic therapy was originally intended for use as single-agent therapy,
and was then combined with chemotherapy,
there is also great potential for use in combination with other biological agents. One very
interesting observation during the past year has
been the results from the BOND2 study (a follow-up of the BOND1 study of cetuximab with
or without irinotecan in irinotecan-refractory
patients). In this study, patients with metastatic
CRC who had progressed on irinotecan-based
therapy were randomized to receive cetuximab
plus bevacizumab or cetuximab plus bevacizumab plus irinotecan13. The combination of

the two biological agents alone led to a remarkable 23% response rate in the refractory setting,
and the addition of irinotecan to the two biological agents led to a 38% response rate. These
results suggest that biological agents can be
combined without undue toxicity and that dual
targeting of various mediators of tumour progression is efficacious in and of itself. However,
the addition of chemotherapy to the biological
agents is likely to achieve an even better result.
In summary, it is clear that bevacizumab
improves the effects of chemotherapy in
patients with metastatic CRC and NSCLC, and
possibly other cancers. The efficacy of singleagent anti-VEGF therapy requires further study,
in both the advanced-disease setting as well as
the adjuvant setting. It is imperative that we
determine the underlying mechanism of action
of anti-VEGF drugs in order to develop predictive markers and better define the optimal use of
these agents, as well as their potential short- and
long-term toxicities.
Lee M. Ellis is at the University of Texas MD
Anderson Cancer Center, PO Box 301402,
Houston, Texas 77230-1402, USA. Peter
Kirkpatrick is at Nature Reviews Drug Discovery.
Correspondence to L.M.E. & P. K.
e-mails: lellis@mdanderson.org;
doi:10.1038/nrd1727
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Ferrara, N. et al. Discovery and development of bevacizumab,

an anti-VEGF antibody for treating cancer. Nature Rev. Drug
Discov. 3, 391400 (2004).
Folkman, J. Tumor angiogenesis: therapeutic implications.
N. Engl. J. Med. 285, 11821186 (1971).
Kim, K. J. et al. Inhibition of vascular endothelial growth
factor-induced angiogenesis suppresses tumour growth in
vivo. Nature 362, 841844 (1993).
Presta, L. G. et al. Humanization of an anti-vascular
endothelial growth factor monoclonal antibody for the
therapy of solid tumors and other disorders. Cancer Res.
57, 45934599 (1997).
Parkin, D. M. et al. Cancer burden in the year 2000. The
global picture. Eur. J. Cancer 37, S4S66 (2001).
Hurwitz, H. et al. Bevacizumab plus irinotecan, fluorouracil,
and leucovorin for metastatic colorectal cancer. N. Engl. J.
Med. 350, 23352342 (2004).
FDA labelling information [online] <http://www.fda.gov/cder/
foi/label/2004/125085lbl.pdf> (2004).
Jain, R. K. Normalization of tumor vasculature: an emerging
concept in antiangiogenic therapy. Science 307, 5862 (2005).
Dias, S. et al. Autocrine stimulation of VEGFR-2 activates
human leukemic cell growth and migration. J. Clin. Invest.
106, 511521 (2000).
Fan, F. et al. Expression and function of vascular endothelial
growth factor receptor-1 on human colorectal cancer cells.
Oncogene 24, 26472653 (2005).
Willett, C. G. et al. Direct evidence that the VEGF-specific
antibody bevacizumab has antivascular effects in human
rectal cancer. Nature Med. 10, 145147 (2004).
Crane, C. H. et al. Preliminary results of a Phase I study of
rhuMab VEGF (bevacizumab) with concurrent radiotherapy
(XRT) and capecitabine (CAP). Proc. ASCO Gastrointestinal
Cancers Symp. A85 (2004).
Saltz, L. B. et al. Interim report of randomized Phase II
trial of cetuximab/bevacizumab/irinotecan (CBI) versus
cetuximab/bevacizumab (CB) in irinotecan-refractory
colorectal cancer. Proc. ASCO Gastrointestinal Cancers
Symp. A169b (2005).
MAY 2005 | S 9
HOT DRUGS | CANCER
Cetuximab
Cetuximab
Cetuximab (Erbitux; ImClone Systems/

Bristol-Myers Squibb) is a monoclonal
antibody that binds to the epidermal growth
factor receptor, which is important in the
growth of many cancers. In February 2004,
it was granted accelerated approval by the
US FDA for the treatment of metastatic
colorectal cancer on the basis of tumour
response rates in Phase II trials.
thereby inhibiting downstream signal transduction2,5,6 (FIG. 1). In preclinical studies,

cetuximab showed promising anticancer
activity both alone and in combination with
traditional cytotoxic drugs2,5,7, prompting its
clinical evaluation in a range of cancers in
which EGFR is thought to have an important
role, including colorectal cancer.
irinotecan had an objective response rate of

22.9%, and those receiving cetuximab alone
had objective response rate of 10.8%. The
median duration of response was 5.7 months
in the combination arm and 4.2 months in
the monotherapy arm, and the median time
to disease progression was 4.1 months in the
combination arm and 1.5 months in the
monotherapy arm6.
Clinical data
Colorectal cancer is one of the most commonly diagnosed cancers, and has been estimated to be the fourth largest cause of cancer
deaths worldwide1. Recently, there have been
important advances in the therapy of this disease, such as the introduction of the cytotoxic
drugs irinotecan (Camptosar; Pfizer) and
oxaliplatin (Eloxatin; Sanofi-Aventis), and the
antibody bevacizumab (Avastin; Genentech/
Roche), which targets tumour angiogenesis.
Nevertheless, a major need remains for novel
agents for treating colorectal cancer, especially
for those patients who fail to respond to
current treatments or who develop resistance
to them.
Cetuximab was studied in trials involving

patients with EGFR-expressing metastatic
colorectal cancer, whose disease had progressed after receiving an irinotecan-containing
regimen6. In a randomized controlled trial
conducted in 329 such patients, patients
received either cetuximab (400 mg per m 2
initial dose, followed by 250 mg per m2
weekly until disease progression or unacceptable toxicity) and irinotecan, or cetuximab
alone6. Patients receiving cetuximab and
Drug properties
Cetuximab is a recombinant, human/mouse

chimeric monoclonal antibody that binds
specifically to the extracellular domain of the
human EGFR2,5,6. Bound antibody competitively inhibits the binding of epidermal
growth factor (EGF) and other ligands to
EGFR, blocking receptor phosphorylation
and activation of receptor-associated kinases,
S10
Cetuximab, used in combination with irinotecan, is approved by the FDA for the treatment
of EGFR-expressing, metastatic colorectal
carcinoma in patients who are refractory to
irinotecan-based chemotherapy6.
Cetuximab administered as a single agent
is approved by the FDA for the treatment of
EGFR-expressing, metastatic colorectal carcinoma in patients who are intolerant to
irinotecan-based chemotherapy6.
Ligand binding
Basis of discovery
Advances in the understanding of the aberrant signalling pathways involved in cancer

have led to considerable efforts to develop
therapies that target specific elements of
these pathways, in the hope that such agents
will have greater activity and higher specificity for tumour cells than traditional cytotoxic drugs. Receptor tyrosine kinases
which have key roles in signalling pathways
that regulate processes such as cell proliferation, apoptosis and angiogenesis have
attracted much attention, as several of them
have been strongly implicated in tumour
growth and progression, including the epidermal growth factor receptor (EGFR) 2,3
(FIG. 1) . The EGFR is expressed in a wide
range of solid tumours, including colorectal
cancers4, which has prompted the development of a number of agents that inhibit its
activity24.
Indications
Ligand (EGF, TGF)
Cetuximab
EGFR
Autophosphorylation
TK
TK
TK
TK
Gefitinib, erlotinib
Activation of signal-transduction
cascades (for example, MAPK)
Cell
proliferation
Apoptosis
Invasion and
metastasis
Angiogenesis
Figure 1 | EGFR and the mode of action of cetuximab. The epidermal growth factor receptor (EGFR)
is one of four members of the erbB family of receptor tyrosine kinases, which consist of an extracellular
domain that can bind ligands, a transmembrane domain and an intracellular tyrosine kinase domain8.
A simplified illustration of the EGFR signal transduction pathway is shown. Binding of a ligand to EGFR
causes receptor dimerization (either with another EGFR monomer or with another member of the erbB
family), leading to tyrosine kinase activation8. The resultant receptor autophosphorylation initiates
signal-transduction cascades involved in cell proliferation and survival8. Cetuximab blocks binding of
ligands to EGFR, thereby inhibiting receptor phosphorylation and downstream events. Agents that
inhibit the tyrosine kinase activity of EGFR have also been developed and, of these, two have been
approved by the FDA so far: gefitinib (Iressa; AstraZeneca) and erlotinib (Tarceva; Genentech/Roche),
for the treatment of advanced non-small-cell lung cancer. MAPK, mitogen-activated protein kinase;
TGF-, transforming growth factor-; TK, tyrosine kinase domain.
| MAY 2005
HOT DRUGS | CANCER
COLORECTAL CANCER | VIEW FROM THE CLINIC

Analysing clinical issues for colorectal cancer
is Richard M. Goldberg, M.D., Professor of
Medicine, Division Chief, Director of Oncology
Services, and Associate Director for Clinical
Research at the Lineberger Comprehensive
Cancer Center, University of North Carolina,
USA. His research interests include clinical
and translational research in gastrointestinal
cancers, particularly colorectal cancer (CRC),
and the development and integration of novel
therapies into treatment programmes. He has
run a number of Phase I, II and III clinical
trials testing new chemotherapy regimens in
patients with advanced malignancies and
gastrointestinal cancers.
What are the current key areas in the pharmacotherapy of CRC?

With neo-adjuvant therapy, one area that shows
promise is investigations into the potential roles
of irinotecan, oxaliplatin, bevacizumab and
cetuximab with radiation and 5-FU in the neoadjuvant management of locally advanced
rectal cancer. Combinations of these therapies
with 5-FU and radiation seem to be pushing the
pathological complete remission rate up from
less than 10% with 5-FU plus radiation to above
25% in some preliminary reports9.
There is likely to be evolution in the optimal adjuvant therapy for stage III colon cancer.
Bevacizumab, cetuximab and other agents
under investigation might well improve cure
rates when combined with chemotherapy.
Current trials, such as NSABP CO-8 and the
Intergroup N-0147, and future studies in
development, will help to address these issues.
For treatment of patients with advanced
disease, a key issue is how to optimize the use
of targeted agents, including those approved
for use (bevacizumab and cetuximab) and
those under investigation (PTK787, IMC18F1,
AZ2171, VEGF-Trap and others). Recently,
results were presented from the BOND2 study,
a randomized Phase II trial which assigned
patients with advanced CRC that was
refractory to irinotecan to bevacizumab in
combination with cetuximab plus or minus
irinotecan10. This study suggests that therapy
with an antibody to VEGF and an antibody
to EGFR can result in tumour shrinkage and
meaningful benefit in many patients, and that
this effect is augmented by the addition of
irinotecan. These data raise the possibility that
combined biological therapies, such as bevacizumab plus cetuximab as used in the
BOND2 trial, might become a viable strategy
for the management of advanced disease
without chemotherapy.
Prevention strategies in CRC have been set

back considerably by the recent toxicity findings
associated with treatment using rofecoxib as a
means of polyp prevention11. The identification
of new prevention strategies has the potential to
make the biggest impact on deaths from CRC.
Early detection methods such as CT colonography and faecal screening for tumour-associated DNA when further refined also have the
potential to reduce the mortality from CRC.
What have been the most significant recent
advances in disease pharmacotherapy?
We now have proof that neo-adjuvant radiation and chemotherapy improves local control
and quality of life in locally advanced rectal
cancer12. The introduction of monoclonal
antibodies targeting EGFR and VEGF into the
therapeutic armamentarium has opened the
door to a new paradigm of therapy. The testing and approval of irinotecan- and oxaliplatinbased therapies for treatment of CRC have led
to median survivals exceeding 20 months
with combinations of chemotherapy with
or without biologicals compared with the
12 months seen with 5-FU or capecitabine
alone13. In addition, various small-molecule
inhibitors of tyrosine kinases are now being
evaluated in clinical trials. Finally, there is a
growing appreciation of the importance of
pharmacogenetic factors that affect drug
response and toxicity, such as the finding that
patients with the UGT1A1 7/7 polymorphism
are more susceptible to toxicity. These data
can be used to influence physician choice of
drug regimens14.
How would you expect the treatment of CRC to
evolve in the next 5 years?
In the next 5 years, I expect 5-year overall
survival for stage II colon cancer to approach
85%, 5-year overall survival for stage III
colon cancer to approach 75%, 5-year overall
survival for stage IV CRC to approach 15%,
and management of advanced disease to
result in median survivals of 3036 months.
These advances should come from continued
improvements in surgical techniques and
staging, more effective chemotherapy and
biological therapy combinations, and the
introduction of new drugs. I would anticipate the approval of 23 new agents for the
management of CRC, including new antiangiogenic agents and possibly small-molecule inhibitors. In general, there will be a
shift in drug development to biologicals from
chemotherapy agents because of the proof
that the cetuximab and bevacizumab trials
have provided that biologicals are useful

tools for treating CRC.
Owing to the availability of novel agents
such as cetuximab and bevacizumab, and
advances in the characterization of tumour
and patient characteristics, in 5 years we
should be on the verge of individualized
chemotherapy/biological therapy prescriptions, an approach that will minimize the toxicity and maximize the activity of treatment
programmes. Nevertheless, it is clear that there
is much to learn before we can optimize the
use of novel therapies. The recent disappointing announcement on the CONFIRM1 trial, in
which the addition of PTK787 to FOLFOX
failed to reach statistical significance in the
predetermined prolongation of time to progression, highlights the uncertainty of new
drug development in this era.
Richard Goldberg is at the University of North
Carolina at Chapel Hill, 27599-7305, USA. Peter
Correspondence to R.M.G. & P.K.
e-mails: richard_m_goldberg@med.unc.edu;
doi:10.1038/nrd1728
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Parkin, D. M. et al. Cancer burden in the year 2000. The

global picture. Eur. J. Cancer 37, S4S66 (2001).
Baselga, J. The EGFR as a target for anticancer therapy
focus on cetuximab. Eur. J. Cancer 37, S16S22 (2001).
Dancey, J. & Sausville, E. A. Issues and progress with
protein kinase inhibitors for the treatment of cancer. Nature
Rev. Drug Discov. 2, 296313 (2003)
Salomon, D. S. et al. Epidermal growth factor-related
peptides and their receptors in human malignancies.
Crit. Rev. Oncol. Hematol. 19, 183232 (1995).
Goldstein, N. I. et al. Biological efficacy of a chimeric
antibody to the epidermal growth factor receptor in a
human tumor xenograft model. Clin. Cancer. Res. 1,
13111318 (1995).
FDA labelling information [online] (cited 6 May 2004)
<http://www.fda.gov/cder/foi/label/2004/125084lbl.pdf>
(2004).
Prewett, M. C. et al. Enhanced antitumor activity of antiepidermal growth factor receptor monoclonal antibody
IMC-C225 in combination with irinotecan (CPT-11) against
human colorectal tumor xenografts. Clin. Cancer Res. 8,
9941003 (2002).
Yarden, Y. & Sliwkowski, M. X. Untangling the ErbB
signalling network. Nature Rev. Mol. Cell Biol. 2, 127137
(2001).
Hofheinz, R.-D. et al. Phase I trial of capecitabine and
weekly irinotecan in combination with radiotherapy for
neoadjuvant therapy of rectal cancer. J. Clin. Oncol. 23,
13501357 (2005).
Saltz, L. B. et al. Interim report of randomized Phase II trial
of cetuximab/bevacizumab/irinotecan (CBI) versus
cetuximab/bevacizumab (CB) in irinotecan-refractory
colorectal cancer. Proc. ASCO Gastrointestinal Cancers
Symp. A169b (2005).
Solomon, S. D. et al. Cardiovascular risk associated with
celecoxib in a clinical trial for colorectal adenoma prevention.
N. Engl. J. Med. 352, 10711080 (2005).
Sauer, R. et al. Preoperative versus postoperative
chemoradiotherapy for rectal cancer. N. Engl. J. Med. 352,
17311740 (2004).
Grothey, A. et al. Overall survival in advanced colorectal
cancer correlates with the proportion of patients treated
with 5-fluorouracil/leucovorin, irinotecan, and oxaliplatin.
J. Clin. Oncol. 22, 12091214 (2004).
Iyer, L. et al. UGT1A1*28 polymorphism as a determinant of
irinotecan disposition and toxicity. Pharmacogenomics J.
2, 4347 (2002).
MAY 2005 | S 1 1
HOT DRUGS | CANCER
Clofarabine
Clofarabine
Clofarabine (Clolar; Genzyme), a purine

nucleoside antimetabolite, was granted
accelerated approval by the US FDA for the
treatment of paediatric patients with relapsed
or refractory acute lymphoblastic leukaemia in
December 2004. It is the first new drug for
paediatric leukaemia to be approved in more
than a decade, and the only one to receive
approval for paediatric use before adult use.
Leukaemias, such as acute lymphoblastic

leukaemia (ALL), are the most common
malignancies in children and adolescents. In
recent decades, major progress has been made
in understanding the pathobiology of acute
leukaemias and the host pharmacogenetic
factors that affect drug metabolism and disposition in these diseases1. In parallel with
these advances, the cure rate of childhood
ALL reached ~80% in some clinical trials in
the 1990s1. Although part of this success can
be credited to the modification of therapy
according to stringently defined risk groups,
the greatest contribution has come from the
optimal use of drugs developed between the
1950s and 1970s. Nevertheless, the prognosis
for children with ALL who do not achieve or
maintain complete remission is very poor,
and so new drugs and therapeutic approaches
are urgently needed.
of total platelet recovery (CRp), defined as

meeting all criteria for CR except for recovery
of platelet counts6.
Six patients (12%) achieved a CR and
four patients (8%) achieved a CRp6. Five
patients achieved a partial response (defined
as complete disappearance of circulating
blasts, an M2 bone marrow (>5% and <25%
blasts), and appearance of normal progenitor
cells or an M1 bone marrow that did not
qualify for CR or CRp)6. Of the fifteen
patients responding, six had post-clofarabine
bone-marrow transplantation, preventing
determination of response duration. In the
other nine responding patients, the response
durations for CR were 43, 50, 82, 93+ and
160+ days, the response duration for CRp
was 32 days, and the response durations for
PR were 7, 16 and 21 days6.
diphosphokinases to the active 5-triphosphate form. Clofarabine triphosphate inhibits

DNA synthesis through its inhibitory action
on both ribonucleotide reductase and DNA
polymerases, and is cytotoxic to rapidly proliferating and quiescent cancer cell types
in vitro2,46. Clofarabine was well tolerated
and showed significant activity in a Phase I
trial in paediatric patients with refractory
and relapsed leukaemia7, prompting its evaluation in Phase II trials in paediatric and
adult patients with ALL and acute myeloid
leukaemia (AML)6,8.
Clinical data
The efficacy and safety of clofarabine (52 mg

per m2 per day intravenously for 5 days,
repeated every 26 weeks following recovery
or return to baseline organ function) were
evaluated in a single-arm study involving 49
paediatric patients (median age, 12 years) with
relapsed or refractory ALL6. Most patients
(46/49) had received 24 prior regimens. The
study endpoints were the rate of complete
remission (CR), defined as no evidence of
circulating blasts or extramedullary disease, an
M1 bone marrow (<5% blasts), and recovery
of peripheral platelet and absolute neutrophil
counts, and complete remission in the absence
Indications
Clofarabine is approved by the FDA for the

treatment of paediatric patients 121 years
old with relapsed or refractory ALL after at
least two prior regimens6. This use is based
on the induction of complete responses 6.
Randomized trials demonstrating increased
survival or other clinical benefit have not
been conducted6.
Basis of discovery
Nucleoside analogues, such as cytarabine, are

an important class of highly effective cytotoxic
drugs for the treatment of leukaemias and
other haematological cancers2. The purine
nucleoside analogues cladribine and fludarabine (FIG. 1a) have shown activity in acute
leukaemias, but at dose levels associated with
prohibitive neurotoxicities2. Both drugs
inhibit DNA synthesis after intracellular conversion to their active triphosphate metabolites, but their primary mode of action is different. Cladribine triphosphate particularly
inhibits ribonucleotide reductase, thereby
decreasing cellular deoxynucleotide pools,
whereas fludarabine triphosphate primarily
inhibits DNA polymerases2. Clofarabine (FIG.
1b) was rationally developed on the basis of
experience with cladribine and fludarabine,
with the aim of improving on their efficacy
and minimizing toxicity2,3.
Drug properties
Clofarabine is converted intracellularly by

deoxycytidine kinase to the 5-monophosphate metabolite, and then by mono- and
S12
NH2
N
NH2
NH2
N
O
HO
N
O
Cl
HO
HO N
O
HO
F N
O
Cl
OH
OH
Cladribine
OH
Fludarabine
OH
Clofarabine
2-chloro-9-(2-deoxy-2-fluoro-D-arabinofuranosyl)-9Hpurin-6-amine;
C10H11ClFN5O3; Mr = 303.68;
CAS number: 123318-82-1
Figure 1 | Clinically used purine nucleoside analogues. a | Chemical structures of cladribine and
fludarabine. b | Clofarabine, a hybrid of cladribine and fludarabine, is also substituted with a halogen at
position 2 of the purine ring, which makes it resistant to deamination by adenosine deaminase2.
Substitution of a fluorine moiety at the 2 carbon in the arabino configuration in clofarabine increases its
stability in acid compared with cladribine and its resistance to cleavage of the glycosidic bond by bacterial
purine nucleoside phosphorylases2,3, which for fludarabine leads to the formation of 2-F adenine, a toxic
compound with no antitumour selectivity.
| MAY 2005
HOT DRUGS | CANCER
ACUTE LEUKAEMIAS | VIEW FROM THE CLINIC

Analysing clinical issues for acute leukaemias
are Ching-Hon Pui, M.D., Director of the
Leukemia/Lymphoma Division of St Jude
Childrens Research Hospital, American
Cancer Society F. M. Kirby Clinical Research
Professor and Professor of Pediatrics at the
University of Tennessee Health Science Center,
USA, and Sima Jeha, M.D., Director of
Developmental Therapeutics in the Leukemia/
Lymphoma Division of St Jude Childrens
Research Hospital and Associate Professor of
Pediatrics at the University of Tennessee Health
Science Center. Dr Puis research interest is
studying the biology and improving the
outcome of childhood leukaemias. Dr Jehas
research interest is developing new agents for
the treatment of leukaemias and lymphomas.
What are the unmet needs in acute leukaemia?

Although cure rates for paediatric ALL in the
past decade reached ~80%, the prognosis for
patients who do not achieve or maintain complete remission remained dismal. Adults with
ALL are more likely than children to have drugresistant disease at presentation and less likely to
tolerate chemotherapy or to comply with treatment plans. Their 5-year survival rates are only
in the range 3040%, despite the frequent use of
haematopoietic stem-cell transplantation1.
Treatment results also remain disappointing for both children and adults with AML:
only about half of the childhood cases, and
only about one-third of cases in patients 18 to
60 years old can be cured with contemporary
treatments9,10. The full extent of the problem
can be better appreciated when one considers
that more than half of all cases of AML affect
patients 60 years of age or older, the vast
majority of whom received only palliative
therapy until recently. This elderly population of AML patients is growing rapidly and
could double by 203010. There is therefore an
urgent need for more effective and well-tolerated treatments for children with AML and
for all adults with acute leukaemia, especially
the elderly.
What progress has been made in recent years in
new drug development for acute leukaemias?
The past decade has seen considerable
progress in defining the molecular pathways
that drive the pathogenesis of acute
leukaemia. These analyses have contributed to
more precise leukaemia classification systems,
allowing more selective conventional therapy,
and have also identified several pivotal genes
or fusion genes whose protein products seem
suitable for targeted therapy, including FLT3,
PTEN, NOTCH1 and RAS1. The paradigm of

such therapy is imatinib mesylate (Gleevec;
Novartis), a potent inhibitor of the BCRABL
tyrosine kinase that is used to treat Philadelphia
chromosome-positive leukaemias. It is well
tolerated, even in the elderly, and can induce
not only clinical but also molecular remissions
in most patients treated11,12.
Although not directed to molecular targets,
several new formulations of existing agents,
such as L-asparaginase, vincristine and cytarabine, have been produced in an effort to
improve the efficacy and reduce the toxicity of
the parent compounds. Epigenetic therapy
with decitabine and 5-azacitidine to reverse
aberrant methylation, or with histone deacetylase inhibitors (SAHA and valproate) to reactivate silenced tumour-suppressor genes, has also
shown promise13. Novel nucleoside analogues
seem to be preferentially effective against
particular subtypes of acute leukaemia: gemcitabine for ALL, 5-azacitidine and decitabine
for myelodysplasia preceding AML, nelarabine
for T-cell ALL and cladribine for AML2,14.
Fludarabine is effective for both lymphoid and
myeloid leukaemia. Finally, clofarabine, which
was rationally designed to incorporate the most
favourable properties of its congeners fludarabine and cladribine2, has now been approved
for use in paediatric patients with relapsed or
refractory ALL6.
What impact will clofarabine have on the
treatment of acute leukaemias?
In addition to providing a valuable treatment
option for paediatric patients with ALL, clofarabine also has substantial activity against
AML7,8,14. In Phase I trials, 1 of 16 adults and 1
of 8 children with AML achieved complete
remission7,14. More encouraging are the results
of Phase II trials in adults with AML. In one
study of 31 adults with relapsed AML, complete remissions were achieved in 6 of 8
patients in first relapse after an initial remission of 12 months or more; in 2 of 11 in first
remission of less than 12 months; and in 5 of
12 in second or subsequent relapse8. In another
study, clofarabine induced complete remission
in 16 of 27 patients (6079 years of age) with
newly diagnosed AML who were considered
unfit for conventional chemotherapy; the drug
was associated with grade 3 toxicities in only 3
patients 15. On the basis of the hypothesis
that clofarabine might modulate cytarabine
triphosphate accumulation and the lack of
overlapping toxicity, Faderl and colleagues
combined this agent with cytarabine in Phase
II adult trials16,17. The combination treatment
was active in myeloid malignancies and well

tolerated. Because clofarabine inhibits the
repair of DNA damage, ongoing studies are
testing this analogue in combination with
DNA-damaging agents, such as topoisomerase
II inhibitors and alkylators18. Taken together,
these preliminary results warrant further
investigation of clofarabine to further define its
role in the treatment of acute leukaemias.
Ching-Hon Pui and Sima Jeha are at St Jude
Children's Research Hospital, 332 North Lauderdale
Street, Memphis, Tennessee 38105-2794, USA.
Peter Kirkpatrick is at Nature Reviews Drug
Discovery. Correspondence to C.H. P. and P.K. e-mails:
ching-hon.pui@stjude.org; p.kirkpatrick@nature.com
doi:10.1038/nrd1729
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Pui, C. H. et al. Acute lymphoblastic leukemia. N. Engl. J.

Med. 350, 15351548 (2004).
Faderl. S. et al. The role of clofarabine in hematologic and
solid malignancies - development of a next-generation
nucleoside analog. Cancer Res. 31 Mar 2005
(doi: 10.1002/cncr.21005).
Montgomery, J. A. et al. Synthesis and biologic activity of
22-fluoro-2-halo derivatives of 9--D-arabinofuranosyladenine.
J. Med. Chem. 35, 397401 (1992).
Parker, W. B. et al. Effects of 2-chloro-9-(2-deoxy-2-fluoro-D-arabinofuranosyl)adenine on K562 cellular metabolism
and the inhibition of human ribonucleotide reductase and
DNA polymerases by its 5-triphosphate. Cancer Res. 51,
23862394 (1991).
Carson, D. A. et al. Oral antilymphocyte activity and induction
of apoptosis by 2-chloro-22-arabino-fluoro-22-deoxyadenosine. Proc. Natl Acad. Sci. USA 89, 29702974 (1992).
FDA labelling information [online], <http://www.fda.gov/
Jeha, S. et al. Clofarabine, a novel nucleoside analog, is
active in pediatric patients with advanced leukemia. Blood
103, 784-789 (2004).
Kantarjian, H. et al. Phase 2 clinical and pharmacologic
study of clofarabine in patients with refractory or relapsed
acute leukemia. Blood 102, 23792386 (2003).
Pui, C. H. et al. Childhood and adolescent lymphoid and
myeloid leukemia. Hematology (Am. Soc Hematol. Educ.
Program) 118145 (2004).
Stone, R. M. et al. Acute myeloid leukemia. Hematology
(Am. Soc. Hematol. Educ. Program) 98117 (2004).
Wassmann, B. et al. A randomized multicenter open label
phase II study to determine the safety and efficacy of
induction therapy with imatinib (Glivec, formerly STI571) in
comparison with standard induction chemotherapy in elderly
(>55 years) patients with Philadelphia chromosome-positive
(Ph+/BCR-ABL+) acute lymphoblastic leukemia (ALL)
(CST1571ADE 10). Ann. Hematol. 82, 716720 (2003).
Thomas, D. A. et al. Treatment of Philadelphia chromosomepositive acute lymphocytic leukemia with hyper-CVAD and
imatinib mesylate. Blood 103, 43964407 (2004).
Egger, G., Liang, G., Aparicio, A. & Jones, P. A. Epigenetics
in human disease and prospects for epigenetic therapy.
Nature 429, 457463 (2004).
Kantarjian, H. M. et al. Phase I clinical and pharmacology
study of clofarabine in patients with solid and hematologic
cancers. J. Clin. Oncol. 21, 11671173 (2003).
Burnett, A. K. et al. A phase 2 evaluation of single agent
clofarabine as first line treatment for older patients with AML
who are not considered fit for intensive chemotherapy.
Blood 104, 248a (2004).
Faderl, S. et al. Results of a phase 1-2 study of clofarabine
in combination with cytarabine (ara-C) in relapsed and
refractory acute leukemias. Blood 105, 940947 (2005).
Faderl, S. et al. Clofarabine plus cytarabine (ARA-C)
combination is active in newly diagnosed patients (pts) =
age 50 with acute myeloid leukemia (AML) and
myelodysplastic syndrome (MDS). Blood 104, 250a (2004).
Faderl, S. et al. Phase I study of clofarabine plus idarubicin
and clofarabine plus idarubicin plus cytarabine (ARA-C) in
patients (PTS) with relapsed and primary refractory acute
myeloid leukemia (AML), myelodysplastic syndrome (MDS),
and myeloid blast phase of chronic myeloid leukemia (CML).
Blood 104, 501a (2004).
MAY 2005 | S 1 3
HOT DRUGS | CANCER
Erlotinib hydrochloride
Erlotinib
Erlotinib hydrochloride (Tarceva; OSI Pharmaceuticals/Genentech/Roche), a member of

a class of targeted anticancer drugs that
inhibit the activity of the epidermal growth
factor receptor, was approved by the US
FDA in November 2004 for the treatment of
advanced non-small-cell lung cancer after
failure of at least one prior chemotherapy
regimen. It is the first such drug to demonstrate an increase in survival in Phase III trials
in patients with advanced non-small-cell
lung cancer.
protein kinases have emerged as key regulators

of all aspects of cancer, and many kinase
inhibitors are now being developed. Agents that
inhibit the activity of cell membrane receptor
tyrosine kinases, such as the epidermal growth
factor receptor (EGFR)3, are among the most
advanced in clinical development.
Basis of discovery
EGFR is part of the ERBB family of receptor

tyrosine kinases, which has four closely related
members: EGFR (ERBB1), HER2 (ERBB2),
HER3 (ERBB3) and HER4 (ERBB4). Each
member consists of an extracellular ligandbinding domain, a transmembrane domain and
an intracellular tyrosine kinase domain4 (FIG. 1a).
Ligand binding to EGFR causes receptor dimerization, either with another EGFR monomer or
with another member of the ERBB family.
Dimerization activates the tyrosine kinase
activity in the intracellular domain, which leads
to receptor autophosphorylation and the initiation of signal-transduction cascades involved in
cell proliferation and survival4.
Activation of EGFR has been implicated in
processes involved in tumour growth and progression, including cell proliferation, inhibition of apoptosis, metastasis and angiogenesis3
(FIG. 1a). EGFR is expressed in various solid
tumours, including 4080% of NSCLCs, providing a strong rationale for testing EGFR
inhibitors in patients with such tumours5,6.
Lung cancer has been estimated to be the leading cause of cancer mortality worldwide1. The
most common form non-small-cell lung
cancer (NSCLC), which accounts for ~75% of
cases is too advanced to be operable in
>50% of patients, and standard first-line
chemotherapy based on platinum agents only
improves survival modestly2. Second-line
treatment options in patients with advanced
NSCLC are limited; docetaxel is the only established choice to be approved by the FDA.
The limited efficacy and lack of specificity
of cytotoxic chemotherapy in solid tumours
such as NSCLC has provided an impetus to
develop therapies that aim to specifically target
cancer cells by modulating the aberrant molecular pathways underlying tumour growth and
progression, in the hope of achieving greater
efficacy with fewer side effects. In particular,
Ligand (EGF, TGF)

EGFR
HN
O
O
Autophosphorylation
TK
TK
TK
TK
N
HCl
Erlotinib hydrochloride
Erlotinib
N-(3-ethynylphenyl)-6,7-bis
(2-methoxyethoxy)-4-quinazolinamine;
C22H23N3O4HCl;
Mr = 429.90;
CAS registry number: 183319-69-9
Activation of signal-transduction
cascades (for example, MAPK)
Cell
proliferation
Apoptosis
Invasion and
metastasis
Angiogenesis
Figure 1 | EGFR signalling and erlotinib. a | Simplified illustration of signal transduction through the
epidermal growth factor receptor3,4 (EGFR). Ligand binding leads to receptor dimerization. This results in
receptor autophosphorylation, which is inhibited by erlotinib. b | Structure of erlotinib hydrochloride.
MAPK, mitogen-activated protein kinase; P, phosphate group; TGF, transforming growth factor; TK,
tyrosine kinase domain.
S14
| MAY 2005
Several possible approaches to targeting

EGFR have been investigated, including
monoclonal antibodies directed against the
extracellular ligand-binding domain, such as
cetuximab (Erbitux; Imclone Systems/BristolMyers Squibb), and small-molecule inhibitors
of the intracellular tyrosine kinase domain3. A
small-molecule EGFR inhibitor, gefitinib
(Iressa; AstraZeneca), was approved by the
FDA for the third-line treatment of patients
with advanced NSCLC in May 2003, and has
now been joined by erlotinib.
Drug properties
Erlotinib (previously known as OSI-774 and

CP-358774; FIG. 1b) is a small molecule that
competes with the binding of ATP to the
intracellular tyrosine kinase domain of EGFR,
thereby inhibiting receptor autophosphorylation and blocking downstream signal transduction79. It showed promising anticancer
effects in various preclinical cancer models7,8,
prompting its clinical evaluation in a range of
cancers, including NSCLC.
Clinical data
Erlotinib hydrochloride (150 mg orally once

daily) was evaluated in a randomized, double
blind, placebo-controlled trial involving 731
patients with locally advanced or metastatic
NSCLC after failure of at least one chemotherapy regimen9. The primary endpoint was
survival, which was significantly longer in
patients receiving erlotinib hydrochloride:
median overall survival in these patients was 6.7
months compared with 4.7 months in patients
receiving placebo9. Median progression-free
survival and tumour response rates in patients
receiving erlotinib hydrochloride were 9.9
weeks and 8.9%, respectively, compared with
7.9 weeks and 0.9%, respectively, in patients
receiving placebo9.
Erlotinib hydrochloride has also been evaluated in combination with platinum-based
chemotherapy (carboplatin and paclitaxel, or
gemcitabine and cisplatin) in two placebocontrolled, randomized trials involving more
than 1,000 first-line patients with locally
advanced or metastatic NSCLC. No clinical
benefit was demonstrated from the addition of
erlotinib hydrochloride in these trials9.
Indications
Erlotinib hydrochloride is approved by the

FDA for the treatment of patients with locally
advanced or metastatic NSCLC after failure of
at least one prior chemotherapy regimen9.
HOT DRUGS | CANCER
NON-SMALL-CELL LUNG CANCER | VIEW FROM THE CLINIC

Analysing clinical issues for targeted
therapies for non-small-cell lung cancer are
John D. Minna, M.D., Professor and Director
of the Hamon Center for Therapeutic
Oncology Research at the University of Texas
Southwestern Medical Center, USA, and
Jonathan Dowell, M.D., Assistant Professor,
Department of Internal Medicine at the
University of Texas Southwestern Medical
Center. Dr Minnas research interest is to
determine all of the molecular abnormalities
leading to the pathogenesis of lung cancer and
to translate this information into new methods
for the diagnosis, prevention, and treatment of
this disease. Dr Dowells research interest is in
clinical and translational trials in lung cancer.
What are the current key issues in the treatment

of non-small-cell lung cancer?
One current key issue in the treatment of
NSCLC is that a plateau seems to have been
reached with regards to the efficacy of standard
chemotherapy. Numerous trials have now
demonstrated that platinum-based chemotherapy combinations are the most effective
treatment in advanced NSCLC. The drug that
is combined with platinum (a taxane, gemcitabine or vinorelbine) seems to affect only the
toxicity profile of the regimen, as all have similar efficacy. Attempts to enhance the activity of
these regimens by adding a third chemotherapy
agent have been uniformly unsuccessful.
Patients vary dramatically in their response
to these standard chemotherapies, and another
major issue is the development of pharmacogenomic tests based on a molecular assessment
of the tumours from individual patients, and
also interindividual variations in drug metabolism, to determine which drugs would be best
for which patients.
A further key to advancing therapy is the
identification of novel molecular targets,
such as EGFR, as well as developing drugs
that effectively act on these targets, such as
erlotinib and gefitinib. In addition, an important current focus is identifying those patients
likely to benefit from these therapies.
What data is available that might help understand patient response to EGFR inhibitors?
The discovery of somatic mutations in the tyrosine kinase domain of EGFR has identified a
subset of NSCLC patients that are exquisitely
sensitive to the EGFR tyrosine kinase inhibitors
(TKIs)1012. However, the randomized singleagent trial of erlotinib indicates that there must
be patients without EGFR mutations that also
derive benefit from drug treatment. Finding
molecular correlates of these responses (such as

EGFR amplification or gene-expression signatures) will be important for identifying the
responding patients prospectively.
William Pao and colleagues have found
identical mutations in EGFR that correlate
with response to erlotinib13. The same clinical
characteristics that are predictive for response
to gefitinib also identify those patients most
likely to benefit from erlotinib. Women,
never-smokers, those with adenocarcinoma
or bronchoalveolar carcinoma, and patients of
East Asian descent are all more likely to
respond to treatment with an EGFR TKI, and
these clinical features correlate with the presence
of an EGFR mutation.
Recently, it was found that in some patients
who have relapsed on erlotinib or gefitinib, their
lung cancers contain, in addition to an EGFR
tyrosine kinase domain mutation, a second
mutation (T790M) that confers the drug resistance14,15. Of interest, tumour cells with the
T790M mutation were still sensitive to another
EGFR TKI, indicating that drugs that overcome
this resistance can be developed.
What might be the best approaches for exploiting targeted therapies in NSCLC?
For many targeted therapies, there is a good
preclinical rationale to suggest that they might
be more effective in the setting of minimal bulk
disease (that is, the adjuvant setting after definitive surgery for early-stage disease). Both of the
available EGFR TKIs (gefitinib and erlotinib)
are currently being investigated in the adjuvant
setting in NSCLC, after the administration of
adjuvant chemotherapy. Of course, the ultimate
in early treatment is use of the drugs for
chemoprevention of the development of lung
cancer in high-risk individuals. There is growing evidence that abnormalities of EGFR signalling occur in smoking-damaged lung
epithelium as a precursor to cancer, and TKIs
as oral agents with little toxicity could have an
important role in such prevention.
With regards to combinations of TKIs with
other drugs in clinically evident disease, all of
the published Phase III trials that have investigated adding a targeted agent to standard
chemotherapy for advanced NSCLC have
failed to demonstrate a benefit for the combination (including four trials with either gefitinib or erlotinib). Two such trials have yet to
be reported one with bexarotene and one
with bevacizumab, and results of these trials
are eagerly anticipated. In the case of the EGFR
inhibitors, the failure of these drugs to enhance
the activity of standard chemotherapy might
have been due to the patient population

studied. These trials were conducted before
the discovery of EGFR mutations that are predictive for response to the EGFR TKIs. Given
the relatively low incidence of these mutations,
the majority of patients included in these trials
probably had wild-type EGFR in their
tumours and would therefore be less likely to
benefit from this therapeutic approach. Trials
in patients with tumour-acquired EGFR
mutations are being planned.
With regards to combining targeted agents,
some preliminary data are available. Roy
Herbst and Alan Sandler conducted a Phase II
trial of erlotinib and bevacizumab in advanced
NSCLC16. The combination was safe and welltolerated, and enough activity was seen to justify
a Phase III comparison with standard chemotherapy in the second-line setting in metastatic
disease.
Jonathan Dowell and John D. Minna are at the
University of Texas Southwestern Medical Center
at Dallas, Dallas, Texas 75390-8593, USA. Peter
Correspondence to J.D.M. and P.K.
e-mails: john.minna@utsouthwestern.edu;
doi:10.1038/nrd1730
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Parkin, D. M. et al. Cancer burden in the year 2000.

The global picture. Eur. J. Cancer 37, S4S66 (2001).
Cersosimo, R. J. Lung cancer: a review. Am. J. Health
Syst. Pharm. 59, 611642 (2002).
Dancey, J. & Sausville, E. A. Issues and progress with
protein kinase inhibitors for the treatment of cancer. Nature
Rev. Drug Discov. 2, 296313 (2003).
Yarden, Y. & Sliwkowski, M. X. Untangling the ErbB signalling
network. Nature Rev. Mol. Cell Biol. 2, 127137 (2001).
Baselga, J. Why the epidermal growth factor receptor? The
rationale for cancer therapy. The Oncologist 7 (S4), 28 (2002).
Salomon, D. S. et al. Epidermal growth factor-related
peptides and their receptors in human malignancies.
Crit. Rev. Oncol. Hematol. 19, 183232 (1995).
Moyer, J. D. et al. Induction of apoptosis and cell cycle arrest
by CP-358,774, an inhibitor of epidermal growth factor
receptor tyrosine kinase. Cancer Res. 57, 48384848 (1997).
Pollack, V. A. et al. Inhibition of epidermal growth factor
receptor-associated tyrosine phosphorylation in human
carcinomas with CP-358,774: dynamics of receptor
inhibition in situ and antitumor effects in athymic mice.
J. Pharmacol. Exp. Ther. 291, 739748 (1999).
FDA label information [online], <http://www.fda.gov/cder/
foi/label/2004/021743lbl.pdf> (2004).
Lynch, T. J. et al. Activating mutations in the epidermal growth
factor receptor underlying responsiveness of non-small-cell lung
cancer to gefitinib. N. Engl J. Med. 350, 21292139 (2004).
Paez, J. G. et al. EGFR mutations in lung cancer:
correlation with clinical response to gefitinib therapy.
Science 304, 14971500 (2004).
Shigematsu, H. et al. Clinical and biological features of
epidermal growth factor receptor mutations in lung cancers.
J. Natl Cancer Inst. 97, 339346 (2005).
Pao, W. et al. EGF receptor gene mutations are common in
lung cancers from never smokers and are associated with
sensitivity of tumors to gefitinib and erlotinib. Proc. Natl
Acad. Sci USA 101, 1330613311 (2004).
Kobayashi, S. et al. EGFR mutation and resistance of nonsmall-cell lung cancer to gefitinib. N. Engl. J. Med. 352,
786792 (2005).
Pao, W. et al. Acquired resistance of lung adenocarcinomas
to gefitinib or erlotinib is associated with a second mutation
in the EGFR kinase domain. PLoS Med. 2, e73 (2005).
Sandler, A. et al. Phase I/II trial evaluating the anti-VEGF
MAb bevacizumab in combination with erlotinib, a
HER1/EGFR-TK inhibitor, for patients with recurrent nonsmall cell lung cancer. J. Clin. Oncol. 22, 14S (2004).
MAY 2005 | S 1 5
HOT DRUGS | CANCER
Pemetrexed disodium
Pemetrexed
In February 2004, pemetrexed disodium

(Alimta; Eli Lilly), an anticancer drug that
targets folate-dependent reactions that are
essential for cell proliferation, became the first
drug to be approved by the US FDA for the
treatment of the rare cancer malignant pleural
mesothelioma. Its accelerated approval for
the second-line treatment of non-small-cell
lung cancer followed in August 2004.
Drugs that target folate-dependent reactions

have a long history in cancer therapy. Aminopterin, which inhibits dihydrofolate reductase
(DHFR), entered clinical use in the late 1940s,
and its less-toxic derivative methotrexate,
which was introduced soon after, is still widely
used to treat various cancers1,2.
Basis of discovery
Reduced folates are essential cofactors in the

enzymatic synthesis of thymidine and purine
nucleotides (FIG. 1a). Methotrexate, which is
closely related to folic acid, has several effects on
folate metabolism that lead to impaired nucleotide production, which ultimately disrupts
DNA synthesis and cell proliferation1.
First, inhibition of DHFR by methotrexate
leads to depletion of the intracellular pool of
reduced folates (FIG. 1a). In addition, and as with
natural folates, the enzyme folyl-polyglutamyl
synthase (FPGS) can attach several glutamate
molecules to methotrexate, which has a number of important effects1. Polyglutamation
decreases cellular efflux, which prolongs drug
action; indeed, the increased propensity of
cancer cells to polyglutamate methotrexate
compared with normal cells might underlie
some of the selectivity of the drug. Furthermore, methotrexate polyglutamates are also
inhibitors of DHFR, and have increased potency
compared with methotrexate as inhibitors of
the folate-dependent enzymes thymidylate synthase (TS) and glycinamide ribonucleotide
formyltransferase (GARFT)1,2 (FIG. 1a).
The utility of methotrexate has led to considerable efforts to develop antifolates that have
a broader spectrum of activity and which circumvent methotrexate resistance1,2. Pemetrexed
is the first of this next generation of antifolates
to be approved as an anticancer drug in the
United States.
methotrexate, its primary site of action was

found to be TS. After being transported into
the cell via the reduced folate carrier, which is
involved in the uptake of physiological folates,
pemetrexed is polyglutamated by FPGS2.
These polyglutamate derivatives (the pentaglutamate is the predominant intracellular
form) are potent inhibitors of TS, and are also
weaker inhibitors of GARFT24. Pemetrexed
and its polyglutamate derivatives also inhibit
DHFR, but with ~1,000-fold less potency than
methotrexate2. As with methotrexate, pemetrexed polyglutamates have an increased intracellular half-life, resulting in prolonged drug
action in malignant cells2,5.
Pemetrexed showed promising therapeutic
effects in preclinical and early clinical studies
in a range of cancers2, including malignant
pleural mesothelioma (MPM) and non-smallcell lung cancer (NSCLC), which have since
become the first indications for which pemetrexed has been studied in Phase III trials6,7.
Clinical data
MPM is a rare cancer that is usually associated

with asbestos exposure. Pemetrexed disodium
(as an intravenous infusion) in combination
with the platinum-based cytotoxic drug cisplatin
was compared with cisplatin alone in a randomized study involving 448 chemotherapy-naive
patients with MPM5,6. To decrease adverse
events, patients were given folic acid and vitamin B12 supplements. Patients receiving both
pemetrexed disodium and cisplatin had a
a
dTMP
significantly increased median overall survival time of 12.1 months, compared with 9.3
months for those receiving cisplatin alone5,6.
NSCLC is the most common form of lung
cancer, which is the leading cause of cancer
mortality worldwide. First-line therapy of
NSCLC is based on platinum agents, and docetaxel is the standard second-line treatment
option. Pemetrexed disodium was compared
with docetaxel in a randomized trial involving
571 patients with locally advanced metastatic
NSCLC after prior chemotherapy; patients
treated with pemetrexed disodium also received
folic acid and vitamin B12 supplements. Pemetrexed disodium did not show superiority over
docetaxel on the primary endpoint of survival,
and there were no statistically significant differences between pemetrexed disodium and docetaxel with respect to the secondary endpoints
such as objective response rate5. However,
pemetrexed disodium had a more favourable
safety profile than docetaxel; for example, it
caused significantly less neutropenia and febrile
neutropenia5,7.
Indications
Pemetrexed disodium in combination with cisplatin is approved by the FDA for the treatment
of patients with MPM whose disease is unresectable or who are otherwise not candidates
for curative surgery5. Pemetrexed disodium as a
single agent is also approved by the FDA for the
treatment of patients with locally advanced or
metastatic NSCLC after prior chemotherapy5.
b
DNA
dUMP
DHF
DHFR
H2N
TS
DNA, RNA
H
N
CO2
O
N
H
Na
7H2O
CO2
Na
HN
5,10-CH2-THF
THF
Purine
nucleotides
10-CHO-THF
GARFT
AlCARFT
Pemetrexed disodium heptahydrate

L-Glutamic
acid, N-[4-[2-(2-amino-4,7-dihydro4-oxo-1H-pyrrolo[2,3-d]pyrimidin-5-yl)ethyl]
benzoyl]-,disodium salt, heptahydrate;
C20H19N5Na2O67H2O;
Mr = 597.49; CAS number: 150399-23-8
GAR
Drug properties
Pemetrexed (FIG. 1b) was discovered through

screening of derivatives of a known GARFT
inhibitor for anticancer activity3. However,
in contrast to the parent compound and
S16
Figure 1 | Folate metabolism and pemetrexed. a | Simplified illustration of some key enzymatic
reactions of folate metabolism, showing enzymes affected by pemetrexed, or its polyglutamates. These
multiple drug effects impair nucleotide synthesis and thereby inhibit cell proliferation. b | Structure of
pemetrexed disodium. AICARFT, aminoimidazole carboxamide ribonucleotide formyltransferase; DHFR,
dihydrofolate reductase; GARFT, glycinamide ribonucleotide formyltransferase; THF, tetrahydrofolate, TS,
thymidylate synthase.
| MAY 2005
HOT DRUGS | CANCER
ANTIFOLATES | VIEW FROM THE CLINIC

Analysing issues that affect the clinical
application of antifolates are David Gibbs, PhD,
FRACP, Medical Oncologist at the Christchurch
Hospital, New Zealand, and Ann Jackman, PhD,
Professor of Biochemical Pharmacology at the
Institute of Cancer Research, UK. Dr Gibbs'
interests are the development of targeted antifolates, and the management of lung cancer
and ovarian cancer. Dr Jackman's interests are
antifolate drug development and overcoming
drug resistance in ovarian cancer by combining
targeted agents with cytotoxic drugs.
What are the key issues that need to be addressed

to maximise the benefits of antifolates?
The key issues in developing the clinical utility
of antifolates are common to the development
of other anticancer drugs: namely, strategies to
reduce normal tissue toxicity, strategies to
abrogate the development of drug resistance
and strategies to predict the patients that are
most likely to benefit from treatment.
There is evidence that clinical response to TS
inhibitors is predicted by levels of TS expression8,9, but this is not yet used as a predictor in
clinical practice. There is also evidence that cell
lines acquire resistance to antifolates, such as
pemetrexed, by downregulation of the reduced
folate carrier (RFC) and FPGS10, which suggests
that measurement of levels of a few key proteins
might allow prediction of antifolate activity.
Normal tissue tolerance limits the dose of
antifolates that can be delivered. Dose-limiting
toxicity can be related to folate levels, and folate
supplementation increases the maximum tolerated dose of methotrexate, lometrexol and
pemetrexed11. Folate supplementation was
found to have no effect on antitumour activity
in animal models and there is no evidence that
folate supplementation reduces anti-tumour
activity in mesothelioma6. However, there is no
a priori reason that folate supplementation will
also result in tumour protection, and this concept needs exploration in further clinical studies.
How well are the mechanisms of the anticancer
activity of antifolates understood, and what are
the implications for their clinical application?
In general, the effect of antifolates depends on
the particular enzymes that are inhibited with
accumulation of enzyme substrate and depletion of enzyme products. However, the mechanism by which these events result in cell death
are not completely understood. The deconvolution of the mechanism of action of antifolates is
further complicated by their metabolism in cells
to higher-order polyglutamate metabolites and
the potential for inhibition of multiple targets.
Of the new generation of antifolates, the

mechanism of action of TS inhibitors has been
studied in the most detail. In general, TS inhibition results in changes in deoxynucleotide
pools, leading to the formation of single and
double-stranded DNA breaks, S-phase arrest
and induction of apoptosis12. However. the
mechanisms by which these effects occur differ
between cell lines, and it is likely that similarly
diverse mechanisms occur in tumours in vivo.
Recent research has indicated that the cyclin
dependent kinase inhibitor p21 modulates TS
inhibitor activity13 and that FAS is required for
the induction of apoptosis in response to TS
inhibitors14. Studies such as these indicate that
much remains to be elucidated about the mechanisms of action of antifolates, and also provide
information that might allow rational choices to
be made in choosing the best cytotoxic drugs
and targeted agents to combine with antifolates.
Studies of resistant cell lines indicate that
resistance to many antifolates occurs by a relatively limited repertoire of mechanisms. These
mechanisms include alterations in the function
of transport proteins such as the RFC, enzymes
responsible for maintaining polyglutamate
homeostasis, upregulation of target enzymes
and changes in intracellular folate pools.
However, this also suggests that if means of
modulating these resistance mechanisms can
be found, sensitivity to these agents might be
maintained or restored.
Finally, an area requiring further understanding, because opinion is divided, is the contribution that the -folate receptor might make
to the antitumour activity of newer antifolates
in those epithelial tumours that overexpress it;
for example, ovarian cancer and mesothelioma.
What do you think will be the impact of new
antifolates in the next 5 years?
In addition to pemetrexed, there are other
agents in clinical development, such as plevitrexed (BGC 9331, a non-polyglutamated TS
inhibitor), OSI-79041 (a liposomal-encapsulated TS inhibitor), nolatrexed (a lipohilic,
non-polyglutamated TS inhibitor) and methylene-10-deazaaminopterin (a DHFR
inhibitor) or licensed for use in some countries
(raltitrexed; a polyglutamated TS inhibitor).
The different classes of newer antifolates have
properties that might be advantageous in some
circumstances. Unlike methotrexate, polyglutamation of raltitrexed and pemetrexed not only
results in increased retention within cells but
also increases affinity for their target enzymes.
Plevitrexed cannot be polyglutamated and OSI79041 does not require polyglutamation for its
activity and thus will retain activity in cells that

have reduced FPGS or increased folylpolyglutamate hydrolase activity. Pemetrexed inhibits
not only TS but also GARFT and this might be
responsible for its greater activity against some
tumour types.
The main recent impact of antifolates in
clinical practice has been the approval of pemetrexed in combination with cisplatin in MPM
and pemetrexed in the second-line treatment of
NSCLC. Both pemetrexed and raltitrexed have
been tested in combination with several cytotoxic agents. It is possible that some of these
combinations might result in clinically relevant
synergy. However, combinations need to be
selected on the basis of an understanding of
mechanisms of action as discussed above.
David Gibbs is at Christchurch Hospital,
Riccarton Avenue, Christchurch, New Zealand.
Ann Jackman is at the Institute of Cancer Research,
15 Cotswold Road, Sutton, Surrey, UK.
Peter Kirkpatrick is at Nature Reviews Drug
Discovery.
Correspondence to D. G & P. K.
e-mails: davidandcatherine@xtra.co.nz;
doi:10.1038/nrd1731
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Takimoto, C. H. New antifolates: pharmacology and clinical

applications. The Oncologist 1, 6881 (1996).
Hanauske, A.-R. et al. Pemetrexed disodium: a novel
antifolate clinically active against multiple solid tumors.The
Oncologist 6, 363373 (2001).
Taylor, E. C. et al. A dideazatetrahydrofolate analogue
lacking a chiral center at C-6, N-[4-[2-(2-amino-3,4 dihydro4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5-yl)ethyl]benzoyl]-Lglutamic acid, is an inhibitor of thymidylate synthase.
J. Med. Chem. 35, 44504454 (1992).
Shih, C. et al. LY231514, a pyrrolo[2,3-d]pyrimidine-based
antifolate that inhibits multiple folate-requiring enzymes.
Cancer Res. 57, 11161123 (1997).
FDA drug approvals list [online], <http://www.fda.gov/
Vogelzang, N. J. et al. Phase III study of pemetrexed in
combination with cisplatin versus cisplatin alone in patients
with malignant pleural mesothelioma. J. Clin. Oncol. 21,
26362644 (2003).
Hanna, N. et al. Randomized phase III trial of pemetrexed
versus docetaxel in patients with non-small-cell lung cancer
previously treated with chemotherapy. J. Clin. Oncol. 22,
15891597 (2004).
Johnston, P. G. et al. Thymidylate synthase gene and
protein expression correlate and are associated with
response to 5-fluorouracil in human colorectal and gastric
tumors. Cancer Res. 55, 1407-1412 (1995).
Farrugia, D. C. et al. Thymidylate synthase expression in
advanced colorectal cancer predicts for response to
raltitrexed. Clin. Cancer Res. 9, 792-801 (2003).
Wang, Y. et al. Decreased expression of the reduced folate
carrier and folypolyglutamate synthetase is the basis for
acquired resistance to the pemetrexed antifolate (LY231514)
in an L1210 murine leukemia cell line. Biochem. Pharmacol.
65, 1163-1170 (2003).
Niyikiza, C. et al. Homocysteine and methylmalonic acid:
markers to predict and avoid toxicity from pemetrexed
therapy. Mol. Cancer Ther. 1, 545-552 (2002).
Touroutoglou N, & Pazdur, R. Thymidylate synthase
inhibitors. Clin. Cancer Res. 2, 227-243 (1996).
Geller, J. I. et al. P21Cip1 is a critical mediator of the cytotoxic
action of thymidylate synthase inhibitors in colorectal
carcinoma cells. Cancer Res. 64, 6296-6303 (2004).
Longley, D. B. et al. The roles of thymidylate synthase and
p53 in regulating Fas-mediated apoptosis in response to
antimetabolites. Clin. Cancer Res. 10, 3562-3571 (2004).
MAY 2005 | S 1 7
HOT DRUGS | CANCER
ONCOLOGY | MARKET INDICATORS
Data adapted from Wood Mackenzies Corporate

Franchise Analysis, 2004. Wood Mackenzie,
Kintore House, 74-77 Queen Street,
Edinburgh EH2 4NS, UK
doi:10.1038/nrd1737
Online links
FURTHER INFORMATION
National Cancer Institute: http://www.nci.nih.gov/
Surveillance, Epidemiology, and End Results:
http://seer.cancer.gov/
Wood Mackenzie: http://www.woodmac.com
Access to this interactive links box is free online.
a 2003
b 2008
882
1,129
3,934
2,338
3,509
8,012
4,226
2,292
6,144
2,583
4,960
29,970
59,271
15,757
2,286
9,367
11,628
1,753
3,394
Colorectal/stomach
Supportive therapies
Others
Lymphoma
Breast cancer
Prostate cancer
5,048
Lung cancer
Ovarian cancer
Leukaemia
Figure 1 | The global oncology market divided by indication. Sales are in US $ million.
Estimated number of new cases in the US (1,000s)
The global oncology market generated sales

of US $30.0 billion in 2003, and is forecast to
increase by 14.6% by 2008, resulting in sales
of US $59.3 billion. FIG. 1 shows the global
market sizes for the key cancer indications.
With the exception of supportive therapies,
the largest market in 2003 was that for drugs
for the treatment of breast cancer, which is
forecast to still be in this position in 2008.
Between 2003 and 2008, the market that is
forecast to grow the most (27.9% compound
annual growth rate) is that for drugs for colorectal and stomach cancers, from US $2.4
billion in 2003 to US $8.0 billion in 2008.
FIG. 2 illustrates the estimated annual number of new cases of the major cancers in the
United States. Breast cancer is the most common cancer in women, and prostate cancer is
the most common cancer in men, but lung
cancer is the leading cause of death, followed by
colon cancer.
TABLE 1 shows the top ten companies in oncology, ranked by sales in 2003, and their predicted rank in 2008.
205.0
189.0
200
169.4
148.3
150
100
60.9
50
30.8
23.3
Colorectal
Breast
Lung
Leukaemia
Lymphoma
Prostate
Ovarian
Figure 2 | Estimated annual number of new cases of selected cancers in the United States in 2002.
Source: Surveillance, Epidemiology, and End Results (SEER), USA.
Table 1 | Top 10 companies in the oncology market by sales

Rank by
2003 sales
Companies
2003 sales
(US $ million)
2003 market
share (%)
Rank by 2008
forecast sales
2008 forecast sales

(US $ million)
2008
market share
Roche and Genentech
4,478
14.9
10,444
17.6
Amgen
3,139
10.5
9,564
16.1
Sanofi-Aventis proforma
3,041
10.1
6,283
10.6
Novartis
2,907
9.7
6,867
11.6
AstraZeneca
2,743
9.2
5,028
8.5
Bristol-Myers Squibb
2,110
7.0
3,084
5.2
Johnson & Johnson
1,666
5.6
1,944
3.3
GlaxoSmithKline
1,635
5.5
10
1,921
3.2
Takeda
1,277
4.3
14
732
1.2
10
Eli Lilly
1,040
3.5
2,871
4.8
Pfizer
1,938
3.3
Top 10
24,036
80.2
49,944
84.3
Others
5,934
19.8
9,327
15.7
Total
29,970
100.0
59,271
100.0
MAY 2005 | S 1 9

Nat Rev Drug Dis - 2005

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Nat Rev Drug Dis - 2005

Загружено:

Авторское право:

Доступные форматы

May 2005 Vol 4 No 5

S6 | Jean-Pierre Issa, Hagop Kantarjian &

S8 | Lee M. Ellis & Peter Kirkpatrick

S10 | Richard Goldberg & Peter Kirkpatrick

S12 | Ching-Hon Pui, Sima Jeha &

S14 | Jonathan Dowell, John D. Minna &

S16 | David Gibbs, Ann Jackman &

elcome to Hot Drugs Cancer 2005, a collection of

2005 Nature Publishing Group

HOT DRUGS | CANCER

Azacitidine (Vidaza; Pharmion), an inhibitor

Myelodysplastic syndromes (MDS) are a heterogeneous group of bone-marrow disorders

A randomized, open-label, controlled trial

Azacitidine is approved by the FDA for the

The primary efficacy endpoint was

Figure 1 | Azacitidine and DNA methylation. a | Azacitidine. b | A family of DNA methyltransferases

2005 Nature Publishing Group

HOT DRUGS | CANCER

EPIGENETIC THERAPIES AND MDS | VIEW FROM THE CLINIC

What are the current key issues in the treatment

reported synergy between hypomethylating

NATURE REVIEWS | DRUG DISCOVERY

activity in AML16, but it could be argued that

List, A. F., Vardiman, J., Issa, J.-P. & DeWitte, T. M.

2005 Nature Publishing Group

HOT DRUGS | CANCER

Bevacizumab (Avastin; Genentech/Roche),

Angiogenesis the development of new blood

The hypothesis that inhibiting angiogenesis

Bevacizumab is a humanized version of the

Decoy effect on VEGF signalling

VEGFR1 and VEGFR2 (FIG. 1), and inhibits the

It has been estimated that colorectal cancer

Bevacizumab, used in combination with intravenous 5-FU-based chemotherapy, is approved

2005 Nature Publishing Group

HOT DRUGS | CANCER

ANTI-VEGF THERAPIES | VIEW FROM THE CLINIC

What impact have anti-VEGF therapies such as

NATURE REVIEWS | DRUG DISCOVERY

plus bevacizumab or cetuximab plus bevacizumab plus irinotecan13. The combination of

Ferrara, N. et al. Discovery and development of bevacizumab,

2005 Nature Publishing Group

HOT DRUGS | CANCER

Cetuximab (Erbitux; ImClone Systems/

thereby inhibiting downstream signal transduction2,5,6 (FIG. 1). In preclinical studies,

irinotecan had an objective response rate of

Cetuximab was studied in trials involving

Cetuximab is a recombinant, human/mouse

Advances in the understanding of the aberrant signalling pathways involved in cancer

Ligand (EGF, TGF)

2005 Nature Publishing Group

HOT DRUGS | CANCER

COLORECTAL CANCER | VIEW FROM THE CLINIC

What are the current key areas in the pharmacotherapy of CRC?

Prevention strategies in CRC have been set

NATURE REVIEWS | DRUG DISCOVERY

have provided that biologicals are useful

Parkin, D. M. et al. Cancer burden in the year 2000. The

2005 Nature Publishing Group

HOT DRUGS | CANCER

Clofarabine (Clolar; Genzyme), a purine

Leukaemias, such as acute lymphoblastic

of total platelet recovery (CRp), defined as