Editorials
3. Verstovsek S, Kantarjian H, Mesa RA, et al: Safety and efficacy of
status, is associated with inferior overall and leukemia-free survival. Leukemia
INCB018424, a JAK1 and JAK2 inhibitor, in myelofibrosis. N Engl J Med
22:756-761, 2008
363(12):1117-1127, 2010
10. Mesa RA: How I treat symptomatic splenomegaly in patients with myelo-
4. Pardanani A, Gotlib JR, Jamieson C, et al: Safety and efficacy of
fibrosis. Blood 113:5394-5400, 2009
TG101348, a selective JAK2 inhibitor, in myelofibrosis. J Clin Oncol 29:789-
11. Steensma DP, Mesa RA, Li CY, et al: Etanercept, a soluble tumor
796, 2011
necrosis factor receptor, palliates constitutional symptoms in patients with
5. Santos FP, Kantarjian HM, Jain N, et al: Phase 2 study of CEP-701, an orally
myelofibrosis with myeloid metaplasia: Results of a pilot study. Blood 99:
available JAK2 inhibitor, in patients with primary or post-polycythemia vera/
2252-2254, 2002
essential thrombocythemia myelofibrosis. Blood 115:1131-1136, 2010
12. Jedidi A, Marty C, Oligo C, et al: Selective reduction of JAK2V617F-
6. Verstovsek S: Therapeutic potential of Janus-activated kinase-2 inhibitors
dependent cell growth by siRNA/shRNA and its reversal by cytokines. Blood
for the management of myelofibrosis. Clin Cancer Res 16:1988-1996, 2010
114:1842-1851, 2009
7. Barosi G, Bergamaschi G, Marchetti M, et al: JAK2 V617F mutational
13. Mesa RA, Kantarjian H, Tefferi A, et al: Functional assessment of
status predicts progression to large splenomegaly and leukemic transformation
performance status in patients with myelofibrosis (MF): Utility and feasibility
in primary myelofibrosis. Blood 110:4030-4036, 2007
of the 6-minute walk test (6MWT). J Clin Oncol 27:376s, 2009 (suppl; abstr
8. Guglielmelli P, Barosi G, Specchia G, et al: Identification of patients with
7083)
poorer survival in primary myelofibrosis based on the burden of JAK2V617F
mutated allele. Blood 114:1477-1483, 2009
9. Tefferi A, Lasho TL, Huang J, et al: Low JAK2V617F allele burden in
DOI: 10.1200/JCO.2010.33.4508; published online ahead of print at
primary myelofibrosis, compared to either a higher allele burden or unmutated
www.jco.org on January 10, 2011

Genetic Predisposition for Chemotherapy-Induced
Neuropathy in Multiple Myeloma
Pamela S. Becker, Division of Hematology, University of Washington, Seattle, WA
See accompanying article on page 797
Peripheral neuropathy is one of the most frequent and debilitat-
In this issue, Johnson et al1 describe the single-nucleotide poly-
ing adverse effects encountered by patients with cancer undergoing
morphisms (SNPs) associated with development of neuropathy after
chemotherapy treatment. For multiple myeloma (MM), every effec-
treatment with thalidomide or vincristine in MM patients. DNA sam-
tive regimen includes at least one agent that can cause neuropathy, a
ples from over 1,500 patients enrolled on Myeloma IX and Stichting
major dose-limiting toxicity in clinical trials and in clinical practice.
Hemato-Oncologie voor Volwassenen Nederland (HOVON 50)/the
Furthermore, neuropathy occurs even before therapy as a conse-
German-speaking Myeloma Multicenter Group (GMMG) high-dose
quence of the disease in approximately 20% of patients with MM, and
3 (HD3) multicenter trials were analyzed for SNPs, and correlations
can even occur in the precursor condition, monoclonal gammopathy
performed for patients who developed neuropathy compared with
of undetermined significance.
those who did not. Myeloma IX was a randomized comparison of
In the past, neuropathy was noted as a complication of the
cyclophosphamide plus VAD (CVAD) versus cyclophosphamide, tha-
vincristine-containing vincristine, doxorubicin, and dexamethasone
lidomide, and dexamethasone (CTD) for younger patients, and CTD
(VAD) regimen, then occurred with the thalidomide-containing reg-
versus melphalan and prednisone for older patients. HOVON-50/
imens and to a lesser extent with lenalidomide, and is now frequent
GMMG-HD3 was a randomized comparison of VAD to thalidomide,
with bortezomib, one of the most active agents in myeloma. The
doxorubicin, and dexamethasone prior to high-dose melphalan. The
incidence of thalidomide-associated neuropathy is 27% to 28%,2,3
study utilized a hypothesis-driven, candidate gene approach to iden-
severe (grade
3) in 6%.2 The incidence of bortezomib-associated
tify SNPs in genes responsible for DNA repair or inflammation of the
neuropathy is 64% as monotherapy,4 and 47% when combined with
nervous system. In the two trials, 31.8% of patients receiving thalido-
melphalan and prednisone,5 severe (grade
3) in 13% with the latter
mide and 33.6% of those receiving vincristine developed neuropathy
regimen. The symptoms can vary and include numbness, dysesthesia,
at a median time of 8 weeks. The risk of neuropathy was not signifi-
paresthesia, hyperalgesia, and severe pain. Both small and large nerve
cantly associated with the clinical features of age, type of immuno-
fibers can be affected.
globulin, stage, interphase fluorescent in situ hybridization
The management of neuropathy in patients receiving chemother-
variant, or response, but was associated with male gender.
apy has largely been supportive. Antidepressants or antiepileptics such
The SNPs associated with thalidomide-induced neuropathy in-
as amitryptiline, gabapentin, and pregabalin have been used to mini-
cluded those in the ATP-binding cassette transporter genes ABCC1
mize painful neuropathy. In addition, pyridoxine, glutamine, or the
and ABCC2, ADME (absorption, distribution, metabolism, excretion)
antioxidants shown to be effective in diabetic neuropathy, alpha lipoic
genes including FMO6 (flavin-containing monooxygenase) and ion
acid or acetyl L-carnitine, may be protective. Previously, it has not
channel gene SLC12A6, SPRR1A (promotes axonal outgrowth), and
been possible to predict who will develop crippling, irreversible symp-
SERPINB2 (induced in injured neurons). In contrast, those SNPs
toms and who will be able to tolerate the antimyeloma drugs with
involved in vincristine-induced neuropathy were in different genes,
minimal toxicity. Fortunately, there is some degree of reversibility
including CAMKK1 (expressed in neurons resistant to oxidative
with discontinuation or dose reduction.
stress), CYP2C8, and CYP2C9 (involved in hepatic drug clearance),
www.jco.org
© 2011 by American Society of Clinical Oncology
783
Information downloaded from jco.ascopubs.org and provided by at CEDARS SINAI MEDICAL CENTER MEDICAL LIBRARY
Copyright © 2011 on March
American 1, 2011
Society from
of
192.231.86.98
Clinical Oncology. All rights reserved.

---

Editorials
Table 1. SNPs With Increased Odds Ratio for Neuropathy
Thalidomide
Vincristine
Bortezomib
Report
Gene
Odds Ratio
Gene
Odds Ratio
Gene
Odds Ratio
Johnson et al1
ABCC1 2.85
CYP3A7 4.31
HOVON-50/GMMG-HD3
SPRR1A 1.72
IRF4 4.54
VAD versus TAD
LIG3 1.94
UGT2A1 3.95
CD86 3.03
NQO1 3.81
DPYD 2.33
HSD3B1 2.44
NQO1 2.04
CYP24A1 3.02
TIMM8A 2.90
CYP2C9 2.28
TNF 1.84
ATR 2.42
ATM 2.65
MGST1 2.39
EME1 2.24
ESR1 2.8
SULT1C1 3.38
Johnson et al1
ABCC2 1.51
ADAMTS1 2.31
Myeloma IX
SFTPC 1.48
SERPINE1 1.90
CVAD versus CTD
CYP2C9 1.54
ADRB2 1.96
CTD versus MP
VANGL1 1.56
TNXB 2.30
PPIA 1.50
SERPINE1 1.88
IGF1R 1.40
FRK 1.94
CTLA4 1.38
CYP11B1 1.91
POLB 1.62
ATM 1.90
ID3 2.73
PSMA4 1.82
Johnson et al1 common to both Myeloma IX
SERPINE 1 1.88, 2.15
and HOVON-50/GMMG-HD3 (Odds
ID3 2.73, 2.45
ratios in this order)
CYP2C9 1.90, 2.36
CAMKK1 1.70, 1.90
CYP2C8 1.74, 2.18
Broyl et al6
Early (1 cycle)
Early (1 cycle)
VAD versus PAD
PPARD 13.62
RDM1 3.65
ALDH1A1 7.62
CASP9 3.59
HEL308 6.67
ALOX12 3.50
PPARD 9.67
ALOX12 3.50
PPARD 9.67
LSM1 4.11
ABCC4 7.15
LTA 4.67
PPARD 8.89
LTA 4.52
SCL10A2 4.30
ABCC5 4.64
Late (2-3 cycles)
Late (2-3 cycles)
CART 4.62
ERCC4 2.74
SIM1 3.30
ERCC4 2.48
JUN 5.00
IFNGR2 2.30
DPYD 3.29
ERCC3 1.26
SLC22A5 4.80
MRE11A 3.27
KIA0274 3.89
ABCC1 3.36
PTGS1 5.40
ABCC1 4.22
FGFR4 3.47
MGMT 3.38
BZRP 2.93
BZRP 3.14
UGT2B7 3.60
NOTE. Genes in bold are either common to both publications Johnson et al1 and Broyl et al,6 or in a gene family common to both.
Abbreviations: HOVON-50/GMMG-HD3, Dutch-Belgian Hemato-Oncology Cooperative Group (HOVON) -50/German-Speaking Myeloma Multicenter Group
(GMMG) -HD3 trial; VAD, vincristine, doxorubicin, and dexamethasone; TAD, thalidomide, doxorubicin, and dexamethasone; CVAD, cyclophosphamide plus
VAD; CTD, cyclophosphamide, thalidomide, and dexamethasone; MP, melphalan and prednisone; PAD, bortezomib, doxorubicin, and dexamethasone.
784
© 2011 by American Society of Clinical Oncology
JOURNAL OF CLINICAL ONCOLOGY
Information downloaded from jco.ascopubs.org and provided by at CEDARS SINAI MEDICAL CENTER MEDICAL LIBRARY
Copyright © 2011 on March
American 1, 2011
Society from
of
192.231.86.98
Clinical Oncology. All rights reserved.

---

Editorials
NFATC2 (a transcriptional activator in T lymphocytes), ID3 (associ-
Genetic information (ie, cytogenetics/interphase fluorescent in
ated with radiation-induced apoptosis), or SLC10A2 (apical sodium-
situ hybridization) already guides treatment decisions for multiple
dependent bile acid transporter). These discrepancies suggest that
myeloma; data from SNP risk profiles for toxicity could be similarly
there may be different pathogenic mechanisms for nerve damage from
incorporated into treatment algorithms if the data were to be compel-
the two agents.
ling. A priori identification of adverse effect risks in given individuals
Since the older regimens that included vincristine and thalido-
would not only lead to broader use of protective agents and prophy-
mide have been replaced with regimens including newer agents, it will
lactic maneuvers including targeting of certain inflammatory path-
be critical to determine whether the findings for each drug are appli-
ways, but would also undoubtedly play a role in selection of drug
cable to the entire drug class; for example, will the same SNPs that are
regimens and drug doses, compatible with the evolving era of person-
associated with thalidomide toxicity have relevance to lenalidomide
alized medicine.
and pomalidomide? Likewise, it will be necessary to learn whether the
However, caution will need to be exercised in translating genetic
SNPs that are relevant to the neurotoxicity of the proteasome inhibi-
risk assessment for toxicity into clinical practice. It will be critical to
tor, bortezomib, will extend to carfilzomib and other drugs in this
examine whether the drugs used for prevention of neuropathy or
class. Moreover, regimens such as bortezomib, thalidomide, and
reduced drug dose for selected patients will diminish the efficacy of
dexamethasone, cisplatin, doxorubicin, cyclophosphamide, and eto-
our best chemotherapy regimens. These decisions may impact sur-
poside, combine at least two drugs that each carries a high risk of
vival and even quality of life due to disease progression. For example,
neuropathy. Furthermore, vincristine use remains widespread in the
the ABC cassette proteins, for which certain SNPs are associated with
treatment of acute lymphoblastic leukemia and lymphoma with
thalidomide- and vincristine-induced neuropathy, regulate intracel-
regimens such as cyclophosphamide, doxorubicin, vincristine, and
lular drug levels, and thus the optimal function of these proteins may
prednisone; etoposide, doxorubicin, vincristine, prednisone, and cy-
be necessary to achieve antitumor effect. It is essential that each of the
clophosphamide; and CVAD alternating with methotrexate and cyt-
SNPs deemed to be reliably associated with neuropathy be correlated
arabine; so the conclusions of this report may have relevance for
prospectively with response to treatment to prevent overzealous
current treatment of other diseases. Confirming that the same corre-
changes in therapy. It therefore may be difficult to put "primum non
lations exist between the SNPs identified in this report and peripheral
nocere" into practice, or "to do no harm" (translated from Hippo-
neuropathy in patients with other conditions such as lymphoma
crates, Of the Epidemics). If not harming nerves becomes paramount,
would constitute validation of this approach.
we may jeopardize treatment response. Future studies should be per-
The most difficult aspect of interpreting the results of correlative
formed to ensure that the benefits of lowering toxicity do not compro-
studies using SNPs is the reproducibility of the same findings in other
mise the significant advances made to prolong life.
studies and patient populations. For example, Broyl et al6 has indepen-
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
dently studied the SNPs associated with the development of neurop-
Although all authors completed the disclosure declaration, the following
athy in patients with multiple myeloma undergoing treatment with
author(s) indicated a financial or other interest that is relevant to the subject
vincristine or bortezomib on the HOVON-65/GMMG-HD4 trial, a
matter under consideration in this article. Certain relationships marked
prospective randomized trial of VAD versus bortezomib, doxorubi-
with a "U" are those for which no compensation was received; those
cin, and dexamethasone. This report identified a nearly completely
relationships marked with a "C" were compensated. For a detailed
distinct set of SNPs compared with Johnson et al1 (Table 1). Broyl et al
description of the disclosure categories, or for more information about
ASCO's conflict of interest policy, please refer to the Author Disclosure
examined SNPs associated with early (after one cycle) or late (after two
Declaration and the Disclosures of Potential Conflicts of Interest section in
to three cycles) development of neuropathy. Because the studies by
Information for Contributors.
Johnson et al and Broyl et al were based on different sets of
Employment or Leadership Position: None Consultant or Advisory
candidate genes derived by different hypotheses, and were applied
Role: None Stock Ownership: None Honoraria: None Research
to patients receiving different drug regimens, it is difficult to rec-
Funding: Pamela S. Becker, Celgene, Millenium Expert Testimony:
oncile the two data sets. However, the most striking finding is the
None Other Remuneration: None
concordant discovery of the same two genes, ABCC1 and DPYD,
REFERENCES
having SNPs associated with thalidomide-induced neuropathy in
1. Johnson DC, Corthals SL, Walker BA, et al: Genetic factors underlying the
the HOVON trial reported by Johnson et al, and with late
risk of thalidomide-related neuropathy in patients with multiple myeloma. J Clin
vincristine-associated neuropathy in the trial reported by Broyl et
Oncol 29:797-804, 2011
2. Glasmacher A, Hahn C, Hoffmann F, et al: A systematic review of phase-II
al (Table 1). ABCC1 is a member of the ATP-binding cassette gene
trials of thalidomide monotherapy in patients with relapsed or refractory multiple
family, for which SNPs in other members (ABCC2, ABCC4,
myeloma. Br J Haematol 132:584-593, 2006
ABCC5) were also identified to be associated with neuropathy risk
3. von Lilienfeld-Toal M, Hahn-Ast C, Furkert K, et al: A systematic review of
in both studies. The other gene, DPYD, which encodes dihydropy-
phase II trials of thalidomide/dexamethasone combination therapy in patients
with relapsed or refractory multiple myeloma. Eur J Haematol. 81:247-252, 2008
rimidine dehydrogenase, is also associated with toxicity in patients
4. Richardson PG, Xie W, Mitsiades C, et al: Single-agent bortezomib in
treated for GI malignancy with fluorouracil.7 A Japanese publica-
previously untreated multiple myeloma: Efficacy, characterization of peripheral
tion also reports a SNP in ABCC2 associated with oxaliplatin-
neuropathy, and molecular correlations with response and neuropathy. J Clin
Oncol 27:3518-3525, 2009
associated neuropathy.8 Thus, the evidence may be greater for a
5. Dimopoulos MA, Mateos MV, Richardson PG, et al: Risk factors for, and
role for these key genes, the ATP-binding cassette gene family
reversibility of, peripheral neuropathy associated with bortezomib-melphalan-
members and DPYD, given that they were identified by both re-
prednisone in newly diagnosed patients with multiple myeloma: Subanalysis of
ports as having SNPs associated with risk of chemotherapy-associated
the phase 3 VISTA study. Eur J Haematol [Epub ahead of print on September 28,
2010]
neuropathy, and independently were shown by others to be associated
6. Broyl A, Corthals SL, Jongen JL, et al: Mechanisms of peripheral neuropathy
with toxicity due to other chemotherapy drugs.
associated with bortezomib and vincristine in patients with newly diagnosed
www.jco.org
© 2011 by American Society of Clinical Oncology
785
Information downloaded from jco.ascopubs.org and provided by at CEDARS SINAI MEDICAL CENTER MEDICAL LIBRARY
Copyright © 2011 on March
American 1, 2011
Society from
of
192.231.86.98
Clinical Oncology. All rights reserved.

---

Editorials
multiple myeloma: A prospective analysis of data from the HOVON-65/GMMG-
8. Mori Y, Katsumata K, Tsuchida A, et al: Single nucleotide polymorphism
HD4 trial. Lancet Oncol 11:1057-1065, 2010
analysis in the GSTP1 and ABCC2 genes about neuropathy by the Oxaliplatin.
7. van Kuilenburg AB, Meijer J, Mul AN, et al: Intragenic deletions and a deep
Gan To Kagaku Ryoho35:2377-2381, 2008
intronic mutation affecting pre-mRNA splicing in the dihydropyrimidine dehydro-
genase gene as novel mechanisms causing 5-fluorouracil toxicity. Hum Genet
DOI: 10.1200/JCO.2010.33.4771; published online ahead of print at
128:529-538, 2010
www.jco.org on January 18, 2011

Treatment of Low-Risk Gestational
Trophoblastic Neoplasia
Carol Aghajanian, Memorial Sloan-Kettering Cancer Center, New York, NY
See accompanying article on page 825
Gestational trophoblastic disease (GTD) spans a heterogeneous
spectrum of diseases that arise in the fetal chorion during pregnancy.
Table 1. FIGO Anatomic Staging for GTN
Included in this definition are hydatidiform moles (partial and com-
Stage
Description
plete), choriocarcinoma, placental site trophoblastic tumors (PSTTs),
I
Disease confined to the uterus
and epithelioid trophoblastic tumors (ETTs).
II
GTN extends outside the uterus, but is limited to the
genital structures (adnexa, vagina, broad ligament)
Hydatidiform moles are benign processes with malignant poten-
III
GTN extends to the lungs, with or without known genital
tial. Malignant transformation occurs in 15% to 20% of complete
tract involvement
hydatiform moles (CHMs) and less than 5% of partial hydatiform
IV
All other metastatic sites
moles (PHMs). In fact, malignant transformation is so rare in PHMs
Abbreviations: FIGO, International Federation of Gynecology and Obstetrics;
that if one is contemplating treatment for this diagnosis, it is prudent
GTN, gestational trophoblastic neoplasia.
to confirm the pathologic diagnosis, to rule out a false-positive human
chorionic gonadotropin (hCG) test, and to make sure that the patient
has not had an incomplete evacuation before proceeding. In the cur-
rent era of early detection of pregnancy, it can be difficult to distin-
In this issue of Journal of Clinical Oncology, Osborne et al4
guish early CHMs from PHMs histologically. Complete and partial
report the results of a randomized phase III trial performed by the
hydatiform moles can be distinguished by performing a p57 immu-
Gynecologic Oncology Group of weekly methotrexate versus
nostain.1 Lack of nuclear p57 staining in villous stromal cells and
pulsed dactinomycin for low-risk GTN. The study accrued patients
cytotrophoblasts confirms diagnosis of CHMs (as opposed to hy-
between 1999 and 2007. In 2002, the FIGO definitions for GTN
dropic abortion or PHMs). CHMs are entirely androgenic, whereas
changed, as did the staging and classification system. The investi-
the p57 gene is paternally imprinted and maternally expressed.
gators chose not to adapt the new definitions for GTN for logistical
According to current International Federation of Gynecology
reasons. They did extend the acceptance of patients with risk scores
and Obstetrics (FIGO) classification,2 hydatiform moles are consid-
from 0 to 4 (June 1999 to June 2002) to 0 to 6 (July 2002 to February
ered to have undergone malignant transformation and therefore meet
2007), and of the 216 evaluable patients, 17 (7.9%) had a risk score
the definition of gestational trophoblastic neoplasm (GTN) if after
5 to 6. The GTN definitions used in the study (
10% decrease in
evacuation there are four values or more indicating an hCG plateau
hCG in three consecutive weekly values;
20% sustained rise in
during a period of at least 3 weeks; a rise of hCG of 10% or greater for
hCG over two consecutive weekly values; persistently elevated
three values or more during a period of at least 2 weeks; or persistence
hCG more than 4 months after initial evacuation; histologically
of hCG 6 months after mole evacuation. Only hydatiform moles that
proven nonmetastatic choriocarcinoma; histologically proven
meet the definition of GTN are subject to staging and risk scoring.
metastatic choriocarcinoma if the metastatic site[s] is one [or
GTN is a term used for hydatiform moles that have undergone malig-
more] of the following: vagina, parametrium, or lung [if no lung
nant transformation; this term is also used for all choriocarcinomas,
lesion is
2 cm]) are significantly different than the current FIGO
including PSTTs and ETTs. PSTTs and ETTs are not subject to staging
definitions, making application of the study results to current
or risk scoring systems and need to be considered separately. Thus, in
practice difficult.
considering treatment of low-risk GTN, one is referring to hydatiform
Several different outpatient chemotherapy regimens have been
moles that have undergone malignant transformation (often re-
used to treat low-risk GTN (Table 3). The variability in primary
ferred to as persistence) and choriocarcinomas, which receive a low
response rates probably, at least in part, represents differences in drug
risk score.
dosages, schedules, routes of administration, and length of therapy
The most recently published FIGO staging and classification
delivered after normalization of hCG, as well as patient selection
systems are detailed in Tables 1 and 2. 3 Use of the FIGO staging system
criteria. Chemotherapy is delivered until hCG values have returned to
is essential for determining initial therapy for patients with GTN to
normal and then at least one cycle is given after the first normal hCG
ensure the best possible outcomes with the least morbidity.
value (practice standards and regimens vary with respect to the
786
© 2011 by American Society of Clinical Oncology
JOURNAL OF CLINICAL ONCOLOGY
Information downloaded from jco.ascopubs.org and provided by at CEDARS SINAI MEDICAL CENTER MEDICAL LIBRARY
Copyright © 2011 on March
American 1, 2011
Society from
of
192.231.86.98
Clinical Oncology. All rights reserved.

---