intrinsic resistance to jak2 inhibition in...
TRANSCRIPT
Cancer Therapy: Preclinical
Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
Anna Kalota1, Grace R. Jeschke1, Martin Carroll1,2, and Elizabeth O. Hexner1,2
AbstractPurpose: Recent results have shown that myeloproliferative neoplasms (MPN) are strongly associated
with constitutive activation of the Janus-activated kinase (JAK)2 tyrosine kinase. However, JAK2 inhibitors
currently approved or under development for treating myeloproliferative neoplasms do not selectively
deplete the malignant clone, and the inhibition of activity of the drug target (JAK2) has not been rigorously
evaluated in the clinical studies. Therefore, in this study we developed an in vitro assay to gain insight into
how effectively JAK2 activity is inhibited in the samples of patients.
Experimental Design: We treated primary cells from normal donors and patients with MPN with
JAK2 inhibitors and measured phosphorylation of downstream targets STAT5 and STAT3 by flow
cytometry. Obtained results were next correlated with JAK2 V617F allele burden and plasma cytokine
level.
Results:We observed a dose-dependent decrease in pSTAT5 and pSTAT3 in ex vivo treated granulocytes.
However, phosphorylation of STAT3 and STAT5 in cells from patients with myelofibrosis was significantly
less inhibited when compared with cells from patients with polycythemia vera, essential thrombocythe-
mia, and normal donors. Sensitivity to inhibition did not correlate with JAK2 V617F clonal burden.
Mixing studies using plasma from patients with myelofibrosis did not transfer resistance to sensitive cells.
Likewise, no single cytokine measured seemed to account for the observed pattern of resistance.
Conclusions: Taken together, these observations suggest that there are cell intrinsic mechanisms
that define a priori resistance to JAK2 inhibition in myelofibrosis, and the lesion is localized upstream
of STAT3 and STAT5. Clin Cancer Res; 19(7); 1729–39. �2013 AACR.
IntroductionPolycythemia vera, essential thrombocythemia, and
myelofibrosis are myeloproliferative neoplasms (MPN)unified by the constitutive activation of the Janus-acti-vated kinase (JAK)2 pathway, conferred most commonlyby a point mutation in the pseudokinase domain of JAK2(JAK2V617F; refs. 1–4). The identification of this muta-tion has prompted wide interest in targeting JAK2 fortherapeutic benefit. Myelofibrosis, whether primary orevolved from polycythemia vera or essential thrombo-cythemia, has a variable but overall poor prognosis.Clinical investigation of JAK2 inhibitors has focused onmyelofibrosis with some unexpected results. Althoughreduction in spleen size and amelioration of constitu-tional symptoms are frequent and can be dramatic, onlymodest changes in mutant clonal burden have been
observed and complete responses by any criteria areexceedingly rare (5–7).
Notably, validation of inhibition of activity of the drugtarget (JAK2) has never been rigorously evaluated in clinicalstudies. JAK family proteins are nonreceptor tyrosinekinases that are associated with cytokine receptors. Cyto-kine engagement with receptor leads to the phosphoryla-tion of JAK2 and the recruitment and phosphorylation ofcytoplasmic signal transducers and activators of transcrip-tion (STAT3 and 5). Upon activation, phosphorylatedSTAT3 and STAT5 (pSTAT3 andpSTAT5) formhomodimersand translocate to the nucleus to activate transcription. InMPN, JAK2with aV617F substitution is constitutively activeand confers cytokine independence/hypersensitivity tohematopoietic tissues. Small-molecule tyrosine kinase inhi-bitors have been shown to be active in both preclinical andlate-phase clinical testing, with analyses of cytokine arraysand measurement of JAK2V617F clonal burden used aspotential biomarkers for response (6, 5). To study theefficacy of JAK2 inhibitors to inhibit JAK signaling in pri-mary patient cells, we usedphospho-specific flow cytometryon whole peripheral blood from patients with MPN andnormal controls to measure phosphorylation of STAT3 andSTAT5, the canonical targets of JAK2. We hypothesized thatthis measurement would be a feasible and informativepharmacodynamic assay for measuring JAK2 signaling inpatients on clinical trials and that differences in signalingmight be observed across the spectrum of disease.
Authors' Affiliations: 1Division of Hematology andOncology and 2Abram-son Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania
Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).
Corresponding Author:ElizabethO. Hexner, University of Pennsylvania, 2West Pavilion, Perelman Center for AdvancedMedicine, 3400 Civic CenterBlvd, Philadelphia, PA 19104. Phone: 215-614-1847; Fax: 215-573-7049;E-mail: [email protected]
doi: 10.1158/1078-0432.CCR-12-1907
�2013 American Association for Cancer Research.
ClinicalCancer
Research
www.aacrjournals.org 1729
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
Materials and MethodsPeripheral blood samples
Peripheral blood samples from patients and normaldonors were collected, annotated, and stored through anInstitutional review board–approved hematologic diseasestissue bank. Previously collected, anonymous samples wereobtained from the Stem Cell and Xenograft Core facility(University of Pennsylvania, Philadelphia, PA).
ReagentsJAK2 inhibitors: CEP701 (lestaurtinib) was provided by
Cephalon Oncology or purchased from LC laboratories.CYT387 and INCB018424 were synthesized by ChemieTekand kindly provided by Dr. Ross Levine (Memorial Sloan-Kettering Cancer Center, New York, NY). Antibodies werepurchased from BD Bioscience and R&D Systems.
Flow cytometryWholeblood sampleswerefixed andpermeabilizedusing
a formaldehyde/triton/methanol method (8, 9) and ana-lyzed by flow cytometry. Briefly, 100 mL of whole bloodwasexposed to varying concentrations of the inhibitor for 15minutes at 37�Cand then stimulated for 20minutes at 37�Cwith10ng/mLgranulocytemacrophage colony-stimulatingfactor (GM-CSF; for pSTAT5 measurement) or 100 ng/mLgranulocyte colony-stimulating factor (G-CSF; for pSTAT3measurement). After stimulation, the samples were fixedwith 4% formaldehyde for 10minutes at room temperatureand permeabilized with 0.1% Triton X-100 for 15 minutesat 37�C. Next, samples were washed twice in cold PBSsupplemented with 4% bovine serum albumin (BSA), trea-ted with cold 100% methanol to enhance epitope avail-
ability, and stored at �20�C. Before analysis, samples werewashed twice in cold PBS supplemented with 4% BSA andthen incubated with directly labeled antibodies at roomtemperature for 30minutes in the dark. Data were acquiredon a BD FACSCalibur using CellQuest Pro software andanalyzed using FlowJo version 9.3.1. For sample of eachpatient, untreated mean fluorescence intensity (MFI) ofpSTAT3 and pSTAT5 was subtracted from treated samplesand MFI for cytokine-stimulated samples was set at 100%.Persistence of pSTAT3 and pSTAT5 in the presence ofinhibitor (X) was calculated as the ratio of mean fluores-cence values of gated events relative to cytokine-stimulatedsample; inhibition is described as 100�X. Statistical anal-ysis was conducted using GraphPad Prism 5.
JAK2 V617F allele burdenMononuclear cells and granulocytes were separated by
density using Ficoll-Paque (Amersham Bioscience). Thegenomic DNA from granulocytes was isolated from cellsusing Gentra Purgene Blood Kit from Qiagen. JAK2 V617Fallele burden was measured using a quantitative real-timePCR (qRT-PCR) assay adapted from Nussenzveig and col-leagues (10) to detect the G!T substitution at position1849 in exon 14 that uses an allele-specific (wild-type vs.mutant) primer with specificity enhanced bymismatch anda locked nucleic acid modification. Allele frequency wascalculated as described by Germer and colleagues (11).
Determination of cytokine level in plasma frompatients and normal donors
Plasma sampleswere obtainedby centrifugationofwholeperipheral blood at 600 � g for 10 minutes and stored at�80�C. Batched frozen samples were analyzed by theHuman Immunology Core (University of Pennsylvania)using a Luminex instrument and the Milliplex MAP Humancytokine 9-plex from Millipore.
Statistical analysisOne-way ANOVA and Student t test were used to analyze
differences between patient and normal donor samples.Correlations were calculated using the Spearman correla-tion coefficient. All calculations weremade using GraphPadPrism 5.
ResultsPhosphorylationof STAT5andSTAT3 isdetectablewithcytokine stimulation and is inhibited in the presence ofCEP701
To more precisely validate target and to define biologicresponse to treatment with JAK2 inhibitors in patientswith MPN, we investigated the level of target inhibition inperipheral blood samples collected from 29 patients withMPN (Table 1) and 6 normal donors. We measured phos-phorylated STAT3 and 5 (pSTAT3 and pSTAT5) as bio-markers for JAK2 inhibition. In our initial studies wedid not observe constitutive basal phosphorylation ofSTAT3 or STAT5 in any samples evaluated (Fig. 1A and
Translational RelevanceThe identification of Janus-activated kinase (JAK)2
mutations in classical, Philadelphia-chromosome neg-ative myeloproliferative neoplasms (MPN) suggested atarget for small-molecule therapy for MPN, with thehope of responses akin to those seen in chronic myeloidleukemia treated with imatinib. Although subsequentstudies of JAK2 inhibitors have proven that they aretherapeutic, responses have been unexpected, and donot in fact meaningfully deplete the clonal burden ofdisease. The mechanism(s) through which JAK2 inhibi-tors exert activity are incompletely understood includinghow effectively they inhibit JAK2 kinase activity in bloodcells. We directly measured downstream inhibition ofSTAT3 and STAT5 phosphorylation in whole blood andfound that across disease subtypes, the response toinhibition varied with myelofibrosis overall resistant.Our findings identify a pattern of resistance that is not inthe classical "escapemutation" paradigm andmay beginto explain the attenuated clinical responses that havebeen observed in the clinic.
Kalota et al.
Clin Cancer Res; 19(7) April 1, 2013 Clinical Cancer Research1730
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
Tab
le1.
Patient
charac
teris
tics
ID#
Reg
istryID
Diagno
sis
Age
Sex
Rac
eWBC
(THO/mL)
%Blasts
HGB
(g/dL)
PLT
(THO/mL)
Cytogen
etics
Treatmen
thistory
V61
7F(%
)Yea
rsfrom
diagno
sis
2211
R12
67Pos
t-PVMye
lofibrosis
74M
W9.8
09.5(t)
4946
,XY[16]
PB,E
SA
890.9
2212
R06
02Polyc
ythe
mia
vera
56F
W4.6
014
.635
646
,XX[26]
PB,H
U,A
SA
6719
.222
22R12
43Prim
arymye
lofibrosis
68F
W5.2
08.5(t)
222
46XX[22]
THAL/PRED
740.7
2223
R10
92Ess
entia
ltho
mboc
ythe
mia
82F
W8.7
012
.860
046
,XX[20]
HU,A
NAG,A
SA,C
HLR
753.4
2273
R06
82Ess
entia
ltho
mboc
ythe
mia
78M
W6.5
012
.640
246
,XY[8]
HU,A
SA
634.3
2283
R06
33Prim
arymye
lofibrosis
46F
W4.1
09.1
245
46,XX[20]
ANAG,A
SA
011
.622
94R13
00Prim
arymye
lofibrosis
65F
W5.5
08.9
129
46,XX[20]
ESA
ND
0.4
2310
R09
73RARS-T
67F
AA
12.8
18.4
1961
46,XX[20]
HU,A
NAG,IFN
,ESA
700.8
2316
R12
31Prim
arymye
lofibrosis
67M
W13
.75
10.7
123
46,XY[20]
ESA
91.5
2317
R13
12Polyc
ythe
mia
vera
78F
W13
.10
14.8
346
ND
PB,H
U15
9.4
2327
R13
19Polyc
ythe
mia
vera
41M
W4.2
014
.417
1ND
PB,A
SA,IF
N79
12.7
2328
R13
20Ess
entia
ltho
mboc
ythe
mia
46M
W5
013
.432
046
,XY[20]
HU
7015
.823
40R12
31Prim
arymye
lofibrosis
67M
W11
.63
9.3
9346
,XY[20]
ESA
111.6
2341
R13
25Prim
arymye
lofibrosis
66M
W18
.10
10.1
1183
46,XY[21]
ESA,H
U,A
SA
730.1
2343
R13
22Polyc
ythe
mia
vera
56M
W12
.10
18.2
443
ND
PB,A
SA
480.1
2359
R11
82Ess
entia
ltho
mboc
ythe
mia
73F
AA
4.9
013
.372
146
,XX[20]
HU,A
SA
791.1
2369
R06
33Prim
arymye
lofibrosis
47F
W4.1
29.7
217
ND
ANAG,A
SA
08.4
2370
R12
69Prim
arymye
lofibrosis
58F
W9.5
013
.489
ND
none
570.6
2371
R13
32Ess
entia
ltho
mboc
ythe
mia
40M
W13
.70
17.1
608
46,X
Y[20]
ASA
911.9
2469
R06
82Ess
entia
ltho
mboc
ythe
mia
78M
W9.4
013
.255
046
,XY[20]
HU,A
SA
544.8
2470
R13
62Polyc
ythe
mia
vera
45F
W4.4
012
.836
846
,XX[25]
ANAG,A
SA,IFN
ND
17.2
2479
R13
08AML(from
pos
t-PVMF)
62M
AA
121.9
35ND(t)
91Com
plex
aPB,H
U,A
SA,IFN
580.8
2487
R06
70Pos
t-PVmye
lofibrosis
57F
W4.7
29.6
133
Com
plex
bPB,H
U,A
SA,IFN
,HSCT
521.3
2517
R13
78Prim
arymye
lofibrosis
63.1
MW
7.6
012
.311
046
,XY[20]
ANAG,H
U,ASA
POS
9.0
2521
R12
36Prim
arymye
lofibrosis
81.8
FW
5.1
2210
.687
46,XX[20]
HU,E
SA
05.0
3013
R15
50Polyc
ythe
mia
vera
43M
W4.3
012
.722
746
,XY[20]
HU,IFN
,PB,A
SA
POS
1.7
3014
R06
39Ess
entia
ltho
mboc
ythe
mia
60.5
FW
70
11.5
1040
ND
ANAG,H
U,A
SA
013
.030
16R15
89Polyc
ythe
mia
vera
59M
AA
6.8
015
304
46,XY[20]
ANAG,H
U,A
SA
6.0
2518
R13
19Polyc
ythe
mia
vera
42M
W3.1
012
.612
2ND
ASA,P
B,IFN
POS
13.0
Abbreviations
:ANAG,a
nagrelide;
ASA,a
spirin;
CHLR
,chloram
bucil;ESA,e
rythropoietin/darbop
oietin;H
SCT,
hematop
oietic
stem
celltran
splantation;
HU,h
ydroxy
urea
;ND,n
oda
ta;P
B,p
hleb
otom
y;POS,p
ositive
but
unkn
ownperce
ntag
e.a47
,XY,þ
der(9
)t(1;9)(q21
;q21
)[1]/4
7,idem
,del(5)(q
21q34
)[19].b
46,XXþd
er(9)t(1;9)(q12
;q21
)[20].
Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
www.aacrjournals.org Clin Cancer Res; 19(7) April 1, 2013 1731
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
Supplementary Fig. S1A; P ¼ 0.48). Therefore, we incorpo-rated cytokine activation into the assay. Unmanipulatedwhole blood from normal donors or patients was stimu-lated with GM-CSF or G-CSF known to activate phosphor-ylation of STAT5 or STAT3, respectively, in humanmyeloidcells (12, 13). Activation was measured in granulocytes bygating on CD15þ cells. There were no differences in cyto-kine responsiveness across disease andnormal samples. Thetotal level of intracellular STAT5 was measured to ensureadequate permeabilization of cells and did not change withstimulation or inhibition (Supplementary Fig. S1B andS1C).
We next used this assay to study the efficacy of JAK2inhibitors (Fig. 1B and C). For these studies, blood fromnormal donors, patients with essential thrombocythemia,polycythemia vera, and patients with the more advanceddisease, myelofibrosis (either primary or evolved frompolycythemia vera or essential thrombocythemia) was firstexposed to inhibitor, primarily lestaurtinib (CEP701) at aconcentration of 10 to 50 mmol/L and then stimulated withGM-CSF or G-CSF to activate phosphorylation of STAT5(Fig. 1B) or STAT3 (Fig. 1C), respectively. The range ofinhibitor concentration was based on mean steady-stateplasma concentrations of 7.7 to 12 mmol/L (1–40 mmol/L)previously described in treated patients for CEP701 (14),and from published pharmacokinetic information onachievable concentrations, when available, forINCB018424 andCYT387 (15).Weobserved adose-depen-dent decrease in cytokine-activated pSTAT3 and pSTAT5 inCD15þ granulocytes from normal donors when treated exvivo with CEP701 (Fig. 1B and C).
Neutrophils from patients with myelofibrosis areintrinsically resistant to JAK2 inhibitors
Next, we sought to measure response to treatment acrossMPN phenotypes. We found that the response to JAK2inhibitor differs among patients withMPN. Although phos-phorylation of STAT5 (Fig. 2A) and STAT3 (Fig. 2B) in bothessential thrombocythemia and polycythemia vera wassimilar and nearly completely abrogated in the presence of20 mmol/L CEP701, pSTAT3 and pSTAT5 in myelofibrosissamples were minimally inhibited (Fig. 2A and B). Inaggregate across samples, the mean inhibition at 20mmol/L in essential thrombocythemia/polycythemia verasamples as measured by pSTAT5 (69%), was significantlymore than that in myelofibrosis samples (36%, P ¼0.015; Fig. 3A and Supplementary Table S1). The responseto inhibition in normal donors (76% inhibited) was com-parable with that observed in essential thrombocythemiaand polycythemia vera samples (pSTAT5, 68.9% inhibited;P¼ 0.90). For pSTAT3, polycythemia vera/essential throm-bocythemia samples were more sensitive (73% inhibited)when compared with myelofibrosis samples (60% inhib-ited) although this was not significant (P ¼ 0.43). Thissuggests that either downstream signaling through STAT3 inmyelofibrosis samples is trulymore responsive to inhibitionthan STAT5 or may reflect a more subtle difference notcaptured in this number of samples. We also noted a
narrower dynamic range for pSTAT3 measurements (Sup-plementary Fig. S1B). Thus, we observed that pSTAT5 andpSTAT3 can be measured in whole blood by phospho-flowcytometry in the presence of exogenous cytokine and thatpharmacologic inhibition is both dose- and MPN-subtypedependent. To extend this observation using 2 additionaltyrosine kinase inhibitors, we repeated these studies withCYT387 and INCB018424, agents currently being investi-gated/approved for the treatment of myelofibrosis. Wefound the samples of myelofibrosis to be similarly lesssensitive to inhibition with these compounds when com-pared with polycythemia vera samples (Fig. 3B). The sam-ples of myelofibrosis exposed to 10 and 20 mmol/L CYT387were significantly more resistant (28% and 36.4% inhibi-tion of pSTAT5, respectively) than polycythemia vera sam-ples (74% and 93% inhibition, P ¼ 0.003 for 10 mmol/L; P¼0.001 for 20mmol/L).Differences in response observed toINCB018424 were also significant (10 mmol/L; myelofibro-sis samples were 42% inhibited, whereas polycythemia verasamples were 72% inhibited, P ¼ 0.025). Taken together,these results suggest that peripheral blood neutrophils frompatients with myelofibrosis were intrinsically resistant toJAK2 inhibitors.
To begin to understand what, if any, coherence mightexist between terminally differentiated neutrophils andmore primitive progenitor compartments with respect tosignaling response, we measured pSTAT5 in CD15þ andCD34þ cells exposed to CYT387 and INCB018424 from apatient with acute myeloid leukemia that had evolved frompost-polycythemia vera myelofibrosis (Fig. 4). We foundthat CD34þ stimulation with GM-CSF results in a moreheterogeneous pSTAT5 signal, likely reflecting differences inGM-CSF receptor expression in this compartment (16), andthat the response to inhibition in CD34þ cells generallymirrors that in CD15þ cells.
The pattern of resistance is not correlated with JAK2V617F allele burden in myelofibrosis
To gain insight into a mechanism by which STAT3 andSTAT5 remain phosphorylated even in the presence ofupstream JAK2 inhibition, we investigated whether wecould detect a correlation between the frequency of JAK2V617F-mutant allele burden in samples from patients withthe mutation and response to drug. Overall, we found norelationship between allele frequency and level of targetinhibition in tested samples (Table 2 and SupplementaryFig. S2). Likewise, a similar (and near complete) pattern ofresistance was observed in a patient withmyelofibrosis whowas negative for the JAK2 V617F mutation. Thus, themutant to wild-type ratio of JAK2 does not seem to directlyaccount for the observed pattern of resistance. We did notethat in cells from patients without myelofibrosis, thereindeed seemed to be a significant direct correlation withhigher levels of JAK2 V617F and persistent signaling (P ¼0.04); although we interpret this cautiously, given thenumbers of samples (n¼ 2) from patients with lower alleleburdens.
Kalota et al.
Clin Cancer Res; 19(7) April 1, 2013 Clinical Cancer Research1732
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
Plasma components do not confer resistance to JAK2inhibitionWe next investigated the possibility that plasma compo-
nents might contribute to or confer resistance of myelofi-brosis samples to JAK2 inhibitors. We mixed cells from apatient with polycythemia vera (sample#: 2518) with plas-ma from patients with myelofibrosis (samples#: 2517 and2521) and evaluated inhibition of pSTAT5 (Fig. 5A–F). For agiven patient, response to CYT387 or INCB018424 as
measured by mean pSTAT5 fluorescence followed a similartrend regardless of plasma source, which suggested thatneither resistance to inhibition nor sensitivity to treatmentwas conferred by plasma components. We also analyzedbasal cytokine levels in plasma samples fromMPN samplesand normal donors. Plasma cytokines that signal throughJAK2 and/or are known to be altered in myelofibrosis weremeasured. No meaningful differences that could directlyexplain the differences in response were observed. For most
Side scatter
CD
15
Total STAT5
phospho S
tat5
Normal Donor 1 A
STAT5 (total)
STAT
5 (
pY
69
4)
10 µmol/L 20 µmol/L 50 µmol/L
GM-CSF (10 ng/mL)
CEP701 –+ + +
Unstimulated GM-CSF (10 ng/mL)
STAT3 (total)
STAT
3 (
pY
70
5)
G-CSF (100 ng/mL)
CEP701 –
– + + +
– 20 µmol/L 50 µmol/L
B
C
0 200 400 600 800 1,000
0
101
102
103
42.1
0 101
102
103
0
101
102
103
0.02 1.07
96.82.13
0 101
102
103
0
101
102
103
1.39 98.1
0.460.036
0 101
102
103
0
101
102
103
1.39 98.1
0.460.04
0 101
102
103
0
101
102
103
0.8 55
431.21
0 101
102
103
0
101
102
103
0 3.4
96.20.3
0 101
102
103
0
101
102
103
0 1.2
98.40.45
0 10 1 10 2 10 3
0
101
102
103
0.3 2.23
87.310.2
0 10 1 10 2 10 3
0
101
102
103
3.55 36.6
53.46.5
0 10 1 10 2 10 3
0
101
102
103
0.1 1.15
899.73
0 10 1 10 2 10 3
0
101
102
103
0.01 1.18
92.76.08
0 101
102
103
0
101
102
103
0.25 0.15
3.995.7
+
Unstained
Figure 1. Measurement of STAT5 and STAT3 phosphorylation in normal peripheral blood granulocytes. A, the gating strategy for isolating granulocytes basedon forward/side scatter properties and expression of CD15 antigen. Total (x-axis) and phospho- (y-axis) STAT5 (B) and STAT3 (C); detection of total STAT3 orSTAT5 served as an internal control for permeabilization. B and C, target inhibition in response to exposure to CEP701 at 10, 20, and 50 mmol/L doses.
Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
www.aacrjournals.org Clin Cancer Res; 19(7) April 1, 2013 1733
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
of the 9 evaluated cytokines (MIP-1b, TNF-a, IL-6, GM-CSF,IFN-g , G-CSF, IFN-a, and Rantes), the observed concentra-tion in patients with MPN was higher than that in normaldonors, consistent with other reports (refs. 6, 17; Supple-mentary Fig. S3). In particular, the basal plasma levels of the2 cytokines also used exogenously in our experiments (G-and GM-CSF) were plotted against response to inhibitor,and there was no significant correlation between basal GM-CSF levels and response to inhibitor (P¼ 0.18, Supplemen-
tary Fig. S3B–S3D). These observations thus fail to show asoluble cell extrinsic factor that explains JAK2 inhibitorresistance in granulocytes from patients withmyelofibrosis.
DiscussionOur findings show that in myelofibrosis, cell intrinsic
properties bestow resistance to JAK2 inhibition, althoughthe mechanism remains unclear. We found that overall
A STAT5
Polycythemia vera (patient# 2317)
100
101
102
103
104
0
101
102
103
0 0.62
96.92.52
100
101
102
103
104
0
101
102
103
1.99 95.6
20.4
100
101
102
103
104
0
101
102
103
0.65 78.8
19.80.768
100
101
102
103
104
0
101
102
103
0.02 4.2
94.91
STAT5 (total)
STAT
5 (
pY
694)
GM-CSF (10 ng/mL)
CEP701+ + +-
- - 10 µmol/L 20 µmol/L
B STAT3
Primary myleofibrosis (patient# 2316)GM-CSF (10 ng/mL)
CEP701
+ + +- - - 10 µmol/L 20 µmol/L
STAT5 (total)
STAT
5 (
pY
694)
G-CSF (100 ng/mL)
CEP701
+ + +- - - 20 µmol/L 50 µmol/L
STAT3 (total)
STAT
3 (
pY
705)
100
101
102
103
104
0
101
102
103
9.92 84.1
4.231.86
100
101
102
103
104
0
101
102
103
0 1.7
94.53.85
100
101
102
103
104
0
101
102
103
6.49 80.1
11.12.39
100
101
102
103
104
0
101
102
103
7.25 60.8
27.14.88
100
101
102
103
104
0
101
102
103
0 0.87
954.13
100
101
102
103
104
0
101
102
103
2.29 56.8
391.94
100
101
102
103
104
0
101
102
103
0.6 34
64.40.9
100
101
102
103
104
0
101
102
103
0 1.98
96.91.15
Polycythemia vera (patient# 2317)
Primary myleofibrosis (patient# 2316)
STAT3 (total)
STAT
3 (
pY
705)
G-CSF (100 ng/mL)
CEP701
+ + +- - - 20 µmol/L 50 µmol/L
10 0 10 1 10 2 10 3 10 4
0
101
102
103
0.01 2.02
92.15.9
10 0 10 1 10 2 10 3 10 4
0
101
102
103
2.67 58.2
36.32.87
10 0 10 1 10 2 10 3 10 4
0
101
102
103
0.01 2.5
92.45.19
10 0 10 1 10 2 10 3 10 4
0
101
102
103
0 0.69
96.52.9
Figure 2. Relative resistance toinhibition of STAT5 and STAT3phosphorylation in myelofibrosis.Representative flow cytometryplots show level of inhibition ofSTAT5 (A) and STAT3 (B)phosphorylation after exposure toCEP701 for patients respectivelydiagnosed with polycythemia vera(top) and primary myelofibrosis(bottom).
Kalota et al.
Clin Cancer Res; 19(7) April 1, 2013 Clinical Cancer Research1734
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
polycythemia vera and essential thrombocythemia samplesare sensitive to inhibition by multiple investigational inhi-bitors, with responses overall comparable with those ofnormal donors. In contrast, in cells from patients withtreatment-na€�ve myelofibrosis, STAT5 remained phosphor-ylated despite treatment with inhibitors, suggesting eitherincomplete upstream inhibition of JAK2 and/or alternativeinputs to STAT5 (and possibly STAT3) that are specific tomyelofibrosis, typically a more advanced and geneticallycomplex disease. Our studies did not support a role forspecific cell extrinsic soluble factors in plasma as majormediators of resistance, nor could we correlate the clonalburden of JAK2 V617F with the resistance phenotype,suggesting that additional acquired features specific tomye-lofibrosis mediated the attenuated response.Our signaling studies focused largely on mature, termi-
nally differentiated neutrophils from patients, an experi-mental design conceived with the intent of developing afeasible peripheral blood-based pharmacodynamic assayfor measuring response to treatment. But whether ourobservation in neutrophils extends to signaling in more
primitive, disease-initiating compartments is also of greatinterest. Our exploratory studies of CD34þ blasts andneutrophils from a patient with acute leukemia evolvedfrom post-polycythemia vera myelofibrosis showed aresponse thatmirrors the response in granulocytes, support-ing the possibility that a signaling phenotype carriesthrough to terminal differentiation, but additional studiesare needed to draw major conclusions about signalingresponses throughout differentiation.Of note, other studieshave confirmed that granulocytes from patients with mye-lofibrosis are derived largely from themalignant clone (18).Although our studies cannot directly distinguish signalingwithin clonally derived granulocytes versus those derivedfrom the premalignant stem cell pool, the signaling patternsinmyelofibrosis tended to be relatively uniform, andwouldbe expected to represent, at least proportionally, clonallyderived cells. In contrast, clonal dominance is not thepredominant pattern in polycythemia vera and essentialthrombocythemia, and this may explain the overlappingsignaling phenotype among polycythemia vera, essentialthrombocythemia, and normal donors (18). Although we
B
0 20 40 600
20
40
60
80
100
120
Dose of CEP701 (µmol/L)
Normal donors
0 10 20 30 40 50 600
50
100
150
Dose of CEP701 (µmol/L)
Normal donors
MF MF
PV/ET PV/ET
pS
TA
T5 (
perc
ent of co
ntr
ol)
pS
TA
T3 (
perc
ent of co
ntr
ol) A
0
40
100
160
0 1 10 20
pS
TA
T5 (
perc
ent of contr
ol)
Dose of INCB018424 (µmol/L)
PV/ET
MF
0
40
80
120
0 1 10 20
pS
TA
T5 (
perc
ent of contr
ol)
Dose of CYT387 (µmol/L)
PV/ET
MF
Figure 3. A,CD15þcells frompatientswithmyelofibrosis are significantlymore resistant to JAK2 inhibition thancells frompatientswith polycythemia vera (PV),essential thrombocythemia (ET), and normal controls. Graphs showmolecular response to various doses of JAK2 inhibitor, CEP701, in normal donors whencompared with samples from patients with polycythemia vera/essential thrombocythemia and myelofibrosis (MF). Molecular response to the drug ispresented as a percentage of STAT5 (left) and STAT3 (right) phosphorylation remaining after exposure to CEP701. Mean fluorescence from cells stimulatedwith GM-CSF for pSTAT5 or G-CSF for pSTAT3 minus background (mean fluorescence of unstimulated cells) was set as a 100%. B, resistance to inhibitioncan be generalized to additional JAK2 inhibitors under clinical development. Response to CYT387 (right) and INCB018424 (left) are plotted as the percentageSTAT5 phosphorylation after exposure to JAK2 inhibitors. MFI from cells stimulated with GM-CSF for pSTAT5 minus background (mean fluorescence ofunstimulated cells) was set as a 100%.
Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
www.aacrjournals.org Clin Cancer Res; 19(7) April 1, 2013 1735
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
100
101
102
103
104
0
101
102
103
0 3.4
96.60
100
101
102
103
104
0
101
102
103
0 87
130
100
101
102
103
104
0
101
102
103
0 21.8
78.20
100
101
102
103
104
0
101
102
103
0 6.6
93.40
STAT
5 (
pY
694)
Side scatter
GM-CSF (10 ng/mL)
INCB018424
CD15+ cells (patient# 2479)
--
+-
+20 µmol/L
+50 µmol/L
STAT
5 (
pY
694)
Side scatter
GM-CSF (10 ng/mL)
INCB018424
CD34+ cells (patient# 2479)
--
+-
+20 µmol/L
+50 µmol/L
100
101
102
103
104
0
101
102
103
0 7
92.90
100
101
102
103
104
0
101
102
103
0 51.3
48.70
100
101
102
103
104
0
101
102
103
0 2.7
97.30
100
101
102
103
104
0
101
102
103
0 2.5
97.50
100
101
102
103
104
0
101
102
103
0 3.4
96.60
100
101
102
103
104
0
101
102
103
0 87
130
STAT
5 (
pY
694)
Side scatter
GM-CSF (10 ng/mL)
CYT387
CD15+ cells (patient# 2479)
--
+-
+20 µmol/L
+50 µmol/L
STAT
5 (
pY
694)
Side scatter
GM-CSF (10 ng/mL)
CYT387
CD34+ cells (patient# 2479)
--
+-
+20 µmol/L
+50 µmol/L
100
101
102
103
104
0
101
102
103
0 7
92.90
100
101
102
103
104
0
101
102
103
0 51.3
48.70
100
101
102
103
104
0
101
102
103
0 21.9
78.20
100
101
102
103
104
0
101
102
103
0 22.8
77.20
100
101
102
103
104
0
101
102
103
0 57.5
42.50
100
101
102
103
104
0
101
102
103
0 59.3
40.70
Figure 4. Coherence between terminally differentiated and progenitor signaling response. Measurement of pSTAT3 and pSTAT5 in CD15þ andCD34þ cells exposed to INCB018424 (top) and CYT387 (bottom) from a patient with acute myeloid leukemia evolved from post-polycythemia veramyelofibrosis.
Kalota et al.
Clin Cancer Res; 19(7) April 1, 2013 Clinical Cancer Research1736
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
found no direct correlation between JAK2 V617F alleleburden and resistance in myelofibrosis, but the samplesfrom patients with essential thrombocythemia or polycy-themia vera with higher JAK2 V617F seemed to be less
sensitive to treatment than samples with lower allele bur-dens. Taken together, one wonders whether clonal domi-nance in a bulk population might influence probability ofindividual cell signaling.
Figure 5. Plasma components donot alter sensitivity to JAK2 inhibitor.Effect of plasma components onresponse to CYT387 (A–C) orINCB018424 (D–F) as measured bySTAT5 phosphorylation andpresented as MFI. Peripheral bloodcells from 2 patients withmyelofibrosis (2521, A and 2517, B)were treated ex vivowith GM-CSF inthe presence or absence of CYT387,as whole, unperturbed blood (WB),as plasma-depleted bloodreconstituted with native plasma(e.g., 2521 þ 2521), or withpolycythemia vera patient (2518)plasma of equivalent volume(e.g., 2521 þ 2518). Likewise,polycythemia vera peripheral bloodwas exposed to CYT387 (C) orINCB018424 (F) as whole blood,native reconstituted blood, or withplasma from either myelofibrosispatient (2518 þ 2517 and 2518 þ2521). Myelofibrosis blood cells(2517, D and 2521, E) were treatedwith GM-CSF with or withoutINCB018424 in native plasma orpolycythemia vera plasma.
CYT387
MF cells in PV plasma
0
20
40
60
80
2517 WB
2517+2517
2517+2518
Mean flu
ore
scence
MF cells in PV plasma
Untreated GM-CSF 1 µmol/L 10 µmol/L
Untreated GM-CSF 1 µmol/L 10 µmol/L Untreated GM-CSF 1 µmol/L 10 µmol/L
Untreated GM-CSF 1 µmol/L 10 µmol/L Untreated GM-CSF 1 µmol/L 10 µmol/L
Untreated GM-CSF 1 µmol/L 10 µmol/L 0
20
40
60
80
2521 W B
2521+2521
2521+2518
Mean flu
ore
scence
CYT387
Mean flu
ore
scence
0
20
40
60
80
2521 WB
2521+2521
2521+2518
INCB018424
A
B E
MF cells in PV plasma
0
20
40
60
80
2517 WB
2517+2517
2517+2518
Mean flu
ore
scence
INCB018424
D
MF cells in PV plasma
PV cells in MF plasma
0
20
40
60
80
2518 WB
2518+2518
2518+2517
2518+2521
CYT387
Mean flu
ore
scence
INCB018424
Mean flu
ore
scence
PV cells in MF plasma
0
20
40
60
80
2518 WB
2518+2518
2518+2517
2518+2521
C F
Table 2. Level of mutant JAK2 V617F allele frequency in granulocytes does not correlate with response toinhibition of phosphoSTAT5
Stratified by JAK2 V617F allele freq Mean allele freq (%) Mean % pSTAT5 at 20 mmol/L n
All patients with JAK2 V617F mutation<50% 17 49.6 5>50% 74 48.6 11
Polycythemia vera/essential thrombocythemia<50% 31 4.05 2>50% 75 35.74 7
Myelofibrosis<50% 8 80.03 3>50% 73 71.02 4
Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
www.aacrjournals.org Clin Cancer Res; 19(7) April 1, 2013 1737
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
Acquired resistance to targeted therapies has been welldescribed (19), although classical escape mutations do notseem to be a major mechanism of resistance to JAK2inhibitors. Perhaps unique features of JAK2 as a target leadto less selective pressure for outgrowthof resistant clones, sothat classical gatekeeper mutations as seen in chronic mye-loid leukemia would not be favored: JAK2 is absolutelyrequired for normal hematopoiesis, and consequentlydefines a more narrow therapeutic window between dis-eased, JAK2 dysregulated and normal cells. Elegant exper-imental engineered ex vivo systems of chronic drug exposurehave likewise shown that resistance mutations do notaccount for persistent (or reactivated) downstream signal-ing in JAK2 V617F cells that grow despite the presence ofinhibitor (20). Rather, signaling in trans of JAK2 by JAK1(and TYK2) seems to mediate a novel mechanism of druginsensitivity that seems to be epigenetically determined atthe JAK2 locus itself. Our data are distinguished from thesein that the cells have not been chronically exposed to drugbut instead are derived from treatment-na€�ve patients andyet our observationsmay have similar underpinnings. Mye-lofibrosis is increasingly recognized as an entity governed byepigenetic dysregulation that confers competitive advantageto the diseased stem cell compartment, an advantage withwhich JAK2 V617F itself may conspire (21, 22), but wherethe mutation alone cannot confer such an advantage (23).
We used pSTAT5 as a surrogate for activation of the JAK2pathway. One limitation of this strategy is that we cannotexclude the possibility that JAK2 was appropriately inhib-ited in myelofibrosis samples, whereas STAT5 remainedactivated in response to other signaling pathways. We werenot able to directly detect JAK2/phospho-JAK2 in primaryMPN samples. Future studies using other strategies willaddress whether novel heterodimers can be detected inprimary cells, and if present, whether they correlate with
persistent activationof STAT3andSTAT5.Ourmethods alsocould not specifically address microenvironmental factorsknown to play an important role in drug response andresistance, which might carry through in progeny of dis-eased stem/progenitor cells (24). Of particular interest iswhether terminally differentiated cell signaling in a clonalpopulation faithfully recapitulates that of more primitiveancestors within the stem/progenitor cell compartment.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: A. Kalota, E.O. HexnerDevelopment of methodology: A. Kalota, M. Carroll, E. O. HexnerAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): A. Kalota, G.R. Jeschke, E.O. HexnerAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): A. Kalota, G.R. Jeschke, M. Carroll, E.O.HexnerWriting, review, and/or revision of the manuscript: A. Kalota, G.R.Jeschke, M. Carroll, E.O. HexnerAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): A. Kalota, E.O. HexnerStudy supervision: A. Kalota, E.O. Hexner
AcknowledgmentsThe authors thank Joy Cannon (StemCell and Xenograft Core, University
of Pennsylvania) for coordination of samples and Gerald Wertheim for helpwith qPCR calculations.
Grant SupportM. Carroll has previously received research funding from Cephalon
Oncology. E.O. Hexner is supported by an NIH grant (K23-HL-093366-01A1) and American Society of Hematology Scholar (fellow) in ClinicalTranslational Research.
The costs of publication of this article were defrayed in part by the pay-ment of page charges. This article must therefore be hereby marked advertise-ment in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received June 18, 2012; revised January 9, 2013; accepted January 30,2013; published OnlineFirst February 5, 2013.
References1. Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, Swanton S,
et al. Acquired mutation of the tyrosine kinase JAK2 in human mye-loproliferative disorders. Lancet 2005;365:1054–61.
2. James C, Ugo V, Le Couedic JP, Staerk J, Delhommeau F, Lacout C,et al. A unique clonal JAK2 mutation leading to constitutive signallingcauses polycythaemia vera. Nature 2005;434:1144–8.
3. Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, Passweg JR,et al. A gain-of-function mutation of JAK2 in myeloproliferative dis-orders. N Engl J Med 2005;352:1779–90.
4. Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ, et al.Activating mutation in the tyrosine kinase JAK2 in polycythemia vera,essential thrombocythemia, and myeloid metaplasia with myelofibro-sis. Cancer Cell 2005;7:387–97.
5. Pardanani A, Gotlib JR, Jamieson C, Cortes JE, Talpaz M, Stone RM,et al. Safety and efficacy of TG101348, a selective JAK2 inhibitor, inmyelofibrosis. J Clin Oncol 2011;29:789–96.
6. Verstovsek S, Kantarjian H, Mesa RA, Pardanani AD, Cortes-FrancoJ, Thomas DA, et al. Safety and efficacy of INCB018424, a JAK1and JAK2 inhibitor, in myelofibrosis. N Engl J Med 2010;363:1117–27.
7. VerstovsekS,MesaRA,Gotlib J, LevyRS,GuptaV,DiPersio JF, et al. Adouble-blind, placebo-controlled trial of ruxolitinib for myelofibrosis. NEngl J Med 2012;366:799–807.
8. Chow S, Minden MD, Hedley DW. Constitutive phosphorylation of theS6 ribosomal protein via mTOR and ERK signaling in the peripheralblasts of acute leukemia patients. Exp Hematol 2006;34:1183–91.
9. Hedley DW, Chow S, Goolsby C, Shankey TV. Pharmacodynamicmonitoring of molecular-targeted agents in the peripheral blood ofleukemia patients using flow cytometry. Toxicol Pathol 2008;36:133–9.
10. NussenzveigRH,SwierczekSI, Jelinek J,GaikwadA, Liu E, VerstovsekS, et al. Polycythemia vera is not initiated by JAK2V617Fmutation. ExpHematol 2007;35:32–8.
11. Germer S, Holland MJ, Higuchi R. High-throughput SNP allele-fre-quency determination in pooled DNA samples by kinetic PCR.Genome Res 2000;10:258–66.
12. Kimura A, Rieger MA, Simone JM, Chen W, Wickre MC, Zhu BM, et al.The transcription factors STAT5A/B regulateGM-CSF-mediated gran-ulopoiesis. Blood 2009;114:4721–8.
13. Zhang H, Nguyen-Jackson H, Panopoulos AD, Li HS, Murray PJ,Watowich SS. STAT3 controls myeloid progenitor growth duringemergency granulopoiesis. Blood 2010;116:2462–71.
14. Levis M, Ravandi F, Wang ES, Baer MR, Perl A, Coutre S, et al. Resultsfrom a randomized trial of salvage chemotherapy followed by lestaur-tinib for patients with FLT3 mutant AML in first relapse. Blood2011;117:3294–301.
Kalota et al.
Clin Cancer Res; 19(7) April 1, 2013 Clinical Cancer Research1738
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
15. Tyner JW,BummTG,Deininger J,WoodL,AichbergerKJ, LoriauxMM,et al. CYT387, a novel JAK2 inhibitor, induces hematologic responsesand normalizes inflammatory cytokines in murine myeloproliferativeneoplasms. Blood 2010;115:5232–40.
16. GibbsKDJr,Gilbert PM, SachsK, Zhao F, BlauHM,Weissman IL, et al.Single-cell phospho-specific flow cytometric analysis demonstratesbiochemical and functional heterogeneity in human hematopoieticstem and progenitor compartments. Blood 2011;117:4226–33.
17. Fleischman AG, Aichberger KJ, Luty SB, Bumm TG, Petersen CL,Doratotaj S, et al. TNFalpha facilitates clonal expansion of JAK2V617Fpositive cells in myeloproliferative neoplasms. Blood 2011;118:6392–8.
18. Moliterno AR, Williams DM, Rogers O, Isaacs MA, Spivak JL. Phe-notypic variability within the JAK2 V617F-positive MPD: roles ofprogenitor cell and neutrophil allele burdens. Exp Hematol 2008;36:1480–6.
19. Shah NP, Nicoll JM, Nagar B, Gorre ME, Paquette RL, Kuriyan J,et al. Multiple BCR-ABL kinase domain mutations confer polyclonalresistance to the tyrosine kinase inhibitor imatinib (STI571) in
chronic phase and blast crisis chronic myeloid leukemia. CancerCell 2002;2:117–25.
20. Koppikar P, Bhagwat N, Kilpivaara O, Manshouri T, Adli M, Hricik T,et al. Heterodimeric JAK-STAT activation as a mechanism of persis-tence to JAK2 inhibitor therapy. Nature 2012;489:155–9.
21. Liu F, Zhao X, Perna F, Wang L, Koppikar P, Abdel-Wahab O, et al.JAK2V617F-mediated phosphorylation of PRMT5 downregulates itsmethyltransferase activity and promotes myeloproliferation. CancerCell 2011;19:283–94.
22. DawsonMA,Bannister AJ,GottgensB, Foster SD, Bartke T,GreenAR,et al. JAK2 phosphorylates histone H3Y41 and excludes HP1alphafrom chromatin. Nature 2009;461:819–22.
23. Li J, Spensberger D, Ahn JS, AnandS,Beer PA,Ghevaert C, et al. JAK2V617F impairs hematopoietic stemcell function in a conditional knock-in mouse model of JAK2 V617F-positive essential thrombocythemia.Blood 2010;116:1528–38.
24. Houthuijzen JM, Daenen LG, Roodhart JM, Voest EE. The role ofmesenchymal stem cells in anti-cancer drug resistance and tumourprogression. Br J Cancer 2012;106:1901–6.
Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
www.aacrjournals.org Clin Cancer Res; 19(7) April 1, 2013 1739
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907
2013;19:1729-1739. Published OnlineFirst February 5, 2013.Clin Cancer Res Anna Kalota, Grace R. Jeschke, Martin Carroll, et al. Intrinsic Resistance to JAK2 Inhibition in Myelofibrosis
Updated version
10.1158/1078-0432.CCR-12-1907doi:
Access the most recent version of this article at:
Material
Supplementary
http://clincancerres.aacrjournals.org/content/suppl/2013/02/05/1078-0432.CCR-12-1907.DC1
Access the most recent supplemental material at:
Cited articles
http://clincancerres.aacrjournals.org/content/19/7/1729.full#ref-list-1
This article cites 24 articles, 9 of which you can access for free at:
Citing articles
http://clincancerres.aacrjournals.org/content/19/7/1729.full#related-urls
This article has been cited by 1 HighWire-hosted articles. Access the articles at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://clincancerres.aacrjournals.org/content/19/7/1729To request permission to re-use all or part of this article, use this link
on February 3, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst February 5, 2013; DOI: 10.1158/1078-0432.CCR-12-1907