Serum Immunoregulatory Proteins as Predictors of Overall ...Treatment with ipilimumab improves overall survival (OS) in patients with metastatic melanoma. Because ipilimumab targets

Microenvironment and Immunology

Serum Immunoregulatory Proteins as Predictorsof Overall Survival of Metastatic MelanomaPatients Treated with IpilimumabYoshinobu Koguchi1, Helena M. Hoen1, Shelly A. Bambina1, Michael D. Rynning2,Richard K. Fuerstenberg2, Brendan D. Curti1, Walter J. Urba1, Christina Milburn3,Frances Rena Bahjat3, Alan J. Korman3, and Keith S. Bahjat1

Abstract

Treatment with ipilimumab improves overall survival (OS)in patients with metastatic melanoma. Because ipilimumabtargets T lymphocytes and not the tumor itself, efficacy maybe uniquely sensitive to immunomodulatory factors present atthe time of treatment. We analyzed serum from patients withmetastatic melanoma (247 of 273, 90.4%) randomly assignedto receive ipilimumab or gp100 peptide vaccine. We quantifiedcandidate biomarkers at baseline and assessed the associationof each using multivariate analyses. Results were confirmed inan independent cohort of similar patients (48 of 52, 92.3%)treated with ipilimumab. After controlling for baseline covari-ates, elevated chemokine (C-X-C motif) ligand 11 (CXCL11)and soluble MHC class I polypeptide–related chain A (sMICA)were associated with poor OS in ipilimumab-treated patients

[log10 CXCL11: HR, 1.88; 95% confidence interval (CI), 1.14–3.12; P ¼ 0.014; and log10 sMICA quadratic effect P ¼ 0.066;sMICA (� 247 vs. 247): HR, 1.75; 95% CI, 1.02–3.01]. Mul-tivariate analysis of an independent ipilimumab-treated cohortconfirmed the association between log10 CXCL11 and OS (HR,3.18; 95% CI, 1.13–8.95; P ¼ 0.029), whereas sMICA was lessstrongly associated with OS [log10 sMICA quadratic effect P ¼0.16; sMICA (�247 vs. 247): HR, 1.48; 95% CI, 0.67–3.27].High baseline CXCL11 and sMICA were associated with poorOS in patients with metastatic melanoma after ipilimumabtreatment but not vaccine treatment. Thus, pretreatmentCXCL11 and sMICAmay represent predictors of survival benefitafter ipilimumab treatment as well as therapeutic targets. CancerRes; 75(23); 5084–92. �2015 AACR.

IntroductionIpilimumab is a humanmonoclonal antibody targeting CTLA-

4 (cytotoxic T lymphocyte antigen-4). CTLA-4 is expressed onactivated T cells, has structural similarities to the costimulatorymolecule CD28, and binds to the same ligands as CD28 albeitwith higher affinity. Binding of CTLA-4 toCD80/CD86 inhibits T-cell activation by limiting IL2 production and expression of theIL2 receptor (CD25; ref. 1). Ipilimumab prevents CTLA-4 frombinding its ligands, thus promoting activation of effector T cellsvia prolonged CD28 signaling (2). In addition, anti-CTLA-4antibodies can deplete intratumoral regulatory T cells, subvertingyet another mechanism of immunosuppression (3).

Ipilimumab improved overall survival (OS) in patients withmetastatic melanoma in a randomized, double-blinded phase IIIclinical trial (4, 5). Patients with metastatic melanoma who

received ipilimumab monotherapy achieved a 28.4% four-yearOS rate (6). Despite the success of ipilimumab, themajority of thepatients on this study died as a consequence of melanoma. Withan increasing number of treatment options available andincreased use of targeted therapies, predictive biomarkers thatidentify those patients most likely to benefit from a specifictreatment are needed (7, 8).

Unlike traditional cancer therapies, immunotherapeutics actprimarily upon cells of the immune system. The requirement forthe immune system as a third-partymediator of the drug's activitysuggests the balance of positive and negative regulators of theimmune response at the time of therapy may be a critical deter-minant of efficacy for any immunotherapy. Cytokines, chemo-kines, and soluble receptors regulate the survival, activity, andlocation of immune effector cells and thus represent potentialplayers in determining drug efficacy. Of particular interest aresoluble factors involved in the recruitment and regulation ofeffector T cells representing the most readily measurable clinicalbiomarkers.

To identify candidate soluble factor(s) predictive of improvedsurvival following ipilimumab treatment, we analyzed pretreat-ment sera from treatment (ipilimumab) and "active control"(gp100 vaccine) patients from the pivotal phase III clinical trialof ipilimumab (4) for a variety of factors and correlated theirlevels with OS. A hypothesis-guided panel of candidate biomar-kers was selected, including biomarkers previously reported toassociate with response to ipilimumab. Each analyte was assessedin univariate and multivariate models for its correlation with OS.

1Earle A. Chiles Research Institute, Providence Cancer Center, Port-land, Oregon. 2R&D Systems, Minneapolis, Minnesota. 3Bristol-MyersSquibb, Redwood City, California.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Keith S. Bahjat, Earle A. Chiles Research Institute,Providence Cancer Center, 4805 NE Glisan Street, 2N83, Portland, OR 97213.Phone: 503-215-7229; Fax: 503-215-6841; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-15-2303

�2015 American Association for Cancer Research.

CancerResearch

Cancer Res; 75(23) December 1, 20155084

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Correlative biomarkers identified in the initial screen were furthervalidated by testing sera from an independent cohort of ipilimu-mab-treated patients at our institution.

Patients and MethodsClinical trials

Detailed information regarding the phase III clinical trial ofipilimumab (NCT00094653) was reported elsewhere (4). Briefly,patientswithmetastaticmelanomahaving failed at least one priortherapy that may have included IL2, dacarbazine, and/or temo-zolomide were enrolled excluding those with ocular melanoma.All patients were HLA-A�0201þ as the restricting element for thegp100 peptides used. All ipilimumab-treated patients receivedipilimumab alone at 3 mg/kg every 3 weeks for 4 treatments. Inthe gp100 group, patients received two peptides (1 mg each),injected subcutaneously as an emulsionwith incomplete Freund'sadjuvant (Montanide ISA-51). Peptide injections were givenimmediately after 90-minute intravenous infusion of placebo.Tumor burden was assessed by the treating physician as previ-ously described (4).

Serum samples were also obtained from patients treated on anexpanded access program at the Earle A. Chiles Research Institute(EACRI cohort). Detailed information regarding this Compas-sionate Use Trial for Unresectable Melanoma with Ipilimumab isavailable elsewhere (NCT00495066). All patients received ipili-mumab alone (3 or 10mg/kg every 3weeks for 4 treatments) withno exclusions for ocular primary melanomas or HLA type.

All patients provided written informed consent and all studieswere carried out in accordance with the Declaration of Helsinkiunder good clinical practice and Institutional Review Boardapproval.

Serum cytokine analysisSerum was collected and stored at �80�C. Chemokine (C-C

motif) ligand 2 (CCL2), CCL3, CCL4, CCL8, CCL18, CCL26,chemokine (C-X-C motif) ligand (CXCL9), CXCL10, CXCL11,CXCL13, and VEGF were measured using a bead-based multi-plexed immunoassay (R&D Systems). Soluble MHC class I poly-peptide–related sequence A (sMICA), sMICB, soluble UL16-bind-ing protein (sULBP)-1, sULBP-2, sULBP-3, and sULBP-4 weremeasured using a custom multiplex bead array (R&D Systems).Bead-based assays were analyzed using the Luminex-based Bio-Plex system (BIO-RAD). Soluble CD25 (sCD25) and solublelymphocyte-activation gene 3 (sLAG-3) were measured by ELISA(R&D Systems). Serum sHLA-G was measured by ELISA (ExbioVestec). Only serum cytokines having statistical significance inunivariate analyses of OS were reported.

Statistical considerationsDifferences in patient baseline characteristics between treat-

ment groups (ipilimumab vs. gp100) or trials (phase III vs.compassionate use) were evaluated using a t test for age and c2

tests for gender, Eastern Cooperative Oncology Group (ECOG)performance status, lactate dehydrogenase (LDH), prior IL2 ther-apy, and prior immunotherapy. Differences in baseline serumbiomarkers between study and treatment groups were tested withWilcoxon rank-sum tests because of skewed distributions.

Analysis of OS was conducted in the phase III trial and sepa-rately in the confirmatory EACRI cohort due to the differences inthe patient populations, study design, and study protocol. In the

phase III trial, differences within treatment group were of primaryinterest and thus tested in separate models. Survival was definedas time frombeginning of ipilimumab treatment to date of death,censoring at date of last follow-up. To calculatemedian follow-uptime, deaths were censored.

Univariate survival analysis was performed for each treatmentsubgroup using Cox proportional hazards regression. Effects ofCXCL11, sMICA, sMICB, sCD25, VEGF, absolute lymphocytecounts (ALC), tumor burden, and effect of LDH [� vs. > upperlimit of normal (ULN)] were shown. Models of quadratic effectsto examine possible nonlinear effects and models of linear effectsof continuous variableswere tested.When the quadratic effectwasnot significant or the linear effect was more strongly significant,main effects model results were reported. For sMICA and VEGF,quadratic effect was significant in somemodels. To present anHR,results are also reported for a categorized variable, with the cutoffpoint determined as the quintile where a threshold effect wasobserved in the phase III trial. Kaplan–Meier plots used fordetermining cutoff points are shown in Supplementary Fig. S1.In multivariate analyses, model results of the remaining variables(other than sMICA), however, are from the model containing thecontinuous form of the variable with the quadratic effect. Con-tinuousmeasureswere approximately lognormal and analyzed aslog10-transformed.

In multivariate analysis of OS, Cox proportional hazardsregression was used to test effects of biomarker candidates onsurvival after controlling for other biomarkers and baselinepatient characteristics. Only CXCL11 and sMICA were included,as they were significant in univariate survival models of theipilimumab group but not the gp100 group. Covariates inmodelsfor both studies were age, gender, ECOG status, prior immuno-therapy, LDH, and ALC. Tumor burden was also included inmultivariatemodel for thephase III trial cohort, but not the EACRIcohort as these data were not captured.

Analyses were performed using SAS 9.3 (SAS Institute Inc.).Forest plots were prepared using Forest Plot Viewer (9) and editedusing Adobe Illustrator. GraphPad Prism was used for depictingsome Kaplan–Meier plots.

ResultsPatient characteristics of phase III study

Demographics for the phase III study were previously reported(4). Briefly, 676patientswere enrolledwith 137 selected to receiveipilimumab monotherapy (treatment group), 136 to receivegp100 monotherapy (control group), and 403 treated with thecombination of these agents. Biomarker analysis was restricted tothe monotherapy groups. Baseline characteristics were similarbetween monotherapy groups (Table 1), except that a higherproportion of patients received prior immunotherapy in thegp100 alone group (P ¼ 0.036; Table 1, column D). Patientswere followed for amedian of 31months (range, 27–43months).OS of the ipilimumab group was 45.6% at 12 months, 33.2% at18 months, and 23.5% at 24 months, with a median OS of 10.1months [95% confidence interval (CI), 8.0–13.8]. OS of thegp100 group was 25.3% at 12 months, 16.3% at 18 months, and13.7% at 24 months with a median OS of 6.4 months (95% CI,5.5–8.7). Analysis of soluble immunomodulatory proteins wasperformed on serum collected prior to treatment. BaselineCXCL11 concentrations were comparable between ipilimumab(median, 38; range, 2–1,027 pg/mL) and gp100 (median, 39;

Immunoregulatory Proteins Predict Outcome of Ipilimumab Treatment

www.aacrjournals.org Cancer Res; 75(23) December 1, 2015 5085

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

range, 2–911 pg/mL) groups. Similarly, baseline levels of sMICAwere consistent between ipilimumab (median, 115; range, 13–1,573 pg/mL) and gp100 (median, 121; range, 13–2,074 pg/mL)groups.

CXCL11, sMICA, and OSUnivariate analysis of ipilimumab-treated patients showed that

a 10-fold increase in CXCL11 was associated with double the riskof death (HR, 2.08; 95% CI, 1.40–3.11; P ¼ 0.0003; Fig. 1),whereas CXCL11 was not associated with OS in the gp100 group(HR, 1.21; 95%CI, 0.87–1.68; P¼ 0.2597). The effect of CXCL11onOS was significantly different for the ipilimumab group versusthe gp100 group (P ¼ 0.040). In the univariate analysis of log10sMICA, higher sMICA was associated with decreased survival inthe ipilimumab group [log10 sMICA quadratic effect P < 0.0001;sMICA (�247 vs. <247): HR, 3.46; 95% CI, 2.16–5.56 with P <0.0001] but not in the gp100 group (log10 sMICA HR, 0.91; 95%CI, 0.61–1.36; P¼ 0.6373). Elevated sMICB, LDH, tumor burden,and sCD25 were all associated with poorer survival regardless oftreatment (Fig. 1). Elevated VEGF was also associated withdecreased survival in both groups, although marginally so forthe ipilimumab-treated group (Fig. 1). Higher numbers of lym-phocytes (ALC) at baseline were associatedwith better OS in bothtreatment groups (Fig. 1). These univariate analyses suggest thatCXCL11 and sMICA are potential predictors ofOS in ipilimumab-treated melanoma patients, whereas sMICB, sCD25, VEGF, LDH,tumorburden, andALC representputative prognostic biomarkers.

Multivariate analyseswere also conducted focusing onCXCL11and sMICA, as these two biomarkers were identified in theunivariate analysis as correlating with ipilimumab but not gp100treatment. Models were used to test the independent effects ofCXCL11 and sMICA after adjusting for each other and the cov-ariates LDH, ALC, tumor burden, age, sex, and ECOG status.

Within the ipilimumab-treated group CXCL11 and LDH, but nottumor burden, ALC, age, sex, or ECOG score, were associatedwithOS (log10 CXCL11 HR, 1.88; 95% CI; 1.14–3.12; P¼ 0.014: LDHHR, 2.99; 95%CI, 1.78–5.02;P<0.0001; Fig. 2A). sMICAwas alsoassociated with OS [log10 sMICA quadratic effect P ¼ 0.0659;sMICA (�247 vs. <247): HR, 1.75; 95% CI, 1.02–3.01 with P ¼0.0420] but less strongly than CXCL11 (Fig. 2A). In the gp100-treated group, only LDH was independently associated with OS(LDH: HR, 2.24; 95% CI, 1.36–3.69; P ¼ 0.0016; Fig. 2B). Thesemultivariate results again suggest that CXCL11 and sMICA arepotential predictive biomarkers of OS in ipilimumab-treatedmelanoma patients, whereas LDH represents a prognostic bio-marker for patients with melanoma irrespective of treatment.

CXCL11 and sMICA in an independent ipilimumab-treatedcohort

We analyzed sera from patients with melanoma (48 of 52,92.3%) collected prior to treatment with ipilimumab in anexpanded access program at our institution (EACRI cohort).When comparing patient characteristics between the ipilimu-mab-treated phase III trial cohort and the EACRI cohort, we foundmore patients in the EACRI cohort with elevated LDH (P ¼0.0007), prior IL2 therapy (P < 0.0001), and prior immunother-apy (P < 0.0001; Table 1, column A, C, and E). This comparisonsuggests that the EACRI cohort included more patients withadvanced disease and poorer prognosis. This discrepancy mayaccount for shorter median survival of the EACRI cohort (8.6months) relative to that of ipilimumab-treated phase III trialcohort (10.1 months). In the EACRI cohort, median follow-upwas 39 months (range, 0.8–40 months).

Univariate analyses showed that elevated pretreatment con-centrations of CXCL11, sCD25, and LDHwere associated with anincreased risk of death (Fig. 3). Similar to phase III study findings,

Table 1. Baseline patient characteristics

A B C D EPhase III trial cohort EACRI cohort Statistics

Demographic orclinical characteristic

Ipilimumabmonotherapy (n ¼ 124)

gp100 monotherapy(n ¼ 123)

Ipilimumabmonotherapy (n ¼ 48)

P: Avs. Ba

P: Avs. Ca

Age, y 0.99 0.51Median 57 57 60Range 23–90 19–88 36–81

Male, % 61.3 54.5 60.4 0.28 0.92ECOG, % 0.85d 0.22d

0 51.6 52.9 41.71 47.6 43.9 52.12 0.8 3.2 6.3

LDH > ULN, % 37.1 36.4 66.0 0.91 0.0007ALC, �109/L 0.27 0.61Median 1.3 1.2 1.3Range 0.4–3.3 0.3–2.8 0.3–4.1

Prior IL2 therapy, % 23.4 25.2 60.4 0.74 <0.0001Prior immunotherapy, %b 39.5 52.9 77.1 0.036 <0.0001CXCL11, pg/mL 0.51 0.0003Median 38 39 14Range 2–1027 2–911 3–153

sMICA, pg/mL 0.99 <0.0001Median 115 121 299Range 13–1573 13–2074 12–2456

Median survival, moc 10.1 6.9 8.6aP values shown from the t test for age, Wilcoxon log-rank test for CXCL11 and sMICA, and c2 tests for all other variables.bIncluding IL2.cMedian survivals times calculated from Kaplan–Meier estimates.dECOG 1–2 versus 0.

Koguchi et al.

Cancer Res; 75(23) December 1, 2015 Cancer Research5086

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

a 10-fold increase in CXCL11 was associated with a 3.7-foldincrease in the risk of death (HR, 3.74; 95% CI, 1.71–8.22; P ¼0.0010). sMICA and VEGF effects were nonlinear, depicted by thethreshold effect as seen in the Kaplan–Meier plots of survival(Supplementary Fig. S2). Elevated sMICAwas also associatedwithincreased risk of death [log10 sMICA quadratic effect: P¼ 0.0244;sMICA (�247 vs. <247): HR, 2.06; 95% CI, 1.06–4.00 with P ¼0.0324; Fig. 3]. Elevated VEGF was associated with decreasedsurvival and sMICB and ALC were not associated with survival.

Multivariate analysis showed that CXCL11 and LDH wereassociated with OS (log10 CXCL11: HR, 3.18; 95% CI, 1.13–8.95; P ¼ 0.0288 and LDH: HR, 2.24; 95% CI, 1.02–4.95; P ¼

0.0457) after controlling for each other, gender, age, ECOG status,and prior immunotherapy (Fig. 4). sMICAmay be associatedwithOS in this cohort [log10 sMICA quadratic effect: P ¼ 0.1589;sMICA (�247 vs. <247): HR, 1.48; 95% CI, 0.67–3.27 with P ¼0.3284; Fig. 4], a result due in part to adjusting for CXCL11 andLDH and somewhat small cohort size. Thus, we confirmed thepredictive association between CXCL11 and OS in ipilimumab-treated melanoma patients but found a weaker associationbetween sMICA and OS in the EACRI cohort. We also confirmedthe association between LDH and OS in the EACRI cohort,compatible with the notion that LDH is a prognostic marker forpatients with metastatic melanoma.

Log10 CXCL11

Biomarker Treatment HR (95% CI) P

Log10 CXCL11

sMICA (≥247 vs. <247)

Log10 sMICA

Log10 sMICB

Log10 sMICB

Log10 sCD25

Log10 sCD25

Log10 VEGF

Log10 ALC

Log10 ALC

Log10 tumor burden

Log10 tumor burden gp100

gp100

gp100

gp100

gp100

gp100

gp100

gp100

ipi 2.08 (1.40–3.11)

1.21 (0.87–1.68)

3.46 (2.16–5.56)*

0.91 (0.61–1.36)

5.56 (3.12–9.91)

2.86 (1.80–4.54)

3.24 (1.20–8.74)

7.75 (3.22–18.67)

1.66 (0.97–2.84)

2.82 (1.06–3.87)§

3.37 (2.19–5.19)

2.79 (1.88–4.16)

0.06 (0.02–0.22)

0.21 (0.09–0.51)

2.24 (1.56–3.21)

2.18 (1.48–3.21) <0.0001

0.01 0.1 1 10

High biomarker level better High biomarker level worse

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

0.0003

0.2597

0.6373

0.0203

0.0630

0.0006

ipi

ipi

ipi

ipi

ipi

ipi

ipi

VEGF (≥ 157 vs. <157)

LDH; > ULN vs. ≤ ULN

LDH; > ULN vs. ≤ ULN

Figure 1.Univariate analysis of biomarker effects on OS for patients from the phase III clinical trial. HR and CI for association with OS of patients treated with ipilimumab (ipi)or gp100. Cox proportional hazards regression was used for univariate analysis of biomarker effects on OS. HR is numerator versus denominator. Among124 patients analyzed, 35 were censored in ipilimumab-treated group. Among 123 patients analyzed, 13 were censored in gp100-treated group. Missing data: LDH,ALC, tumor burden 1–2 missing in gp100 group; CXCL11, sCD25, VEGF 5 missing in gp100 group, 11 missing in ipilimumab group. � , in quadratic effectsmodel of ipilimumab group, (log10 sMICA)2 P < 0.0001. x, in gp100 group, (log10 VEGF)2 P ¼ 0.0002.

Immunoregulatory Proteins Predict Outcome of Ipilimumab Treatment

www.aacrjournals.org Cancer Res; 75(23) December 1, 2015 5087

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Log10 CXCL11

Log10 tumor burden

Log10 ALC

Age

ECOG; 1-2 vs 0

Sex: male vs. female

Prior immunotherapy:yes vs. no

sMICA (≥247 vs. <247):

LDH; > ULN vs. ≤ ULN

Log10 CXCL11

Log10 sMICA

Log10 tumor burden

Log10 ALC

Age

ECOG; 1-2 vs 0

Sex: male vs. female

Prior immunotherapy:yes vs. no

1.29 (0.80–2.07)

1.51 (0.99–2.32)

1.00 (0.98–1.02)

1.21 (0.79–1.84)

1.45 (0.93–2.26)

0.46 (0.14–1.50)

2.24 (1.36–3.69)

0.94 (0.60–1.47)

1.01 (0.70–1.44)

0.9419

0.3731

0.0016

0.1966

0.1050

0.7731

0.9741

0.83 (0.50–1.40)

1.43 (0.84–2.43)

0.99 (0.97–1.01)

1.10 (0.65–1.86)

1.48 (0.95–2.32)

0.25 (0.06–1.08)

2.99 (1.78–5.02)

1.75 (1.02–3.01)*

1.88 (1.14–3.12)

0.4915

0.1918

0.1800

0.7242

0.0865

0.0639

<0.0001

0.0420

0.0141

0.0569

0.2956

LDH; > ULN vs. ≤ ULN

Biomarker HR (95% CI) P

BiomarkerA

1 10High biomarker level better High biomarker level worse

B

HR (95% CI) P

0.1 1 10High biomarker level better High biomarker level worse

0.1 1 10High biomarker level better High biomarker level worse

Figure 2.Multivariate analysis of biomarker effects onOS for patients from the phase III clinical trial.HR and CI for association of potential biomarkerwith OS of patients treated with ipilimumab (A) orgp100 (B). Cox proportional hazards regressionwas used for multivariate analysis of biomarkereffects on OS. Among the 113 total patientsanalyzed, 34 were censored in ipilimumab-treatedgroup. Among total 115 patients analyzed,13 were censored in gp100-treated group. � , inquadratic effects model of ipilimumab group,(log10 sMICA)2 P ¼ 0.0659.

Koguchi et al.

Cancer Res; 75(23) December 1, 2015 Cancer Research5088

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Kaplan–Meier survival curvesTo illustrate the effects of CXC11 and sMICA on OS in the

phase III trial, Kaplan–Meier survival plots are shown (Fig. 5)for the biomarker high or low groups on the basis of selectedcutoff points (the median of CXC11; 35 pg/mL, the 80thpercentile of sMICA; 247 pg/mL, Fig. 5A and B). Because ofthe quadratic association of sMICA with survival, the 80thpercentile was the cutoff point chosen on the basis of theapproximate threshold value seen in the sMICA quintile plot(Supplementary Fig. S1). Kaplan–Meier survival plots for theEACRI cohort are also shown using the same cutoff points asused for the phase III study data plot (Fig. 5C and D). Thedistribution of baseline CXCL11 was lower and the distributionof sMICA levels was higher in the confirmatory cohort (Table 1,column E). Nonetheless, both cutoff points successfully dichot-omize patients treated with ipilimumab into patients with pooror better OS.

DiscussionWe found that high baseline serum CXCL11 and sMICA were

associated with poor OS in patients with metastatic melanomatreated with ipilimumab but not in patients treated with a"control" gp100 vaccine. This association was validated in anindependent cohort of ipilimumab-treated melanoma patients,strongly suggesting that measurement of pretreatment serumCXCL11 and sMICA levels may identify patients most likely to

benefit from ipilimumab. Because a minority of patients withmelanoma benefit from ipilimumab, avoiding treatment of theserefractory patients would reduce exposure to inefficient therapy,eliminate their risks for adverse effects and lower overall costs oftherapy.

Potential predictors of responsiveness to ipilimumab treatmenthave been identified in previous reports. For instance, elevatedlevels of several candidate biomarkers [e.g., C-reactive protein(CRP; refs. 10, 11), erythrocyte sedimentation rate (ESR; ref. 12),LDH (10–14), S100 protein (12), sCD25 (15), and VEGF (16)]were thought to associate with reduced benefit following ipili-mumab treatment. In contrast, increased baseline ALC wereassociated with improved OS upon ipilimumab treatment (10–14, 17). Interestingly, each of these candidate biomarkers were atsome time reported as prognostic biomarkers for melanoma (18,19). We were fortunate to have serum samples from the phase IIIstudy comparing an ipilimumab-treated cohort with a controlcohort allowing us to differentiate predictive versus prognosticbiomarkers (20). We also used samples from a second indepen-dent cohort of ipilimumab-treated patients to validate our basicfindings, which together with multivariate analyses, diminishedthe possibility of coincidental influence from other covariates.These analyses have demonstrated the predictive nature ofCXCL11 and sMICA for patients treated with ipilimumab, andthe prognostic nature of sMICB, VEGF, sCD25, LDH, and ALC forall patients with melanoma independent of treatment withipilimumab.

0.1 1 10

High biomarker level better High biomarker level worse

Biomarker HR (95% CI) P

Log10 CXCL11 3.74 (1.71–8.22)

2.06 (1.06–4.00)*

1.31 (0.74–2.29)

6.68 (1.36–32.75)

2.36 (1.21–4.63)§

2.11 (1.04–4.28)

0.58 (0.16–2.07) 0.3989

0.0376

0.0121

0.0192

0.3533

0.0324

0.0010

Log10 sMICB

Log10 sCD25

Log10 ALC

VEGF(≥157 vs. <157)

LDH;> ULN vs. ≤ ULN)

sMICA(≥247 vs. <247)

Figure 3.Univariate analysis of the ipilimumab-treated EACRI cohort. HR and CI forassociation of potential biomarkerwith OS of patients treated withipilimumab. Cox proportional hazardsregression was used for univariateanalysis of biomarker effects on OS.HR is numerator versus denominator.Among 48 total patients analyzed, 8were censored. � , in quadraticeffects model, (log10 sMICA)2

P ¼ 0.0244. x, (log10 VEGF)2

P ¼ 0.0200.

Immunoregulatory Proteins Predict Outcome of Ipilimumab Treatment

www.aacrjournals.org Cancer Res; 75(23) December 1, 2015 5089

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Our data revealed a strong association between elevatedserum CXCL11 protein and reduced OS upon ipilimumabtreatment in patients with metastatic melanoma. CXCL11,along with CXCL9 and CXCL10, binds to CXCR3, a criticalchemokine receptor for directional migration of TH1 and cyto-toxic T cells (21). In contrast to our findings with serumCXCL11, tissue expression of CXCL11 mRNA correlates withT-cell infiltration into tumors and improved prognosis (22).Microarray analysis of mRNA expression in melanoma tissuesfrom ipilimumab-treated patients also associated baseline andposttreatment tissue expression of CXCL11 with presence of Tcells in tumor and favorable clinical responses (23). The dis-crepancy between these results and ours may be attributed tosample source (tissue vs. serum) and/or assay targets (mRNAvs. protein). Paired analysis of mRNAs and proteins in tissueand serum might provide evidence to resolve this discrepancy.As CXCL11 has distinct immunoregulatory functions, in con-trast to immunostimulatory functions mediated by CXCL9 andCXCL10 (24, 25), we prefer the following nonmutually exclu-sive explanations of how CXCL11, even in the presence ofCXCL9 and CXCL10, may limit T-cell effector function as apart of negative feedback loop: (i) disrupting chemokine gra-dients for directional migration of T cells (26), (ii) preventingCXCR3–CXCL9/10 interaction by promoting receptor internal-ization (24), (iii) suppressing T-cell responses through induc-tion and/or recruitment of regulatory T cells (25, 27), and (iv)promoting growth and metastasis of tumors expressing CXCR3(28–31).

Elevated levels of sMICA and sMICB have been reported inseveral types of malignant diseases and implicated in cancerimmunoevasion (32). While the cellular stress response in trans-formed cells induces expression of membrane-bound MICA/MICB, the cleavage of these molecules produces soluble formsof MICA and MICB capable of inhibiting interactions betweenmembrane-bound MICA/MICB and NKG2D, thus desensitizingthe activation signal through NKG2D in effector T cells andnatural killer (NK) cells (33). The cleavage of membrane-boundMICA/MICB is promoted by themetalloprotease ADAM10 and isenhanced within hypoxic tumor environments (34). As hypoxiaalso promotes expression of immune inhibitory molecules (e.g.,PD-L1 and LAG-3) and favors accumulation of regulatoryimmune cells (35), sMICA may be a biomarker that reflects thisimmunosuppressive tumor environment.

The greatest significance of these findings may ultimately be inthe identification of CXCL11 and sMICA as immunotherapeutictargets. Agents that inhibit the immunosuppressive activity ofCXCL11 and sMICA without preventing interaction of the recep-tors (CXCR3 and NKG2D, respectively) with immunopotentiat-ing ligands (CXCL9/CXCL10 and membrane-bound MICA,respectively)mayhave therapeutic activity in patientswith cancer,either alone or with other agents, including chemotherapeutics,radiation, or other immunotherapeutics. To explore this idea,future studies should address whether CXCL11 and sMICA direct-ly interfere with ipilimumab-enabled effector T cells or aremerelyelevated as a result of the immunosuppressive environment foundin patients refractory to ipilimumab therapy.

Biomarker HR (95% CI) P

Log10 CXCL11 3.18 (1.13–8.95) 0.0288

0.3284

0.0457

0.5918

0.9404

0.8977

0.9407

0.23821.73 (0.70–4.30)

0.97 (0.44–2.16)

1.00 (0.97–1.03)

0.97 (0.44–2.14)

0.67 (0.16–2.88)

2.24 (1.02–4.95)

1.48 (0.67–3.27)*

Log10 ALC

ECOG;1-2 vs. 0

Prior immunotherapy:yes vs. no

Age

Sex:male vs. female

LDH;> ULN vs. ≤ULN)

sMICA (≥247 vs. <247)

0.1 1 10High biomarker level better High biomarker level worse

Figure 4.Multivariate analysis of theipilimumab-treated EACRI cohort. HRand CI for association of potentialbiomarker with OS of patients treatedwith ipilimumab. Cox proportionalhazards regression was used formultivariate analysis of biomarkereffects on OS. HR is numeratorversus denominator. Among47 total patients analyzed, 8 werecensored. � , in quadratic effectsmodel, (log10 sMICA)2 P ¼ 0.1589.

Cancer Res; 75(23) December 1, 2015 Cancer Research5090

Koguchi et al.

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Disclosure of Potential Conflicts of InterestY. Koguchi, F.R. Bahjat, H.M.Hoen, andA.J. Korman have ownership interest

in a pending patent. C. Milburn is an employee of Bristol-Myers Squibb. F.R.Bahjat is the co-founder of NeurAlexo, VP Pharmalocalogy at Oncovir andreports receiving other commercial research support from Oncovir, BristolMyers Squibb and is also a consultant/advisory board member of Bristol MyersSquibb. A.J. Korman has ownership interest in Bristol-Myers Squibb. K.S. Bahjatreports receiving commercial research grant from Bristol-Myers Squibb. Nopotential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: Y. Koguchi, C. Milburn, A.J. Korman, K.S. BahjatDevelopment of methodology: C. Milburn, K.S. BahjatAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y. Koguchi, S.A. Bambina, R.K. Fuerstenberg,B.D. Curti, A.J. Korman, K.S. BahjatAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Y. Koguchi, H.M. Hoen, B.D. Curti, F.R. Bahjat,K.S. BahjatWriting, review, and/or revision of the manuscript: Y. Koguchi, H.M. Hoen,M.D. Rynning, R.K. Fuerstenberg, B.D. Curti, W.J. Urba, C. Milburn, F.R. Bahjat,A.J. Korman, K.S. Bahjat

Administrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): H.M. Hoen, K.S. BahjatStudy supervision: B.D. Curti, A.J. Korman, K.S. Bahjat

AcknowledgmentsThe authors thank Gwen Kramer for assistance with sample preparation,

Michael Gough, Marka Crittenden, and Will Redmond for helpful discussions,and the dedication and skill of the clinical research team at the ProvidenceCancer Center.

Grant SupportThese studies were supported by institutional funding from the Providence

Portland Medical Foundation and a sponsored-research grant from Bristol-Myers Squibb (BMS).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received August 19, 2015; revised September 16, 2015; accepted September23, 2015; published online December 1, 2015.

Figure 5.Kaplan–Meier curves for OS according to pretreatment CXCL11 or sMICA status. Curves for OS obtained by applying selected cut points for CXCL11 (A and C) andsMICA (B and D) to the phase III trial cohort (A and B) or the EACRI cohort (C and D). Numbers of subjects at risk at each 10-month interval are listed beloweach graph. The difference in treatment effect [ipilimumab (ipi) or gp100] according to CXCL11 or sMICA concentration is represented in the top right ofeach graph (log-rank).

www.aacrjournals.org Cancer Res; 75(23) December 1, 2015 5091

Immunoregulatory Proteins Predict Outcome of Ipilimumab Treatment

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

References1. Wolchok JD, Hodi FS, Weber JS, Allison JP, Urba WJ, Robert C, et al.

Development of ipilimumab: a novel immunotherapeutic approach forthe treatment of advanced melanoma. Ann N Y Acad Sci 2013;1291:1–13.

2. Rudd CE, Taylor A, Schneider H. CD28 and CTLA-4 coreceptor expressionand signal transduction. Immunol Rev 2009;229:12–26.

3. Simpson TR, Li F, Montalvo-Ortiz W, Sepulveda MA, Bergerhoff K, Arce F,et al. Fc-dependent depletion of tumor-infiltrating regulatory T cells co-defines the efficacy of anti-CTLA-4 therapy against melanoma. J Exp Med2013;210:1695–710.

4. Hodi FS,O'Day SJ,McDermott DF,Weber RW, Sosman JA,Haanen JB, et al.Improved survivalwith ipilimumab in patients withmetastaticmelanoma.N Engl J Med 2010;363:711–23.

5. Page DB, Postow MA, Callahan MK, Allison JP, Wolchok JD. Immunemodulation in cancer with antibodies. Annu Rev Med 2014;65:185–202.

6. Wolchok JD, Weber JS, Maio M, Neyns B, Harmankaya K, Chin K, et al.Four-year survival rates for patients with metastatic melanoma whoreceived ipilimumab in phase II clinical trials. Ann Oncol 2013;24:2174–80.

7. Sharma P, Allison JP. The future of immune checkpoint therapy. Science2015;348:56–61.

8. Curti BD, Urba WJ. Clinical deployment of antibodies for treatment ofmelanoma. Mol Immunol 2015;67:18–27.

9. Boyles AL, Harris SF, Rooney AA, Thayer KA. Forest plot viewer: a newgraphing tool. Epidemiology 2011;22:746–7.

10. Wilgenhof S, FourDuS, Vandenbroucke F, EveraertH, Salmon I, Li�enardD,et al. Single-center experience with ipilimumab in an expanded accessprogram for patients with pretreated advanced melanoma. J Immunother2013;36:215–22.

11. Simeone E, Gentilcore G, Giannarelli D, Grimaldi AM, Carac�o C, CurviettoM, et al. Immunological and biological changes during ipilimumabtreatment and their potential correlation with clinical response and sur-vival in patients with advancedmelanoma. Cancer Immunol Immunother2014;63:675–83.

12. Kelderman S, Heemskerk B, van Tinteren H, van den Brom RRH, HospersGAP, van den Eertwegh AJM, et al. Lactate dehydrogenase as a selectioncriterion for ipilimumab treatment in metastatic melanoma. CancerImmunol Immunother 2014;63:449–58.

13. DiGiacomoAM,Danielli R, Calabr�oL, Bertocci E,Nannicini C,GiannarelliD, et al. Ipilimumab experience in heavily pretreated patients with mel-anoma in an expanded access program at the University Hospital of Siena(Italy). Cancer Immunol Immunother 2010;60:467–77.

14. Delyon J, Mateus C, Lefeuvre D, Lanoy E, Zitvogel L, Chaput N, et al.Experience in daily practice with ipilimumab for the treatment ofpatients with metastatic melanoma: an early increase in lymphocyteand eosinophil counts is associated with improved survival. Ann Oncol2013;24:1697–703.

15. Hannani D, V�etizou M, Enot D, Rusakiewicz S, Chaput N, Klatzmann D,et al. Anticancer immunotherapy by CTLA-4 blockade: obligatory contri-bution of IL-2 receptors and negative prognostic impact of soluble CD25.Cell Res 2015;25:208–24.

16. Yuan J, Zhou J, Dong Z, Tandon S, Kuk D, Panageas KS, et al. Pretreatmentserum VEGF is associated with clinical response and overall survival inadvanced melanoma patients treated with ipilimumab. Cancer ImmunolRes 2014;2:127–32.

17. Ku GY, Yuan J, Page DB, Schroeder SEA, Panageas KS, Carvajal RD, et al.Single-institution experience with ipilimumab in advanced melanomapatients in the compassionate use setting. Cancer 2010;116:1767–75.

18. Mouawad R, Spano J-P, Khayat D. Old and new serological biomarkers inmelanoma: where we are in 2009. Melanoma Res 2010;20:67–76.

19. Rochet NM, Kottschade LA, Grotz TE, Porrata LF, Markovic SN. Theprognostic role of the preoperative absolute lymphocyte count and abso-lute monocyte count in patients with resected advanced melanoma. Am JClin Oncol 2015;38:252–8.

20. Simon R. Clinical trials for predictive medicine: new challenges andparadigms. Clin Trials 2010;7:516–24.

21. Van Raemdonck K, Van den Steen PE, Liekens S, Van Damme J, Struyf S.CXCR3 ligands in disease and therapy. Cytokine Growth Factor Rev2014;26:311–27.

22. Galon J, Angell HK, Bedognetti D,Marincola FM. The continuum of cancerimmunosurveillance: prognostic, predictive, and mechanistic signatures.Immunity 2013;39:11–26.

23. Ji R-R, Chasalow SD, Wang L, Hamid O, Schmidt H, Cogswell J, et al. Animmune-active tumor microenvironment favors clinical response to ipi-limumab. Cancer Immunol Immunother 2011;61:1019–31.

24. Sauty A, Colvin RA, Wagner L, Rochat S, Spertini F, Luster AD. CXCR3internalization following T cell-endothelial cell contact: preferential role ofIFN-inducible T cell chemoattractant (CXCL11). J Immunol 2001;167:7084–93.

25. Zohar Y, Wildbaum G, Novak R, Salzman AL, Thelen M, Alon R, et al.CXCL11-dependent induction of FOXP3-negative regulatory T cells sup-presses autoimmune encephalomyelitis. J Clin Invest 2014;124:2009–22.

26. Griffith JW, Sokol CL, Luster AD. Chemokines and chemokine receptors:positioning cells for host defense and immunity. Annu Rev Immunol2014;32:659–702.

27. KochMA, Tucker-Heard G, Perdue NR, Killebrew JR, Urdahl KB, CampbellDJ. The transcription factor T-bet controls regulatory T cell homeostasis andfunction during type 1 inflammation. Nat Immunol 2009;10:595–602.

28. Billottet C, Quemener C, Bikfalvi A. CXCR3, a double-edged sword intumor progression and angiogenesis. Biochim Biophys Acta 2013;1836:287–95.

29. Robledo MM, Bartolome RA, Longo N, Rodríguez-Frade JM, Mellado M,Longo I, et al. Expression of functional chemokine receptors CXCR3 andCXCR4 on human melanoma cells. J Biol Chem 2001;276:45098–105.

30. Monteagudo C, Martin JM, Jorda E, Llombart-Bosch A. CXCR3 chemokinereceptor immunoreactivity in primary cutaneous malignant melanoma:correlation with clinicopathological prognostic factors. J Clin Pathol2007;60:596–9.

31. Longo-Imedio MI, Longo N, Trevi~no I, L�azaro P, S�anchez-Mateos P.Clinical significance of CXCR3 and CXCR4 expression in primary mela-noma. Int J Cancer 2005;117:861–5.

32. Le Bert N, Gasser S. Advances in NKG2D ligand recognition and responsesby NK cells. Immunol Cell Biol 2014;92:230–6.

33. Raulet DH,Gasser S, Gowen BG, DengW, JungH. Regulation of ligands forthe NKG2D activating receptor. Annu Rev Immunol 2013;31:413–41.

34. Baginska J, Viry E, Paggetti J, Medves S, Berchem G, Moussay E, et al. Thecritical role of the tumor microenvironment in shaping natural killer cell-mediated anti-tumor immunity. Front Immun 2013;4:490.

35. Labiano S, Palazon A, Melero I. Immune response regulation in the tumormicroenvironment by hypoxia. Semin Oncol 2015;42:378–86.

Cancer Res; 75(23) December 1, 2015 Cancer Research5092

Koguchi et al.

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

2015;75:5084-5092. Cancer Res   Yoshinobu Koguchi, Helena M. Hoen, Shelly A. Bambina, et al.   of Metastatic Melanoma Patients Treated with IpilimumabSerum Immunoregulatory Proteins as Predictors of Overall Survival

  Updated version

  http://cancerres.aacrjournals.org/content/75/23/5084

Access the most recent version of this article at:

  Material

Supplementary

  http://cancerres.aacrjournals.org/content/suppl/2015/10/14/0008-5472.CAN-15-2303.DC1

Access the most recent supplemental material at:

  Cited articles

  http://cancerres.aacrjournals.org/content/75/23/5084.full#ref-list-1

This article cites 35 articles, 6 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/75/23/5084.full#related-urls

This article has been cited by 4 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

  [email protected]

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://cancerres.aacrjournals.org/content/75/23/5084To request permission to re-use all or part of this article, use this link

on January 18, 2021. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from