Efficacy and Safety of Avelumab Treatment in Patients With Advanced Unresectable Mesothelioma: Phase 1b Results From the JAVELIN Solid Tumor Trial

Key Points

Question  Does avelumab have antitumor activity in patients with previously treated, unresectable malignant mesothelioma?

Findings  In this phase 1b study, 53 patients with unresectable mesothelioma and previous platinum and pemetrexed treatment received intravenous avelumab every 2 weeks. The confirmed objective response rate was 9%, with 17% of patients free of disease progression at 12 months; 9% of patients experienced grade 3 or higher treatment-related adverse events.

Meaning  Avelumab has clinical activity and acceptable safety in patients with previously treated, unresectable malignant mesothelioma.

Importance  Patients with malignant mesothelioma whose disease has progressed after platinum and pemetrexed treatment have limited options. Anti–programmed cell death 1 (PD-1) antibodies have antitumor activity in this disease, but little is known about the activity of anti–programmed cell death ligand 1 (PD-L1) antibodies in patients with mesothelioma.

Objective  To assess the efficacy and safety of avelumab in a cohort of patients with previously treated mesothelioma.

Design, Setting, and Participants  Phase 1b open-label study (JAVELIN Solid Tumor) in patients with unresectable mesothelioma that progressed after platinum and pemetrexed treatment, enrolled at 25 sites in 3 countries between September 9, 2014, and July 22, 2015.

Interventions  Participants received avelumab, 10 mg/kg, every 2 weeks until disease progression, unacceptable toxic effects, or withdrawal from the study.

Main Outcomes and Measures  Prespecified end points included confirmed best overall response based on Response Evaluation Criteria In Solid Tumors, version 1.1; duration of response; progression-free survival (PFS); overall survival (OS); PD-L1 expression–based analyses; and safety.

Results  Of 53 patients treated with avelumab, the median age was 67 (range, 32-84) years; 32 (60%) were male. As of December 31, 2016, median follow-up was 24.8 (range, 16.8-27.8) months. Twenty patients (38%) had 3 or more previous lines of therapy (median, 2; range, 1-8). The confirmed objective response rate (ORR) was 9% (5 patients; 95% CI, 3.1%-20.7%), with complete response in 1 patient and partial response in 4 patients. Responses were durable (median, 15.2 months; 95% CI, 11.1 to not estimable months) and occurred in patients with PD-L1–positive tumors (3 of 16; ORR, 19%; 95% CI, 4.0%-45.6%) and PD-L1–negative tumors (2 of 27; ORR, 7%; 95% CI, 0.9%-24.3%) based on a 5% or greater PD-L1 cutoff. Disease control rate was 58% (31 patients). Median PFS was 4.1 (95% CI, 1.4-6.2) months, and the 12-month PFS rate was 17.4% (95% CI, 7.7%-30.4%). Median OS was 10.7 (95% CI, 6.4-20.2) months, and the median 12-month OS rate was 43.8% (95% CI, 29.8%-57.0%). Five patients (9%) had a grade 3 or 4 treatment-related adverse event, and 3 (6%) had a grade 3 or 4 immune-related, treatment-related adverse event. There were no treatment-related deaths.

Conclusions and Relevance  Avelumab showed durable antitumor activity and disease control with an acceptable safety profile in a heavily pretreated cohort of patients with mesothelioma.

Trial Registration  ClinicalTrials.gov identifier: NCT01772004

Malignant mesothelioma is a rare cancer with a poor prognosis, estimated to cause 43 000 deaths worldwide each year.¹ Pleural mesothelioma is the most common form and accounts for 80% to 85% of cases, peritoneal mesothelioma accounts for 10% to 15% of cases, and other sites account for less than 5%.² Among histologic subtypes, epithelial mesotheliomas are most common and are associated with longer life expectancy than less common sarcomatoid or biphasic (mixed) mesotheliomas.^3,4

For patients with unresectable malignant pleural mesothelioma, treatment with cisplatin plus pemetrexed is the current first-line standard of care.⁴ In a randomized phase 3 trial, median overall survival (OS) among patients who received cisplatin plus pemetrexed vs cisplatin alone was 12.1 months (95% CI, 10.0-14.4) vs 9.3 months (95% CI, 7.8-10.7) (P = .02), respectively.⁵ Data for second-line treatment are limited, and there is no US Food and Drug Administration–approved treatment in this setting. Second-line chemotherapy options listed in guidelines include gemcitabine, vinorelbine, and pemetrexed,⁴ and median OS for these agents administered as monotherapy or combination therapy across various studies has ranged from 4.1 to 10.9 months.^6-14 Thus, clinical studies of new treatment options are needed.

Antibodies targeting programmed cell death 1 (PD-1) or its ligand, programmed cell death ligand 1 (PD-L1), induce durable remissions in various tumor types.¹⁵ Programmed cell death ligand 1 is expressed by approximately 40% to 60% of mesotheliomas.^16-18 The anti–PD-1 antibodies pembrolizumab and nivolumab (with or without ipilimumab) have been added to National Comprehensive Cancer Network guidelines as second-line treatment options for malignant pleural mesothelioma.⁴ In a phase 1b trial of pembrolizumab in 25 patients with PD-L1–positive tumors in whom standard therapy had failed or who were ineligible for standard therapy, 2 (8%) received pembrolizumab as first-line treatment, 15 (60%) received pembrolizumab as second-line treatment, and 8 (32%) received pembrolizumab as later-line treatment. The objective response rate (ORR) was 20%, median progression-free survival (PFS) was 5.4 months, and OS was 18.0 months.¹⁹ In a randomized phase 2 study of 125 patients treated with nivolumab alone or in combination with ipilimumab as second- or third-line therapy, the ORR was 19% for nivolumab alone and 28% for nivolumab in combination with ipilimumab, median PFS was 4.0 months (95% CI, 2.8-5.7) for nivolumab alone and 5.6 months (95% CI, 3.2-8.4) for combination therapy, and median OS was 13.6 months (95% CI, 6.7–not reached) for nivolumab and not reached (95% CI, not reached) for combination therapy after a median follow-up of 15 months.²⁰

Avelumab is a human IgG1 monoclonal antibody that targets PD-L1.²¹ Unlike other available PD-1/PD-L1 antibodies, avelumab retains a wild-type Fc region that, in preclinical models, can induce antitumor activity via adaptive and innate effector cells, including against mesothelioma cells.^16,22 Avelumab has been approved in various countries for the treatment of metastatic Merkel cell carcinoma and in North America for locally advanced or metastatic urothelial carcinoma that progressed during or after platinum-containing chemotherapy.^23,24 The safety and efficacy of avelumab have been investigated in a large, multicohort phase 1 study: JAVELIN Solid Tumor (see the full trial protocol in Supplement 1). In the initial dose-escalation part, avelumab was administered at doses up to 20 mg every 2 weeks, and the maximum tolerated dose was not reached.²¹ Based on a combination of clinical, pharmacokinetic, and pharmacodynamic data,²¹ avelumab administered at a dosage of 10 mg/kg every 2 weeks was selected for further evaluation in expansion cohorts, including patients with non–small cell lung cancer, urothelial carcinoma, gastric cancer, and ovarian cancer.^25-28

Herein, we report phase 1b data from the JAVELIN Solid Tumor trial in the cohort of patients with previously treated, unresectable mesothelioma.

JAVELIN Solid Tumor is an ongoing international, phase 1, open-label trial assessing the clinical activity, pharmacokinetics, and safety profile of avelumab in patients with metastatic solid tumors. In this phase 1b dose-expansion cohort (Figure 1), eligible patients had unresectable pleural or peritoneal mesothelioma and progressive disease after platinum and pemetrexed treatment, administered either in combination or as monotherapy. Patients were not permitted to have received previous treatment with a T-cell–targeting immune checkpoint inhibitor (eg, ipilimumab, tremelimumab, or anti–PD-1 or anti–PD-L1 antibodies). For full eligibility criteria, see the eMethods in Supplement 2. The trial was conducted in compliance with the ethical principles of the Declaration of Helsinki²⁹ and the International Council on Harmonisation Good Clinical Practice Guidelines.³⁰ The protocol was approved by the institutional review board or independent ethics committee at each center (eMethods in Supplement 2). All patients provided written informed consent before their enrollment.

Avelumab (EMD Serono, a business of Merck KGaA) was administered intravenously over 60 minutes at 10 mg/kg every 2 weeks until disease progression, unacceptable toxic effects, or any other protocol-specified criterion for withdrawal occurred (eMethods in Supplement 2).²⁵ A premedication regimen of an antihistamine (diphenhydramine or equivalent) and acetaminophen was administered 30 to 60 minutes before all avelumab infusions. Dose modifications (reductions or increases) were not permitted.

Safety was assessed at each biweekly visit; assessments included adverse events (AEs), physical examination, clinical laboratory tests (hematology, hepatic panels, and serum chemistry), and documentation of concurrent medications. Adverse events and laboratory abnormalities were classified and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.03.³¹ Serious AEs were defined as untoward events that were life threatening, required hospitalization, and resulted in disability, congenital anomaly, or death. Immune-related AEs were identified using a prespecified list of Medical Dictionary for Regulatory Activities³² terms.

Clinical activity was assessed by investigators using Response Evaluation Criteria in Solid Tumors, version 1.1³³ to determine the best overall response. Radiographic tumor assessments were done at baseline and then every 6 weeks. For patients who had a partial response (PR) or complete response (CR), a confirmatory computed tomographic or magnetic resonance imaging scan was performed no sooner than 28 days after the first documented response. Programmed cell death ligand 1 expression was assessed using a proprietary immunohistochemistry assay (Dako PD-L1 IHC 73-10 pharmDx; Dako).²⁶ In this report, PD-L1–positive status was defined prospectively using a cutoff of 5% or greater expression on tumor cells; other PD-L1 cutoffs were also evaluated.^25,27

Primary end points for the JAVELIN Solid Tumor trial were dose-limiting toxic effects occurring within the first 3 weeks of treatment in the dose-escalation part (reported previously)²¹ and the confirmed best overall response adjudicated by an independent review committee in specified expansion cohorts (not including mesothelioma). Prespecified end points assessed in the current cohort included confirmed best overall response based on investigator assessment, immune-related best overall response, duration of response, PFS, OS, evaluation of PD-L1 expression, and safety.

A sample size of 50 patients was planned to provide point estimates and 95% Clopper-Pearson CIs (eg, for an ORR of 10%, a 95% CI of 3.3%-21.8%) in the case of 5 responders. Safety and efficacy were analyzed in all patients who received 1 dose or more of avelumab. Change in the sum of target lesion diameters from baseline was evaluated in patients with baseline tumor assessments and at least 1 postbaseline assessment. Time-to-event end points were estimated with the Kaplan-Meier method, and 95% CIs for the median were calculated using the Brookmeyer-Crowley method.³⁴P values for association between categorical variables were determined by using the Fisher exact test (2-sided). An exploratory landmark analysis of OS by using the Kaplan-Meier method was performed based on best response achieved by 3 months among patients remaining in the study. Statistical analyses were performed using SAS software (version 9.1.3 or higher, SAS Institute Inc) or R, (version 2.10.1 or higher, The R Foundation).

Between September 9, 2014, and July 22, 2015, 53 patients were enrolled at 25 sites in 3 countries and treated with avelumab (Figure 1). Median (range) age was 67 (32-84) years (Table 1). All patients had previously received systemic anticancer treatment, including a platinum agent and pemetrexed, with a median (range) of 2 (1-8) previous lines of systemic therapy; 35 patients (66%) had received 2 or more lines, and 20 patients (38%) had received 3 or more lines. Median (range) duration of treatment was 2.8 (0.9-28.1) months, and patients received a median (range) of 6 (2-59) doses of avelumab. Median (range) follow-up based on all enrolled patients, defined as the time between start of study treatment to analysis cutoff date (December 31, 2016), was 24.8 (16.8-27.8) months. At data cutoff, 4 patients (8%) continued receiving the study treatment. Reasons for treatment discontinuation were disease progression (29 patients [55%]), AEs (14 [26%]), death (2 [4%]; attributed to disease progression and unrelated pneumonia), withdrawal of consent (1 [2%]), and other (3 [6%]; referral to hospice, physician’s discretion, and treatment break after CR).

Of 53 patients, 1 (2%) had a confirmed CR (eFigure 1 in Supplement 2) and 4 (8%) had a confirmed PR, resulting in an ORR of 9% (95% CI, 3.1%-20.7%) (eTable 1 in Supplement 2). Response was ongoing in 3 patients, and the median (95% CI) duration of response was 15.2 (11.1–not estimable) months (eFigure 2A in Supplement 2). In addition, 26 patients (49%) had a best response of stable disease, resulting in a disease control rate of 58%. In patients with 1 previous line of therapy, the ORR was 6% (1 of 18; 95% CI, 0.1%-27.3%), those with 2 previous lines had an ORR of 13% (2 of 15; 95% CI, 1.7%-40.5%), and those with 3 previous lines of therapy had an ORR of 10% (2 of 20; 95% CI 1.2%-31.7%). Of 48 patients whose data were assessable for change in tumor size, 24 (50%) had a reduction of any level (eFigures 2B and 2C in Supplement 2). Median (95% CI) PFS was 4.1 (1.4-6.2) months, and the PFS rate at 6 months was 38.0% (95% CI, 24.2%-51.7%) and at 12 months was 17.4% (95% CI, 7.7%-30.4%) (Figure 2A). Median (95% CI) OS was 10.7 (6.4-20.2) months, and the 12-month OS rate was 43.8% (95% CI, 29.8%-57.0%) (Figure 2B). Median (95% CI) OS in patients with 1 previous line of therapy was 8.5 (4.2 to not estimable) months; in patients with 2 previous lines of therapy, 10.7 (3.6 to not estimable) months; and in patients with 3 or more previous lines of therapy, 11.2 (6.4-20.2) months (eFigure 3 in Supplement 2). Based on an exploratory landmark analysis of patients remaining on study at 3 months (46 patients), OS beyond 3 months was longer in patients who had an objective response (CR or PR) or stable disease compared with those who did not (eFigure 4 in Supplement 2).

Expression of PD-L1 was evaluable in 43 patients (81%) (Table 1). With use of a PD-L1 expression cutoff of 5% or greater to define PD-L1–positive or PD-L1–negative subgroups, ORRs were 19% (3 of 16 patients; 95% CI, 4.0%-45.6%) vs 7% (2 of 27 patients; 95% CI, 0.9%-24.3%) (P = .34). The median (95% CI) PFS in patients with PD-L1–positive or PD-L1–negative tumors was 5.3 (1.4-17.8) months vs 1.7 (1.4-8.3) months, and 6-month PFS rates (95% CI) were 37.5% (14.1%-61.2%) vs 42.0% (23.1%-59.8%). The median (95% CI) OS in patients with PD-L1–positive or PD-L1–negative tumors was 20.2 (4.9 to not estimable) months vs 10.2 (3.8-21.0) months, and 12-month OS rates (95% CI) were 72.5% (42.1%-88.8%) vs 40.7% (22.5%-58.2%) (eTable 2 in Supplement 2). Analyses using a cutoff of 1% or greater yielded similar ORR, PFS, and OS findings (eTable 2 and eFigure 5 in Supplement 2). Of 5 patients who had an objective response, 3 had PD-L1–positive tumors and 2 had PD-L1–negative tumors (based on both the 5% and 1% cutoffs). At higher PD-L1 cutoffs, data were not informative because too few patients had PD-L1–positive tumors (2 patients with 25% or greater cutoff, 5 with 50% or greater cutoff, and 8 with 80% or greater cutoff).

All 53 patients had an AE of any grade, of which 43 (81%) were treatment-related AEs (TRAEs) (Table 2). The most common TRAE was an infusion-related reaction, occurring in 19 patients (36%); all were grade 1 or 2 and occurred at the first or second infusion. Other TRAEs occurring in 10% or more of patients were chills (8 patients [15%]), fatigue (8 [15%]), and pyrexia (6 [11%]). Five patients (9%) had a grade 3 or greater TRAE: pneumonitis, colitis, blood creatine phosphokinase increase, hypoalbuminemia, and type 1 diabetes mellitus occurring in 1 patient each (2%). Twelve patients (23%) had an immune-related TRAE of any grade, and those occurring in more than 1 patient were pneumonitis (3 patients [6%]), colitis (2 [4%]), hyperthyroidism (2 [4%]), and hypothyroidism (2 [4%]); 3 patients (6%) had a grade 3 or greater immune-related TRAE (Table 2). Twenty-two patients (42%) had serious AEs, which were related to treatment in 5 patients: pneumonitis (2 patients), increased blood creatine phosphokinase (1), colitis (1), and type 1 diabetes mellitus (1). Four patients (8%) had an unrelated AE that led to death (disease progression, 3 patients; pneumonia, 1). Seventeen patients (32%) discontinued treatment after an AE, including 10 (19%) with a TRAE: infusion-related reaction (4 patients), colitis (2), increased blood creatine phosphokinase (1), hyperthyroidism (1), pneumonitis (1), and type 1 diabetes mellitus (1).

In this phase 1b cohort of patients with unresectable mesothelioma who were heavily pretreated, including receipt of previous platinum and pemetrexed treatment in all patients, avelumab showed durable clinical activity and long-term disease control. The ORR was 9%, including responses achieved irrespective of previous lines of therapy, and median duration of response was 15.2 months. Furthermore, the disease control rate was high (58%), with a 6-month PFS rate of 38.0%, a 12-month PFS rate of 17.4% (median, 4.1 months), and a 12-month OS rate of 43.8% (median, 10.7 months). Achievement of an objective response or stable disease by 3 months was associated with longer OS, suggesting that achieving stable disease is clinically meaningful in recurrent mesothelioma. In patients with PD-L1–positive tumors, the ORR was 19%, the 6-month PFS rate was 37.5% (median, 5.3 months), and the 12-month OS rate was 72.5% (median, 20.2 months). These findings are encouraging in a disease setting where treatment options are limited.

Although cross-trial comparisons should be interpreted with caution because of differences in patient characteristics and treatment histories, efficacy findings for avelumab in this heavily pretreated population appear at least comparable to second-line options listed in National Comprehensive Cancer Network guidelines.⁴ In an observational study of retreatment with pemetrexed-based chemotherapy in patients with response or stable disease lasting 3 months or longer with previous treatment, the median PFS was 3.8 months, the 12-month PFS rate was 8%, and the median OS was 10.5 months.⁶ In a single-center study of second- or third-line vinorelbine or gemcitabine monotherapy, the median PFS was 1.6 to 1.7 months and the median OS was 4.9 to 5.4 months.¹¹ In a single-center study of gemcitabine combined with vinorelbine in pretreated patients, the median time to progression was 2.8 months, the 6-month PFS rate was 31%, and the median OS was 10.9 months.¹² In the phase 2 study of second- or third-line nivolumab, the 12-month OS rate with nivolumab monotherapy was 51% (median, 13.6 months) and with the combination of nivolumab plus and ipilimumab was 58% (median not reached).²⁰ Finally, findings with avelumab in the PD-L1–positive subgroup are comparable to data reported for pembrolizumab in a phase 1b study performed exclusively in patients with PD-L1–positive tumors, which yielded a 6-month PFS rate of 46% (median, 5.4 months) and 12-month OS rate of 63% (median, 18.0 months).¹⁹

Avelumab showed an acceptable safety profile that compares favorably with chemotherapy. The incidence of grade 3 or 4 TRAEs was low (9%), with no treatment-related deaths. In studies of anti–PD-1 antibodies, grade 3 or 4 TRAEs occurred in 20% of patients treated with pembrolizumab, 13% of patients treated with nivolumab monotherapy, and 26% of patients treated with nivolumab plus ipilimumab, with 3 treatment-related deaths reported for the combination.^19,20 Although the incidence of an infusion-related reaction with avelumab (any grade in 36%) is high compared with other tumor types,^25,26 no patient in the mesothelioma cohort had a grade 3 or greater infusion-related reaction.

Findings from this study should be interpreted in the context of its early-phase design, particularly the small number of patients enrolled, which limits data robustness and reduces the potential for subgroup analyses. In addition, the single-arm, nonrandomized design prevents any direct comparison with alternative treatment options in this disease setting.

Avelumab showed clinically meaningful antitumor activity in a heavily pretreated population of patients with pleural or peritoneal mesothelioma. Although the study and its findings should be interpreted in context as a small phase 1b cohort, the level of long-term disease control, the duration of OS, and the safety profile suggest that avelumab could be a potential new therapeutic option for patients with mesothelioma. Further studies of avelumab or other anti–PD-1 or anti–PD-L1 antibodies in this disease, including combinations with other agents, are warranted.

Accepted for Publication: September 8, 2018.

Published Online: January 3, 2019. doi:10.1001/jamaoncol.2018.5428

Open Access: This article is published under the JN-OA license and is free to read on the day of publication.

Corresponding Author: Raffit Hassan, MD, Thoracic and GI Malignancies Branch, National Cancer Institute, National Institutes of Health, Building 37, Room 5116, Bethesda, MD 20892 (hassanr@mail.nih.gov).

Author Contributions: Dr Hassan had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Hassan, Thomas, von Heydebreck, Chin, Gulley.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Hassan, Nemunaitis, Patel, Dowlati, Vaishampayan, Chin, Gulley.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: von Heydebreck, Chin.

Obtained funding: Chin.

Administrative, technical, or material support: Hassan, Patel, Dowlati, von Heydebreck, Chin, Gulley.

Supervision: Hassan, Patel, Dowlati, Kochuparambil, Taylor, Vaishampayan, Verschraegen, Gulley.

Conflict of Interest Disclosures: Dr Hassan reported receiving research funding from Aduro Biotech, Bayer, and Morphotek. Dr Nemunaitis reported receiving consulting, advisory, and speaking fees from Amgen, AstraZeneca, and Symvivo; being a founder and member of the board of directors at Gradalis and serving as the Director of the phase 1 research program for US Oncology (USON); receiving compensation for expert testimony from Foundation Medicine; serving in a leadership role for Symvivo; receiving speaking fees from Amgen and AstraZeneca; receiving research funding, having stock or other ownership, patents, royalties, and other intellectual property from Gradalis; and receiving travel, accommodation, and expenses from Symvivo, Gradalis, Amgen, and AstraZeneca. Dr Patel reported receiving speaking fees from Celgene, Exelixis, Genentech/Roche, and Taiho Pharmaceutical. Dr Bennouna reported receiving consulting and advisory fees, honoraria, and speaking fees from AstraZeneca, Bristol-Meyers Squibb, Boehringer, Merck, and Roche; research funding from AstraZeneca and Merck KGaA; and travel, accommodation, and expenses from AstraZeneca and Roche. Dr Chen reported receiving honoraria from Celgene. Dr Dowlati reported receiving fees for consulting and advisory roles for AbbVie/Stemcentrx and ARIAD and research funding from Amgen, Bristol-Meyer Squibb, EMD Serono, Lilly/ImClone, MedImmune, and OncoMed. Dr Taylor reported receiving consulting and advisory fees and honoraria from Blueprint, Bristol-Myers Squibb, Eisai, Loxo, and Novartis; speaking fees from Bristol-Myers Squibb and Eisai; and travel, accommodation, and expenses from Blueprint, Bristol-Myers Squibb, Eisai, and Loxo. Dr Powderly reported receiving consulting and advisory fees from AstraZeneca/MedImmune, Bristol-Myers Squibb, Curis, Genentech/Roche, and TopAlliance BioSciences; serving in leadership roles for BioCytics and Carolina BioOncology Instititute; receiving speaking fees from Bristol-Myers Squibb, Dendreon, Genentech/Roche, and Merck; patents, royalties, and other intellectual property from Biocytics; stocks and other ownership interests in BioCytics, Bluebird Bio, Carolina BioOncology Institute, Juno Therapeutics, Kite Pharma, Lion Biotechnologies, and ZIOPHARM Oncology; and research funding from AbbVie, AstraZeneca, Bristol-Myers Squibb, Corvus Pharmaceuticals, Curis, EMD Serono, Genentech/Roche, Incyte, Lilly/ImClone, Macrogenics, Seattle Genetics, and TopAlliance BioSciences. Dr Vaishampayan reported receiving consulting and advisory fees from Astellas Pharma, Bayer, Bristol-Myers Squibb, Exelixis, Genentech/Roche, and Pfizer; honoraria from Bayer, Bristol-Myers Squibb, Exelixis, Genentech/Roche, Janssen, Novartis, Pfizer, and Sanofi; speaking fees from Bayer, Bristol-Myers Squibb, Exelixis, Genentech/Roche, Pfizer, and Sanofi; and research funding from Astellas Pharma, Bristol-Myers Squibb, Exelixis, Novartis, and Pfizer. Dr Gulley reported receiving research funding from Astellas Medivation, Bavarian Nordic, Celgene, EMD Serono, and Pfizer. No other disclosures were reported.

Funding/Support: This study was funded by Merck KGaA, Darmstadt, Germany, and is part of an alliance between Merck KGaA and Pfizer, Inc. Drs Hassan, Thomas, and Gulley received research funding from the National Cancer Institute Center for Cancer Research.

Role of the Funder/Sponsor: Merck KGaA, Darmstadt, Germany, led the design and conduct of the study, monitored it, and provided the study drug. Merck KGaA, with the authors, also participated in the collection, management, analysis, and interpretation of the data. Employees of Merck KGaA are authors of this manuscript and were involved in reviewing and approving the manuscript as well as the decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 3.

Additional Contributions: The authors thank the patients and their families, investigators, coinvestigators, and the study teams at each of the participating centers and at Merck KGaA (Darmstadt, Germany), EMD Serono Research & Development Institute (a business of Merck KGaA, Darmstadt, Germany), and Quintiles. Medical writing support was provided by ClinicalThinking and funded by Merck KGaA and Pfizer.