TRIAL DESIGN

This was a randomized, open-label, phase 3 trial of nivolumab plus ipilimumab followed by nivolumab monotherapy versus sunitinib monotherapy. Randomization (in a 1:1 ratio) was performed with a block size of 4 with stratification according to IMDC risk score (0 vs. 1 or 2 vs. 3 to 6) and geographic region (United States vs. Canada and Europe vs. the rest of the world).

Nivolumab and ipilimumab were administered intravenously at a dose of 3 mg per kilogram over a period of 60 minutes and 1 mg per kilogram over a period of 30 minutes, respectively, every 3 weeks for four doses (induction phase), followed by nivolumab monotherapy at a dose of 3 mg per kilogram every 2 weeks (maintenance phase). Sunitinib was administered at a dose of 50 mg orally once daily for 4 weeks of each 6-week cycle. No dose reductions were allowed for nivolumab or ipilimumab. Dose delays for adverse events were permitted in both groups. Patients treated with nivolumab plus ipilimumab had to discontinue both nivolumab and ipilimumab if they had a treatment-related adverse event during the induction phase that required discontinuation, and they could not continue on to nivolumab maintenance therapy. Detailed discontinuation criteria are shown in the Supplementary Appendix, available with the full text of this article at NEJM.org.

A November 2017 protocol amendment, after the primary end point had been met, permitted crossover from the sunitinib group to the nivolumab-plus-ipilimumab group. Nivolumab, ipilimumab, and sunitinib were provided by the sponsors, except when sunitinib was procured as a local commercial product in certain countries.

TRIAL OVERSIGHT

This trial was approved by the institutional review board or ethics committee at each site and was conducted according to Good Clinical Practice guidelines, defined by the International Conference on Harmonisation. All the patients provided written informed consent that was based on the Declaration of Helsinki principles. A data and safety monitoring committee reviewed efficacy and safety. The trial was designed by the authors in collaboration with the sponsors (Bristol-Myers Squibb and Ono Pharmaceutical). Bristol-Myers Squibb collected and analyzed the data with the authors. A data confidentiality agreement was in place between Bristol-Myers Squibb and the investigators. The authors vouch for the completeness and accuracy of the data and analyses and for the adherence of the trial to the protocol (available at NEJM.org). The development of the first draft of the manuscript was led by the first author and the last three authors; all the authors contributed to drafting the manuscript and provided final approval to submit the manuscript for publication. Medical writing support, funded by Bristol-Myers Squibb, was provided by PPSI.

END POINTS AND ASSESSMENTS

The coprimary end points were the objective response rate, progression-free survival, and overall survival among intermediate- and poor-risk patients. The objective response rate was defined as the percentage of patients having a confirmed best response of complete response or partial response according to RECIST, version 1.1, on the basis of assessment by an independent radiology review committee. Progression-free survival was defined as the time from randomization to first RECIST-defined progression or death. Overall survival was defined as the time from randomization to death.

Secondary end points included the objective response rate, progression-free survival, and overall survival, all in the intention-to-treat population; and the incidence rate of adverse events among all treated patients. Exploratory end points included the objective response rate, progression-free survival, and overall survival, all among favorable-risk patients. Additional exploratory end points included outcomes according to the level of tumor programmed death ligand 1 (PD-L1) expression (≥1% vs. <1%), as assessed at a central laboratory with the use of the Dako PD-L1 IHC 28-8 pharmDx test, and health-related quality of life on the basis of the score on the National Comprehensive Cancer Network Functional Assessment of Cancer Therapy–Kidney Symptom Index (FKSI-19) (see the Supplementary Appendix), both in intermediate- and poor-risk patients.^17,18 FKSI-19 scores range from 0 to 76, with higher scores indicating fewer symptoms.

Disease assessments were performed with computed tomography or magnetic resonance imaging at baseline, 12 weeks after randomization, continuing every 6 weeks for the first 13 months, and then every 12 weeks until progression or treatment discontinuation. After progression or treatment discontinuation, patients were followed for safety and survival. Adverse events were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0.¹⁹ Patients in both groups were allowed to continue therapy after initial investigator-assessed, RECIST-defined progression if they had clinical benefit without disabling toxic effects. Patients discontinued trial therapy on evidence of further progression, defined as an additional 10% or greater increase in tumor burden volume from the time of initial progression (including all target lesions and new measurable lesions) according to investigator assessment.

EFFICACY

Coprimary End Points in Intermediate- and Poor-Risk Patients

Nivolumab plus ipilimumab had a significant overall survival benefit over sunitinib; the 12-month overall survival rate was 80% (95% confidence interval [CI], 76 to 84) with nivolumab plus ipilimumab versus 72% (95% CI, 67 to 76) with sunitinib, and the 18-month overall survival rate was 75% (95% CI, 70 to 78) versus 60% (95% CI, 55 to 65) (hazard ratio for death, 0.63; 99.8% CI, 0.44 to 0.89; P<0.001). The median overall survival was not reached (95% CI, 28.2 months to not estimable) with nivolumab plus ipilimumab versus 26.0 months (95% CI, 22.1 to not estimable) with sunitinib (Fig. 1A).

Open in a separate window

Figure 1

Overall Survival and Progression-free Survival among IMDC Intermediate- and Poor-Risk Patients

Progression was defined according to the Response Evaluation Criteria in Solid Tumors, version 1.1. For progression-free survival, the between-group difference did not meet the prespecified threshold (P = 0.009) for statistical significance. IMDC denotes International Metastatic Renal Cell Carcinoma Database Consortium, NE not estimable, and NR not reached.

The coprimary end point of objective response rate was 42% (95% CI, 37 to 47) with nivolumab plus ipilimumab versus 27% (95% CI, 22 to 31) with sunitinib (P<0.001), with complete responses in 40 patients (9%) versus 5 patients (1%) (Table 2). Of all intermediate- and poor-risk patients, 81% of those treated with nivolumab plus ipilimumab and 70% of those treated with sunitinib had a duration of response of at least 1 year, and the median duration of response was not reached (95% CI, 21.8 months to not estimable) and 18.2 months (95% CI, 14.8 to not estimable), respectively (Table 2, and Fig. S2 in the Supplementary Appendix). Rates of investigator-assessed objective response were consistent with rates of independently assessed objective response (Table S1 in the Supplementary Appendix).

Table 2

Antitumor Activity in IMDC Intermediate- and Poor-Risk Patients.^*

Variable	Nivolumab plus Ipilimumab (N = 425)	Sunitinib (N = 422)
Confirmed objective response rate — % (95% CI)†	42 (37-47)‡	27 (22-31)‡
Confirmed best overall response — no. (%)†
Complete response	40 (9)‡§	5 (1)‡§
Partial response	137 (32)	107 (25)
Stable disease	133 (31)	188 (45)
Progressive disease	83 (20)	72 (17)
Unable to determine or not reported	32 (8)	50 (12)
Median time to response (range) — mo	2.8 (0.9-11.3)	3.0 (0.6-15.0)
Median duration of response (95% CI) — mo	NR (21.8-NE)	18.2 (14.8-NE)
Patients with ongoing response — no./total no. (%)	128/177 (72)	71/112 (63)

Open in a separate window

^*NE denotes not estimable, and NR not reached.

^†Response was assessed according to Response Evaluation Criteria in Solid Tumors, version 1.1, by an independent radiology review committee.

^‡P<0.001 for the difference between groups.

^§The analysis of the between-group difference in complete response was exploratory.

For the coprimary end point of progression-free survival, the median was 11.6 months (95% CI, 8.7 to 15.5) with nivolumab plus ipilimumab and 8.4 months (95% CI, 7.0 to 10.8) with sunitinib (Fig. 1B). The between-group difference did not meet the prespecified threshold (P = 0.009) for statistical significance (hazard ratio for disease progression or death, 0.82; 99.1% CI, 0.64 to 1.05; P = 0.03).

Overall survival favored nivolumab plus ipi limumab over sunitinib across subgroups (Fig. 2). Similarly, the objective response rate was higher with nivolumab plus ipilimumab than with sunitinib in all subgroups (Fig. S3 in the Supplementary Appendix).

Open in a separate window

Figure 2

Subgroup Analysis of Overall Survival among IMDC Intermediate- and Poor-Risk Patients

Patients with intermediate risk had an IMDC score of 1 or 2, and those with poor risk had a score of 3 to 6. IMDC risk scores are defined by the number of the following risk factors present: a Karnofsky performance-status score of 70 (on a scale from 0 to 100, with lower scores indicating greater disability; patients with a performance-status score of <70 were excluded from the trial), a time from initial diagnosis to randomization of less than 1 year, a hemoglobin level below the lower limit of the normal range, a corrected serum calcium concentration of more than 10 mg per deciliter (2.5 mmol per liter), an absolute neutrophil count above the upper limit of the normal range, and a platelet count above the upper limit of the normal range. Bone, liver, lung, and lymph-node metastases were not protocol-prespecified subgroups. PD-L1 denotes programmed death ligand 1.

EXPOSURE AND SAFETY

The median duration of treatment in all patients who received a trial drug was 7.9 months (95% CI, 6.5 to 8.4) with nivolumab plus ipilimumab and 7.8 months (95% CI, 6.4 to 8.5) with sunitinib. A total of 79% of the patients received all four doses of ipilimumab with nivolumab. Among the 547 patients treated with nivolumab plus ipilimumab, nivolumab dose delays occurred in 319 (58%), and ipilimumab dose delays occurred in 148 (27%). Among the 535 patients treated with sunitinib, dose delays occurred in 315 (59%), and dose reductions occurred in 283 (53%). A total of 157 of 550 patients (29%) in the nivolumab-plus-ipilimumab group and 129 of 546 patients (24%) in the sunitinib group were treated beyond initial investigator-assessed, RECIST-defined progression, as permitted according to the protocol.

Treatment-related adverse events of any grade occurred in 509 of 547 patients (93%) treated with nivolumab plus ipilimumab and 521 of 535 patients (97%) treated with sunitinib (Table 3). Grade 3 or 4 events occurred in 250 patients (46%) and 335 patients (63%) in the respective groups. Treatment-related adverse events leading to discontinuation occurred in 118 of 547 patients (22%) in the nivolumab-plus-ipilimumab group and 63 of 535 patients (12%) in the sunitinib group. Eight deaths in the nivolumab-plus-ipilimumab group and four deaths in the sunitinib group were reported to be treatment-related (Table 3). Of the 436 patients treated with nivolumab plus ipilimumab who had a treatment-related select (immune-mediated) adverse event (includes skin, endocrine, gastrointestinal, pulmonary, hepatic, and renal categories), 152 (35%) received high-dose glucocorticoids (≥40 mg of prednisone per day or equivalent).

QUALITY OF LIFE

The rate of completion of the FKSI-19 questionnaire exceeded 80% in both groups during the first 6 months. The mean baseline FKSI-19 score (a quality-of-life metric) was similar in the two groups among patients with intermediate or poor risk (60.1 for nivolumab plus ipilimumab and 59.1 for sunitinib); the mean change from baseline was greater in the nivolumab-plus-ipilimumab group than in the sunitinib group at each assessment during the first 6 months (P<0.001) (Fig. 3). The pattern-mixture model and the mixed-model repeated-measures approach indicated a significant difference in favor of nivolumab plus ipilimumab, which substantiated the descriptive results (not shown).

Open in a separate window

Figure 3

Health-Related Quality of Life in IMDC Intermediate- and Poor-Risk Patients

Scores on the National Comprehensive Cancer Network Functional Assessment of Cancer Therapy–Kidney Symptom Index (FKSI-19) range from 0 to 76, with higher scores indicating fewer symptoms. Only time points for which data were available for five or more patients are shown. The number at risk shows the number of randomly assigned patients who were in the trial at each respective time point. I bars indicate standard errors.

Supported by Bristol-Myers Squibb and Ono Pharmaceutical. The authors received no financial support or compensation for publication of this manuscript. Patients treated at the Memorial Sloan Kettering Cancer Center were supported in part by a Memorial Sloan Kettering Cancer Center Support Grant/Core Grant (P30 CA008748).

Dr. Motzer reports receiving grant support and consulting fees from Pfizer, Novartis, Eisai, and Exelixis and grant support from Genentech/Roche; Dr. Tannir, receiving grant support, consulting fees, and honoraria from Bristol-Myers Squibb, receiving grant support from Epizyme and Mirati Therapeutics, receiving grant support, consulting fees, and honoraria from and serving on an advisory board for Exelixis and Novartis, receiving consulting fees and honoraria from Argos Therapeutics and Pfizer, and receiving consulting fees and honoraria from and serving on an advisory board for Eisai, Nektar, and Oncorena; Dr. McDermott, receiving honoraria and consulting fees from Bristol-Myers Squibb, Pfizer, Merck, Novartis, Eisai, Exelixis, Array BioPharma, and Genentech BioOncology and grant support from Prometheus Laboratories; Dr. Melichar, receiving advisory board fees and lecture fees from Bristol-Myers Squibb, Merck, Merck Sharp & Dohme, Roche, Novartis, Astellas Pharma, Pfizer, and Bayer; Dr. Choueiri, receiving grant support, paid to his institution, consulting fees, and advisory fees from AstraZeneca, Bristol-Myers Squibb, Exelixis, Genentech, Merck, Novartis, Peloton Therapeutics, Pfizer, Roche, and Eisai, grant support, paid to his institution, from Tracon Pharmaceuticals, and consulting fees and advisory fees from Bayer, Cerulean Pharma, Foundation Medicine, GlaxoSmithKline, Prometheus Laboratories, and Corvus Pharmaceuticals; Dr. Plimack, receiving grant support, paid to her institution, from Acceleron Pharma, AstraZeneca, Aveo Pharmaceuticals, Bristol-Myers Squibb, Dendreon, Eli Lilly, GlaxoSmithKline, Merck, Peloton Therapeutics, and Pfizer, receiving consulting fees from Acceleron Pharma, Astellas Pharma, AstraZeneca, Bristol-Myers Squibb, Clovis Oncology, Dendreon, Eli Lilly, Exelixis, Genentech/Roche, Glaxo-SmithKline, Horizon Pharma, Inovio Pharmaceuticals, Merck, Novartis, Pfizer, and SynerGene Therapeutics, receiving fees for the development of educational presentations from Bristol- Myers Squibb, Merck, Novartis, and Roche, holding a pending patent (14/588,503) on methods for screening patients with muscle-invasive bladder cancer for neoadjuvant chemotherapy responsiveness, and holding a pending patent (15/226,474) on the combination of an immunomodulatory agent with PD-1 or PD-L1 checkpoint inhibitors in the treatment of cancer; Dr. Barthélémy, receiving advisory board fees from Bristol-Myers Squibb, Pfizer, Novartis, Roche, Ipsen, and Sanofi; Dr. Porta, receiving consulting fees and lecture fees from Bristol-Myers Squibb, Novartis, Ipsen, Eisai, and EUSA Pharma, grant support, consulting fees, and lecture fees from Pfizer, and consulting fees from Janssen; Dr. George, receiving consulting fees from AstraZeneca, Exelixis, and Janssen, grant support and consulting fees from Bayer, Bristol-Myers Squibb, Corvus Pharmaceuticals, Novartis, and Pfizer, and grant support from Celldex Therapeutics and Merck; Dr. Powles, receiving grant support and honoraria from AstraZeneca and Roche, grant support from Novartis, and honoraria from Bristol-Myers Squibb, Merck, and Pfizer; Dr. Donskov, receiving grant support from Pfizer and Novartis; Dr. Kollmannsberger, receiving advisory board fees and honoraria from Bristol-Myers Squibb and Pfizer; Dr. Gurney, receiving advisory board fees from Bristol-Myers Squibb, Merck Sharp & Dohme, Roche, and AstraZeneca and grant support and advisory board fees from Pfizer; Dr. Hawkins, receiving lecture fees from Novartis, Ipsen and Pfizer, lecture fees and travel support from Bristol-Myers Squibb, and royalties for a patent (WO 90/05144) on the phage antibody patent family collectively known as Winter II; Dr. Ravaud, receiving grant support, advisory board fees, and travel support from Pfizer, advisory board fees and travel support from Bristol-Myers Squibb, travel support from Novartis and Merck Sharp & Dohme, and advisory board fees from Roche; Dr. Grimm, receiving grant support, consulting fees, and lecture fees from Novartis and Bristol-Myers Squibb, consulting fees and lecture fees from Pfizer, Bayer HealthCare, and AstraZeneca, lecture fees from Astellas Pharma, Hexal, and Apogepha, and consulting fees from Intuitive Surgical, Sanofi-Aventis, Amgen, Merck Sharp & Dohme, and MedUpdate; Dr. Bracarda, receiving consulting fees, lecture fees, and travel support from Pfizer, Astellas Pharma, Bristol-Myers Squibb, Roche, and Janssen, travel support from Novartis and Bayer, consulting fees and travel support from Ipsen, and consulting fees from EUSA Pharma and Merck Sharp & Dohme; Dr. Barrios, receiving grant support and consulting fees from Novartis, Boehringer Ingelheim, GlaxoSmith-Kline, Roche, and Pfizer, grant support from Amgen, AstraZeneca, Taiho Pharmaceutical, Eli Lilly, Sanofi, Mylan, Merrimack Pharmaceuticals, Merck, AbbVie, Astellas Pharma, BioMarin Pharmaceutical, Bristol-Myers Squibb, Daiichi Sankyo, Abraxis BioScience, Asana BioSciences, AB Science, Medivation, Exelixis, ImClone, LEO Pharma, Millennium Pharmaceuticals, and consulting fees from Eisai; Dr. Tomita, receiving grant support and honoraria from Astellas Pharma and Pfizer, grant support from AstraZeneca, honoraria from Bristol-Myers Squibb, honoraria and consulting fees from Novartis, grant support, honoraria, and consulting fees from Ono Pharmaceutical, and consulting fees from Taiho Pharmaceutical; Dr. Rini, receiving grant support, paid to his institution, and consulting fees from Bristol-Myers Squibb; Dr. Chen, being employed by Bristol-Myers Squibb; Drs. Mekan, McHenry, and Wind-Rotolo and Mr. Doan, being employed by and owning stock in Bristol-Myers Squibb; Dr. Sharma, owning stock in Jounce Therapeutics, Kite Pharma, Constellation Pharmaceuticals, and Neon Therapeutics and receiving consulting fees from Merck and Astellas Pharma; and Dr. Escudier, receiving honoraria from Bayer, Novartis, Pfizer, Exelixis, and Roche.

We thank the patients and their families for making this trial possible; Paul Gagnier, the initial medical monitor; Jennifer Mc-Carthy, the CheckMate 214 protocol manager; the staff at Dako for collaborative development of the PD-L1 IHC 28-8 pharmDx assay; and Jennifer Granit and Lawrence Hargett of PPSI (a Parexel company) for professional medical writing and editorial assistance with an earlier version of the manuscript.