Adjuvant Therapy of Nivolumab Combined With Ipilimumab Versus Nivolumab Alone in Patients With Resected Stage IIIB-D or Stage IV Melanoma (CheckMate 915)

Abstract

PURPOSE Ipilimumab and nivolumab have each shown treatment benefit for high-risk resected melanoma. The phase III CheckMate 915 trial evaluated adjuvant nivolumab plus ipilimumab versus nivolumab alone in patients with resected stage IIIB-D or IV melanoma. PATIENTS AND METHODS In this randomized, double-blind, phase III trial, 1,833 patients received nivolumab 240 mg once every 2 weeks plus ipilimumab 1 mg/kg once every 6 weeks (916 patients) or nivolumab 480 mg once every 4 weeks (917 patients) for ≤ 1 year. After random assignment, patients were stratified by tumor programmed death ligand 1 (PD-L1) expression and stage. Dual primary end points were recurrence-free survival (RFS) in randomly assigned patients and in the tumor PD-L1 expression-level < 1% subgroup. RESULTS At a minimum follow-up of approximately 23.7 months, there was no significant difference between treatment groups for RFS in the all-randomly assigned patient population (hazard ratio, 0.92; 95% CI, 0.77 to 1.09; P = .269) or in patients with PD-L1 expression < 1% (hazard ratio, 0.91; 95% CI, 0.73 to 1.14). In all patients, 24-month RFS rates were 64.6% (combination) and 63.2% (nivolumab). Treatment-related grade 3 or 4 adverse events were reported in 32.6% of patients in the combination group and 12.8% in the nivolumab group. Treatment-related deaths were reported in 0.4% of patients in the combination group and in no nivolumab-treated patients. CONCLUSION Nivolumab 240 mg once every 2 weeks plus ipilimumab 1 mg/kg once every 6 weeks did not improve RFS versus nivolumab 480 mg once every 4 weeks in patients with stage IIIB-D or stage IV melanoma. Nivolumab showed efficacy consistent with previous adjuvant studies in a population resembling current practice using American Joint Committee on Cancer eighth edition, reaffirming nivolumab as a standard of care for melanoma adjuvant treatment.