First-Line Nivolumab Plus Ipilimumab in Advanced NSCLC: 4-Year Outcomes From the Randomized, Open-Label, Phase 3 CheckMate 227 Part 1 Trial

Introduction: In CheckMate 227, nivolumab plus ipilimu mab prolonged overall survival (OS) versus chemotherapy in patients with tumor programmed death-ligand 1 (PD-L1) greater than or equal to 1% (primary end point) or less than 1% (prespecified descriptive analysis). We report re sults with minimum 4 years’ follow-up. Methods: Adults with previously untreated stage IV or recurrent NSCLC were randomized (1:1:1) to nivolumab plus ipilimumab, nivolumab, or chemotherapy (PD-L1 1%); or to nivolumab plus ipilimumab, nivolumab plus chemotherapy, or chemotherapy (PD-L1 <1%). Efficacy included OS and other measures. Safety included timing and management of immune-mediated adverse events (AEs). A post hoc analysis evaluated efficacy in patients who dis continued nivolumab plus ipilimumab due to treatment related AEs (TRAEs). Results: After 54.8 months’ median follow-up, OS remained longer with nivolumab plus ipilimumab versus chemo therapy in patients with PD-L1 greater than or equal to 1% (hazard ratio ¼ 0.76; 95% confidence interval: 0.65–0.90) and PD-L1 less than 1% (0.64; 0.51–0.81); 4-year OS rate with nivolumab plus ipilimumab versus chemotherapy was 29% versus 18% (PD-L1 1%); and 24% versus 10% (PD L1 <1%). Benefits were observed in both squamous and nonsquamous histologies. In a descriptive analysis, efficacy was improved with nivolumab plus ipilimumab relative to nivolumab (PD-L1 1%) and nivolumab plus chemotherapy (PD-L1 <1%). Safety was consistent with previous reports. The most common immune-mediated AE with nivolumab plus ipilimumab, nivolumab, and nivolumab plus chemo therapy was rash; most immune-mediated AEs (except endocrine events) occurred within 6 months from start of treatment and resolved within 3 months after, mainly with systemic corticosteroids. Patients who discontinued nivo lumab plus ipilimumab due to TRAEs had long-term OS benefits, as seen in the all randomized population. Conclusions: At more than 4 years’ minimum follow-up, with all patients o