RÉSUMÉ
OBJECTIVES: CheckMate 227 (NCT02477826) evaluated first-line nivolumab-plus-ipilimumab versus chemotherapy in patients with metastatic nonsmall cell lung cancer (NSCLC) with programmed death ligand 1 (PD-L1) expression ≥ 1% or < 1% and no EGFR/ALK alterations. However, many patients randomized to chemotherapy received subsequent immunotherapy. Here, overall survival (OS) and relative OS benefit of nivolumab-plus-ipilimumab were adjusted for potential bias introduced by treatment switching. MATERIALS AND METHODS: Treatment-switching adjustment analyses were conducted following the NICE Decision Support Unit Technical Support Document 16, for CheckMate 227 Part 1 OS data from treated patients (database lock, July 2, 2019). Inverse probability of censoring weighting (IPCW) was used in the base-case analysis; other methods were explored as sensitivity analyses. RESULTS: Of 1166 randomized patients, 391 (PD-L1 ≥ 1%) and 185 (PD-L1 < 1%) patients received nivolumab-plus-ipilimumab; 387 (PD-L1 ≥ 1%) and 183 (PD-L1 < 1%) patients received chemotherapy, with 29.3-month minimum follow-up. Among chemotherapy-treated patients, 169/387 (43.7%; PD-L1 ≥ 1%) and 66/183 (36.1%; PD-L1 < 1%) switched to immunotherapy poststudy. Among treated patients, median OS was 17.4 months with nivolumab-plus-ipilimumab versus 14.9 months with chemotherapy (hazard ratio [HR], 0.80; 95% confidence interval [CI], 0.68-0.95) in the PD-L1 ≥ 1% subgroup and 17.1 versus 12.4 months (HR, 0.62; 95% CI, 0.49-0.80) in the PD-L1 < 1% subgroup. After treatment-switching adjustment using IPCW, the HR (95% CI) for OS for nivolumab-plus-ipilimumab versus chemotherapy was reduced to 0.68 (0.56-0.83; PD-L1 ≥ 1%) and 0.53 (0.40-0.69; PD-L1 < 1%). Sensitivity analyses supported the robustness of the results. CONCLUSION: Treatment-switching adjustments resulted in a greater estimated relative OS benefit with first-line nivolumab-plus-ipilimumab versus chemotherapy in patients with metastatic NSCLC.