Objectives Our objectives were to evaluate the cost effectiveness of pembrolizumab compared with standard-of-care (SoC) platinum-based chemotherapy as first-line treatment in patients with metastatic non-small-cell lung cancer (NSCLC) that expresses high levels of programmed death ligand-1 (PD-L1) [tumour proportion score (TPS) ≥50%], from a US third-party public healthcare payer perspective. Methods We conducted a partitioned-survival model with a cycle length of 1 week and a base-case time horizon of 20 years. Parametric models were fitted to Kaplan–Meier estimates of time on treatment, progression-free survival and overall survival from the KEYNOTE-024 randomized clinical trial (patients aged ≥18 years with stage IV NSCLC, TPS ≥50%, without epidermal growth factor receptor (EGFR)-activating mutations or anaplastic lymphoma kinase (ALK) translocations who received no prior systemic chemotherapy) and validated with long-term registry data. Quality-adjusted life-years (QALYs) were calculated based on EuroQoL-5 Dimensions (EQ-5D) utility data collected in the trial. Costs ($US, year 2016 values) for drug acquisition/administration, adverse events and clinical management were included. Costs and outcomes were discounted at 3% per year. A series of deterministic and probabilistic sensitivity analyses were performed to test the robustness of the results. Results In the base-case scenario, pembrolizumab resulted in an expected gain of 1.31 life-years (LYs) and 1.05 QALYs and an incremental cost of $US102,439 compared with SoC. The incremental cost per QALY gain was $US97,621/QALY and the incremental cost per LY gain was $US78,344/LY. Conclusions Pembrolizumab is projected to be a cost-effective option compared with SoC platinum-based chemotherapy as first-line treatment in adults with metastatic NSCLC expressing high levels of PD-L1.

Multiple immunotherapy and chemotherapy combinations are approved for the management of advanced NSCLC which have not been directly compared in randomized clinical trials. This study indirectly compared the effectiveness of pembrolizumab + chemotherapy versus atezolizumab + chemotherapy+/-bevacizumab for previously untreated non-squamous NSCLC patients without EGFR/ALK aberrations. A matching-adjusted indirect comparison (MAIC) was conducted using individual patient data (IPD) from KEYNOTE-021 Cohort G (KN021 G) (pembrolizumab + carboplatin + pemetrexed; N = 59) and KEYNOTE-189 (KN189) (pembrolizumab + pemetrexed + platinum chemotherapy; N = 410) and published aggregate data from IMpower 130 (atezolizumab + carboplatin + nab-paclitaxel; N = 451) and IMpower 150 (atezolizumab + carboplatin + paclitaxel + bevacizumab; N = 356). To adjust for cross-trial differences in baseline characteristics, data from patients randomized to pembrolizumab + chemotherapy in KN021 G/KN189 were reweighted to match the baseline characteristics of patients randomized to atezolizumab + chemotherapy from IMpower 130 or atezolizumab + chemotherapy + bevacizumab from IMpower 150. Outcomes included overall survival (OS), blinded independent review-assessed progression-free survival (PFS) and objective response rate (ORR). OS and PFS follow-up were truncated to the trial with shorter follow-up. Sensitivity analyses were conducted without truncation of follow-up of OS and PFS. After matching for cross-trial differences, the effective sample size of pembrolizumab + chemotherapy was 428 and 389 for the IMpower 130 and IMpower 150 comparisons, respectively. The estimated HRs (95 % CIs) of pembrolizumab + chemotherapy versus atezolizumab + chemotherapy were 0.80 (0.67,0.95) and 0.79 (0.67,0.93) with regard to OS and PFS, respectively. For pembrolizumab + chemotherapy versus atezolizumab + chemotherapy + bevacizumab, the estimated HR (95 % CIs) was 0.86 (0.72,1.03) for OS and 0.81 (0.68,0.96) for PFS. For ORR, the estimated risk ratio (95 % CI) and the risk difference (95 % CI) was 0.9 (0.8,1.1) and -3.5 % (-10.0,3.1) for pembrolizumab + chemotherapy versus atezolizumab + chemotherapy, respectively, and 0.8 (0.7,0.9) and -12.2 % (-19.6,-4.8) for pembrolizumab + chemotherapy versus atezolizumab + chemotherapy + bevacizumab, respectively. Findings were consistent across sensitivity analyses for both outcomes. MAIC results showed a significantly better OS and PFS for pembrolizumab + chemotherapy compared with atezolizumab + chemotherapy and a significantly better PFS for pembrolizumab + chemotherapy compared with atezolizumab + chemotherapy + bevacizumab.