MO01.18 An indirect Comparison of Pembrolizumab+Chemo vs Ipilimumab+Nivolumab as First-Line Therapies in Patients with PD-L1 TPS≥1% Metastatic NSCLC

Multiple anti-PD-1/PD-L1 first-line treatment options are approved for advanced non-small cell lung cancer (NSCLC), including pembrolizumab+chemotherapy studied in KEYNOTE 021G (KN021G), KEYNOTE-189 (KN189) and KEYNOTE-407 (KN407) and nivolumab+ipilimumab studied in CheckMate 227 Part 1A. A comprehensive understanding of the potential differences between these treatment options is required to inform clinical decision making. In the absence of head-to-head trial data, this analysis indirectly compared the effectiveness of pembrolizumab+chemotherapy versus nivolumab+ipilimumab for the first-line treatment of metastatic NSCLC patients with a PD-L1 tumor proportion score (TPS) ≥1%. A matching-adjusted indirect comparison (MAIC) was conducted using pooled individual patient data (IPD) from the intention-to-treat (ITT) population of KN021G, KN189 and KN407 restricted to patients with stage IV disease (N=816; database cut-off dates: 8/19/2019, 5/20/2019, and 5/9/2019, respectively) and published aggregated data of nivolumab+ipilimumab from the CheckMate 227 Part 1A (N=793; database cut-off date: 7/2/2019). Since platinum-doublet chemotherapy was the comparator in all trials it was used as the anchor for the MAIC. Study designs were similar for all three trials. To adjust for cross-trial differences in baseline characteristics, data from KN021G/KN189/KN407 were re-weighted to match the baseline characteristics of CheckMate 227 Part 1A (including age, sex, region, smoking status, ECOG status, histology, metastasis, PD-L1 expression). Outcomes of interest included overall survival (OS), progression-free survival (PFS) and objective response rate (ORR). Assessment of PFS and ORR was based on blinded independent review. Base case analyses were restricted to the patient population with PD-L1 TPS≥1%, with sub-group analyses in TPS≥50% and 1–49% sub-groups. The effective sample size of the KN021G/KN189/KN407 population after adjusting for cross trial differences was 456. The estimated HR (95% CIs) of pembrolizumab+chemotherapy versus nivolumab+ipilimumab was 0.80 (0.59, 1.09) and 0.53 (0.41, 0.68) for OS and PFS, respectively, which favored pembrolizumab+chemotherapy. For ORR, the estimated risk ratio (95% CI) was 1.78 (1.32, 2.39) for pembrolizumab+chemotherapy versus nivolumab+ipilimumab and the risk difference was 25% (15, 36). After matching, the landmark 1-year OS rate was 70.95% versus 62.40% for pembrolizumab+chemotherapy and nivolumab+ipilimumab, respectively, and the 2-year OS rate was 49.22% versus 39.79%, respectively. Findings were consistent across PD-L1 TPS 1–49% and TPS≥50% sub-groups. These MAIC results show that the pembrolizumab+histology-specific platinum-doublet chemotherapy option leads to a greater clinical benefit than nivolumab+ipilimumab in patients with PD-L1≥1% and in different PD-L1 TPS sub-groups. Given the lack of head-to-head studies, these analyses may inform clinical and formulary decision making for prioritizing treatments.