Abstract
PURPOSE: Pembrolizumab demonstrated antitumor activity in programmed death ligand 1 positive (combined positive score (CPS) ≥ 1) gastric/gastroesophageal junction cancer in KEYNOTE-059 (third line or beyond), KEYNOTE-061 (second line), and KEYNOTE-062 (first line). We characterized efficacy and safety of pembrolizumab monotherapy in Japanese patients across several lines of therapy in these studies. METHODS: This analysis was conducted in 34 patients from KEYNOTE-059 cohort 1 (all pembrolizumab), including 13 patients with CPS ≥ 1, 65 patients with CPS ≥ 1 from KEYNOTE-061 (pembrolizumab, n = 27; chemotherapy, n = 38), and 70 patients with CPS ≥ 1 from KEYNOTE-062 (pembrolizumab, n = 38; chemotherapy, n = 32). Overall survival (OS), progression-free survival (PFS), objective response rate (ORR), and safety were evaluated. RESULTS: In KEYNOTE-059, ORR with pembrolizumab was 9%, median PFS was 2 months, and median OS was 10 months. In KEYNOTE-061, median OS was 12 months with pembrolizumab versus 10 months with chemotherapy (hazard ratio (HR), 0.67; 95% confidence interval (CI), 0.39-1.15). Median PFS (pembrolizumab vs. chemotherapy) was 2 months versus 4 months (HR, 1.21; 95% CI, 0.69-2.13); ORR was 7% versus 18%. In KEYNOTE-062, median OS was 20 months with pembrolizumab versus 18 months with chemotherapy (HR, 0.76; 95% CI, 0.43-1.33). Median PFS (pembrolizumab vs. chemotherapy) was 6 months versus 7 months (HR, 1.03; 95% CI, 0.61-1.74); ORR was 29% versus 34%. CONCLUSIONS: The current analysis provides valuable information that anti-PD-1 therapies are worthy of further assessment for gastric cancer. TRIAL REGISTRATION: ClinicalTrials.gov: NCT02335411 (KEYNOTE-059), NCT02370498 (KEYNOTE-061), and NCT02494583 (KEYNOTE-062).