BackgroundImmune checkpoint inhibitors (ICIs) combined with chemotherapy have become the first-line treatment of metastatic gastric and gastroesophageal adenocarcinomas (GEACs). This study aims to figure out the optimal combined positive score (CPS) cutoff value.MethodsWe searched for randomized phase III trials to investigate the efficacy of ICIs plus chemotherapy for metastatic GEACs compared with chemotherapy alone. Pooled analyses of hazard ratios (HRs) based on PD-L1 expression were performed.ResultsA total of six trials (KEYNOTE-062, KEYNOTE-590, KEYNOTE-859, ATTRACTION-04, CheckMate 649, and ORIENT-16) were included, comprising 5,242 patients. ICIs plus chemotherapy significantly improved OS (HR: 0.79, 95% CI 0.72–0.86 in global patients; HR: 0.75, 95% CI 0.57–0.98 in Asian patients) and PFS (HR: 0.74, 95% CI 0.68–0.82 in global patients; HR: 0.64, 95% CI 0.56–0.73 in Asian patients) compared with chemotherapy alone. The differences in OS (ratio of HR: 1.05, 95% CI 0.79–1.40; predictive value: − 5.1%) and PFS (ratio of HR: 1.16, 95% CI 0.98–1.36; predictive value: − 13.5%) were not statistically significant between the global and Asian patients. Subgroup analyses indicated that the optimal CPS threshold was at ≥ 5 for OS and ≥ 10 for PFS with the highest predictive values.ConclusionsThe benefit derived from ICIs plus chemotherapy is similar between Asian and global GEAC patients. However, those with a PD-L1 CPS < 5 or CPS < 10 may not have significant benefits from ICIs therapy. Therefore, it is advisable to routinely assess PD-L1 expression in GEAC patients considered for ICIs treatment.