Efficacy and safety of immune checkpoint inhibitor-based regimens for advanced gastric/gastroesophageal junction adenocarcinoma: a Bayesian network meta-analysis of randomized controlled trials

Background Direct head-to-head evidence comparing different immune checkpoint inhibitor (ICI)-based regimens for advanced gastric or gastroesophageal junction adenocarcinoma remains limited. This study aimed to synthesize direct and indirect evidence to evaluate efficacy and safety of relevant immunotherapy regimens and to provide references for clinical practice. Methods Four electronic databases, including PubMed, Embase, Cochrane Library and Web of Science were searched up to July 2026. The primary endpoints were overall survival (OS) and progression-free survival (PFS). Secondary endpoints included objective response rate (ORR) and the incidence of grade ≥3 treatment-related adverse events (TRAEs). The risk of bias was assessed using the RoB 2.0 tool. A Bayesian random-effects network meta-analysis was performed, and surface under the cumulative ranking curve (SUCRA) values were calculated for treatment ranking. A prespecified subgroup analysis was conducted according to PD-L1 combined positive score (CPS). Results Ten randomized controlled trials involving 7, 728 patients were included. The primary analysis was restricted to the eight first-line RCTs (n = 6, 643) to satisfy the transitivity assumption. Compared with chemotherapy alone, ICI-containing regimens significantly improved OS (HR = 0.776, 95% CI 0.712-0.845, P < 0.0001), PFS (HR = 0.705, 95% CI 0.624-0.797, P < 0.0001), and ORR (OR = 1.598, 95% CI 1.394-1.832, P < 0.0001). A modest but statistically significant increase in grade ≥3 TRAEs was detected with ICI addition (OR = 1.143, 95% CI 1.026-1.274, P = 0.016). Survival benefits varied according to PD-L1 CPS; significant improvements were observed in patients with CPS ≥5 and CPS 1-4, whereas the benefit did not reach statistical significance in those with CPS <1. SUCRA ranking suggested that PD-1/CTLA-4 bispecific antibody plus chemotherapy had the highest probability of being the optimal OS strategy. Conclusions In patients with advanced gastric or gastroesophageal junction adenocarcinoma, ICI-containing regimens significantly improved OS, PFS, and ORR over chemotherapy alone, with greater benefit at higher PD-L1 CPS and a modest but significant increase in grade ≥3 TRAEs that should be weighed against the survival benefit. Treatment rankings derived from indirect evidence carry considerable uncertainty and should not be interpreted as definitive superiority; confirmation in prospective head-to-head trials is warranted.