Impact of immune checkpoint inhibitors as neoadjuvant/perioperative therapy for locally advanced resectable gastric/gastroesophageal junction adenocarcinoma: a systematic review and meta-analysis

Background Immune checkpoint inhibitor (ICI)-based neoadjuvant/perioperative strategies are expanding for locally advanced resectable gastric/gastroesophageal junction (G/GEJ) adenocarcinoma, but the evidence base differs substantially across strategies, and improved pathological response has not consistently translated into survival benefit. We aimed to synthesize strategy-specific comparative and prospective single-arm evidence. Methods PubMed, Web of Science, Embase, the Cochrane Library, and major oncology conference proceedings were searched from inception to August 3, 2026. RCTs and prospective single-arm trials evaluating ICI-containing treatment initiated before surgery were included. Five strategies were evaluated: ICI plus chemotherapy (ICI+CT), ICI plus chemotherapy and anti-angiogenic therapy (ICI+CT+A), ICI plus chemoradiotherapy (ICI+CRT), ICI plus chemotherapy and anti-HER2 therapy (ICI+CT+H), and dual-ICI therapy (ICI+ICI). Random-effects comparative meta-analyses were conducted separately by strategy, while single-arm analyses provided descriptive absolute estimates. Results A total of 74 trials, including 16 RCTs and 58 prospective single-arm trials, were included. In randomized comparisons, ICI+CT was associated with improved pCR (RR, 3.01; 95% CI, 2.19-4.15) and MPR (RR, 1.59; 95% CI, 1.36-1.85) and, in perioperative trials, OS (HR, 0.81; 95% CI, 0.70-0.93) and EFS (HR, 0.74; 95% CI, 0.66-0.83) compared with chemotherapy. ICI+CT+A was associated with improved pCR (RR, 3.43; 95% CI, 1.85-6.37) and MPR (RR, 1.40; 95% CI, 1.12-1.76), but survival evidence was limited to one phase II RCT reporting OS. Comparative evidence for ICI+CRT was limited to one small RCT. Evidence for ICI+CT+H and ICI+ICI was derived mainly from small biomarker-selected noncomparative studies. Conclusions Evidence differed substantially across the five strategies. ICI+CT was associated with improved pCR and MPR and, in perioperative RCTs, OS and EFS; it had the largest and most mature randomized evidence base. ICI+CT+A was associated with improved pCR and MPR, whereas comparative evidence for ICI+CRT, ICI+CT+H, and ICI+ICI remained limited. Further comparative trials are required, particularly for safety and long-term survival. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420251271789 , identifier CRD420251271789.