Comparative effectiveness of perioperative strategies for resectable gastric and gastroesophageal junction cancer: a Bayesian network meta-analysis.

Abstract

Perioperative strategies for resectable gastric and gastroesophageal junction (GEJ) adenocarcinomas are continuously evolving, with recent regimens, particularly those incorporating immunotherapy, showing promising results, although their comparative efficacy remains uncertain. We conducted a Bayesian network meta-analysis of randomized controlled trials (RCTs) published between January 2004 and March 2025 that compared perioperative treatments involving chemotherapy, radiotherapy, immunotherapy, or targeted agents. Five outcomes were analyzed: overall survival (OS), progression-free survival (PFS), R0 resection, pathological complete response (pCR), and major pathological response (MPR). A Bayesian random-effects model was applied to estimate hazard ratios (HRs) and odds ratios (ORs), and surface under the cumulative ranking curve (SUCRA) values were used for treatment ranking. A total of 25 RCTs involving 11,317 patients were included. Neo/Peri DOS/DOX, comprising neoadjuvant or perioperative docetaxel-oxaliplatin-S-1 (DOS) or docetaxel-oxaliplatin-capecitabine (DOX), ranked highest for OS and PFS, showing significant survival benefits over both surgery alone and adjuvant chemotherapy. Regimens combining perioperative chemotherapy with PD-1/PD-L1 inhibitors (Neo/Peri CT+PD1/PDL1) achieved the greatest improvement in pCR and MPR, although their survival benefit was limited to comparisons with surgery alone. None of the regimens significantly improved R0 resection. The findings were robust across sensitivity analyses, with no major inconsistencies detected. In conclusion, DOS/DOX demonstrated superior survival outcomes and may represent a leading perioperative option, while PD-1/PD-L1-based combinations improved early pathological responses but require further validation with mature survival data.