Association between autoimmune gastritis and gastric cancer adjusted for Helicobacter pylori infection: A systematic review and meta-analysis

Background: The association between autoimmune gastritis and gastric adenocarcinoma remains controversial, and Helicobacter pylori infection is a major potential confounder. Aims: To evaluate the association between autoimmune gastritis and gastric adenocarcinoma after controlling for H. pylori infection. Methods: PubMed, Embase and Scopus were searched from inception to 13 April 2025 for observational studies reporting autoimmune gastritis, H. pylori infection and gastric adenocarcinoma. Data were synthesised using random-effects meta-analysis, subgroup and adjusted analyses by H. pylori status, and a multilevel arm-based logistic regression model when compatible study-specific estimates were unavailable. Study quality was assessed using Joanna Briggs Institute tools. Results: Twenty-one studies were included, although only four contributed to conventional meta-analysis. After adjustment for H. pylori , no significant association between autoimmune gastritis and gastric adenocarcinoma was observed (OR = 1.18, 95% CI 0.61–2.26; I² = 86%). Subgroup analyses were also non-significant. However, a multilevel model including nine studies showed a weak but significant association independent of H. pylori (OR = 1.27, 95% CI 1.02–1.57). Among H. pylori -negative patients with autoimmune gastritis, gastric adenocarcinoma incidence reached 98.4 per 100,000 person-years. Conclusions: Autoimmune gastritis may be modestly associated with gastric adenocarcinoma after adjustment for H. pylori status. The association observed in the multilevel arm-based model, together with the higher-than-expected gastric adenocarcinoma incidence reported in longitudinal autoimmune gastritis cohorts, is compatible with a possible independent contribution of autoimmune gastritis to gastric carcinogenesis. These findings should be considered hypothesis-generating given the substantial heterogeneity, potential exposure misclassification, and residual confounding. PROSPERO registration: CRD420251275536