Gastric atrophy and gastric cancer: a meta-analytical assessment of risk and the influence of topographical distribution

Gastric atrophy is a precursor in the Correa cascade leading to gastric cancer, yet the precise magnitude of risk remains uncertain, particularly in low-incidence Western populations. We performed a systematic review and meta-analysis of observational studies published between 2010 and 2024 to evaluate gastric cancer risk in gastric atrophy, with stratification by anatomical distribution. Ten studies comprising 21 006 patients were included. Pooled analysis demonstrated a markedly increased risk of gastric cancer in gastric atrophy, with a standardised incidence ratio of 20.7 (95% confidence interval: 14.0-30.2). Subgroup analysis confirmed significantly higher risk with extensive gastric atrophy compared with antral-predominant disease (pooled odds ratio: 0.27), supporting current guideline stratification. We also reviewed studies excluded from pooled analysis, including several large European cohorts that initially appeared eligible but failed on methodological grounds. Common limitations included inadequate distinction between gastric atrophy and intestinal metaplasia, lack of stratification between autoimmune and Helicobacter pylori-associated atrophy, and use of Operative Link staging systems without atrophy-specific outcomes. These exclusions highlight the fragility of the Western evidence base and its limited ability to inform surveillance recommendations. Our findings extend prior meta-analyses by incorporating topography-specific risk estimates and explicitly identifying evidence gaps in Western populations. Gastric atrophy confers a markedly increased risk of gastric cancer, particularly when extensive. Standardised biopsy mapping and atrophy-specific reporting are needed to refine surveillance strategies and tailor guidelines to regional risk profiles. Current surveillance recommendations for distal gastric atrophy in low-risk Western populations should be applied cautiously, given the evidence and absence of robust prospective data.