Multimodal Risk Stratification and Surveillance for Gastric Cancer in Chronic Atrophic Gastritis and Gastric Intestinal Metaplasia: Integrating Histology, Endoscopy, Biomarkers, and Artificial Intelligence

Authors: Gao, Shujie; Zhao, Yiqing; Ji, Huihong; Lu, Qingfang; Zhang, Yuqi; Liu, Mengru; Wang, Junping

Source journal: Clinical and experimental gastroenterology (2026-01-01)

DOI: 10.2147/CEG.S643684

Abstract

Gastric cancer remains a major global health burden, and chronic atrophic gastritis (CAG) and gastric intestinal metaplasia (GIM) are important premalignant conditions. The risk of progression, however, varies considerably between individuals, making reliable risk stratification central to decisions on surveillance. Histological staging remains the most established approach. In a recent meta-analysis of prospective cohorts, Operative Link on Gastritis Assessment (OLGA) stages III-IV and OLGIM (Operative Link on Gastric Intestinal Metaplasia Assessment) stages III-IV were associated with substantially higher risks of gastric cancer than stages 0-II, with risk ratios of 32.31 and 12.38, respectively. Endoscopic assessment adds information on the distribution of mucosal abnormalities and can improve biopsy targeting; an EGGIM score >4 has shown 95.7% sensitivity and 91.4% specificity for identifying OLGIM stages III-IV. Helicobacter pylori eradication reduces subsequent gastric cancer risk, but surveillance remains necessary in patients with advanced atrophy, extensive GIM, or previous gastric neoplasia. Serological biomarkers may help identify individuals who should undergo endoscopy, although variation in assay thresholds and population characteristics limits their use for definitive risk staging. Artificial intelligence is also being explored to improve diagnostic consistency. In a multicentre pathology study, a deep-learning model achieved external-test AUCs of 0.884 for intestinal metaplasia and 0.877 for atrophy. Current evidence therefore favours combining these approaches according to their clinical roles rather than treating them as interchangeable tools: histology remains the basis of risk staging, endoscopy refines mucosal assessment and biopsy, biomarkers are most useful for triage, and artificial intelligence may support interpretation. Whether such integrated strategies improve surveillance efficiency or clinical outcomes still requires prospective, externally validated studies.

Read the full text

Related topics

Related open access articles

Abstract and bibliographic data from PubMed (U.S. National Library of Medicine). Copyright of the abstract belongs to the original publisher or authors.