Gastric Dysplasia and Surveillance in Familial Adenomatous Polyposis

Background Familial adenomatous polyposis (FAP) is associated with colorectal and extracolonic malignancy risk. Although duodenal surveillance is well established for establishing risk of malignancy, monitoring gastric dysplasia using esophagogastroduodenoscopy (EGD) is less clearly characterized as a mechanism for risk management. We aimed to characterize gastric pathology findings and factors associated with gastric dysplasia in patients with FAP. Methods We performed a retrospective cohort study of adults with clinically or genetically confirmed FAP using electronic health records from MedStar Health, a multihospital health system in the Washington, DC, metropolitan area, between 2000 and 2025. No patients in the confirmed cohort (N=188) had attenuated FAP. Patients were initially identified through an electronic health record search using ICD-10 coding for FAP and subsequently confirmed by medical chart review based on a documented adenomatous polyposis coli mutation or classic clinical criteria (>100 colorectal adenomas). Kaplan-Meier methods estimated incident gastric dysplasia after index EGD. Fisher exact test assessed the association between gastric polyp presence and dysplasia. Results Among 188 identified patients, 150 had confirmed FAP, and 93 had EGD follow-up. Fundic gland polyps were identified in 64 (69%) of 93 patients, and gastric dysplasia was identified in 26 (28%) of 93 patients. Seven patients had gastric dysplasia at index EGD; the remaining 86 entered the incident time-to-event analysis, of whom 19 developed incident gastric dysplasia during a median 5 years of follow-up. Estimated cumulative incidence was 0.0%, 2.1%, 4.3%, and 18.5% at 1, 3, 5, and 10 years, respectively. Patients without incident gastric dysplasia were censored at their last EGD; therefore, not all patients were seen through 10 years of follow-up. Gastric dysplasia occurred in 26 of 72 patients with gastric polyps and in 0 of 21 without polyps (36.1% vs 0.0%; Fisher exact test, P <.001). No gastric adenocarcinoma was observed. Conclusions Gastric dysplasia was common among patients with FAP undergoing EGD surveillance and was significantly associated with gastric polyp presence. These findings support longitudinal endoscopic assessment of patients with FAP, incorporating gastric polyp burden and histology. The prevalence of gastric polyps in patients with FAP and the association between presence of gastric polyps and incidence of gastric dysplasia are consistent with results from a fuller analysis from this study.