S90 Predictors of Biliary Malignancy Among GLP-1 Receptor Agonist Users

Introduction: Biliary malignancies, including gallbladder and bile duct cancers, are rare but aggressive tumors often diagnosed at advanced stages. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are widely used for obesity and type 2 DM; however, their link with biliary neoplasia remains unclear. While early safety concerns raised questions about hepatobiliary risks, emerging evidence suggests potential anti-inflammatory and metabolic benefits. This study assessed the association between GLP-1 RA use and biliary malignancy, identifying independent predictors among users, using data from a large United States cohort. Methods: We conducted a retrospective cohort analysis using the All of Us Research Program, which integrates electronic health records, surveys, and biomarkers to support precision medicine. Adults with obesity or type 2 DM (2016 to 2022) were included; those with pancreatitis, CKD stage 3–5, prior biliary cancer or a personal/family history of multiple endocrine neoplasia were excluded. Biliary malignancy was defined by electronic health records diagnoses of gallbladder and intrahepatic/extrahepatic bile duct cancers. A 1:1 nearest-neighbor propensity score matching compared GLP-1 RA users and non-users. Multivariable logistic regression adjusted for demographics, comorbidities (smoking, metabolic dysfunction-associated steatotic liver disease, gallstones, autoimmune disease, HIV), personal/family history of cancers and medications (statins, insulin, DPP-4 inhibitors, SGLT2 inhibitors, metformin, sulfonylureas, TZD). A secondary model examined predictors among GLP-1 RA users. Results: Of 219,416 eligible individuals, 17,190 (7.8%) were GLP-1 RA users. Across the cohort, biliary malignancy occurred in 276 individuals (0.126%). Among 34,380 matched participants, there was no significant association between GLP-1 RA use and biliary cancers [adjusted OR (aOR): 1.14; 95% CI: 0.39–3.35; P = 0.814]. Within GLP-1 users, gallstones (aOR: 7.42; 95% CI: 2.59–21.28; P < 0.001) and HIV infection (aOR: 7.35; 95% CI: 1.60–33.84; P = 0.011) were strong independent predictors. SGLT2 inhibitor use showed a suggestive protective trend (aOR: 0.16; 95% CI: 0.02–1.21; P = 0.075). Conclusion: GLP-1 RA use was not associated with increased biliary malignancy risk, but gallstones and HIV strongly predicted risk among users. The protective signal from SGLT2 inhibitors, though non-significant, prompts the need for further research. These findings support targeted surveillance in high-risk GLP-1 RA users and suggest SGLT2 inhibitors as promising candidates for future prevention studies.