Glucagon-like peptide-1 receptor agonists and risk of pancreatic cancer in high-risk individuals: A multicenter real-world analysis.

126

Background: Individuals with benign pancreatic neoplasms, such as intraductal papillary mucinous neoplasms, and those with hereditary cancer susceptibility carry a substantially increased risk of pancreatic cancer. Emerging evidence suggests that glucagon-like peptide-1 receptor agonists (GLP-1RAs) may have protective effects against pancreatic carcinogenesis. We therefore evaluated the association between GLP-1RA use and pancreatic cancer risk among high-risk individuals using a large real-world database. Methods: We conducted a retrospective cohort study using the TriNetX Global Collaborative Network, comprising de-identified electronic health records from over 140 healthcare organizations worldwide. Adults (≥18 years) with diabetes mellitus and high-risk pancreatic conditions, defined as benign pancreatic neoplasms or documented hereditary cancer susceptibility, were identified. Patients receiving GLP-1RAs were compared with non-users receiving alternative non–GLP-1RA diabetes therapies. Propensity score matching (1:1) was performed based on demographics, metabolic comorbidities, pancreatic risk factors, and baseline medications. The primary outcome was incident pancreatic cancer. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated using Cox proportional hazards models. Results: After 1:1 propensity score matching, GLP-1RA users and non-users were well balanced across baseline characteristics in both the benign pancreatic neoplasm and hereditary cancer susceptibility cohorts. Over a mean follow-up of approximately 2 years, GLP-1RA use was associated with a significantly lower risk of incident pancreatic cancer. Among individuals with benign pancreatic neoplasms, pancreatic cancer occurred in 21 of 456 GLP-1RA users compared with 42 of 456 non-users (HR, 0.54; 95% CI, 0.32–0.91; log-rank P = 0.019). Similarly, among patients with hereditary cancer susceptibility, pancreatic cancer occurred in 11 of 958 GLP-1RA users versus 26 of 958 non-users (HR, 0.46; 95% CI, 0.23–0.93; log-rank P = 0.027). Conclusions: In this large real-world analysis of high-risk individuals with diabetes mellitus, GLP-1RA use was associated with a significantly lower risk of pancreatic cancer. These findings suggest a potential protective association in populations at increased baseline risk. Prospective studies are warranted to confirm these observations and to elucidate underlying biological mechanisms.