Abstract
Introduction and Objective: Concerns have been raised about potentially harmful cancer-related effects of certain medications used to manage type 2 diabetes (T2D). However, there is limited evidence from head-to-head comparisons between commonly used second-line T2D medications with respect to incident cancer risk.
Methods: In a retrospective target trial emulation, we compared glucagon-like peptide-1 receptor agonists (GLP-1RA), sodium-glucose cotransporter 2 inhibitors (SGLT2i), dipeptidyl peptidase-4 inhibitors (DPP4i), and sulfonylureas with respect to incident cancer risk among adults with T2D and moderate CVD risk. We used de-identified administrative claims from the OptumLabs Data Warehouse and the Medicare fee-for-service 100% sample to identify adults ≥21 years with T2D between 2014-2021 with an estimated 1-5% annualized risk of a major adverse cardiovascular event who initiated these drugs. We then compared rates of any incident cancer (excluding non-melanoma skin cancer) and each cancer type using propensity score inverse probability of treatment weighted Cox proportional hazards models.
Results: We identified 40,956 patients starting GLP-1RA (mean age 65.1 ± 8.3 years, 51.1% male), 52,050 starting SGLT2i (mean age 65.1 ± 8.4 years, 50.6% male), 74,314 starting DPP4i (mean age 65.44 ± 8.5 years, 50.9% male), and 187,960 starting sulfonylurea (mean age 65.4 ± 8.5 years, 51.0% male). Treatment with GLP-1RA was associated with a 12% higher risk of incident cancer compared to SGLT2i (HR 1.12, 95% CI 1.04-1.22), 12% higher risk compared to DPP4i (HR 1.12, 95% CI 1.05-1.20), and 15% higher risk compared to sulfonylurea (HR 1.15, 95% CI 1.08-1.22). This increase in risk was driven primarily by higher incidence of melanoma, male reproductive, and female breast cancers.
Conclusion: This increased risk of some cancers observed with GLP-1RA warrants further investigation and cautious use among individuals at risk for these malignancies.