Denosumab and the Risk of Diabetes in Patients Treated for Osteoporosis.

Importance: Denosumab, a humanized monoclonal antibody against receptor activator of nuclear factor κB ligand (RANKL), is a widely used antiresorptive medication for osteoporosis treatment. Recent preclinical studies indicate that inhibition of RANKL signaling improves insulin sensitivity, glucose tolerance, and β-cell proliferation, suggesting that denosumab may improve glucose homeostasis; however, whether denosumab reduces the risk of incident diabetes remains unclear. Objective: To evaluate whether denosumab use is associated with a lower risk of developing diabetes in patients with osteoporosis. Design, Setting, and Participants: This nationwide, propensity score-matched cohort study used administrative data from Taiwan's National Health Insurance Research Database. Adult patients who received denosumab for osteoporosis therapy in Taiwan between 2012 and 2019 were included. To eliminate the inherent bias from confounding by indication, the patients were categorized into a treatment group (34 255 patients who initiated denosumab treatment and adhered to it) and a comparison group (34 255 patients who initiated denosumab treatment but discontinued it after the initial dose) according to the administration status of the second dose of denosumab. Propensity score matching was performed to balance patient characteristics and to control for confounders. Exposure: Treatment with denosumab. Main Outcomes and Measures: The primary outcome was incident diabetes requiring treatment with antidiabetic drugs. A Cox proportional hazards model was used to estimate the hazard ratio (HR) for incident diabetes. Data were analyzed from January 1 to November 30, 2023. Results: After propensity score matching, 68 510 patients were included (mean [SD] age, 77.7 [9.8] years; 57 762 [84.3%] female). During a mean (SD) follow-up of 1.9 (1.6) years, 2016 patients developed diabetes in the treatment group and 3220 developed diabetes in the comparison group (incidence rate, 35.9 vs 43.6 per 1000 person-years). Compared with the comparison group, denosumab treatment was associated with a lower risk of incident diabetes (HR, 0.84; 95% CI, 0.78-0.90). Several sensitivity analyses also demonstrated similar results of lower diabetes risk associated with denosumab treatment. Conclusions and relevance: The results from this cohort study indicating that denosumab treatment was associated with lower risk of incident diabetes may help physicians choose an appropriate antiosteoporosis medication for patients with osteoporosis while also considering the risk of diabetes.