Fracture incidence after denosumab discontinuation: Real-world data from a large healthcare provider

Background: Vertebral fractures (VF) upon Denosumab (DMAB) discontinuation were first described as a distinct phenomenon in 2015, yet the magnitude of this event remains undetermined. Objectives: To estimate fracture risk after DMAB discontinuation, in a real-world setting. Methods: The computerized database of a 2.3-million members' state-mandated health organization was utilized to detect osteoporotic patients with at least two DMAB dispenses. Treatment discontinuation was defined as a refill gap of 3 months or more, while the discontinuation date was defined as an anticipated missed purchase date. Fractures were identified by an osteoporosis registry and individually adjudicated by an expert's review. Fractures occurring within one year from discontinuation among DMAB discontinuers (DD) and from the 2nd year of treatment onwards for persistent users (PU) were included. Results: A total of 1500 DD (92% females, mean +/- SD age = 71.8 +/- 9.5y), and 1610 PU (91%, 71.7 +/- 8.8) were identified. At baseline, the groups were comparable in fracture- history, bisphosphonate exposure, smoking, and bone density. Multiple VF occurred in 12 (0.8%) DD vs. 2 (0.1%) PU (p = 0.006). The overall rate of fractures per 100 patient-years of follow-up was significantly higher in DD than PU (RR 3.2, 95% CI 2.2-4.8), as well as the rate of VF (RR 4.7, 95% CI 2.3-9.6) and multiple VF (RR 14.6, 95% CI 3.3-65.3, effect size 1.06). Conclusions: Patients who discontinue DMAB are at greater risk of major OP fractures than those who persist with treatment. Same is true for clinical multiple vertebral fractures, yet the incidence of the latter was low. These findings demonstrate a need for greater awareness and thoughtful management of DMAB discontinuation.