Randomized trial compares fracture prevention and cardiovascular outcomes in osteoporosis therapies, suggesting differences in effectiveness.
Background: Romosozumab and teriparatide are approved osteoanabolic therapies for osteoporosis, but direct real-world comparisons remain limited. In this study, we evaluated comparative effectiveness and selected safety outcomes. Methods: Female patients with osteoporosis aged ≥ 50 years who initiated romosozumab or teriparatide were identified in the TriNetX Global Network from January 2020 through December 2025. Outcomes included any osteoporotic fracture, spine or hip fracture, major adverse cardiovascular events (MACE; a nonfatal composite of acute myocardial infarction and stroke), and all-cause mortality, which was analyzed separately. Propensity score matching balanced measured baseline characteristics. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox regression. Results: After 1:1 propensity score matching, each matched cohort included 8616 patients, with follow-up administratively limited to 3 years. Romosozumab initiation was associated with lower risks of any osteoporotic fracture (445 vs. 832 events; HR, 0.68; 95% CI, 0.61–0.77) and spine or hip fracture (341 vs. 682 events; HR, 0.63; 95% CI, 0.55–0.72). The primary analysis also showed a lower estimated risk of MACE (145 vs. 255 events; HR, 0.80; 95% CI, 0.65–0.98), but this association was not consistent across sensitivity analyses. All-cause mortality did not differ significantly (81 vs. 126 events; HR, 0.96; 95% CI, 0.72–1.27). Conclusions: Romosozumab initiation was associated with lower fracture incidence than teriparatide initiation. The cardiovascular and mortality findings do not establish superiority, equivalence, or cardioprotection and should be interpreted cautiously.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: