Osteoporosis drugs for prevention of clinical fracture in white postmenopausal women: a network meta-analysis of survival data.

Abstract

By Bayesian random effects network meta-analysis stratified by prevalent vertebral fracture (PVF), we conclude that different effective drugs should be used to prevent fragility fractures according to postmenopausal women with or without PVF and that there are two drugs (i.e., parathyroid hormone (1-84) and abaloparatide) less tolerated than placebo. No studies have compared various osteoporosis drugs in postmenopausal women (PMW) either with or without prevalent vertebral fracture (PVF). We aimed to compare them in the two different subgroups. We searched different databases to select relevant studies. We performed Bayesian random effects network meta-analysis to synthesize hazard ratio (HR) and 95% confidence interval (CI) for clinical fracture stratified by PVF and to synthesize risk ratio (RR) for tolerability and vertebral fracture. We included 33 trials involving 79,144 PMW. In the PVF ≥ 50% subgroup, teriparatide (HR 0.39, 95% CI 0.28-0.57), romosozumab (HR 0.49, 95% CI 0.29-0.75), risedronate (HR 0.62, 95% CI 0.50-0.79), zoledronate (HR 0.67, 95% CI 0.47-0.96), and alendronate (HR 0.69, 95% CI 0.47-0.97) reduced clinical fracture risk. In the other subgroup, abaloparatide (HR 0.56, 95% CI 0.33-0.92), romosozumab (HR 0.67, 95% CI 0.47-0.95), and denosumab (HR 0.68, 95% CI 0.50-0.85) reduced clinical fracture risk. Five drugs reduced vertebral fracture risk in the PVF ≥ 50% subgroup whereas seven did in the other subgroup. All drugs did not increase withdrawal risk except for parathyroid hormone (1-84) (PTH) (RR 1.9, 95% CI 1.4-2.6) and abaloparatide (RR 1.6, 95% CI 1.2-2.3). Different effective drugs should be used to prevent fragility fractures according to PMW with or without PVF, and romosozumab is the only one which can reduce clinical and vertebral fractures in both of the two populations. PTH and abaloparatide are less tolerated than placebo whereas the eight other drugs assessed in the study have the same tolerability as placebo.