Meta-Analysis of Clinical Fracture Risk Reduction of Antiosteoporosis Drugs: Direct and Indirect Comparisons and Meta-Regressions

Antiosteoporotic drug (AOD) trials have variabilities in duration and fracture risks. This study evaluated AOD's versus controls regarding reduction in relative rates and rate differences in vertebral and hip fractures and comparative costs.Primary randomized controlled trials of antiosteoporotic drugs in postmenopausal women with documentation of vertebral fracture rates or hip fracture rates were extracted from meta-analyses and PubMed through February 2021. Direct and indirect meta-analyses and meta-regressions analyzed the fracture reductions.There were 24 randomized controlled trials of drug versus placebo (73 862 women) and 10 randomized controlled trials of drug versus drug. The reductions in the relative rates of vertebral fractures were significant for antiresorptive (alendronate, risedronate, zoledronate, denosumab, and raloxifene) and anabolic (teriparatide, abaloparatide, and romosozumab) drugs. Denosumab, teriparatide, and abaloparatide were more effective in reducing vertebral fracture rates than oral bisphosphates (all P < .05) but were not more effective in reducing vertebral fracture rates than zoledronate. The reductions in hip fracture rates were significant for alendronate, denosumab, and zoledronate (all P < .05), without significant differences among drugs. Anabolic drugs did not show significant hip fracture rate reduction. Meta-regression of rate differences enabled the calculation of costs per vertebral fracture prevented, which were estimated at >$100 000 for anabolic drugs and between $2289 and $28 947 for antiresorptive drugs. Many direct drug versus drug trials were underpowered to demonstrate benefits of one drug over another.This study suggests goal-directed, cost-effective therapies relative to patient risk for vertebral and hip fractures. Anabolic drugs are better at preventing vertebral fractures than oral bisphosphonates. Anabolic drugs are not superior to zoledronate or denosumab and are substantially more expensive. When comparing drugs that prevented hip fractures, there was no statistical benefit of any drug.