Network meta-analysis compares osteoporosis drugs for bone density in men

BackgroundEvidence from randomized controlled trials (RCTs) on pharmacological management of primary osteoporosis in men is fragmented, making direct comparisons across drug classes difficult, and head-to-head evidence remains limited. We performed a network meta-analysis (NMA) to compare the relative efficacy and safety of commonly used anti-osteoporotic agents in men.MethodsWe systematically searched PubMed, Web of Science, and the Cochrane Library from inception to December 2025 for RCTs enrolling men with primary osteoporosis. Eligible interventions included oral bisphosphonates (OBP), intravenous bisphosphonates (IBP), abaloparatide (ABA), denosumab (DEN), teriparatide (TER), and odanacatib (ODN), compared with placebo/usual care (PLA/CTRL) or active treatments. Primary efficacy outcomes were changes in bone mineral density (BMD) at the lumbar spine (LS), total hip (TH), and femoral neck (FN), preferentially extracted at the time point closest to 12 months (range: 6–24 months). Safety outcomes were any adverse events (AEs) and serious adverse events (SAEs). Pairwise meta-analyses were conducted in Stata 18.0, and NMA in R 4.3.1 under a consistency framework. Treatments were ranked using SUCRA. The protocol was registered in PROSPERO (CRD420261279572).ResultsTwenty-one RCTs (n = 4,409) were included. Compared with PLA/CTRL, TER and ABA significantly improved FN-BMD (TER: MD 2.97, 95% CI 0.74–5.20; ABA: MD 2.83, 95% CI 0.11–5.55), and OBP also showed benefit (MD 1.61, 95% CI 0.15–3.07). For TH-BMD, significant gains were observed for OBP (MD 2.25, 95% CI 1.70–2.81), ABA (MD 2.13, 95% CI 1.43–2.83), and TER (MD 1.12, 95% CI 0.35–1.89); OBP showed a greater increase than TER (MD 1.13, 95% CI 0.32–1.95). For LS-BMD, ABA (MD 7.31, 95% CI 4.25–10.37), TER (MD 6.97, 95% CI 4.53–9.42), and OBP (MD 3.43, 95% CI 1.84–5.02) were superior to PLA/CTRL, and both ABA and TER outperformed OBP (ABA vs OBP: MD 3.88, 95% CI 0.44–7.33; TER vs OBP: MD 3.55, 95% CI 1.08–6.01). Most head-to-head comparisons across active agents were inconclusive with wide uncertainty. Node-splitting suggested no significant local inconsistency (P > 0.05). Safety comparisons for AEs and SAEs were largely imprecise; SUCRA rankings should therefore be interpreted cautiously.ConclusionsIn men with primary osteoporosis, treatment effects on BMD differ by skeletal site: ABA and TER yield larger gains at the lumbar spine, whereas OBP demonstrates robust improvements at the total hip; evidence for femoral neck improvement is clearer for TER and ABA versus PLA/CTRL. However, the extent to which these BMD improvements translate into fracture risk reduction remains uncertain because direct fracture endpoint data were limited in the included RCTs. Safety evidence also remains limited due to imprecision, and SUCRA rankings should therefore be interpreted alongside effect estimates and their uncertainty. Further high-quality, long-term RCTs focusing on fracture outcomes and drug-specific adverse events are warranted.