Abstract

Background No clinical trials have compared osteoporosis drugs with incident fractures as the primary outcome. We compared the anti-fracture efficacy of teriparatide with risedronate in patients with severe osteoporosis. Methods In this double-blind, double-dummy trial, we enrolled post-menopausal women with at least two moderate or one severe vertebral fracture and a bone mineral density T score of less than or equal to −1·50. Participants were randomly assigned to receive 20 μg of teriparatide once daily plus oral weekly placebo or 35 mg of oral risedronate once weekly plus daily injections of placebo for 24 months. The primary outcome was new radiographic vertebral fractures. Secondary, gated outcomes included new and worsened radiographic vertebral fractures, clinical fractures (a composite of non-vertebral and symptomatic vertebral), and non-vertebral fractures. This study is registered with ClinicalTrials.gov (NCT01709110) and EudraCT (2012-000123-41). Findings We enrolled 680 patients in each group. At 24 months, new vertebral fractures occurred in 28 (5·4%) of 680 patients in the teriparatide group and 64 (12·0%) of 680 patients in the risedronate group (risk ratio 0·44, 95% CI 0·29–0·68; p<0·0001). Clinical fractures occurred in 30 (4·8%) of 680 patients in the teriparatide group compared with 61 (9·8%) of 680 in the risedronate group (hazard ratio 0·48, 95% CI 0·32–0·74; p=0·0009). Non-vertebral fragility fractures occurred in 25 (4·0%) patients in the teriparatide group and 38 (6·1%) in the risedronate group (hazard ratio 0·66; 95% CI 0·39–1·10; p=0·10). Interpretation Among post-menopausal women with severe osteoporosis, the risk of new vertebral and clinical fractures is significantly lower in patients receiving teriparatide than in those receiving risedronate. Funding Lilly.