Clinical Trials Express: fracture risk reduction with denosumab in Japanese postmenopausal women and men with osteoporosis: denosumab fracture intervention randomized placebo controlled trial (DIRECT)

Abstract

Context: Denosumab 60 mg sc injection every 6 months for 36 months was well tolerated and effective in reducing the incidence of vertebral, nonvertebral, and hip fracture in predominantly Caucasian postmenopausal women with osteoporosis.

Objective: The objective of this phase 3 fracture study was to examine the antifracture efficacy and safety of denosumab 60 mg in Japanese women and men with osteoporosis compared with placebo.

Design and setting: A randomized, double-blind, placebo-controlled trial with an open-label active comparator as a referential arm was conducted.

Patients: Subjects were 1262 Japanese patients with osteoporosis aged 50 years or older, who had one to four prevalent vertebral fractures.

Intervention: Subjects were randomly assigned to receive denosumab 60 mg sc every 6 months (n = 500), placebo for denosumab (n = 511), or oral alendronate 35 mg weekly (n = 251). All subjects received daily supplements of calcium and vitamin D.

Main outcome measure: The primary endpoint was the 24-month incidence of new or worsening vertebral fracture for denosumab vs placebo.

Results: Denosumab significantly reduced the risk of new or worsening vertebral fracture by 65.7%, with incidences of 3.6% in denosumab and 10.3% in placebo at 24 months (hazard ratio 0.343; 95% confidence interval 0.194-0.606, P = .0001). No apparent difference in adverse events was found between denosumab and placebo during the first 24 months of the study.

Conclusion: These results provide evidence of the efficacy and safety of denosumab 60 mg sc every 6 months in Japanese subjects with osteoporosis.

Trial registration: ClinicalTrials.gov NCT00680953.

Figure 1.

Disposition of study subjects. *, Three subjects were excluded due to lack of efficacy data after IP administration; **, one subject was excluded due to lack of efficacy data after IP administration. FAS, full analysis set.

Figure 2.

Cumulative incidence of vertebral fracture. The cumulative incidence of new or worsening vertebral fracture (A) and new vertebral fracture (B) in all subjects in the denosumab and placebo groups. For panels A and B, the percentages given were incidence by the Kaplan-Meier estimate over a 24-month treatment period. The HR (95% CI) and P value of new or worsening vertebral fracture (A) for the denosumab vs the placebo group was 0.343 (0.194, 0.606) and P = .0001. The HR (95% CI) and P value of a new vertebral fracture (B) for the denosumab vs placebo group was 0.260 (0.129, 0.521) and P < .0001. In the subgroup analysis in men at 24 months, the incidence of new or worsening vertebral fracture in the denosumab and placebo groups was 0% and 12.5%, respectively, and that of new vertebral fracture in the denosumab and placebo groups was 0% and 8.3%, respectively. The P value of a new or worsening vertebral fracture and new vertebral fracture for the denosumab vs placebo group was P = .0748 and P = .1478, respectively. The HRs (95% CI) were not estimated because there were no subjects with a vertebral fracture in the denosumab group.

Figure 3.

Mean BMD percentage changes from baseline. Mean BMD percentage changes from baseline over a 24-month treatment period at the lumbar spine (L1–L4) (A), total hip (B), femoral neck (C), and distal one third radius (D) are shown. a, P < .0001 based on the Student's t test at each time point for the denosumab vs placebo group; b, a referential comparison, P < .01 based on the Student's t test at each time point for the alendronate vs placebo group; c, P < .05 based on the Student's t test at each time point for the denosumab vs alendronate group. The bars show 95% CIs of the mean values at each time point. A central vender performed all BMD analyses. Abnormal vertebrae, such as those with an abnormality, fracture, or artifact, were excluded from analyses.

Figure 4.

Median BTM percentage changes from baseline. Median BTM changes from baseline over a 24-month treatment period for serum CTX-1 (A) and serum BSAP (B) are shown. a, P < .0001 based on the Wilcoxon rank-sum test for the denosumab vs placebo group; b, as a referential comparison, P < .01 based on the Wilcoxon rank-sum test at each time point for the alendronate vs placebo group; c, P < .05 based on the Wilcoxon rank-sum test at each time point for the denosumab vs alendronate group. The bars show the interquartile range of the percentage changes from baseline at each time point.