Combining fracture outcomes in phase 3 trials of osteoporosis : An analysis of the effects of denosumab in postmenopausal women

Summary
This paper explores use of metrics that combine fracture outcomes that add power to phase 3 studies and provide a surrogate outcome for regulatory agencies.

Introduction
The aim of this study was to develop an analytic framework that would combine information from all fracture outcomes (including radiographic vertebral fractures) in phase 3 studies to provide a metric for the assessment of treatment efficacy.

Methods
Data from the phase 3 study of denosumab were used as an exemplar comparing the effects of active intervention with placebo on the risk of all fractures associated with osteoporosis. Fracture outcomes were assigned utility weights drawn from the published literature and applied to age-specific health state values of the general population. For each fracture outcome in each arm of the study, cumulative disutility was computed to serve as the principal end point. The hypothesis tested was that treatment with denosumab results in a significant reduction in mean fracture-related disutility.

Results
Treatment with denosumab was associated with significantly lower utility loss compared with placebo. For patients treated with denosumab, mean utility loss was 42% less than with placebo (4.5 vs. 7.5 QALYs/1000 patient years, respectively, p < 0.001).

Conclusions
Denosumab significantly decreased utility loss. The use of metrics that combine fracture outcomes may provide added power to phase 3 studies and provide a surrogate outcome for regulatory agencies.