Romosozumab in patients who experienced an on-study fracture: post hoc analyses of the FRAME and ARCH phase 3 trials

Summary

Post hoc analysis of FRAME and ARCH revealed that on-study nonvertebral and vertebral fractures by Month 12 were less common in women initially treated with romosozumab versus placebo or alendronate. Recurrent fracture risk was also lower in romosozumab‑treated patients, and there were no fracture‑related complications. Results support continuing romosozumab treatment post‑fracture.

Purpose

Post hoc analysis evaluating efficacy and safety of romosozumab, administered in the immediate post‑fracture period, in the FRAME and ARCH phase 3 trials.

Methods

In FRAME (NCT01575834) and ARCH (NCT01631214), postmenopausal women with osteoporosis were randomized 1:1 to romosozumab 210 mg monthly or comparator (FRAME, placebo; ARCH, alendronate 70 mg weekly) for 12 months, followed by antiresorptive therapy (FRAME, denosumab; ARCH, alendronate). In patients who experienced on-study nonvertebral or new/worsening vertebral fracture by Month 12, we report the following: fracture and treatment‑emergent adverse event (TEAE) incidence through 36 months, bone mineral density changes (BMD), and romosozumab timing. Due to the sample sizes employed, meaningful statistical comparisons between treatments were not possible.

Results

Incidence of on-study nonvertebral and vertebral fractures by Month 12 was numerically lower in romosozumab- versus comparator-treated patients (FRAME, 1.6% and 0.5% versus 2.1% and 1.6%; ARCH, 3.4% and 3.3% versus 4.6% and 4.9%, respectively). In those who experienced on-study nonvertebral fracture by Month 12, recurrent nonvertebral and subsequent vertebral fracture incidences were numerically lower in patients initially treated with romosozumab versus comparator (FRAME, 3.6% [2/56] and 1.8% [1/56] versus 9.2% [7/76] and 3.9% [3/76]; ARCH, 10.0% [7/70] and 5.7% [4/70] versus 12.6% [12/95] and 8.4% [8/95], respectively). Among those with on-study vertebral fracture by Month 12, recurrent vertebral and subsequent nonvertebral fracture incidences were numerically lower with romosozumab versus comparator (FRAME, 0.0% [0/17] and 0.0% [0/17] versus 11.9% [7/59] and 8.5% [5/59]; ARCH, 9.0% [6/67] and 7.5% [5/67] versus 15.0% [15/100] and 16.0% [16/100], respectively). In patients with fracture by Month 12, no fracture‑related complications were reported in romosozumab-treated patients. BMD gains were numerically greater with romosozumab than comparators.

Conclusion

Data suggest support for the efficacy and safety of continuing romosozumab treatment following fracture.

Trial registrations

NCT01575834; NCT01631214.