Romosozumab versus denosumab in long‐term users of glucocorticoids: A pilot randomized controlled trial

Abstract Objective To compare the efficacy of romosozumab (ROMO) and denosumab (DEN) in prevalent long‐term glucocorticoid (GC) users. Methods Adult patients receiving oral prednisolone (≥5 mg/day) with high risk of fracture were randomized to receive subcutaneous ROMO (210 mg monthly) or DEN (60 mg 6‐monthly) for 12 months, followed by DEN for two more doses. The primary end point was the change in spine bone mineral density (BMD) from Months 0 to 12. Secondary end points included changes in BMD of the spine/hip/femoral neck and bone turnover markers at various time points and adverse events. Results Seventy patients (age 62.6 ± 9.1 years; 96% women; median prednisolone dose 5.0 mg/day; duration of therapy 10.7 ± 7.4 years) were enrolled, and 63 completed the study. At Month 12, the spine BMD increased significantly in both ROMO (+7.3% ± 4.5%; p < 0.001) and DEN (+2.3% ± 3.1%; p < 0.001) groups. The absolute spine BMD gain from Months 0 to 12 was significantly greater in ROMO‐treated patients ( p < 0.001). Although the total hip BMD at Month 12 also increased significantly in the ROMO (+1.6% ± 3.3%; p = 0.01) and DEN groups (+1.6% ± 2.6%; p = 0.003), the absolute BMD gain was not significantly different between the groups. At Month 24, the spine BMD continued to increase in both the ROMO (+9.7% ± 4.8%; p < 0.001) and DEN group (+3.0% ± 3.0%; p < 0.001) compared to baseline, and the absolute BMD gain remained significantly greater in ROMO‐treated patients. The total hip BMD continued to increase in both groups (ROMO +2.9% ± 3.7%; p < 0.001; DEN +2.2% ± 3.4%; p = 0.001), but the changes from baseline were similar. Injection site reaction was more frequently reported in ROMO‐treated patients. Conclusion ROMO was superior to DEN in raising the spine BMD at Month 12 in chronic GC users. After switching to DEN, ROMO‐treated patients continued to gain spine BMD to a greater extent than DEN until Month 24.