Summary: In a small randomised trial of 39 patients with rheumatic disease newly starting high-dose glucocorticoids, median lumbar spine bone mineral density rose by 8.6% at 12 months with romosozumab, 3.3% with denosumab and fell by 0.4% with a bisphosphonate. Fractures were not assessed.
PICO Summary
| Element | Detail |
|---|---|
| Population | 39 patients with rheumatic disease, previously untreated for osteoporosis, newly starting prednisolone 15 mg/day or more; randomised trial, Japan (UMIN000037239). |
| Intervention | Romosozumab (n=11). |
| Comparison | Denosumab (n=14) or a bisphosphonate, risedronate (n=14). |
| Outcome | Median (25th to 75th percentile) percent change in lumbar spine BMD at 12 months: romosozumab 8.6% (3.1 to 12.4), denosumab 3.3% (1.5 to 6.2), bisphosphonate -0.4% (-3.4 to 1.1). Bone-specific alkaline phosphatase rose slightly with romosozumab, while P1NP and osteocalcin fell in all groups, less so with romosozumab. Bone resorption markers fell in all groups. |
Expert Commentary
This small randomised comparison suggests romosozumab is a strong candidate for glucocorticoid-induced osteoporosis, and I read it as promising but preliminary. In 39 patients newly starting prednisolone at 15 mg/day or more, median lumbar spine BMD rose by 8.6% at 12 months with romosozumab, 3.3% with denosumab and fell by 0.4% with a bisphosphonate. Bone-formation markers behaved as expected for a sclerostin inhibitor, with smaller falls than in the other groups, and resorption markers decreased in all arms. The important limitation is that BMD is a surrogate and the trial, with 11 to 14 patients per arm, was not designed to show fracture reduction or to establish cardiovascular safety in people with rheumatic disease. Can I use this with my patients? Not yet as routine practice. For patients at high fracture risk who are starting high-dose glucocorticoids and have no cardiovascular contraindication, it supports considering an anabolic option within established guidance and with specialist input. I would like to see a larger trial with fractures and cardiovascular events as endpoints.
References
Kawazoe M, Kaneko K, Masuoka S, et al. Comparison of efficacy of romosozumab with denosumab and risedronate in patients newly initiating glucocorticoid therapy. J Clin Endocrinol Metab. 2025;110(8):e2778-e2786. doi:10.1210/clinem/dgae810. PMID: 39552045.
