Optimizing Denosumab Therapy in Rheumatic Diseases: Considerations for Disease Activity, Glucocorticoid Use, and Sequential Treatment

To the Editor:

We read with great interest the article by Furukawa et al, titled “Long-Term Effectiveness and Safety of Denosumab for Osteoporosis in Patients With Rheumatic Diseases.”¹ The authors’ comprehensive evaluation of bone mineral density (BMD) changes over 7 years in patients with rheumatic diseases (RDs) provides critical insights into the long-term efficacy and safety of denosumab in this high-risk population. Their findings, particularly the sustained BMD improvement despite glucocorticoid (GC) use and high retention rate (68.1% at 7 years), significantly advanced our understanding of osteoporosis management in RDs. However, we would like to raise several points for further discussion and clarification to enhance the impact of this valuable study.

First, this study included a diverse cohort of patients with RDs (eg, rheumatoid arthritis, systemic lupus erythematosus, and polymyalgia rheumatica). Although this reflects real-world clinical practice, the potential differential effects of denosumab across these conditions, given the variations in disease activity, chronic inflammation, and concomitant therapies, have not been explored. For instance, disease-modifying antirheumatic drugs and biologics can modulate receptor activator of nuclear factor-κB ligand (RANKL) expression independently of GCs.^2-4 Stratified analyses by specific RDs or inflammatory markers (eg, C-reactive protein, interleukin 6) could elucidate whether the efficacy of denosumab is influenced by the underlying disease activity.

Second, the observation that GC doses ≥ 5 mg/day initially attenuated BMD improvement is clinically important. However, this study did not account for cumulative GC exposure or the potential modifying effects of GC-sparing agents (eg, hydroxychloroquine and biologics). A subanalysis comparing patients on stable GC regimens vs those with dose fluctuations might clarify whether GC variability affects denosumab’s effectiveness. Additionally, the Cox proportional hazards model for an 8% BMD increase could be expanded to include the interactions between GC dose and disease duration.

Third, the reported incidence of osteonecrosis of the jaw (ONJ; 2/165 [1.2%]) and serious infections (24/165 [14.5%]) warrants further attention. Although the authors attributed infections to GC/immunosuppressant use, a deeper analysis of risk factors (eg, diabetes, prior fractures, or concomitant biologic use) could help identify high-risk subgroups. Further, the absence of atypical femoral fractures is reassuring but merits caution because of the limited sample size for femoral BMD assessments (n = 48 at year 1). Longer follow-up and larger cohorts are required to confirm this finding.

Fourth, the study highlights BMD loss after denosumab discontinuation; however, data on alternative therapies (eg, bisphosphonates and teriparatide) were limited. Given the recent guidelines emphasizing sequential therapy to prevent rebound fractures,^5-7 a protocol for transitioning patients off denosumab, particularly in RD populations, would be invaluable. For example, could early bisphosphonate overlap mitigate bone loss in these patients?

Finally, prospective studies should compare denosumab head-to-head with other antiresorptive agents (eg, zoledronate) in RDs. Fracture risk assessment tools (eg, World Health Organization Fracture Risk Assessment Tool [FRAX]) should be integrated to personalize treatment decisions, and multidisciplinary strategies (eg, dental screening for ONJ and vaccination protocols) should be explored to optimize safety.

In conclusion, Furukawa et al made a seminal contribution by demonstrating the long-term benefits of denosumab in RD-associated osteoporosis.¹ Addressing these issues could further refine its use in this vulnerable population. We commend the authors for their rigorous work and look forward to future research based on these findings.

• Copyright © 2026 by the Journal of Rheumatology