Long-term Effect of Denosumab on Bone Microarchitecture as Assessed by Tissue Thickness-Adjusted Trabecular Bone Score (TBS) in Postmenopausal Women with Osteoporosis: Results from the Freedom and Open-Label Extension (OLE)

Abstract

TBS is an indirect measure of bone microarchitecture assessed on lumbar spine (LS) DXA scans and provides information beyond bone mineral density (BMD). TBS algorithm has been updated to account for regional soft tissue noise in DXA images. This retrospective analysis applied updated built-in tissue thickness-adjusted TBS algorithm to investigate the long-term effect of denosumab on bone microarchitecture in FREEDOM and OLE. This analysis included 279 postmenopausal women with LS or total hip BMD T-score <−2.5 and ≥−4.0 who completed the FREEDOM DXA sub study and continued in the OLE study: 150 women received denosumab 60 mg subcutaneously every 6 months for 3 years and same-dose open-label denosumab for 7 years (long-term denosumab group); 129 women received placebo for 3 years and open-label denosumab for 7 years (crossover denosumab group). BMD and TBS were assessed on LS DXA scans, blinded from treatment allocation, at FREEDOM: baseline, Month 1, and Years 1, 2, and 3; and at OLE: baseline, Years 1, 2, 3, 5, and 7. Baseline characteristics were similar between groups. Long-term denosumab led to significant and progressive increases in TBS over 10 years of treatment (Figure). A similar trend was observed in the crossover group during 7 years of denosumab therapy. In the long-term denosumab group, % of patients with normal microarchitecture (TBS >1.074) increased from 26.1% at baseline to 53.2% up to Year 10, and % of patients with degraded (TBS ≤1.027) or partially degraded (1.027< TBS≤1.074) microarchitecture decreased from 48.6% to 29.1% and from 25.4% to 17.7%, respectively (P < 0.0001; TBS thresholds equivalent to 1.230 and 1.310 for the classical TBS algorithm corrected for body mass index). A similar improvement in bone microarchitecture was observed in the crossover group from OLE baseline up to OLE Year 7 (P < 0.0001). Over the course of long-term denosumab treatment, TBS changes were largely unrelated to LS BMD changes: r2 was 0.05 from baseline to Year 10 in the long-term group and 0.28 from OLE baseline to OLE Year 7 in the crossover group. Up to 10 years of denosumab treatment significantly and progressively improved TBS assessment of bone microarchitecture independently of BMD in postmenopausal women with osteoporosis.