Romosozumab Improves Tissue Thickness-Adjusted Trabecular Bone Score in Women With Osteoporosis and Diabetes

Abstract

Context: Trabecular bone score (TBS), a gray-level texture index derived from lumbar spine (LS) dual-energy x-ray absorptiometry (DXA) scans, is decreased in patients with diabetes and is associated with increased fracture risk, independent of areal bone mineral density (aBMD), but potentially influenced by abdominal fat tissue.

Objective: Evaluate effect of romosozumab (210 mg monthly) for 12 months followed by alendronate (70 mg weekly) for 24 months vs alendronate alone (70 mg weekly) for 36 months on LS aBMD and TBS in women with type 2 diabetes (T2D) enrolled in the ARCH study.

Methods: This post hoc analysis included women from ARCH who had T2D at baseline and LS DXA scans at baseline and ≥1 postbaseline visit (romosozumab-to-alendronate, n = 165; alendronate-to-alendronate, n = 195). aBMD and TBS (determined by an updated tissue thickness-adjusted TBS algorithm [TBSTT]) were assessed on LS DXA scans at baseline and ≥1 postbaseline visit (months 12, 24, and 36).

Results: Romosozumab led to significantly greater gains in LS aBMD and TBSTT at month 12 vs alendronate, and the greater gains with romosozumab were maintained after transition to alendronate and persisted significantly at months 24 and 36 vs alendronate alone. TBSTT percentage changes weakly correlated to LS aBMD percentage changes from baseline to month 36 (romosozumab-to-alendronate, R2 = 0.1493; alendronate-to-alendronate, R2 = 0.0429).

Conclusion: In postmenopausal women with osteoporosis and T2D, 12 months of romosozumab followed by 24 months of alendronate vs alendronate alone significantly improved LS aBMD and TBSTT (independently of abdominal fat) and to a greater extent. Hence, romosozumab may improve bone strength in patients with T2D.

Trial registration: ClinicalTrials.gov-NCT01631214.

Keywords: DXA; anabolic; bone mineral density; fracture risk; osteoporosis; type 2 diabetes (T2D).

Figure 1.

Study design and diabetes analysis set. Eligibility criteria for ARCH included age 55 to 90 years and 1 of the following: T-score ≤−2.5 at the total hip or femoral neck and either ≥1 moderate or severe vertebral fracture or ≥2 mild vertebral fractures or T-score ≤−2.0 at the total hip or femoral neck and either ≥2 moderate or severe vertebral fractures or a fracture of the proximal femur sustained 3 to 24 months before randomization. Of the 195 patients who received blinded alendronate, 99.5% (194/195) had T2D and 0.5% (1/195) had T1D; of the 165 patients who received blinded romosozumab, 98.8% (163/165) had T2D and 1.2% (2/165) had T1D. ^aHad a medical history of hyperglycemia/new-onset diabetes mellitus with only narrow scope terms at baseline. ^bPatients with BMI >38 kg/m² or <15 kg/m² were excluded as that was out of the range for a proper TBS assessment; patients were also excluded if the DXA system or scan acquisition mode used for aBMD assessment was not compatible with TBS computation. Abbreviations: aBMD, areal bone mineral density; BMD, bone mineral density; BMI, body mass index; DXA, dual-energy x-ray absorptiometry; LS, lumbar spine; PO, orally; QM, monthly; QW, weekly; SC, subcutaneously; T1D, type 1 diabetes; T2D, type 2 diabetes; TBS, trabecular bone score; TBS_TT, tissue thickness–adjusted TBS.

Figure 2.

Percentage change from baseline to months 12, 24, and 36 by visit and treatment group for (A) LS aBMD and (B) LS TBS_TT in the T2D subgroup. n = number of patients with aBMD and TBS_TT measurements. The LS aBMD and TBS_TT data were analyzed based on a repeated measures model adjusting for treatment, presence of severe vertebral fracture at baseline, visit, treatment-by-visit interaction, baseline aBMD, or TBS value as fixed effects, with DXA machine type and baseline aBMD or TBS value-by-machine type interaction as covariates. Abbreviations: aBMD, areal bone mineral density; CI, confidence interval; Diff, percentage change from baseline for romosozumab treatment group minus percentage change from baseline for alendronate treatment group; DXA, dual-energy x-ray absorptiometry; LS, lumbar spine; PO, orally; QM, monthly; QW, weekly; SC, subcutaneously; T2D, type 2 diabetes; TBS, trabecular bone score; TBS_TT, tissue thickness–adjusted trabecular bone score.

Figure 3.

Percentage of patients by LS TBS_TT risk category at baseline and months 12, 24, and 36 in the alendronate-to-alendronate and romosozumab-to-alendronate groups in the subpopulation with T2D in ARCH (A) over time and (B) by TBS_TT risk category at each time point. n = number of women with mean calibrated TBS_TT measurements at baseline and months 12, 24, and 36. *P < .001 vs baseline based on Bhapkar's test for homogeneity. Abbreviations: LS, lumbar spine; T2D, type 2 diabetes; TBS_TT, tissue thickness–adjusted trabecular bone score.

Figure 4.

Relationship between LS aBMD and TBS_TT percentage change from baseline to months 12, 24, and 36 in (A) the alendronate alone group and (B) the romosozumab-to-alendronate group. Data analyzed for patients with aBMD and TBS_TT measurements at baseline and months 12, 24, or 36. Pearson correlation: R² = 0.0025 at month 12, R² = 0.0107 at month 24, and R² = 0.0429 at month 36 in the alendronate alone group; R² = 0.0399 at month 12, R² = 0.0751 at month 24, and R² = 0.1493 at month 36 in the romosozumab-to-alendronate group. Percentage of values are shown in each quadrant. Abbreviations: aBMD, areal bone mineral density; LS, lumbar spine; R², correlation coefficient; TBS_TT, tissue thickness–adjusted trabecular bone score.