Diabetes Mellitus and the Benefit of Antiresorptive Therapy on Fracture Risk

Abstract

Type 2 diabetes (T2D) is associated with increased risk of fractures. However, it is unclear whether current osteoporosis treatments reduce fractures in individuals with diabetes. The aim of the study was to determine whether presence of T2D influences the efficacy of antiresorptive treatment for osteoporosis using the Foundation for the National Institutes of Health (FNIH)-American Society for Bone and Mineral Research (ASBMR)-Study to Advance Bone Mineral Density (BMD) as a Regulatory Endpoint (SABRE) cohort, which includes individual patient data from randomized trials of osteoporosis therapies. In this study we included 96,385 subjects, 6.8% of whom had T2D, from nine bisphosphonate trials, two selective estrogen receptor modulator (SERM) trials, two trials of menopausal hormone therapy, one denosumab trial, and one odanacatib trial. We used Cox regression to obtain the treatment hazard ratio (HR) for incident nonvertebral, hip, and all fractures and logistic regression to obtain the treatment odds ratio (OR) for incident morphometric vertebral fractures, separately for T2D and non-DM. We used linear regression to estimate the effect of treatment on 2-year change in BMD (n = 49,099) and 3-month to 12-month change in bone turnover markers (n = 12,701) by diabetes status. In all analyses, we assessed the interaction between treatment and diabetes status. In pooled analyses of all 15 trials, we found that diabetes did not impact treatment efficacy, with similar reductions in vertebral, nonvertebral, all, and hip fractures, increases in total hip and femoral neck BMD, and reductions in serum C-terminal cross-linking telopeptide (CTX), urinary N-telopeptide of type I collagen/creatinine (NTX/Cr) and procollagen type 1 N propeptide (P1NP) (all interactions p > 0.05). We found similar results for the pooled analysis of bisphosphonate trials. However, when we considered trials individually, we found a few interactions within individual studies between diabetes status and the effects of denosumab and odanacatib on fracture risk, change in BMD or bone turnover markers (BTMs). In sum, these results provide strong evidence that bisphosphonates and most licensed antiresorptive drugs are effective at reducing fracture risk and increasing BMD irrespective of diabetes status. © 2022 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Keywords: ANTIRESORPTIVES; BIOCHEMICAL MARKERS OF BONE REMODELING; CLINICAL TRIALS; DXA; OSTEOPOROSIS.

Fig. 1

Forest plots showing the effects of treatment on fracture risk in T2D (solid circle) and non‐DM (solid square). a. vertebral fractures, b. non‐vertebral fractures and c. all fractures. The p values for T2D status*treatment interaction for each trial, the overall effects, and the p value for heterogeneity of T2D status*treatment interaction across trials are all shown. *unestimable: no fractures in active and/or placebo groups in T2D.

Fig. 2

Forest plots showing the effects of treatment on change in BMD and BTM in T2D (solid circle) and non‐DM (solid square). a. change in total hip BMD, b. change in femoral neck BMD, c. change in lumbar spine BMD and d. change in CTX, e. change in P1NP, f. change in NTX/Cr. The p values for T2D status*treatment interaction for each trial, the overall effects, and the p value for heterogeneity of T2D status*treatment interaction across trials are all shown.