Advanced glycation end-products and sclerostin are elevated across diabetes phenotypes and associated with vertebral fractures in type 2 diabetes

BACKGROUND: To investigate the associations of circulating advanced glycation end-products (AGEs; carboxymethyllysine and pentosidine) and sclerostin with bone mineral density (BMD) and vertebral fractures in adults with type 1 diabetes mellitus (T1DM), type 2 diabetes mellitus (T2DM), and latent autoimmune diabetes in adults (LADA). METHODS: In this cross-sectional case–control study, 76 patients with T1DM, 91 with T2DM, and 40 with LADA were compared with 85 non-diabetic controls. Anthropometric variables, biochemical parameters, and bone turnover markers (osteocalcin and β-CTX) were assessed. Serum AGEs and sclerostin were quantified by ELISA. Areal BMD was measured by dual-energy X-ray absorptiometry (DXA) at the lumbar spine (L1–L4) and femoral neck. Vertebral fractures were identified on spine radiographs using Genant’s semiquantitative method. Multivariable analyses were performed to determine independent predictors of vertebral fractures. RESULTS: Serum carboxymethyllysine, pentosidine, and sclerostin concentrations were significantly higher, while osteocalcin and β-CTX were significantly lower, in all diabetes groups versus controls (all p < 0.001). Vertebral fractures were present in 14.49% of participants with T1DM, 34.2% with T2DM, and 27.5% with LADA. In fracture-stratified analyses within the T2DM group, participants with vertebral fractures were older and had higher carboxymethyllysine levels than those without fractures. In multivariable analysis restricted to the T2DM group, age (AOR 1.079; 95% CI 1.019–1.143; p = 0.009) and pentosidine (AOR 1.848; 95% CI 1.030–3.316; p = 0.040) were independently associated with vertebral fractures, whereas carboxymethyllysine was not. CONCLUSION: Elevated AGEs and sclerostin together with suppressed bone turnover markers were observed across diabetes phenotypes and are consistent with altered bone metabolism. In T2DM, age and pentosidine were independently associated with prevalent vertebral fractures; however, given the cross-sectional design, these associations should not be interpreted as evidence of a direct mechanistic pathway. Longitudinal studies are needed to clarify whether these biomarker patterns reflect diabetes severity, skeletal fragility, or both. CLINICAL TRIAL NUMBER: Not applicable.