Umeå universitets logga

It is important to identify patients at highest risk of fractures. The aim of the present study was to compare the separate and combined performances of three major bone-related genetic rick scores (GRSs) for prediction of hip, wrist and vertebral fractures as well as of trabecular and cortical bone mass separately.Recently as many as 1,103 independent single nucleotide polymorphisms (SNP) associated with estimated bone mineral density of the heel (eBMD) were identified. Using these SNPs, we developed a weighted GRS for eBMD (eBMDGRS) and determined if it contributes information for fracture prediction beyond two previously developed GRSs for femur neck BMD (FN-BMDGRS, 49 SNPs) and lumbar spine BMD (LS-BMDGRS, 48 SNPs). Associations between the three GRSs and wrist (ncases = 1,037; ncontrols = 2,854), hip (ncases = 1,106; ncontrols = 2,602) and radiographic vertebral (ncases = 288; ncontrols = 1,187) fractures were evaluated in the UFO-fracture and MrOS Sweden cohorts. Associations between the three GRSs and trabecular and cortical bone parameters analysed by HRpQCT in the distal radius (n=426) were evaluated in the MrOS Sweden cohort.Although all three GRSs were significantly associated with all three fracture types, eBMDGRS was the strongest predictor of wrist and vertebral fractures while the strengths of the associations for eBMDGRS and FN-BMDGRS with hip fracture risk were similar. In combined GRSs models, eBMDGRS was the only significant predictor of wrist (Odds ratio [OR] = 1.46; 95% CI 1.33-1.59 per SD increase) and vertebral (OR = 1.32; 95% CI 1.16-1.51) fractures while both eBMDGRS and FN-BMDGRS were significant independent predictors of hip fracture risk (eBMDGRS OR = 1.22; 95% CI 1.11-1.34; FN-BMDGRS OR = 1.16; 95% CI 1.06-1.27).eBMDGRS was the major GRS contributing to prediction of trabecular BMD (Variance explained BMD 13.7%) while both eBMDGRS and FN-BMDGRS contributed information for prediction of cortical bone area (eBMDGRS 4.6%; FN-BMDGRS 4.1%).In conclusion, eBMDGRS is the only independent GRS for prediction of wrist and vertebral fractures while both eBMDGRS and FN-BMDGRS contribute independent information for prediction of hip fractures. We propose that eBMDGRS captures unique information about trabecular bone useful for prediction of wrist and vertebral fractures. The findings in the present study may facilitate personalized medicine to predict different fracture types as well as cortical and trabecular bone mass separately.