Integrative analysis of genetic and clinical risk factors for bone loss in a Korean population.

PURPOSE: The relative contribution of genetic and clinical factors for bone loss is not well known. This study aimed to investigate the annualized percentage change in total hip bone mineral density (BMD) and the genetic and clinical risk factors for bone loss in a Korean prospective cohort study over a 6-year period.

METHODS

We included 645 men aged ≥50 years and 683 postmenopausal women who had repeated BMD testing between 2007 and 2014. The association between covariates and annualized percentage change in hip BMD was analyzed through the multivariate linear regression analysis. A total of 2614 single-nucleotide polymorphisms (SNPs) from 23 known BMD-related candidate genes and genome-wise association study were investigated.

RESULTS

Hip bone loss increased more rapidly in women than in men with advancing age. Hip bone loss in men increased with lean mass (LM) loss (%/year) (P < 0.001) and current smoking (P = 0.024) and decreased with increasing waist circumference (WC) (P < 0.001), alcohol consumption (P = 0.049), and increase in red blood cell counts (P = 0.031). Decreasing WC (P = 0.009), LM loss (%/year) (P < 0.001), and years since menopause ≤ 3 years (P = 0.003) significantly correlated with hip bone loss in women aged 45-59 years. Hip bone loss in women aged ≥60 years increased with advancing age (P = 0.012), alcohol consumption (P = 0.028), LM loss (%/year) (P = 0.031), and fat mass loss (%/year) (P < 0.001) and decreased with increasing WC (P = 0.025). LRP5 rs498830 (β = 0.127, P = 0.007) and TNFSF11 rs7325635 (β = 0.146, P = 0.001) were the top SNPs related to hip bone loss in men and postmenopausal women, respectively. However, none of the SNPs were associated with hip bone loss after Benjamini-Hochberg adjustment.

CONCLUSION

In this study, decreasing WC and LM were significant risk factors for hip bone loss in both men and women. Those factors were also identified that had sex-specific or age-specific effects on hip bone loss. None of the SNPs were associated with hip bone loss after multiple testing adjustments. The understanding of the modifiable factors contributing to bone loss has been broadened, and this may have implications such as in developing individualized preventive strategy. Further studies are needed to better predict the risk for bone loss in men and women.