Prediction of Future Osteoporotic Fracture Occurrence by Genetic Profiling: A 6-Year Follow-Up Observational Study.

Abstract

Heredity is an important risk factor for osteoporotic fracture, but it remains unclear whether genetic factors improve the predictability of future fracture occurrence. To compare an integration model of genetic profiling with the current model for predicting future fracture occurrence. A retrospective observational cohort study. Postmenopausal women aged 45-93 years who were untreated (n = 117), hormone-treated (n = 491), or bisphosphonate (BP)-treated (n = 415), with a mean 6.1-year follow-up. The main outcome was incident fractures. Ninety-five single nucleotide polymorphisms were genotyped. We calculated the Korean-specific genetic risk score 35 (GRS35) from 35 single nucleotide polymorphisms associated with osteoporosis-related traits at the baseline visit. Osteoporotic fracture occurred more frequently in the highest GRS35 tertile group than in the lower two tertile groups after adjustments for confounders (hazard ratio [HR], 1.73; 95% confidence interval [CI], 1.17-2.55). The associations of the GRS35 with incident fracture were only significant in the BP group (HR, 2.25; 95% CI, 1.28-3.95) and not in the untreated (HR, 1.26; 95% CI, 0.34-4.66) and hormone-treated (HR, 1.21; 95% CI, 0.62-2.36) groups. Integration of the GRS35 into the current model further improved its predictability for incident fracture occurrence by 6.3% (P = .010). Genetic profiling can more accurately predict future fracture risk, especially in individuals taking BPs.