In the elderly, the rate of bone loss at the femoral neck (ΔBMD) is a risk factor for hip fracture and mortality. There is evidence that the variation in ΔBMD is partially determined by genetic factors. This study sought to develop a genetic profiling of BMD-associated variants, and to define the association between the genetic profiling and the ΔBMD.
Sixty-eight BMD-associated SNPs from genomewide association studies (GWAS) were genotyped in 863 women and 527 men aged from 60 years who were participants of the Dubbo Osteoporosis Epidemiology Study. A genetic risk score (GRS) was constructed for each individual by summing the product of regression coefficient [associated with BMD from GWAS] and the number of minor risk alleles for each SNP. ΔBMD, expressed as annual percent change-in-BMD, was determined by linear regression analyis for each individual who had at least two femoral neck BMD measurements. The relationship between GRS and ΔBMD was analyzed by the standard multiple linear regression model.
The mean of ΔBMD was -0.66% (SD 1.58%) for women and -0.57% (SD 1.40%) for men. In women, one unit greater in GRS was associated with a 0.015 g/cm2 (SE 0.006) increase in baseline BMD, after adjusting for age and weight. More importantly, each unit increase in GRS was associated with 0.23% (SE 0.09) lower rate of loss (P=0.019), and this association was independent of age and baseline BMD. GRS, age and baseline BMD collectively accounted for 2.4% of variance of ΔBMD. In men, there was no significant association between either baseline BMD or ΔBMD and GRS.
These data demonstrate for the first time that genetic profiling is associated with femoral neck ΔBMD, independent of baseline BMD and age. This finding suggests that genetic profiling can be used as an additional means for evaluating bone loss in an individual.