QW2: Multiple polygenic scores improve bone mineral density prediction in an independent sample of Caucasian women
POSTER PRESENTATION (Video):
Presenter: Xiangxue Xiao, MSPH
Authors: Xiangxue Xiao, MSPH; Qing Wu, MD, ScD
MENTOR: Qing Wu, MD, ScD
Purpose of the study: To determine if multiple polygenic scores improve bone mineral density prediction over single genetic risk score (GRS) in an independent sample of Caucasian female subjects.
Study design: Based on summary statistics of four genome-wide association studies (GWAS) related to two osteoporosis-associated traits, namely bone mineral density (BMD) and heel quantitative ultrasound (eBMD), four GRSs were derived for 1205 individuals in the Genome-Wide Scan for Female Osteoporosis Gene Study (GWSFO). The effect of each GRS on BMD variation was assessed by using multivariable linear regression, with adjustment for conventional risk factors. For eBMD, the GRS that explained the most variance in BMD was selected to be entered into a multi-score model. Elastic net regularized regression was used to develop the multi-score model, which estimated the joint effect of two GRSs (GRS_BMD and GRS_eBMD) on BMD variation.
Results: With the same clinical risk factors having been adjusted for, the model that included GRS_BMD performed best by explaining 29.14% of the variance in BMD; in contrast, the single-score model that included GRS_eBMD explained only 18.53% of BMD variance. The model that includes both GRS_BMD and GRS_ eBMD, as well as the clinical risk factors, aggregately explained 31.12% in BMD variation. Compared to the single GRS models, the multi-score model explained significantly more variance in BMD.
Conclusions: The multi-polygenic score model provided significant improvement in explaining BMD variation compared to single score models.