Abstract Context Vitamin D inadequacy, assessed by 25-hydroxyvitamin D [25(OH)D], affects around 50% of adults in the United States and is associated with numerous adverse health outcomes. Blood 25(OH)D concentrations are influenced by genetic factors that may determine how much vitamin D intake is required to reach optimal 25(OH)D. Despite large genome-wide association studies (GWASs), only a small portion of the genetic factors contributing to differences in 25(OH)D levels has been discovered. Objective Therefore, knowledge of a fuller set of genetic factors could be useful for risk prediction of 25(OH)D inadequacy, personalized vitamin D supplementation, and prevention of morbidity and mortality from deficient 25(OH)D. Design Using PRSice and weights from published African- and European-ancestry GWAS summary statistics, ancestry-specific polygenic scores (PGSs) were created to capture a more complete set of genetic factors. Patients or Other Participants Participants (European ancestry n=9,569, African ancestry n=2,761) came from three cohort studies. Main Outcome Measure(s) Blood concentrations of 25(OH)D. Results The PGS for African ancestry was derived using all input SNPs (a p-value cut-off of 1.0) and had an R 2 of 0.3%; for European ancestry, the optimal PGS used a p-value cut-off of 3.5×10 −4 in the target/tuning dataset and had an R 2 of 1.0% in the validation cohort. Those with highest genetic risk had 25(OH)D that was 2.8-3.0 ng/ml lower than those with lowest genetic risk (p=0.0463 to 3.2×10 −13 ), requiring an additional 467 to 500 IU of vitamin D intake to maintain equivalent 25(OH)D. Conclusions PGSs are a powerful predictive tool that could be leveraged for personalized vitamin D supplementation to prevent the negative downstream effects of 25(OH)D inadequacy.

Ancestry-specific polygenic scores and SNP heritability of 25(OH)D in African- and European-ancestry populations