Genome-Wide Approach to Measure Variant-Based Heritability of Drug Outcome Phenotypes

Pharmacogenomic studies have successfully identified variants-typically with large effect sizes in drug target and metabolism enzymes-that predict drug outcome phenotypes. However, these variants may account for a limited proportion of phenotype variability attributable to the genome. Using genome-wide common variation, we measured the narrow-sense heritability ( $h_{\text{SNP}}^2$ ) of seven pharmacodynamic and five pharmacokinetic phenotypes across three cardiovascular drugs, two antibiotics, and three immunosuppressants. We used a Bayesian hierarchical mixed model, BayesR, to model the distribution of genome-wide variant effect sizes for each drug phenotype as a mixture of four normal distributions of fixed variance (0, 0.01%, 0.1%, and 1% of the total additive genetic variance). This model allowed us to parse $h_{\text{SNP}}^2$ into bins representing contributions of no-effect, small-effect, moderate-effect, and large-effect variants, respectively. For the 12 phenotypes, a median of 969 (range 235-6,304) unique individuals of European ancestry and a median of 1,201,626 (range 777,427-1,514,275) variants were included in our analyses. The number of variants contributing to $h_{\text{SNP}}^2$ ranged from 2,791 to 5,356 (median 3,347). Estimates for $h_{\text{SNP}}^2$ ranged from 0.05 (angiotensin-converting enzyme inhibitor-induced cough) to 0.59 (gentamicin concentration). Small-effect and moderate-effect variants contributed a majority to $h_{\text{SNP}}^2$ for every phenotype (range 61-95%). We conclude that drug outcome phenotypes are highly polygenic. Thus, larger genome-wide association studies of drug phenotypes are needed both to discover novel variants and to determine how genome-wide approaches may improve clinical prediction of drug outcomes.