Health disparity in pharmacogenomics: African American SNPs and drug metabolism

As pharmacogenomics moves from bench-top to bed-side, African Americans have been left out of the advances personalized medicine holds. To date, African Americans have been largely absent from the numerous pharmacogenetic studies that identify predictive single nucleotide polymorphisms (SNPs) that are being used clinically to improve care. As we move toward more comprehensive bioinformatics and highthroughput methods of evaluating the genome for functionally relevant SNPs that affect drug phenotypes, we are also producing a growing health disparity in the translation of these findings into the clinic for African Americans. Current genomic methods are trying to reach beyond genome–wide association in hopes of finding biological plausibility to genetic findings. This search has led to the use of expression quantitative trait loci (eQTLs) as a tool in pharmacogenetic association studies. SNPs labeled as eQTLs have intrinsic biological plausibility since they are significantly associated with changes in gene expression. However, no liver eQTL studies have been conducted in African Americans. By using liver-specific eQTLs, we will discover SNPs that affect drug metabolizing enzymes (DMEs), which will have wide-spread scientific and clinical impact; given that most drugs currently prescribed undergo some form of hepatic metabolism. Because of the increase genetic diversity found in African Americans, they may carry population specific SNPs can never be found by pharmacogenomics studies in populations of European-descent alone. Without African American focused pharmacogenomics studies we risk excluding African Americans from personalized medicine. We plan to use primary hepatocytes to conduct gene expression studies both before and after drug enzyme induction to identify genes that are differentially expressed. Drug enzyme induction is driven by increased gene transcription; therefore, by focusing on the genes that are differentially expressed after induction we can pin-point the drivers (i.e. genes) that regulate drug metabolizing enzymes. We will then look at SNPs within and in close proximity to these genes to determine if any are associated with differences in gene expression (dubbed DI-eQTLs). We will then look to see if these DI-eQTLs are associated with pharmacokinetic measures in the same set of cell cultures, affectedly preforming a pharmacokinetic and a pharmacogenetic study in the same individual. We hypothesize that drug enzyme induction will provide us with DI-eQTLs that are strongly associated to drug pharmacokinetics, a clinically relevant phenotype. To validate these findings we will query publically available data to look for an enrichment of our DI-eQTLs within the most strongly associated SNPs in pharmacogenomic GWAS data. The association between genotype, gene expression and drug pharmacokinetics in African Americans has never been conducted before and will provide an invaluable resource for pharmacogenomics and clinical pharmacology, as well as clinical translation in this understudied population.