Several associations between drug responses and clinically relevant genetic variants have already been identified with GWAS. In some cases in which the effect sizes are large, significant results have been found with as few as 51 cases and 282 controls6. (In this case, patients had different reactions to the lipid-lowering drug simvastatin.) Although physicians must weigh the costs and benefits of using pharmacogenetic testing to guide prescription and dosage decisions for individual patients, these findings suggest that the small samples that have already been collected from under-represented populations could yield leads that have not been identified in populations of European ancestry.
