Whole-genome sequencing in pharmacogenetics

Thomas J Urban

Sections:

Continuing rapid advances in DNA-sequencing technology now allow the field of pharmacogenetics to move from candidate genes and genome-wide investigation of common genetic variation, to nearly complete representation of all variation in the human genome through whole-genome sequencing (WGS). There is much cause for optimism regarding the application of WGS to drug response traits, particularly for studies of therapeutic drug response and rare, severe adverse drug reactions (sADRs). As genomic sequencing becomes a more routine part of clinical practice, we should expect an acceleration of pharmacogenetic discovery, and an increase in the uptake of pharmacogenetic information in clinical decision-making.

From pharmacogenetics to pharmacogenomics: genome-wide studies of drug response

Over the past several years, we have witnessed an explosion of novel genetic discoveries related to all aspects of human health, including therapeutic and adverse responses to drugs. The advent of genome-wide association studies (GWAS) in 2005 was followed quickly by the discovery of now over 1000 common genetic variants (i.e., those with minor allele frequencies >3–5%) showing unambiguous associations with hundreds of complex human traits, in just the 8 years since [101]. The preponderance of significant variants discovered by GWAS have been those associated with human disease risk, but a number of strong genetic associations with drug response traits have been identified, and these have generally been among the most striking GWAS discoveries in terms of the magnitude of their genetic effects. Even as the human genetics community began publicly lamenting the underwhelming impact of GWAS findings on clinical disease prediction, and remarking on the problem of ‘missing heritability’ of risk for common diseases unaccounted for by common variants [1], the arena of pharmacogenetics was often singled out as an exception [2]. GWAS of drug response traits are quite exceptional in having provided a number of clinically significant genetic predictors of drug outcomes, with estimated effect sizes sometimes orders of magnitude greater than those seen for any human disease [3,4]. Despite this, GWAS of drug response traits currently represent only a small fraction of the total number of GWAS performed to date [101].