Integrating polygenic risk scores with clinical drug development

This PhD proposal will build on the development of polygenic risk scores (PRS) as predictors of disease risk to assess their utility across the drug-development value chain from early phase studies to randomised clinical trials, focussing on the selection of patient population [1].

Our understanding of clinical disorders has been transformed by the success in identifying genetic variants associated with disease risk. Genome-wide association studies (GWAS) have showed that the genetic architecture underlying these disorders is highly polygenic with hundreds or thousands of variants contributing to genetic risk. Typically, numerous variants have strong statistical evidence of association with disease, but individually confer only a small effect. Across the genome, these associated variants can be combined into a polygenic risk score, which provides an individual-level measure of genetic liability for a trait. PRS can be used for risk stratification to predict clinical outcomes, and when combined with other biomarkers can attain clinical utility. In this PhD proposal, we will investigate how polygenic risk scores can be used within drug development, to strengthen evidence supporting drug development in preclinical and early phase studies, and to improve study design in phase 3 randomised control trials (RCT).

The importance of genetics in drug development is well-recognised, as drug targets with genetic association support are substantially more likely to be successful in early and late-stage studies, and to lead to an approved drug [2]. The existing utility of genetic associations is based on evidence at a single genetic variant. In this PhD proposal we will extend beyond a single variant to the genome-wide support as captured by polygenic risk scores (PRS), and assess how PRS can be integrated within drug development and assessment pipelines. The student will undertake a series of projects covering steps in drug development, from target choice to trial design,...