Currently, there is persistent trial-and-error in the prescribing of medications, which contributes to delayed therapeutic benefit, avoidable adverse events, and healthcare inefficiency. While pharmacogenomics has identified variants of interest, real-world validation remains sparse, and is often underpowered. Large-scale biobanks such as UK Biobank provide uniquely powered, longitudinal, genetically characterised cohorts to test drug-gene interactions across diverse populations and healthcare settings. This research will highlight current inequalities of prescribing the above medications within the UK, and also enable future precision prescribing that reduces risk, and ultimately improves outcomes for common, high-burden conditions while optimising healthcare resources.
In particular, we will initially focus our research into cardiometabolic and psychiatric drugs and hormone replacement therapy. The research questions are as follows:
(i) What are the demographic, sociodemographic, lifestyle factors, as well as co-occurrence of medication factors which affect uptake for the above specified medications?
(ii) Which genetic variants predict likelihood of being prescribed specific medications?
(iii) Which genetic markers influence drug efficacy, adherence, switching, and adverse reactions?
(v) How do demographic and socioeconomic factors interact with genetic variation to shape treatment outcomes?
Our objectives are to
(i) characterise medication use patterns and prescribing disparities,
(ii) perform genome-wide association studies (GWAS) of drug utilisation and response,
(iii) model gene-drug interactions using surrogate (e.g., blood pressure or lipid reduction) and clinical (e.g., myocardial infarction, stroke, psychiatric relapse) endpoints,
(iv) identify biologically plausible mechanisms through systems biology and AI pathway analysis, and
(v) construct predictive models of treatment outcomes for potential clinical application.