Molecular pharmacoepidemiology (MPE) is defined as the study of how genes alter the effectiveness and safety of medications in large populations using real-world data. This novel research field applies methods from the areas of epidemiology and causal inference to answer questions from the realm of pharmacogenetics. The UK Biobank presents an ideal data source for MPE studies due to the availability of large-scale genomic data and electronic health records (EHRs).
In the proposed project, we will use whole-genomic sequencing data linked to EHRs for a series of MPE studies to assess whether genetic variants can affect the effectiveness and safety of commonly prescribed medications, with a primary focus on cardiometabolic drugs. For example, we will assess whether (i) genetic variants of p-glycoprotein may affect the risk of stroke and bleeding associated with direct oral anticoagulants, (ii) genetic variants of pancreatic beta-cell K+ channels and cytochrome P450 2C9 (CYP2C9) may affect the risk of hypoglycemia and major adverse cardiovascular events (MACE) associated with sulfonylureas, and (iii) genetic variants of CYP2D6 and beta-1 adrenoreceptors may affect the risk of MACE and bradyarrhythmia associated with beta-blockers.
The studies will apply state-of-the-art pharmacoepidemiologic methods at the level of design (e.g., active-comparator new-user design, prevalent new-user design) and data analysis (e.g., propensity score-based methods). Effect estimates for the drug-event associations of interest will be assessed among patients with and without the specific genetic variants. Stratum-specific results will be compared to identify potential effect modifications by genetic variants.
Overall, the results of this project will boost our understanding regarding the impact of genetic variability on drug effectiveness and safety. Considering the scarcity of evidence in the area, our project will aid precision medicine, thereby directly improving patient care.
