(A-130) Prevalence of Drug-Gene Interactions Following Acute Myocardial Infarction: An Observational Study Using Data From the UK Biobank

Background: Common genetic variants can alter the function of drug-metabolizing enzymes resulting in adverse clinical consequences. Therapeutic recommendations (e.g. dose or drug amendments) for ‘actionable’ drug-gene interactions (DGIs) involving germline genotypes have been developed by international committees.

Objectives: To estimate the prevalence of actionable DGIs in patients following incident acute myocardial infarction (AMI).

Methods: This was a retrospective cohort study of individuals recruited to the UK Biobank who experienced their first AMI following their initial assessment centre visit and before 31st December 2017. Individuals with missing genotype information or primary care prescription data were excluded. We identified actionable DGIs from The Clinical Pharmacogenetics Implementation Consortium (CPIC) and The Dutch Pharmacogenetics Working Group. Medicine data was extracted for the 2 years following each individual’s incident AMI. Allelic information was extracted for: CYP2C9*2 (rs1799853), *3 (rs1057910), CYP2C19*2 (rs4244285), *17 (rs12248560), CYP3A5*3 (rs776746), CYP2D6*3 (rs35742686), *4 (rs3892097), *9 (rs5030656), *10 (rs1065852), *41 (rs28371725), SLCO1B1*5/*15 (rs4149056), TPMT*3B (rs1800460), *3C (rs1142345), and *3A (rs1800460). Metaboliser phenotypes were determined using activity scores or CPIC standard diplotype-phenotype tables. We describe the number of DGIs experienced by each individual during the 2 years post incident AMI. Furthermore, we estimate those interactions predicted to have a ‘substantial’ effect, defined as DGIs due to variant homozygous/compound heterozygous genotypes.

Results: Of approximately 500,000 people recruited to the UK Biobank, we identified 6,820 with an AMI during the study period. The cohort had a mean age of 60 years (SD=7) and 72.2% were male. A total of 883 (8.9%) individuals were found to have at least one DGI in the first 2 years post incident AMI, with 227 (3.3%) having multiple DGIs. The most common DGIs in this cohort were atorvastin-SLCO1B1 (n=442), clopidogrel-CYP2C19 (n=324), and simvastatin-SLCO1B1. Evidence of substantial DGIs were found in 4.1% of the cohort, with the most common being clopidogrel-CYP2C19 (n=103) and amitriptyline-CYP2D6.

Conclusions: This study demonstrates that DGIs are common for individuals with a recent diagnosis of AMI, highlighting a potential medicines optimisation gap for this high-risk group. Indeed, the inclusion of medication information along with genetic data in healthcare systems will support efforts to improve the effectiveness and safety of medicines.