Joel Ås: Genomic Analysis of Adverse Drug Reactions

Abstract

Adverse drug reactions (ADRs) pose a significant global challenge, leading to substantial costs, suffering, and even loss of life. Genetic factors can play a role in determining a patient's response to the drug treatments and predicting ADRs. While many genetic associations with ADRs have been identified, there are still numerous ADRs suspected to have genetic components.

In Paper I, the collection and curation strategies for ADR cases in the Swedegene biobank are established, presenting a cohort of 2,550 ADR-cases. Paper II presents the association between genetic variations in human leukocyte antigen (HLA) genes and the development of pancreatitis as a response to azathioprine treatment in patients with Crohn's disease. Paper III reports on an international collaboration to investigate the genetic aetiology of atypical femur fractures (AFF) during bisphosphonate treatment. The study found that previously identified genetic variants did not replicate, and --- as the cohort is the largest of its kind --- provides valuable insights into common genetic factors of AFF. Paper IV examines the genetic associations with central nervous system (CNS) toxicity as an ADR to antimicrobial drugs, identifying correlations with three genes linked to suicide and schizophrenia, although the biological connection remains unclear. Finally, Paper V presents a methodology for the experimental design of ADR studies by analysing the known protein interactions of drugs and proteins associated with ADRs. This approach aims to mitigate the impact of competing genetic correlations by identifying common protein interactions to validate the inclusion of drugs and ADRs in the study. These interactions are then ranked based on importance to the selected drugs and ADRs and used to propose genetic targets of interest. 

Overall, the findings of these studies contribute to the understanding of genetic predispositions to ADRs and provide a novel approach for data-driven experimental design for phenotype and genetic target selection.