Search

Background: There are approximately 12 million statin users in the United Kingdom. Reports of statin intolerance occurs between 7 and 29% of users, manifesting as muscle ache, fatigue or more seriously, muscle breakdown leading to myopathy. Creatine phosphokinase (CK) levels are used as a biomarker of statin-induced muscle damage. Non-adherence or discontinuation of therapy is a common result of intolerance and can result in negative cardiovascular disease-related outcomes. Aim: This thesis attempts to identify trends in record-linked medical data in a Scottish Caucasian cohort (GoDARTS) that best represent statin intolerance in order to study associated genetic factors. Methods: Prescribing trends such as switching or discontinuation of statin therapy were examined, and thresholds created to select true cases of intolerance. Information on CK levels was gathered from medical records and appropriate test results were utilized. Genotypic data was gathered for the variants and genetic regions of interest using a variety of methods including chip-based genotyping followed by imputation, TAQMAN genotyping, and exome sequencing. Subsequently hypothesis-based association analyses were conducted, including linear and logistic regressions, followed by meta-analyses, regional GWAS followed by a regional meta –analysis. Results: The phenotypes of statin intolerance were validated both internally and externally. Previously reported missense variants in LILRB5 (Asp247Gly) and CKM (Glu83Gly) were replicated and shown to be associated with CK levels irrespective of statin usage in the GoDARTS cohort and the clinical trial setting (JUPITER). Further, the CKM variant was also associated with inducibility of CK at times of tissue injury. The Asp247 genotype in LILRB5 was associated with increased risk of statin intolerance, and was replicated in associations with non-compliance to statin therapy and the development of myalgia in the JUPITER trial. The association with myalgia showed a stratified effect based on therapy (statin or placebo), with those on placebo showing the genotype effect. Further, the variant was also associated with increased risk of statin-induced myositis, cases of which had been clinically adjudicated and exome sequenced for the PREDICTION-ADR consortium. Further exploration of the LILR gene region showed an association with variants in LILRB2 (His20Arg and Val235Met) which were in strong LD with each other but were not in linkage with the variant in LILRB5. Stratified analysis revealed that the risk for carriers of the LILRB2 variants was increased depending on the genotype carried at the LILRB5 variant. Conclusions: This study characterises novel genetic factors associated with statin intolerance impacting adherence. The findings point to the immunomodulatory effects of statins. The results suggest that true statin-induced myalgia and non-specific myalgia are distinct, with a possible role for the immune system in their development. The findings encourage further investigation into the immune-physiology of statin-induced muscle damage and identifies genetically susceptible groups who are more likely to be statin intolerant.