Abstract 14547: Genetically Predicted Inhibition of HMG-CoA Reductase Reduces Risk of DNMT3A Clonal Hematopoiesis

Abstract

Introduction: Clonal hematopoiesis (CH) is associated with increased morbidity and mortality from cancer and cardiovascular diseases including heart failure and ischemic heart disease. At present no medications are known to protect from CH. However, leukemia cells with equivalent mutations show sensitivity to statin-induced cell death. Aims: To evaluate the association of LDL-C lowering genetic variants at HMGCR (the target of statins) with CH risk using two-sample Mendelian randomization. Methods: We investigated whether lower LDL-C genetically predicted by 6 variants at HMGCR associates with CH risk. Secondary exposures included a polygenic instrument of 74 LDL-C-lowering genome-wide variants. Genetic associations for LDL-C derived from a genome-wide association study (GWAS) meta-analysis of circulating LDL-C levels by the Global Lipid Genetics Consortium (N≤188,578 individuals of European ancestry). Outcome associations for CH were from a GWAS in 368,526 European individuals in UK Biobank. Results: Genetically predicted reduction in HMG-CoA reductase activity equivalent to a 38.7mg/dL lower circulating LDL-C level was associated with reduced odds of overall CH (OR 0.75, 95% CI 0.58-0.97, P = 0.027) or DNMT3A CH (OR 0.64, 95% CI 0.48-0.86, P = 0.003) but not TET2 CH (OR 1.12, 95% CI 0.63-2.00, P = 0.70). Associations were consistent across all 6 HMGCR variants in sensitivity analyses. Co-localization plots showed evidence of co-localization of LDL-C with DNMT3A CH and overall CH at the HMGCR locus. Associations were stronger for CH with variant allele fraction ≥10%, and, statins suppressed colony formation of mutant DNMT3A cells in vitro , suggesting statins may limit clonal expansion. In a secondary analysis to explore LDL-C dependence, LDL-C lowering predicted by the polygenic instrument did not associate with overall or DNMT3A CH but associated with increased odds of TET2 CH (OR 1.13, 95% CI 1.01-1.27, P = 0.03). Findings were replicated in a non-European cohort. Conclusions: Genetically predicted lifelong reduction of HMG-CoA reductase activity was associated with reduced risk of overall and DNMT3A -mutant CH, likely by LDL-C-independent mechanisms. This provides a strong rationale for clinical trials testing the effects of statins in CH.