Abstract Body (Do not enter title and authors here): Background: Coronary Artery Disease (CAD) is a complex heterogeneous disease with the greatest mortality and morbidity burden. While statins have been standard of care to reduce risk of first coronary event by approximately 20% per 1.0 mmol/L reduction of low-density lipoprotein (LDL) cholesterol, if statins were more efficacious in certain subgroups of the population, it might affect clinical decision-making. Motivated by this, we developed the largest meta-GWAS of statin benefit and derived polygenic risk score models to identify subgroups with the most efficacious response. Methods: We first performed a multi-ethnic meta-GWAS using REGENIE on a total of 61,097 individuals from All of Us and Genetic Epidemiology Research on Adult Health and Aging Cohort on the statin response, defined as difference between the most recent pretreatment level and earliest on-treatment LDL-C before and after statin initiation. We then optimized polygenic risk score (PRS) models on 70% of statin users within UK Biobank (UKBB; N=2190) and validated these models in an unseen 30% of UKBB participants. Results: The meta-GWAS identified genome-wide significant variants in CELSR2, CELSR3, APOB, LDLR, APOE, and SLCO1B1. 1 SD increase of statin benefit PRS was found to be associated with 0.03 mmol/L of LDL (95% CI: -0.01, 0.08) lowered. 1 SD increase of statin benefit PRS was also found to be associated with a 2.4% (95% CI: 0.5, 4.2) decrease in risk in CAD. Conclusions: We demonstrate that a meta-GWAS and subsequent PRS can identify a subgroup that may benefit from increased LDL lowering and lowered CAD risk, albeit with a small effect size. Increased power is needed to identify genetic architecture underlying differential response to statin within a diverse population.