Personalized polygenic profiling based on the genetic architecture of lipid metabolism in the Russian population

Objective Elevated cholesterol levels are associated with the risk of the most socially significant cardiovascular diseases, such as atherosclerosis, ischemic heart disease, and myocardial infarction. Methods In this study, we sought to study the genetic architecture of lipid metabolism by conducting genome-wide association studies of total cholesterol, LDL-C, and HDL-C levels in a sample of the Russian population ( n = 8,732) who were not carriers of variants linked to familial hypercholesterolemia and did not take lipid-regulating agents. Based on the detected associations and machine learning methods, several polygenic score models were constructed for each lipid type, and the link between polygenic scores and atherosclerosis, ischemic heart disease, and myocardial infarction was examined in an additional sample of patients diagnosed with either of these diseases ( n = 3,954). Results A meta-analysis of the results of the conducted genome-wide association studies showed that total cholesterol and LDL-C were largely associated with the same variants located in the HMGCR, CERT1, POLK, ANKDD1B, APOC3, BCL3, CBLC, BCAM, NECTIN2, TOMM40, APOE, APOC1, APOC4, and SMARCA4 genes, while HDL-C was associated with variants located in the LPL, ALDH1A2, LIPC, and CETP genes. Men and women differed in genetic predictors of lipid levels, with variants in the SMARCA4 and LDLR genes associated with cholesterol levels only in women. Conclusion The models developed in this study consider age and a modifiable factor, BMI. Therefore, the personalized polygenic profiling approach presented in this study enables life-long CVD risk assessment.