Abstract 03: LDL-C Genetic Risk Score Predicts Hyperlipidemia and Modifies Lipid Trajectory Over the Life Course: The Bogalusa Heart Study

Abstract

Introduction: Variability in lipid levels is decreased in childhood compared to adulthood. However, few studies have assessed whether genetic information may help to predict hyperlipidemia beyond measured lipid values at this critical developmental stage. Hypothesis: We hypothesized that a low-density lipoprotein cholesterol (LDL-C) genetic risk score (GRS) would predict hyperlipidemia and LDL-C over the life course, even after adjustment for childhood measures. Methods: The analysis included 651 Bogalusa Heart Study participants (63% women, 31% African American, baseline age=9.8 + 3.1 years, median follow-up=40 years) with genome-wide association study (GWAS) data and at least one childhood and one adulthood measure of LDL-C. A weighted LDL-C GRS was constructed using loci from previous GWAS meta-analyses. Hyperlipidemia was defined as an LDL-C > 130 mg/dL or statin use. Cox proportional hazards regression models examined the associations between GRS and hyperlipidemia. Linear and mixed linear regression models were employed to examine the associations of GRS with adulthood LDL-C and life-course LDL-C trajectory, respectively. All models adjusted for age, sex, and childhood LDL-C. Results: Among participants of European ancestry, increasing GRS tertile conferred strong, dose-response increases in the hazards of hyperlipidemia ( Figure ). Similarly, the highest GRS tertile was associated with a 20 mg/dl increased LDL-C level in middle-aged adults compared to the lowest tertile (P<0.0001). Each tertile increase in GRS was also associated with 5.5 mg/dL larger 10-year increase in LDL-C (interaction-P=0.04). No associations were observed among participants of African ancestry. Conclusions: A LDL-C GRS was associated with adulthood LDL-C phenotypes independently of childhood LDL-C values in participants of European but not African ancestry. These findings highlight a need for increased genomics research in diverse populations and suggest a predictive utility of genetic information in childhood.