Abstract 10103: Polygenic Scores and Longitudinal Trajectories of Cardiometabolic Phenotypes

Abstract

Introduction: Genetic predisposition via polygenic scores (PGS) is strongly associated with single measurements of systolic blood pressure (SBP) and low-density lipoprotein cholesterol (LDL-C), but it is unclear how PGS are associated with their longitudinal trajectories and how these trajectories are associated with risk of coronary heart disease (CHD). Hypothesis: PGS are associated with classes of longitudinal trajectories, and these trajectories are associated with CHD. Methods: Analyses were conducted in 8,453 white participants from the Atherosclerosis Risk in Communities study. SBP and LDL-C measurements were adjusted for medication use. Latent class models were used to group participants into longitudinal trajectory classes based on SBP and LDL-C values from 3-5 visits over up to 27 years. PGS were created by multiplying publicly available weights with participant genotype dosages. The association between PGS and trajectory classes for SBP and LDL-C was examined using logistic regression. We then examined the association of trajectory classes with incident myocardial infarction (MI) and CHD using cox proportional hazard models. Results: Participants were assigned to 4 classes of SBP and 2 classes of LDL-C trajectories ( Figure ). The SBP PGS was significantly associated with SBP classes 2 (OR = 1.28 per standard deviation change in the PGS, P = 4.8E-7) and 4 (OR = 1.7, P < 2E-16) relative to class 3, and with class 4 relative to class 2 (OR= 1.3, P = 2.0E-6). The LDL-C PGS was significantly associated with class 2 (OR = 1.63, P < 2E-16). There were 147 CHD and 102 MI events during 7 additional years of follow up. SBP class 2 and LDL-C class 2 were significantly associated with greater risk of incident MI. Conclusions: Our results highlight longitudinal trajectory classes as a tool for contextualizing the genomics of cardiometabolic phenotypes and the association of these phenotypes with CHD.