Abstract 17635: Proteomic Signatures of Diabetes-Related Coronary Heart Disease: The Atherosclerosis Risk in Communities Study

Abstract

Introduction: The proteomic alterations that characterize coronary heart disease (CHD) in patients with type 2 diabetes are largely uncharacterized. Hypothesis: There are unique proteomic signatures of diabetes-related CHD. Methods: We quantified 4,955 plasma proteins using an aptamer-based assay (SomaLogic v4.0) in Atherosclerosis Risk in Communities (ARIC) Visit 2 (1990-92) participants. Incident CHD as defined as either myocardial infarction or fatal CHD. We used Cox regression to estimate the hazard ratios for the proteins and CHD associations, stratified by baseline diabetes status. We used Bonferroni-corrected p -values to adjust for multiple comparisons. Results: We included 9,939 individuals free of CHD (mean age: 57±5.7 years, 58% women, 23% Black adults, 14 % with diabetes). Over a median 25 years follow-up, there were 1,281 CHD events (330 among individuals with diabetes). Among 1,426 individuals with diabetes, after adjustment for confounders, 21 proteins were associated with incident CHD (p<10 -5 ) (Figure, Panel A). Of these proteins, 16 were also associated with prevalent diabetes and 11 exhibited a significant interaction by diabetes status (p<0.05/21 - Figure, Panel B). The top 6 proteins uniquely associated with diabetes-related CHD include inactive tyrosine-protein kinase 7, semaphorin-4D, scavenger receptor class A member 5, kremen protein 1, and cartilage intermediate layer protein 2, and semaphorin-6A. Among those without diabetes, of the 8 proteins significantly associated with CHD (p<10 -5 ), growth differentiation factor-15 was the only one that overlapped with proteins identified among individuals with diabetes. Conclusions: A large-scale proteomics analysis identified 20 novel proteomic markers unique to diabetes-related CHD. Our findings have the potential to shed more light on biological pathways linking diabetes and CHD.