Abstract 13308: Complementary Investigations of Plasma Proteomics Identify Novel Protein Associations With Incident Diabetes and Glucose Metabolism

Introduction: Type 2 diabetes (T2D) is a cardiovascular disease risk equivalent and likely results from broad metabolic changes, which high throughput proteomics have helped to unravel. Prior studies are limited by proteomic coverage, cross sectional design, and lack of physiologic phenotyping. Hypothesis: Complementary proteomic studies of incident T2D and physiologic responses to an intravenous glucose tolerance test (IVGTT) will identify novel proteins with roles in glucose homeostasis and future risk of T2D. Methods: Cardiovascular Health Study (CHS) and HERITAGE study participants without diabetes underwent SomaScan ® profiling of 4,776 plasma proteins. HERITAGE participants underwent IVGTT, from which insulin sensitivity index (S I ), acute insulin response to glucose (AIR G ), and glucose effectiveness (S G ) were derived. We used Cox regression to test protein associations with 18-year incident T2D in CHS, and multivariable linear regression to test protein associations with IVGTT measures in HERITAGE. Results: In CHS (N = 2631, 74 ± 5 years, 62% female, 14% Black), 57 proteins were significantly associated with incident T2D after comprehensive covariate and multiple testing adjustment. Of these, 44, 9, and 8 were associated with S I , AIR G , and S G respectively in HERITAGE (N = 752, 35 ± 14 years, 55% female, 38% Black) (Figure). Notable findings include beta-glucuronidase, which associated with increased T2D risk (HR 1.46 per SD increase in log 2 protein level) and lower S G , suggesting a role in insulin-independent glucose disposal, and thrombospondin-2, which associated with increased T2D risk (HR 1.26 per SD), lower AIR G , and not with S I , indicating that it may be a marker of pancreatic dysfunction. Conclusions: By integrating proteomics from two complementary prospective cohorts using different but related outcomes, we identified 34 novel protein-T2D associations, and characterized their relationship with physiologic axes of glucose metabolism.