Diabetes Genetic Clusters and Clinical Outcomes in American Indians.

Diabetes has a large medical and public health impact in American Indians. Studies have used genetic data to distinguish type 1 diabetes (T1D) and type 2 diabetes (T2D) and uncover biologic mechanisms underlying T2D clinical heterogeneity. We applied a T1D polygenic score (PS) to 3,084 American Indians (mean age 56 years, 58% women, 39% diabetes). We also calculated partitioned PS for eight clusters of T2D-associated variants and evaluated their association with 20 cardiometabolic traits and five clinical outcomes. The profile of T1D PS for individuals with diabetes was consistent with T2D. A total T2D PS was significantly associated with early age of T2D onset (P = 3.5 × 10-11). Partitioned PS for T2D clusters were significantly associated with cardiometabolic traits for the obesity cluster (increased measures of body fat and total triglycerides but lower HDL cholesterol), while the lipodystrophy cluster was associated with increased fasting insulin, waist-to-hip ratio, triglycerides, and blood pressure, and lower body fat percentage and HDL cholesterol. T2D clusters were not associated with cardiovascular and kidney outcomes. Our findings support a relationship of cluster-specific T2D partitioned PS with cardiometabolic traits described in other populations, but there are opportunities for developing improved clustering methods using genetic variation from American Indians. ARTICLE HIGHLIGHTS: Diabetes is highly prevalent in American Indians and a main cause of morbidity and mortality, and its clinical heterogeneity can be uncovered using genetic data. We are interested in identifying type 1 versus type 2 diabetes in American Indians with diabetes and identifying biological mechanisms for type 2 diabetes that are related to clinical outcomes using genetic data. Using genetic data, we found a high probability of participants having type 2 diabetes. We identified similar associations of type 2 diabetes genetic clusters, that are related to biological mechanisms, with cardiometabolic traits as previously described in other populations, but no associations of genetic clusters with cardiovascular and kidney outcomes. Our findings support type 2 diabetes as the main cause of diabetes in our American Indian cohort and provide insights into improvements using genetic data to uncover type 2 biological mechanisms.