Polygenic Susceptibility to Diabetes and Poor Glycemic Control in Stroke Survivors

Abstract

ABSTRACTImportanceType 2 diabetes mellitus is a highly heritable disease with numerous identified genetic risk variants. However, the concrete role of these variants in the clinical care of stroke patients remains poorly understood.ObjectiveTo evaluate whether higher polygenic susceptibility to type 2 diabetes mellitus (PSD) is associated with worse glycemic control and higher risk of cardiovascular events in stroke survivors.Design, Setting, and ParticipantsWe conducted a 3-stage genetic association study. First, we used a cross-sectional design and data from the UK Biobank to evaluate the relationship between PSD and glycemic control (enrollment took place between 2006 and 2010). Second, we used a cross-sectional design and data from the All of Us Research Program to replicate associations identified in the prior stage (enrollment between 2018 and 2022). Third, we used a prospective design and data from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial to evaluate the relationship between PSD and post-stroke acute cardiovascular events (enrollment between 1997 and 2001). The present analyses were performed between May 2022 and August 2023.ExposuresLow, intermediate, and high PSD modeled through categories of a polygenic risk score that included up to 462 independent DNA sequence variants associated with type 2 diabetes mellitus at genome-wide levels (p<5}10-8).Main Outcomes and MeasuresHemoglobin A1c levels, treatment-resistant diabetes mellitus (defined as hemoglobin A1c ≥ 7.0% despite antidiabetic treatment), and risk of post-stroke acute cardiovascular events (stroke, myocardial infarction, or cardiovascular death).ResultsStage 1 included 5,670 stroke survivors (mean age 61, 41% females), including 1,215 (21%) with known diabetes. Compared to stroke survivors with low PSD, those with high PSD had 35% higher hemoglobin A1c (beta 0.35, standard error [SE] 0.034; p <0.001) and 3 times the risk of having treatment-resistant diabetes (OR 3.40, 95%CI 2.52-6.24; p <0.001). These results pertaining to hemoglobin A1c remained significant when evaluating diabetics and non-diabetics separately (both tests p<0.05). Stage 2 replicated these results in 2,012 stroke survivors, including 447 (22.2%) with diabetes (mean age 64, 52% females, p <0.05 for all tests). Stage 3 included 1,750 stroke survivors (mean age 68, 34.8% females), including 441 (25.2%) with diabetes. Compared to stroke survivors with low PSD, those with high PSD had 50% higher risk of post-stroke cardiovascular events (hazard ratio 1.53, 95%CI 1.03, 2.28, p < 0.05).Conclusions and RelevanceAmong stroke survivors enrolled in 3 different studies, a higher PSD was associated with poorer glycemic control and higher risk of treatment-resistant diabetes and post-stroke vascular events. Given that millions of Americans are receiving diabetes-related polygenic risk data from direct-to-consumer companies, further research is needed to determine whether precision medicine strategies based on this information can improve the clinical management of these patients.KEY POINTSQuestionDoes polygenic susceptibility to type 2 diabetes mellitus lead to worse glycemic control and higher risk of cardiovascular events in stroke survivors?FindingsHigher polygenic susceptibility to type 2 diabetes mellitus, modeled via polygenic risk scoring, was associated with higher levels of Hemoglobin A1c, higher risk of treatment resistant diabetes mellitus, and higher risk of recurrent cardiovascular events in stroke survivors.MeaningPolygenic susceptibility to type 2 diabetes is associated with worse clinical trajectories in stroke survivors. Given that millions of Americans are receiving diabetes-related polygenic risk data from direct-to-consumer companies, further research is needed to understand how this information could help post-stroke clinical management.