Unmasking subclinical cardiac dysfunction in type 2 diabetes through genetic and epigenetic profiling
TL;DR
This study provides proof-of-concept evidence that integrating T2D-derived polygenic and methylation risk scores with detailed clinical phenotyping captures distinct biological information related to subclinical myocardial dysfunction.
Abstract
Diabetic Cardiomyopathy (DiabCM) develops in patients with Type-2-Diabetes Mellitus (T2D) and is characterized by cardiac dysfunction independent of ischemic heart disease or systemic hypertension. Its underlying mechanisms remain poorly defined with likely genetic and non-genetic mechanisms including epigenetic modifications. Combining T2D-derived polygenic risk scores (PRS) and methylation risk scores (MRS) with key clinical characteristics may aid risk stratification for myocardial dysfunction among patients with T2D. We analyzed 173 deeply phenotyped participants from the CARDIATEAM discovery study, grouping them by the severity of myocardial dysfunction determined by echocardiography. Both PRS and MRS show significantly different distribution between severity groups ( p -value = 0.02 and 1.29e-05, respectively) with significant odds ratios (ORs) at ≥ 80th and ≥ 90th percentiles: PRS (2.23 [95%CI 1.03–4.92] and 2.82 [95%CI 1.02–8.34]), and MRS (4.52 [95%CI 2.01–10.83] and 3.49 [95%CI 1.22–11.02]) respectively. Combining individuals with PRS and MRS above the 80th percentile showed a higher OR (6.48). The combined PRS + MRS model demonstrated the best discriminative ability (AUC = 0.753), while the interaction model did not enhance performance (AUC = 0.752). Adding PRS and MRS to ten baseline covariates (age, sex and eight cardiometabolic variables) slightly increased bootstrap-validated sensitivity/specificity, from 0.819/0.878 to 0.827/0.885. MRS increased step-wise from non-diabetic low-risk to diabetic high-risk clusters, whereas PRS primarily distinguished diabetic from non-diabetic status. Exploratory pathway analysis of MRS-mapped genes identified enrichment for phosphatase signalling and oxidative-stress pathways implicated in diabetic cardiomyopathy. This study provides proof-of-concept evidence that integrating T2D-derived polygenic and methylation risk scores with detailed clinical phenotyping captures distinct biological information related to subclinical myocardial dysfunction. MRS showed stronger association, including after adjustment for clinical covariates and in sensitivity analyses restricted to individuals with T2D, whereas the PRS added limited discriminatory value within the T2D population. These findings are exploratory and require validation in larger independent cohorts.