Abstract 4350904: Comprehensive Plasma Transcriptomics Profiling Identifies a Small Set of Circulating MicroRNA Biomarkers to Distinguish Hypertrophic Cardiomyopathy from Other Cardiomyopathies with Left Ventricular Hypertrophy

Abstract

Introduction: Hypertrophic cardiomyopathy (HCM) is the most common genetic cardiac disease. It can be challenging to distinguish HCM from other cardiomyopathies with left ventricular hypertrophy (LVH), including hypertensive LVH, transthyretin amyloid cardiomyopathy (ATTR-CM), and aortic stenosis (AS). Hypothesis: A small set of plasma microRNAs discriminate between HCM and other cardiomyopathies that cause LVH. Methods: In this multicenter case-control study, plasma transcriptomics profiling was performed in cases with HCM and controls with hypertensive LVH, ATTR-CM, and AS. Half of the cases enrolled earlier in each disease group were categorized as the training set and the remaining half as the test set. We specified microRNAs that were significantly (univariable P<0.05) upregulated or downregulated in HCM compared to hypertensive LVH in both the training and test sets (comparison #1). We performed the same comparison between HCM and ATTR-CM (comparison #2) as well as between HCM and AS (comparison #3). We identified microRNAs that were consistently upregulated or downregulated in HCM throughout all 3 comparisons. Then, we identified microRNAs independently associated with HCM (multivariable P<0.05) after adjusting for clinical parameters that were significantly different between HCM and controls. Using the selected microRNAs, a logistic regression model to distinguish HCM from controls was developed in the training set. We calculated an area under the receiver-operating-characteristics curve (AUROC) in the test set. We performed subgroup analyses comparing HCM with each of the controls. Results: We analyzed 2,656 microRNAs in patients with HCM (n=294), hypertensive LVH (n=321), ATTR-CM (n=167), and AS (n=38). After adjusting for 12 parameters that were significantly different between HCM and controls, 3 microRNAs were selected ( Figure 1 ). The logistic regression model using the 3 microRNAs had an AUROC of 0.95 (95% confidence interval [CI] 0.93-0.97) with a sensitivity of 0.95 and a specificity of 0.90. ( Figure 2 ). In the subgroup analysis, the model had AUROCs of 0.98 (95% CI, 0.96-0.99) for HCM vs. hypertensive LVH, 0.94 (95% CI, 0.90-0.98) for HCM vs. ATTR-CM, and 0.82 (95% CI, 0.71-0.92) for HCM vs. AS ( Figure 3 ). Conclusions: Our comprehensive plasma transcriptomics profiling identified a small set of circulating microRNAs that distinguish HCM from other cardiomyopathies with LVH independently of potential confounders.