Circulating miRNAs and Risk of Sudden Death in Patients With Coronary Heart Disease

Abstract

Objectives: This study evaluated whether plasma miRNAs were specifically associated with sudden cardiac and/or arrhythmic death (SCD) in a cohort of patients with coronary heart disease (CHD), most of whom were without primary prevention implantable cardioverter-defibrillators.

Background: Novel biomarkers for sudden death risk stratification are needed in patients with CHD to more precisely target preventive therapies, such as implantable cardioverter-defibrillators. miRNAs have been implicated in regulating inflammation and cardiac fibrosis in cells, and plasma miRNAs have been shown to predict cardiovascular death in patients with CHD.

Methods: We performed a nested case control study within a multicenter cohort of 5,956 patients with CHD followed prospectively for SCD. Plasma levels of 18 candidate miRNAs previously associated with cardiac remodeling were measured in 129 SCD cases and 258 control subjects matched on age, sex, race, and left ventricular ejection fraction.

Results: miR-150-5p, miR-29a-3p, and miR-30a-5p were associated with increased SCD risk (odds ratios and 95% confidence intervals: 2.03 [1.12 to 3.67]; p = 0.02; 1.93 [1.07 to 3.50]; p = 0.02; 0.55 [0.31 to 0.97]; p = 0.04, respectively, for third vs. first tertile miRNA level). Unfavorable levels of all 3 miRNAs was associated with a 4.8-fold increased SCD risk (1.59 to 14.51; p = 0.006). A bioinformatics-based approach predicted miR-150-5p, miR-29a-3p, and miR-30a-5p to be involved in apoptosis, fibrosis, and inflammation.

Conclusions: These findings suggest that plasma miRNAs may regulate pathways important for remodeling and may be useful in identifying patients with CHD at increased risk of SCD.

Keywords: coronary heart disease; microRNAs; risk prediction; sudden death.

Conflict of interest statement

There are no conflicts for any of the authors related to design, execution or analysis of the experiments presented in this manuscript. CMA reports being a consultant for MyoKardia and Sanofi and receiving grants from St Jude Medical, National Institutes of Health, Abbott, and Roche Diagnostics outside the submitted work.

Copyright © 2020 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1:. miRNAs and Risk of Sudden Cardiac and/or Arrhythmic Death.

Conditional Logistic Regression Model adjusted for prior MI, NYHA class, and history of diabetes

Figure 2. (Central Illustration): miRNAs and Risk of Sudden Cardiac and/or Arrhythmic Death

Conditional Logistic Regression Model adjusted for prior MI, NYHA class, history of diabetes miR-150-5p, miR-29a-5p, and miR-30a-5p. Levels above the median for each of the miRNAs is associated with risk of SCD. When combined into a multi-marker risk score, unfavorable levels of all three miRNAs is associated with a 4.8 fold increased risk of SCD.

Figure 3:. Network interactions with experimentally determined targets of miR-150-5p, miR-29a-3p, and miR-30a-5p.

Gene targets (ovals) of the three miRNAs (orange rectangles) are connected by strong (thick) and weak (thin) evidence types according to miRTarBase. This subnetwork view focuses on 160 targets that either of have strong evidence for interacting with only one miRNA (green, radial layout), weak evidence to two or more miRNAs (blue, interstitial layout), or a mix of both (green, interstitial layout).

Figure 4:. Enrichment analysis for ontology terms and pathways related to miR-150-5p, miR-29a-3p, and miR-30a-5p targets.

Combing results for Gene Ontology: Biological Process (gold), Jensen Disease Ontology and OMIM (blue), and WikiPathways (gray), the bar graph shows the enrichment score calculated as −log(p value) * Z score.