Predicting adverse ventricular remodelling and poor outcome after acute myocardial infarction by a distinctive miRNA signature

Abstract

Abstract Introduction and aim Adverse ventricular remodelling and heart failure (HF) are frequent complications of acute myocardial infarction (MI), but no reliable methods are currently available for prediction. MicroRNAs (miRNAs) can be used as biomarkers and they play a crucial role in multiple diseases onset and evolution. The aim of our study was to identify and validate a panel of serum miRNAs with differential expression in MI and test their prognostic ability to predict chronic HF. Methods Sequential approach with several models: 1) Initial selection of miRNAs by a 762 miRNA hybridization array with samples from 12 patients with acute MI and 10 age and sex-matched controls; 2) Validation of selected miRNAs in a prospective study with consecutive enrolment of patients with type 1 MI at the Cardiac ICU of a tertiary hospital; two independent cohorts were included (2013–2014 and 2017–2019) with clinical and CV imaging assessment at baseline and long-term follow-up; serum miRNAs and other biomarkers (BNP, NT-proBNP, troponin, etc) were studied; the composite HF-related outcome included HF hospitalizations and CV mortality; 30 age and sex-adjusted controls were studied for miRNAs comparison; miRNA signature study was completed with 3) in vitro model of hypoxia ±nutrient supply deprivation in H9C2 cardiomyocytes, 4) in vivo rat model of ischemia-reperfusion and 5) permanent coronary occlusion. All procedures were approved by the Ethics Committee (refs. 175/13 & 061/16), the Animals Ethics Committee (ref. 298/16) and complied with the Declaration of Helsinki. All patients provided written informed consent. Results 14 miRNAs were initially identified with highest differentiation between MI and controls. In the validation study, 311 patients with MI were enrolled, mean age 60±15, 81% male; ST elevation MI 65%, GRACE risk score 144±45, Killip class ≥II 23%, LVEF<50% at discharge 46% and NT-proBNP 3091±5997 pg/ml. miR-210, miR-21, miR-23 and miR-221 discriminated between patients with more severe form of MI presentation by Killip class (figure 1-B); in addition, GRACE risk score statistically correlated with miR-21, miR-221, miR-27, miR-93, miR-23, miR-148, miR-107, miR-210 and miR-144. Among markers of acute HF, miR-21, miR-23, miR-27, miR-210 and miR-221 significantly correlated with NT-proBNP (figure 1-C) and BNP levels. In parallel, miR-210, miR-21, miR-23, miR-106, miR-221, miR-27 and miR-93 correlated with left ventricular ejection fraction (LVEF). Finally, miR-21, miR-23, miR-210 and miR-221 associated with patients' long-term survival free of long-term HF-related events. Both in vitro and in vivo experimental models served as confirmation of the importance of these outlined miRNAs in cardiomyocyte response to ischemic stress and ventricular remodelling. Conclusions We identified a signature of microRNAs associated with adverse ventricular remodelling and poor long-term outcome after MI, that could serve as biomarkers involved in HF progression. Figure 1 Funding Acknowledgement Type of funding source: Public grant(s) – National budget only. Main funding source(s): Instituto de Salud Carlos III