Circulating microRNA signatures are associated with epicardial coronary dysfunction in patients underwent invasive physiological evaluation: The ANFIBIO Study

Abstract Background Circulating plasmatic microRNAs (miRNAs) are novel biomarkers with potential applications in chronic coronary syndromes (CCS). Nevertheless, their association with specific patterns of coronary dysfunction remains insufficiently evaluated. The objective of this study is to identify specific plasma miRNA signatures related to epicardial and/or microvascular coronary dysfunction. Methods The ANFIBIO study is a prospective, multicenter cohort study that enrolled consecutive patients presenting with chest pain of presumed coronary origin. Patients underwent invasive physiological evaluation, including measurement of fractional flow reserve (FFR) and index of microvascular resistance (IMR), were recruited. Based on FFR and IMR values, patients were categorized into four groups: a) Normal coronary indices; b) Isolated microvascular dysfunction; c) Isolated epicardial dysfunction; and d) Combined microvascular and epicardial dysfunction. miRNA profiling was performed using RT-qPCR. To complement the individual miRNA analysis, an unsupervised clustering approach was employed to identify patient subgroups based on miRNA expression profiles (miRNotypes). Results 128 patients were included with a mean age of 67.5-years, and 45 patients (35.2%) were women. The patient distribution across study groups was: 48 patients (37.5%) with normal coronary indices, 30 patients (23.4%) with isolated microvascular dysfunction, 25 patients (19.5%) with isolated epicardial dysfunction and 25 patients (19.5%) with combined dysfunction. Plasma miRNA signatures in case of microvascular dysfunction were similar to patients with normal indices but differed significantly from those with isolated epicardial dysfunction. After adjusting for cardiovascular risk factors and medications, miR-21-5p and miR-92a-3p were independently associated with epicardial dysfunction. While individual miRNAs had limited discriminatory capacity for epicardial dysfunction (AUC from 0.506 to 0.653), incorporating a 3-miRNA signature (miR-17-5p, miR-21-5p and miR-92a-3p) into a clinical model (CM) based on existing electronic health data significantly improved patient classification (NRI = 0.576, p-value 0.001; IDI = 0.082, p-value = 0.001). A trend toward higher discrimination capacity was observed (AUC CM = 0.674 vs. AUC CM+miRNA signature = 0.752; DeLong's test p-value = 0.064). Unsupervised clustering revealed three distinct plasma miRNotypes associated with specific patterns of coronary microvascular and epicardial dysfunction. Conclusions Circulating miRNA profiles differ between patterns of coronary dysfunction in patients with suspected CCS. Plasma miRNA-based molecular phenotyping could complement existing clinical tools aiding in the identification of epicardial coronary dysfunction.