Circulating Relaxin-1 Level Is a Surrogate Marker of Myocardial Fibrosis in HFrEF

Introduction: Relaxin-1 (RLN1) has emerged as a possible therapeutic target in myocardial fibrosis due to its anti-fibrotic effects. Previous randomized clinical trials investigated therapeutic role of exogenous relaxin in patients with acute-on-chronic heart failure (HF) and failed to meet clinical endpoints. Here, we aimed to assess endogenous, circulating RLN1 levels in patients with heart failure with reduced ejection fraction (HFrEF) of ischemic origin. Furthermore, we analyzed relation of RLN1 and left ventricular diastolic function, left and right ventricular fibrosis, and invasive hemodynamic measurements. Unique feature of our study is the availability of ex vivo human myocardial tissue. Methods: Human myocardial samples were available from the Transplantation Biobank of the Heart and Vascular Center at Semmelweis University after local ethical approval and informed consent of all participants (n = 47). Tissue was collected immediately after heart explantations; peripheral blood was collected before induction of anesthesia. Myocardial sections were stained for Masson's trichrome and Picrosirius red staining to quantify fibrosis. Medical records were analyzed (ECG, anthropometry, blood tests, medication, echocardiography, and invasive hemodynamic measurements). Results: Average RLN1 levels in HFrEF population were significantly higher than measured in age and gender matched healthy control human subjects (702 ± 283 pg/ml in HFrEF vs. 44 ± 27 pg/ml in control n = 47). We found a moderate inverse correlation between RLN1 levels and degree of myocardial fibrosis in both ventricles (r = -0.357, p = 0.014 in the right ventricle vs. r = -0.321, p = 0.028 in the left ventricle with Masson's trichrome staining). Parallel, a moderate positive correlation was found in left ventricular diastolic function (echocardiography, E/A wave values) and RLN1 levels (r = 0.456, p = 0.003); a negative correlation with RLN1 levels and left ventricular end-systolic diameter (r = -0.373, p = 0.023), and diastolic pulmonary artery pressure (r = -0.894, p < 0.001). RLN1 levels showed moderate correlation with RLN2 levels (r = 0.453, p = 0.0003). Conclusion: Increased RLN1 levels were accompanied by lower myocardial fibrosis rate, which is a novel finding in our patient population with coronary artery disease and HFrEF. RLN1 can have a biomarker role in ventricular fibrosis; furthermore, it may influence hemodynamic and vasomotor activity via neurohormonal mechanisms of action. Given these valuable findings, RLN1 may be targeted in anti-fibrotic therapeutics and in perioperative care of heart transplantation.