Circulating microRNAs as markers for prediction of heart failure medical treatment outcome

Abstract

Purpose: Chronic heart failure (HF) is a major worldwide health care problem. Despite significant improvement of mortality and morbidity with medical therapy, not all patients respond well to the treatment. This lack of response has not been adequately explored and currently no biomarker is available to indicate which patient will respond medical treatment non-responders. As circulating microRNAs have been described as potential biomarkers in various heart diseases, we explored the role circulating microRNAs as potential biomarkers of HF and as predictors of treatment outcome. Methods: MicroRNA arrays were performed from plasma of 9 healthy controls, 10 HF treatment Non-responders, and 10 HF treatment Responders at 6-month follow up. HF treatment was defined as death, re-hospitalization for HF, and/or reduction of left ventricular end-systolic volume of <10% in 6 months despite standard HF medication therapy. Then we selected 15 microRNAs with significant changes for further validation by quantitative real-time PCR assays with the plasma samples from another 17 healthy controls, 13 Non-responders and 13 Responders at 6-month follow up. Finally, we determined the diagnostic accuracy of these candidate microRNAs. Results: Our results showed that 3 of 5 microRNAs previously reported in HF patients (miR -423-5p, -320a, -17, -92a, and -22) failed to predict the treatment outcome. We identified 7 fibrosis- and inflammation-related microRNAs (including miR -27b, -221, -222, -140-3p, -155, -29b, and -31) were altered in HF patients. More importantly, they were able to predict the HF patients as receiver operator characteristics (ROC) curve analysis demonstrated higher predictive power of these 7 microRNAs within the HF registry, when comparing to healthy controls (AUC >0.87, P 0.90 for Responder and AUC<0.55 for Responder at 6-month follow up, P<0.05) and the Wilcoxon matched pairs test (P<0.05) confirmed the ability of these 7 microRNAs to discriminate the treatment Non-responders and Responders [PCY1] Conclusion: We identified 7 microRNAs (miR -27b, -221, -222, -29b, -31, -140-3p, and -155) as the effective biomarkers for HF and they possess a high discriminatory ability to distinguish HF treatment Non-responders from Responders. Our studies suggest that these 7 microRNAs are putative effective biomarkers of HF and of predicting treatment Non-responders.