Prediction of Major Adverse Cardiovascular Events in Patients With Hypertrophic Cardiomyopathy Using Proteomics Profiling
Yuichi J. Shimada,Yoshihiko Raita,Lusha W. Liang,Mathew S. Maurer,Kohei Hasegawa,Michael A. Fifer,Muredach P. Reilly,Yuichi J. ShimadaYoshihiko RaitaLusha W. LiangMathew S. MaurerKohei HasegawaMichael A. FiferMuredach P. ReillyDivision of Cardiology,Department of Medicine,Columbia University Irving Medical Center,New York,NY. (Y.J.S.,L.W.L.,M.S.M.,M.P.K.)Irving Institute for Clinical and Translational Research,Columbia University Irving Medical Center,New York,NY. (M.P.K.)Cardiology Division,Department of Medicine,Massachusetts General Hospital,Harvard Medical School,Boston,MA.(Y.J.S.,M.A.F.)Department of Emergency Medicine,Massachusetts General Hospital,Harvard Medical School,Boston,MA. (Y.R.,K.H.)
DOI: https://doi.org/10.1161/CIRCGEN.121.003546
2022-10-13
Abstract:Hypertrophic cardiomyopathy often causes major adverse cardiovascular events (MACE), for example, arrhythmias, stroke, heart failure, and sudden cardiac death. Currently, there are no models available to predict MACE. Furthermore, it remains unclear which signaling pathways mediate MACE. Therefore, we aimed to prospectively determine protein biomarkers that predict MACE in hypertrophic cardiomyopathy and to identify signaling pathways differentially regulated in patients who subsequently develop MACE. In this multi-centre prospective cohort study of patients with hypertrophic cardiomyopathy, we conducted plasma proteomics profiling of 4979 proteins upon enrollment. We developed a proteomics-based model to predict MACE using data from one institution (training set). We tested the predictive ability in independent samples from the other institution (test set) and performed time-to-event analysis. Additionally, we executed pathway analysis of predictive proteins using a false discovery rate threshold of <0.001. The study included 245 patients (n=174 in the training set and n=71 in the test set). Using the proteomics-based model to predict MACE derived from the training set, the area under the receiver-operating-characteristic curve was 0.81 (95% CI, 0.68–0.93) in the test set. In the test set, the high-risk group determined by the proteomics-based predictive model had a significantly higher rate of developing MACE (hazard ratio, 13.6 [95% CI, 1.7–107]; P =0.01). The Ras-MAPK (mitogen-activated protein kinase) pathway was upregulated in patients who subsequently developed MACE (false discovery rate<1.0×10 -7 ). Pathways involved in inflammation and fibrosis—for example, the TGF (transforming growth factor)-β pathway—were also upregulated. This study serves as the first to demonstrate the ability of proteomics profiling to predict MACE in hypertrophic cardiomyopathy, exhibiting both novel (eg, Ras-MAPK) and known (eg, TGF-β) pathways differentially regulated in patients who subsequently experience MACE.
genetics & heredity,cardiac & cardiovascular systems