Abstract
<h3>Purpose</h3> Treatments for right heart failure are supportive, with no direct therapies for right ventricular (RV) contractility. Using machine learning (ML), we sought to identify whether clinical RV dysfunction is reflected by impaired contractility in myocytes. Using this model, we then characterized the properties of an ideal sarcomere enhancing drug to restore RV contractility. <h3>Methods</h3> In a cohort of 20 patients with heart failure with reduced ejection fraction (HFrEF), we performed unsupervised ML on right heart catheterization data to obtain phenotypes of RV function. We next collected indices of myofilament mechanical function using a force-length control apparatus to assess permeabilized RV myocytes from hearts explanted from those same patients. We compared indices of myofilament function using a Mann-Whitney Test to identify a corresponding myofilament phenotype concordant with clinical RV dysfunction. <h3>Results</h3> We performed Gaussian Mixture Modeling, a form of unsupervised ML, on four clinical indices of RV function, namely right atrial pressure (RAP), RAP-to-pulmonary capillary wedge pressure ratio (RA/PCWP), pulmonary artery pulsatility index (PAPi), and pulmonary vascular resistance (PVR). Mixture modeling revealed one group of patients with good RV function (n=6) and another with poor RV function (n=14), specifically elevated RAP (p=0.0028) and RAP/PCWP (p=0.0005) and depressed PAPi (p=0.0005). In the group with poor RV function, we found altered myofilament mechanics, namely depressed calcium-activated maximum stress (T<sub>max</sub>, p=0.048) and depressed actin-myosin cross-bridge detachment (p=0.050) but no alterations in other myofilament measures. We hypothesized that the optimal sarcomere enhancer to augment RV contractility would be one that augmented RV T<sub>max</sub> without altering actin-myosin kinetics. One such drug is EMD57033. We tested EMD57033 in permeabilized RV myocytes in a subset of 10 patients and found that it improved myofilament contractility selectively in patients with depressed RV T<sub>max</sub> (group-drug interaction p=0.006). <h3>Conclusion</h3> We used unsupervised machine learning solely on clinical indices of RV function to identify specific myofilament abnormalities indicative of RV dysfunction. EMD57033 selectively remedied the myofilament abnormalities observed in this RV dysfunctional group.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
