Risk Prediction in Patients With Heart Failure With Preserved Ejection Fraction Using Gene Expression Data and Machine Learning.

Liye Zhou,Haitao Yang,Hongmei Yao,Zhi Li,Ruiling Fang,Yuehua Cui,Hongyan Cao,Zhifei Guo,Yongqing Wu,Bijue Wang

doi:10.3389/fgene.2021.652315

Abstract

Heart failure with preserved ejection fraction (HFpEF) has become a major health issue because of its high mortality, high heterogeneity, and poor prognosis. Using genomic data to classify patients into different risk groups is a promising method to facilitate the identification of high-risk groups for further precision treatment. Here, we applied six machine learning models, namely kernel partial least squares with the genetic algorithm (GA-KPLS), the least absolute shrinkage and selection operator (LASSO), random forest, ridge regression, support vector machine, and the conventional logistic regression model, to predict HFpEF risk and to identify subgroups at high risk of death based on gene expression data. The model performance was evaluated using various criteria. Our analysis was focused on 149 HFpEF patients from the Framingham Heart Study cohort who were classified into good-outcome and poor-outcome groups based on their 3-year survival outcome. The results showed that the GA-KPLS model exhibited the best performance in predicting patient risk. We further identified 116 differentially expressed genes (DEGs) between the two groups, thus providing novel therapeutic targets for HFpEF. Additionally, the DEGs were enriched in Gene Ontology terms and Kyoto Encyclopedia of Genes and Genomes pathways related to HFpEF. The GA-KPLS-based HFpEF model is a powerful method for risk stratification of 3-year mortality in HFpEF patients.

Highlights

Heart failure (HF) is the leading cause of death and disability worldwide among older adults (Manolis et al, 2019)
We further explored the differentially expressed genes (DEGs) between the good-outcome and poor outcome groups using significance analysis of microarrays (Tusher et al, 2001) and conducted Gene Ontology (GO) enrichment analysis and the Kyoto Encyclopedia of the Genes and Genomes (KEGG) pathway analysis based on the DEGs using KOBAS software1 (Ai and Kong, 2018)
Using the gene expression data of heart failure with preserved ejection fraction (HFpEF) patients, this study explored five machine learning methods and one conventional logistic regression model to predict the survival status of patients with HFpEF

Summary

Introduction

Heart failure (HF) is the leading cause of death and disability worldwide among older adults (Manolis et al, 2019). Over 50% of patients with HF exhibit heart failure with preserved ejection fraction (HFpEF; Komajda et al, 2011; Rich et al, 2018), and the prevalence of HFpEF is increasing relative to heart failure with reduced ejection fraction (HFrEF) at an Machine Learning Prediction of HFpEF alarming rate of 1% per year (Monika et al, 2018). HFpEF patients have a poor prognosis, and the 5-year mortality rate of HFpEF is as high as 50% (Shah et al, 2017). While the mortality rate of HFrEF has significantly decreased over the past few years because of specific HFrEF treatments (Loh et al, 2013), no effective treatment has been identified for HFpEF patients (Shah et al, 2014). With an aging population worldwide, the emerging epidemic of HFpEF requires urgent attention to determine methods for faster disease risk assessment and to predict clinical outcomes to guide therapy, monitoring, and patient management

Objectives

Methods

Results

Conclusion