Machine learning based on biomarker profiles identifies distinct subgroups of heart failure with preserved ejection fraction

Abstract

The lack of effective therapies for patients with heart failure with preserved ejection fraction (HFpEF) is often ascribed to the heterogeneity of patients with HFpEF. We aimed to identify distinct pathophysiologic clusters of HFpEF based on circulating biomarkers.

Impaired pulmonary vasomotor control in exercising swine with multiple comorbidities

“…In the present study, CKD was induced by embolizing ~ 3/4 of the kidneys and resulted in a 30% decrease in renal function mimicking an early stage of CKD. Our findings, which indicate that PVD is already present in swine after 5-6 months of exposure to multiple comorbiditiesincluding CKD-are in accordance with two studies using unsupervised phenomapping of a large group of patients clinically diagnosed with HFpEF [47,61]. This phenomapping resulted in three or four main phenotypes, in which the presence of CKD clustered with RV dilation and high pulmonary pressures in the group with the worst prognosis [47,61].…”

Endotyping in Heart Failure ― Identifying Mechanistically Meaningful Subtypes of Disease ―

“…Woolley et al examined a panel of 363 proteomic biomarkers from 429 patients with HFpEF and used unsupervised ML to identify 4 distinct endotypes with the following clinical characteristics: a younger group with lower N-terminal-proB-type natriuretic peptide (NT-proBNP) levels, an older group with CKD, a group with multiple comorbidities, and a group with significant coronary artery disease (CAD). 20 Interestingly, the clinical characteristics of the endotypes were very similar in this study using proteomics data compared with the Shah study using clinical phenotyping. 3 The group with the highest prevalence of CKD was associated with worse outcomes in both studies.…”

Omics, machine learning, and personalized medicine in heart failure with preserved ejection fraction: promising future or false hope?

“…in this study and Sanders‐van Wijk et al . in the PROMIS‐HFpEF cohort provide some external validity to the reported findings 8,9 . Second, the replicability of such a study is inherently difficult.…”

“…This 'phenomapping' approach uses big data-based unsupervised clustering methods to group patients in an unbiased manner. First pioneered by Shah et al 7 8 represents an important step forward in better phenomapping patients with HFpEF using deeper proteomics analysis. By including 429 patients from the BIOSTAT-CHF cohort, the authors quantified 363 protein biomarkers encompassing cardiovascular, immune response, and oncology panels.…”

“…However, the consistency in the observations made by Woolley et al in this study and Sanders-van Wijk et al in the PROMIS-HFpEF cohort provide some external validity to the reported findings. 8,9 Second, the replicability of such a study is inherently difficult. Using PCA aids in data dimensionality reduction but reproducing the created principal components is impossible without the resultant vector.…”