HFrEF subphenotypes based on 4210 repeatedly measured circulating proteins are driven by different biological...

Area of Science:

• Proteomics
• Biomarker Discovery
• Machine Learning in Medicine

Background:

• Heart failure with reduced ejection fraction (HFrEF) is a complex condition with significant mortality.
• Heterogeneity in HFrEF necessitates novel approaches for subtyping and understanding disease mechanisms.
• Circulating proteins offer a window into dynamic biological processes in HFrEF.

Purpose of the Study:

• To identify distinct protein-based subphenotypes in HFrEF using serial proteomic data.
• To investigate the underlying dynamic biological mechanisms driving these subphenotypes.
• To explore opportunities for personalized treatment strategies in HFrEF.

Main Methods:

• Serial blood sampling from 382 HFrEF patients over a median of 2.1 years.
• Aptamer-based multiplex proteomic analysis of 4210 circulating proteins.
• Unsupervised machine learning to derive subphenotypes based on proteomic profiles.

Main Results:

• Four distinct protein-based HFrEF subphenotypes were identified.
• Subphenotypes differed in age, ejection fraction, and prevalence of chronic renal failure.
• Protein subsets driving subphenotype allocation were linked to oxidative stress, inflammation, and extracellular matrix organization.
• Subphenotypes 2 and 3 showed significantly worse prognosis compared to subphenotype 1.

Conclusions:

• Four distinct circulating-protein based subphenotypes exist in HFrEF.
• These subphenotypes are characterized by unique protein profiles, clinical features, and prognoses.
• Understanding these subphenotypes can inform personalized therapeutic approaches for HFrEF.