Heart failure & cardiomyopathies
Harnessing the Plasma Proteome to Predict Mortality in Heart Failure Subpopulations
Authors' conclusion
PRSs can provide valid, accurate, and dynamic prognostic estimates for patients with HF. This approach has the potential to improve longitudinal monitoring of patients and facilitate personalized care.
Study context
- Population
- Individuals with heart failure (HF) with reduced ejection fraction (HF with reduced ejection fraction (HFrEF); ejection fraction <40%) and preserved ejection fraction (HFpEF; ejection fraction ≥50%)training/validation counts per HF type · HFrEF training 1,247 · HFrEF validation 762 · HFpEF training 725 · HFpEF validation 785
- Predictors
- 17- and 14-protein models for HFrEF and HFpEF
- Predicted outcome
- all-cause mortality within 1 year
Methodological details3 independent studies, machine learning, SomaScan Assay measuring ≈5000 proteins
- 3 independent studies
- Machine learning
- SomaScan Assay measuring ≈5000 proteins
Model performance
As reported in the abstract
C-index or AUC
HFpEF PRS (validation)
C-index 0.79
HFpEF PRS (validation)
AUC 0.82
MAGGIC (HFpEF validation)
C-index 0.71
MAGGIC (HFpEF validation)
AUC 0.74
HFrEF PRS (validation, vs MAGGIC)
C-index 0.76
HFrEF PRS (validation, vs MAGGIC)
AUC 0.83
MAGGIC (HFrEF validation)
C-index 0.75
MAGGIC (HFrEF validation)
AUC 0.84
HFpEF PRSPRS vs NT-proBNP
C-index 0.76
HFpEF PRSPRS vs NT-proBNP
AUC 0.81
NT-proBNP (HFpEF)PRS vs NT-proBNP
C-index 0.72
NT-proBNP (HFpEF)PRS vs NT-proBNP
AUC 0.76
HFrEF PRSPRS vs NT-proBNP
C-index 0.75
HFrEF PRSPRS vs NT-proBNP
AUC 0.78
NT-proBNP (HFrEF)PRS vs NT-proBNP
C-index 0.73
NT-proBNP (HFrEF)PRS vs NT-proBNP
AUC 0.77
- HFpEF PRS performed significantly better ( P ≤0.1) than MAGGIC and NT-proBNP
- HFrEF PRS performed comparably to MAGGIC
Further findings
1-year mortality, PRS vs NT-proBNP
- HFrEF PRS significantly better C-Index than NT-proBNP ( P =0.026)
PRS composition
- Known HF pathophysiology biomarkers
- 93%
- Novel proteins
- 7%
Longitudinal PRS (HFrEF: n=396; 1107 samples; HFpEF: n=175; 350 samples)
- HFrEF and HFpEF proteomic risk scores (PRSs) were higher and increased more over time in individuals who experienced a fatal event during follow-up
Original publication
Harnessing the Plasma Proteome to Predict Mortality in Heart Failure Subpopulations. Circ Heart Fail. Published 7 March 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →