Incident heart failure in chronic kidney disease: proteomics informs biology and risk stratification

Area of Science:

• Biomarkers and proteomics
• Cardiovascular disease
• Nephrology

Background:

• Heart failure (HF) is a significant complication for individuals with chronic kidney disease (CKD).
• Existing risk prediction models, such as the Pooled Cohort equations to Prevent Heart Failure (PCP-HF), perform poorly in CKD populations.
• There is a critical need for novel therapeutic targets and improved risk stratification for HF in CKD.

Purpose of the Study:

• To identify novel circulating protein biomarkers associated with incident heart failure (HF) in patients with chronic kidney disease (CKD).
• To explore potential causal relationships and druggable targets for HF in CKD using proteomic data.
• To develop and validate enhanced multi-protein risk models for HF prediction in CKD, comparing their performance against established scores.

Main Methods:

• SomaScan proteomic analysis of 4638 proteins in 2906 participants from the Chronic Renal Insufficiency Cohort (CRIC) study, with validation in the Atherosclerosis Risk in Communities (ARIC) study.
• Primary outcome: 14-year incident HF. Secondary outcomes: 4-year HF, HF with reduced ejection fraction, and HF with preserved ejection fraction.
• Application of Mendelian randomization and Gene Ontology for causality assessment, and comparison of novel multi-protein models against the PCP-HF risk score.

Main Results:

• Over 200 proteins were initially associated with incident HF in CKD patients (P < 1 × 10-5) after adjusting for estimated glomerular filtration rate.
• After covariate adjustment, including N-terminal pro-B-type natriuretic peptide, 17 proteins remained significantly associated with HF.
• Four druggable protein targets (FCG2B, IGFBP3, CAH6, ASGR1) were identified through Mendelian randomization.
• A 48-protein model demonstrated superior predictive performance (C-statistic 0.790 in CRIC) compared to the PCP-HF model (C-statistic 0.703) for incident HF.

Conclusions:

• Large-scale proteomics has successfully identified novel circulating protein biomarkers and potential mediators of HF in the context of CKD.
• The developed proteomic risk models offer improved accuracy for predicting HF in CKD patients compared to the existing PCP-HF risk score.
• These findings highlight promising avenues for developing new preventative therapies and refining risk stratification strategies for HF in CKD.