Related Experiment Video
Updated: Apr 14, 2026
Author Spotlight: Advancing the Analysis of Plasma Extracellular Vesicle Proteome for Cardiovascular Biomarker Studies
Published on: January 31, 2025
Proteomic phenotyping with machine learning for cardiovascular outcomes in haemodialysis: insights from the AURORA
Madonna Salib1, Sophie Girerd1,2, Florence Pinet3
1Inserm, Centre d'Investigations Cliniques-1433, and Inserm U1116, CHRU Nancy, F-CRIN INI-CRCT, Université de Lorraine, 54500 Vandoeuvre-Les-Nancy, France.
Insights
Machine learning identified four patient phenotypes in hemodialysis patients. One phenotype, linked to toll-like receptors (TLRs) signaling, showed increased risk for cardiovascular death and mortality, suggesting personalized treatment approaches.
Area of Science:
- Nephrology
- Proteomics
- Machine Learning
- Cardiovascular Medicine
Background:
- Cardiovascular (CV) disease is a leading cause of mortality in patients undergoing hemodialysis, yet clinical trials have shown neutral results.
- Personalized treatment strategies are needed to improve CV outcomes in this high-risk population.
- Understanding distinct patient profiles is crucial for tailoring interventions.
Purpose of the Study:
- To identify distinct biological phenotypes in hemodialysis patients using proteomic data and machine learning.
- To investigate the association of these phenotypes with cardiovascular outcomes.
Main Methods:
- Unsupervised machine learning (clustering) was applied to plasma protein biomarkers from 382 hemodialysis patients (derivation cohort).
- A decision tree model was built to predict cluster membership and assess CV outcomes in 389 patients (validation cohort).
- Key prognostic factors were adjusted for in multivariable analyses.
Main Results:
- Four distinct phenotypes were identified: 'cytokine storm signalling', 'toll-like receptors (TLRs) signalling', 'inflammation and fibrosis', and a 'reference phenotype'.
- The 'TLRs signalling' phenotype was significantly associated with increased risk of CV death (HR=1.65), all-cause mortality (HR=1.43), and major adverse CV events (MACE) (HR=1.48) in the validation cohort.
- These phenotypes represent distinct biological mechanisms contributing to patient risk.
Conclusions:
- Machine learning analysis of proteomic data successfully identified four biologically distinct phenotypes in hemodialysis patients.
- The 'TLRs signalling' phenotype is associated with adverse CV outcomes, highlighting its clinical relevance.
- These identified phenotypes offer potential targets for personalized therapies to improve CV prognosis in hemodialysis.
Aims:
Cardiovascular (CV) trials have yielded neutral results in haemodialysis. A better understanding of patient profiles is needed to personalize treatment strategies in order to improve CV outcomes in this setting. This study sought to identify biological phenotypes based on proteomic data using machine learning approaches in patients undergoing haemodialysis.
Methods And Results:
A clustering analysis using 253 plasma protein biomarkers was performed in 382 patients (machine learning derivation analysis) from the AURORA trial, which tested the effect of rosuvastatin on CV outcomes in patients on haemodialysis. A decision tree was subsequently constructed to predict cluster membership and assess its association with CV outcomes in another subset of the trial (n = 389 patients, validation analysis). Four phenotypes were identified, namely 'cytokine storm signalling', 'toll-like receptors (TLRs) signalling', 'multiple pathways related to inflammation and fibrosis' phenotypes, as well as a 'reference phenotype' which exhibited the least biological abnormalities. In multivariable analysis of the validation study, after adjusting for key prognostic factors, the TLRs phenotype was significantly associated with CV death, all-cause mortality, and MACE (HR = 1.65 [1.13-2.41], 1.43 [1.03-1.98], and 1.48 [1.04-2.10], respectively).
Conclusion:
Using unsupervised machine learning on proteomic data, we identified four mechanistic biological phenotypes involving cytokine storm and TLRs signalling, inflammation and fibrosis. These biological phenotypes may contribute to CV prognosis and pave the way for personalized therapy in haemodialysis.
More Related Videos
08:51Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
Published on: September 20, 2024
09:21Author Spotlight: Generating Neuronal Phenotypic Profiles - A Protocol to Culture and Image Human Midbrain Dopaminergic Neurons
Published on: July 7, 2023
Related Concept Videos
01:20Blood Studies for Cardiovascular System I: Cardiac Biomarkers
The essential diagnostic tools for detecting myocardial necrosis and monitoring individuals suspected of having acute coronary syndrome (ACS) include:
Troponins
Troponins, particularly cardiac troponins I and T, are the most precise and sensitive markers of myocardial injury. They are detectable within 4-6 hours of myocardial injury and remain...
01:19Blood Studies for Cardiovascular System II: CRP, Hcy, and Cardiac Natriuretic Peptide Markers
These markers indicate stress or strain on the heart muscle:
Natriuretic Peptides (BNP)
Cardiac myocytes produce these hormones in response to ventricular stretching...
01:29Pharmacogenomics: Identification of New Drug Targets