Related Experiment Video
Updated: Apr 29, 2026
Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
Published on: September 20, 2024
Refining biomarker-based clustering of cardiovascular inflammatory phenotypes in HIV using Recursive Feature
Rachel Mac Cann1,2,3, Dana Alalwan3, Gurvin Saini3
1School of Medicine, University College Dublin, Belfield, Dublin, Ireland.
Insights
This study refines biomarker clustering to identify inflammatory patterns linked to cardiovascular disease (CVD) in people with HIV. An individual additive model best revealed these associations, improving CVD risk prediction.
Area of Science:
- Immunology
- Cardiovascular Science
- Biostatistics
Background:
- People with HIV have increased non-communicable disease risk, particularly cardiovascular disease (CVD), due to chronic inflammation.
- Existing biomarker patterns linked to CVD in people with HIV lack clarity on predictive combinations.
- A refined framework is needed to better capture inflammatory patterns associated with cardiovascular phenotype (CVP) in people with HIV.
Purpose of the Study:
- To develop and evaluate a recursive feature addition (RFA) framework for enhancing biomarker-driven clustering.
- To identify optimal biomarker combinations for predicting cardiovascular phenotype (CVP) in people with and without HIV.
- To improve understanding of inflammatory profiles associated with CVD risk in the context of HIV.
Main Methods:
- Three RFA models were developed and compared: cumulative biological relevance, individual biomarker evaluation, and greedy forward-backward selection.
- Models utilized an initial 24-marker panel and 31 exploratory biomarkers.
- Assessment included principal component analysis (PCA), cluster stability, biological coherence, and association with CVP and ASCVD risk.
Main Results:
- All RFA models yielded three distinct biomarker-derived clusters.
- The individual additive model (Model 2) showed the most effective separation of inflammatory profiles, incorporating 11 additional biomarkers.
- Cluster 3 in Model 2 exhibited heightened immune activation, highest CVP prevalence (11%), and strongest CVP association (aOR 2.3).
Conclusions:
- A recursive feature addition (RFA) framework, particularly using a stepwise, individual biomarker additive model, optimizes unsupervised clustering.
- This approach effectively reveals additional associations between inflammatory patterns and cardiovascular phenotype (CVP).
- The findings enhance the ability to identify inflammatory drivers of CVD in people with and without HIV.
Background:
People living with HIV remain at elevated risk for a number of non-communicable diseases, including cardiovascular disease (CVD), driven in part by chronic inflammation. While prior studies have identified inflammatory biomarker patterns linked to CVD in people with HIV, it remains unclear which combinations of biomarkers most effectively predict clinical outcomes. We aimed to develop and evaluate a framework for refining biomarker-based clustering approaches to better capture inflammatory patterns associated with a cardiovascular phenotype (CVP) in people with HIV.
Methods:
We developed and evaluated three recursive feature addition (RFA) models to enhance biomarker-driven clustering of people with and without HIV. Using a 24-marker initial panel of biomarkers chosen for their links to clinical CVP in people with HIV, we compared three models for selective inclusion of 31 additional, exploratory biomarkers: (1) a stepwise additive model evaluating biomarkers cumulatively based on biological relevance; (2) a stepwise additive model evaluating biomarkers individually; and (3) a greedy forward-backward selection model. Each model was assessed using principal component analysis (PCA), cluster stability, biological coherence and association with a CVP and 10-year Atherosclerotic Cardiovascular Disease (ASCVD) risk.
Results:
All three RFA models generated three, biomarker-derived clusters. Post RFA cluster biomarker composition, model stability and clinical associations of these clusters differed across models. The individual additive model (Model 2) produced the most distinct separation of inflammatory profiles, incorporating 11 additional biomarkers, including, GDF-15, IFN-λ2 and Thrombopoietin). In this model, Cluster 3 was characterised by heightened innate and adaptive immune activation, the highest CVP prevalence (11%) and the strongest association with CVP (adjusted odds ratio (aOR) 2.3, 95% CI 1.04-5.09).
Conclusion:
We demonstrate that an RFA framework using a stepwise, additive model evaluating biomarkers individually to enhance clustering profiles provides optimal unsupervised clustering of exploratory biomarkers to reveal additional associations between inflammatory patterns and CVP in people with and without HIV.
