Integrating Imaging-Derived Clinical Endotypes with Plasma Proteomics and External Polygenic Risk Scores Enhances...

Area of Science:

• Cardiology
• Genetics
• Medical Imaging

Background:

• Coronary microvascular disease (CMVD) is a significant cause of ischemic heart disease, yet it is underdiagnosed.
• Development of risk prediction models for CMVD has been hindered by limited large-scale genome-wide association studies (GWAS).
• Significant genetic overlap exists between CMVD and coronary artery disease (CAD), offering potential for polygenic risk score (PRS) development.

Purpose of the Study:

• To develop and evaluate CMVD polygenic risk score (PRS) models using GWAS data.
• To integrate PRS, plasma proteomics, and perfusion PET imaging for CMVD risk prediction using machine and deep learning.
• To establish a novel unsupervised endotyping framework for CMVD based on myocardial blood flow imaging.

Main Methods:

• Developed CMVD PRS by selecting variants from CMVD GWAS and applying weights from CAD GWAS.
• Integrated plasma proteomics, clinical measures, and PRS into machine and deep learning models for risk prediction.
• Created an unsupervised endotyping framework using perfusion PET-derived myocardial blood flow data to identify patient subgroups.

Main Results:

• Integrated multimodal data (genetics, proteomics, imaging) significantly improved CMVD risk prediction.
• The novel imaging-based endotyping framework revealed distinct CMVD patient subgroups.
• This stratification approach achieved AUROCs between 0.65 and 0.73 per class, outperforming traditional models.

Conclusions:

• The study presents the first imaging-based endotyping for CMVD, integrating genetic and proteomic data for risk prediction.
• Multimodal data integration and imaging-based stratification offer a framework for more precise and personalized CMVD diagnosis.
• This approach captures CMVD's underlying heterogeneity, paving the way for tailored patient management.