Pericoronary Radiomics Signature for Non-Culprit Lesion Progression and Revascularization Decision in NSTE-ACS

Area of Science:

• Cardiovascular Imaging
• Radiomics
• Artificial Intelligence in Medicine

Background:

• Non-ST-elevation acute coronary syndrome (NSTE-ACS) poses a significant risk due to non-culprit lesions (NCLs).
• Early identification of high-risk NCLs is crucial for effective patient management and risk stratification.

Purpose of the Study:

• To develop a coronary CT angiography (CCTA)-based radiomics model of pericoronary adipose tissue (PCAT) for identifying high-risk NCLs in NSTE-ACS patients.
• To assess the model's ability to predict major adverse cardiovascular events (MACE) and non-calcified plaque progression.

Main Methods:

• A prospective cohort of 542 NSTE-ACS patients was analyzed using baseline CCTA.
• Radiomic features of PCAT were extracted to build a radiomics signature (Rad model) using machine learning.
• A combined clinic-radiomics model was developed and validated for predicting 4-year MACE and plaque progression.

Main Results:

• The PCAT radiomics model independently predicted MACE (HR, 1.988; p < 0.001).
• The combined model showed superior discrimination for 4-year MACE compared to the clinical model alone (AUC, 0.793 vs. 0.703; p < 0.05).
• A higher baseline Rad model score was significantly associated with annualized non-calcified plaque volume progression (β, 0.477; p < 0.001).

Conclusions:

• A CCTA-based PCAT radiomics model effectively identifies patients with NSTE-ACS at higher risk for future MACE.
• This radiomics approach is linked to accelerated plaque progression, offering a potential non-invasive tool for individualized risk stratification.