Low-Density Lipoprotein Cholesterol 4: The Notable Risk Factor of Coronary Artery Disease Development

Area of Science:

• Cardiology
• Biochemistry
• Data Science

Background:

• Coronary artery disease (CAD) is a leading global cause of mortality, often progressing asymptomatically.
• Early risk assessment in asymptomatic individuals is critical for timely intervention.
• Identifying novel biomarkers for CAD risk stratification is a significant clinical need.

Purpose of the Study:

• To investigate the association between low-density lipoprotein cholesterol (LDL-C) subfractions and CAD.
• To develop and evaluate machine learning models for predicting CAD risk.
• To identify key predictors for CAD development in a diverse patient cohort.

Main Methods:

• Recruitment of 356 CAD patients and 164 non-CAD controls diagnosed via coronary angiography.
• Quantification of LDL-C subfractions using the Lipoprint system and analysis of clinical data.
• Development and comparison of six machine learning models for CAD risk prediction.

Main Results:

• Elevated levels of triglycerides, LDLC-3, LDLC-4, LDLC-5, LDLC-6, and total small and dense LDL-C were observed in CAD patients.
• Male sex, older age, higher BMI, smoking, drinking, hypertension, and diabetes mellitus were significant clinical risk factors.
• LDLC-3, LDLC-4, and LDLC-5 were identified as significant lipid-based risk factors, with LDLC-4 being a primary predictor in the XGboost model (AUC 0.945).

Conclusions:

• Machine learning models, especially XGboost, demonstrate high accuracy in predicting CAD risk.
• Specific LDL-C subfractions, particularly LDLC-4, are valuable biomarkers for identifying individuals at high risk of CAD.
• These predictive models can facilitate early screening and preventive strategies for asymptomatic populations, potentially averting severe cardiovascular events.