Elevated Lipoprotein(a) Independently Increases Risk for Short-Term Atherosclerotic Cardiovascular Events in Machine...

Area of Science:

• Cardiology
• Preventive Medicine
• Biomarkers

Background:

• Lipoprotein(a) [Lp(a)] is an underutilized biomarker in predicting short-term atherosclerotic cardiovascular disease (ASCVD) risk.
• Current risk prediction models often do not fully incorporate Lp(a) levels.

Purpose of the Study:

• To investigate the contribution of Lp(a) to short-term ASCVD event prediction.
• To develop and validate machine learning (ML) models incorporating Lp(a) for predicting ASCVD events using real-world data.

Main Methods:

• Utilized a large claims database (731,983 individuals) to analyze the association of Lp(a) with incident ASCVD and all-cause mortality.
• Developed novel ML models to predict 1, 2, and 3-year incident ASCVD events, validated in an independent cohort (53,930 patients).

Main Results:

• Elevated Lp(a) levels were independently associated with increased ASCVD events and all-cause mortality.
• ML models incorporating Lp(a) demonstrated robust discrimination for 1, 2, and 3-year ASCVD prediction (C-statistic: 0.83-0.84).
• Lp(a) showed a greater contribution to 1-year ASCVD prediction than traditional risk factors like smoking and diabetes.

Conclusions:

• Lp(a) is a significant predictor of short-term ASCVD risk.
• Assessing Lp(a) levels can help identify patients who may benefit from intensified preventive therapies for imminent ASCVD risk.