Subclinical Atherosclerosis Risk Can Be Predicted in Female Patients With Systemic Lupus Erythematosus Using...

Area of Science:

• Cardiovascular Health
• Metabolomics
• Systemic Lupus Erythematosus (SLE) Research

Background:

• Cardiovascular disease (CVD) is a primary cause of mortality in women with systemic lupus erythematosus (SLE).
• Accelerated atherosclerosis in SLE patients is often undetected by current CVD risk scores.
• There is a need for improved CVD risk prediction in this vulnerable population.

Purpose of the Study:

• To develop and validate a novel, female-focused predictive atherosclerosis risk signature.
• To utilize serum metabolites and machine learning for enhanced risk prediction in SLE.
• To improve cardiovascular risk assessment and management strategies for women with SLE.

Main Methods:

• Serum metabolomics analysis was performed on female SLE patients.
• Machine learning pipelines were employed to identify predictive metabolite signatures.
• Developed signatures were validated in independent SLE cohorts using vascular ultrasound and plaque status.
• A refined 5-feature score was developed to predict subclinical plaque and progression.

Main Results:

• A 35-metabolite/5-clinical trait signature significantly outperformed existing CVD risk scores and routine lipid profiles.
• The validated atherosclerosis risk signature accurately predicted plaque status (AUC=0.79) and progression (AUC=0.79).
• A refined 5-feature score achieved high accuracy (AUC=0.84) in stratifying atherosclerosis risk and identified distinct risk subgroups.

Conclusions:

• The developed atherosclerosis risk score shows potential for improved CVD risk assessment in female SLE patients.
• This score could enhance clinical management strategies for cardiovascular health in SLE.
• Further validation in diverse cohorts is recommended to support broader clinical application.