A Latent Autoantibody Axis Associated with Vascular Vulnerability in Ischemic Stroke: Integrated Statistical and...

Area of Science:

• Cardiovascular Research
• Immunology
• Biomarker Discovery

Background:

• Ischemic stroke is a leading cause of death and disability globally.
• Identifying individuals at high vascular risk requires improved strategies.
• The role of serum autoantibodies in predicting stroke risk is not fully understood.

Purpose of the Study:

• To investigate the predictive value of specific serum autoantibodies (anti-PDCD11, anti-DNAJC2, anti-PAI-1) for ischemic stroke.
• To explore the integrative biological significance of these autoantibodies beyond established clinical risk factors.
• To assess the performance of machine-learning models incorporating antibody markers for stroke risk prediction.

Main Methods:

• Analysis of 833 participants (stroke patients and controls).
• Quantification of serum autoantibodies and construction of logistic regression and machine-learning models.
• Evaluation of model performance using cross-validation, calibration, and reclassification indices.
• Exploration of biological structures using PCA, unsupervised clustering, and SHAP analysis.

Main Results:

• Clinical models showed excellent stroke discrimination (AUC ~0.92).
• Addition of antibody markers did not significantly improve overall model performance or reclassification.
• SHAP analysis indicated antibody markers are influential contributors, and PCA revealed a dominant antibody component associated with stroke risk.
• Unsupervised clustering identified a high-risk subgroup with elevated antibody levels.

Conclusions:

• A latent antibody axis is associated with vascular vulnerability in stroke.
• Autoantibody profiling captures integrated biological signals of vascular and immunological stress.
• While not enhancing global prediction, autoantibodies may improve biological characterization of stroke susceptibility, requiring prospective validation.