Spatially resolved proteomic signatures of atherosclerotic carotid artery disease

Atherosclerotic plaque rupture remains a leading cause of adverse cardiovascular events, yet the molecular drivers of plaque vulnerability are incompletely understood. To address this challenge, we developed an integrated approach that combines histomorphology-guided spatial proteomics with machine learning to map protein signatures across spatially distinct plaque subregions. Our analysis revealed that vulnerability signatures concentrate in the necrotic core and fibrous cap subregions, and are significantly enriched for ossification, inflammation, cholesterol metabolism, and extracellular matrix degradation pathways. When comparing the vulnerability status across subregions, we found that the necrotic core has the most distinctive vulnerability-associated proteome, with 454 proteins significantly altered between stable and vulnerable states. We identified a mechanistic link between inflammation and oxidative stress, PCSK9 upregulation, and vascular smooth muscle cell dysfunction in vulnerable plaques. This finding suggests arterial PCSK9 as a therapeutic target beyond its established role in hepatic lipid metabolism. By employing machine learning, we developed and independently validated a seven-protein tissue panel (receiver operating characteristic- area under the curve = 0.86) and found a 12-protein serum panel to predict plaque vulnerability status. Thus, plaque vulnerability signatures are spatially concentrated in specific subregions and highlight actionable biomarkers and therapeutic targets for advanced carotid artery disease.