Persistent Circulating Proteomic Signatures Distinguish Magnetic Resonance Angiography-Defined Carotid Plaque Vulnerability

Background Carotid artery luminal stenosis alone may not predict the risk of future neurovascular events. Plaque vulnerability is emerging as an independent predictor of higher risk for a neurovascular event. We evaluated blood proteomics, paired with high-resolution cross-sectional imaging preoperatively. From preoperative and postoperative blood samples, our objective was to identify blood proteins that distinguish patients with vulnerable and non-vulnerable carotid artery plaques. Materials and methods We conducted an exploratory, single-center proteomic study of 27 consecutive patients undergoing carotid endarterectomy with preoperative 3.0T magnetic resonance angiography with vessel wall imaging. Plaques were classified as vulnerable when magnetic resonance angiography with vessel wall imaging identified at least one high-risk feature: thin fibrous cap, lipid-rich necrotic core, ulceration, intraplaque hemorrhage, or inflammation. Serum samples were analyzed preoperatively and postoperatively using a high-throughput, untargeted proteomics technology (Olink). Differential protein abundance between vulnerable and non-vulnerable groups was assessed separately at each time point. Results For both time points, the vulnerable group demonstrated a blood proteomic signature involving diacylglycerol signaling and oxidative stress proteins, including diacylglycerol kinase iota and peroxiredoxin-like 2B, with postoperative vulnerability additionally associated with platelet activation (Ras guanyl releasing protein 2), cytoskeletal remodeling (metastasis suppressor 1), and ribonucleic acid/ribosome-related proteins (Shwachman-Bodian-Diamond syndrome protein, RNA-binding protein with multiple splicing 2, zinc ribbon domain-containing protein 2). Non-vulnerable plaques showed a higher abundance of proteins involved in cellular maintenance and deoxyribonucleic acid repair. Furthermore, the persistence of higher diacylglycerol kinase iota and peroxiredoxin-like 2B in the vulnerable group and X-ray repair cross-complementing protein 2 in the non-vulnerable group at both time points suggests stable, patient-level systemic differences detectable in blood, despite plaque removal. Conclusions Serum proteomics identified distinct systemic pathways associated with cross-sectional imaging carotid plaque vulnerability, highlighting diacylglycerol signaling, oxidative stress, and platelet activation as candidate mechanisms and new biomarker candidates for risk stratification, symptom surveillance, and targeted medical interventions.