Researchers have identified new plasma proteomic signatures that could significantly improve the assessment of atherosclerosis burden, a condition that often progresses silently for years before leading to severe cardiovascular events such as myocardial infarction or stroke. An international team of scientists, primarily from Boston and Munich, presented their findings at the recent ASHG Conference in Boston, outlining how these signatures may serve as effective biomarkers for cardiovascular risk.
The team utilized advanced machine learning techniques, specifically CatBoost, to analyze plasma proteomes from the UK Biobank involving 44,788 participants. They focused on Olink Explore 3072, which includes data on 2,920 proteins, to derive four distinct proteomic signatures. These were categorized into biologically-informed protein sets: the whole proteome (WholeProteome; n=2,920), proteins related to genetic predisposition for atherosclerosis (Genetic; n=402), those implicated in atherogenesis (Mechanistic; n=680), and proteins enriched in arterial tissue (Arterial; n=248).
The study found that these signatures could robustly differentiate individuals with established atherosclerotic disease from matched controls, achieving a receiver operating characteristic area under the curve (ROC-AUC) of up to 0.91 (95% confidence interval: 0.89–0.93). Furthermore, among 41,200 individuals without baseline atherosclerosis, all four signatures showed strong associations with the occurrence of major adverse cardiovascular events over a median follow-up period of 13.7 years. Notably, the WholeProteome signature indicated a hazard ratio (HR) of 1.70 for each standard deviation increase, demonstrating its potential to enhance risk discrimination.
Implications for Cardiovascular Health
The correlation between signature levels, the number of clinically affected vascular beds, and carotid ultrasound-measured plaque burden underscores their predictive power. The researchers successfully validated these findings in external cohorts, including the KORA S4 (n=1,361) and KORA-Age1 (n=796) studies. Their longitudinal analyses revealed that individuals with steeper annual increases in signature trajectories often had higher baseline risk factors or experienced subsequent major adverse cardiovascular events.
The scientists also noted consistent associations between the proteomic signatures and polygenic risk scores for coronary artery disease, highlighting their ability to capture genetically influenced atherosclerotic burden.
These findings indicate that proteomic signatures can effectively reflect the burden of atherosclerosis and enhance cardiovascular risk prediction in asymptomatic individuals. The researchers suggest that plasma proteomics may provide a scalable and accessible alternative to traditional imaging techniques for identifying subclinical atherosclerosis, ultimately supporting preventive strategies for cardiovascular diseases.
This innovative approach to assessing atherosclerosis burden could revolutionize how healthcare professionals monitor cardiovascular health and implement preventive measures, paving the way for improved patient outcomes in the future.