Blood Protein Signatures Refine Cardiovascular Risk Prediction in Cohort Study

Protein risk scores built from a small panel of blood-borne proteins can flag people whose atherosclerosis is silently progressing and sharpen the accuracy of standard cardiovascular risk prediction, according to a cohort study published in PLOS Medicine. The research, led by Yani Huang, Xiangfeng Lu and colleagues, suggests that measuring circulating proteins may help clinicians identify which individuals need earlier or more intensive prevention.

Subclinical atherosclerosis refers to early artery damage that has not yet caused symptoms, typically assessed by coronary artery calcium (CAC) scanning or detection of carotid plaque. Existing risk stratification tools, the authors note, inadequately identify individuals with subclinical atherosclerosis who could benefit from early intervention. The study set out to develop protein risk scores (ProtRSs) for the progression of these early changes and to test whether they add predictive value beyond traditional risk assessment for atherosclerotic cardiovascular disease (ASCVD).

The analysis drew on a community cohort of 903 individuals who underwent repeated assessments of subclinical atherosclerosis over six years, from 2008–2009 to 2013–2014. Using least absolute shrinkage and selection operator (LASSO) regression, the researchers selected biomarkers from a panel of 117 proteins to construct separate scores for CAC progression and for carotid plaque progression. Each score contained 14 proteins. These compact signatures explained 17.1% of the variance in CAC progression and 18.4% of the variance in plaque progression.

A key consistency check strengthened the findings: the effect sizes of the selected proteins on subclinical atherosclerosis progression showed strong concordance with their effects on incident ASCVD, with a correlation greater than 0.75. In other words, proteins associated with worsening artery disease were also associated with later cardiovascular events, supporting the biological plausibility of the scores.

To test clinical usefulness, the team evaluated the scores in a case-cohort of 1,835 individuals followed from 2007–2008 to 2021, with a median follow-up of 10.4 years. Each standard deviation increase in the CAC protein risk score was associated with a hazard ratio of 1.36 (95% confidence interval 1.20–1.54) for ASCVD risk, while each standard deviation increase in the plaque score carried a hazard ratio of 1.38 (95% CI 1.23–1.54). These associations were assessed on top of a guideline-recommended traditional risk score.

Adding the protein scores to the traditional risk score yielded a net reclassification improvement of 0.127, which was statistically significant (p < 0.001).

The reclassification results were concrete. Among individuals classified as intermediate risk by traditional assessment, 18.6% were moved into the high-risk category, a group with a 10-year ASCVD event rate of 10.9%, indicating a need for preventive interventions. Meanwhile, 32.8% of those the traditional score labeled high risk were identified as having an even higher event rate, 13.7%, suggesting they may require intensified management.

The authors conclude that protein risk scores for coronary calcification and carotid plaque progression may provide additional value over traditional risk assessment, highlighting their potential in refining risk stratification and guiding targeted intervention strategies.

The study’s main limitation, as the authors state, is that the predictive value of the protein risk scores warrants further large-scale external validation. The cohort was a single community-based population, and the scores were developed and tested within it, so performance in other populations remains to be established.

Subject of Research: Plasma protein risk scores for subclinical atherosclerosis progression and cardiovascular risk prediction

Article Title: Plasma proteomic signatures of subclinical atherosclerosis progression for cardiovascular risk stratification: A cohort study

Article References: Huang, Y., Gao, R., Li, J., Yang, B., Huang, K., Chen, S., Cao, J., Liu, X., Shen, C., Yu, L., Zhao, L., Li, Y., Hu, D., Huang, J., Lu, B., Gu, D., Liu, F., & Lu, X. (2026). Plasma proteomic signatures of subclinical atherosclerosis progression for cardiovascular risk stratification: A cohort study. PLOS Medicine, 23(10), e1005265. https://doi.org/10.1371/journal.pmed.1005265

Image Credits: AI Generated

DOI: 10.1371/journal.pmed.1005265

Keywords: proteomics, subclinical atherosclerosis, coronary artery calcium, carotid plaque, cardiovascular risk prediction, risk reclassification, cohort study, LASSO regression, ASCVD, biomarkers, preventive cardiology, PLOS Medicine