Abstract Body (Do not enter title and authors here): Background: Platelet hyperreactivity is a predictor of increased risk for cardiovascular (CV) events; however, its measurement remains technically challenging. The assessment of readily measurable circulating biomarkers offers a practical strategy to identify individuals with platelet hyperreactivity and increased CV risk.

Objective: Our aim was to develop a PROtein Signature of PlatelEt Reactivity (PROSPER) that correlates with platelet activity and CV outcomes.

Methods: We used a cohort of healthy individuals (n=40) with measurement of platelet aggregation in response to submaximal ADP, collagen, epinephrine, and arachidonic acid. Additionally, targeted proteomic profiling of plasma was performed via Olink Explore Cardiometabolic panel, consisting of 384 proteins. We developed PROSPER by integrating platelet aggregation measurements with proteomic data using a boosted forest regressor model. PROSPER was validated in an independent cohort of participants (n=33) with matched platelet aggregation and proteome measures. To evaluate its clinical relevance, PROSPER was calculated for 50,689 participants in the UK Biobank (UKB). Associations between PROSPER and incident CV events were assessed using Cox proportional hazards models, adjusting for age, sex, race, ethnicity, diabetes, and body mass index.

Results: In a cohort of healthy individuals (mean age 29±9.2 years, 46% female, 55% White) we developed PROSPER (Figure 1a), a score consisting of 40 unique plasma proteins. In an independent validation cohort (n=33), PROSPER was significantly associated with a composite of platelet activity (r=0.53, p=0.002; Figure 1b). We then applied PROSPER to 50,689 participants in the UKB without established CV disease (Figure 1c). Over a mean follow-up of 13.6 years, higher PROSPER scores were associated with increased risk of CV death or myocardial infarction (MI) (Figure 1d). Compared to individuals in the lowest PROSPER quintile, those in the highest quintile had a significantly elevated risk of CV death or MI (adjHR 1.62, 95% CI 1.35-1.92; Figure 1e)

Conclusions: We developed a circulating protein signature (PROSPER) that reflects heightended platelet activity and is independently associated with CV risk. This proteomic score has the potential to serve as a diagnostic tool and may enable a more personalized approach to CV risk reduction.