Sex-specific proteomic signatures improve cardiovascular risk prediction for the general population without cardiovascular disease or diabetes

Abstract Importance Accurate prediction of 10-year major adverse cardiovascular events (MACE) is crucial for effective cardiovascular disease prevention and management. Objective To evaluate whether adding sex-specific proteomic profiles to the SCORE2 model enhances 10-year MACE risk prediction in the large UK Biobank (UKB) cohort. Design, Setting, and Participants Data from 47,382 UKB participants, aged 40 to 69 years without prior cardiovascular disease or diabetes, were utilized. The cohort was randomly divided into derivation (70%) and validation (30%) sets. Exposures Proteomic profiling of plasma samples was conducted using the Olink Explore 3072 platform, measuring 2,923 unique proteins, of which 2,085 could be used. Sex-specific Least Absolute Shrinkage and Selection Operator (LASSO) regression was used for biomarker selection. Main Outcomes and Measures The primary outcome was 10-year MACE incidence, defined as cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke. Model performance was assessed by changes in Harrell’s C-index, net reclassification index (NRI), and integrated discrimination index (IDI). Results During 10-year follow-up, 2,163 participants experienced MACE. Overall, 18 proteins were selected by LASSO regression, with 5 of them identified in both sexes, 7 only in males, and 6 only in females. Incorporating these proteins, significantly improved the C-index of the SCORE2 model from 0.713 to 0.778 ( P □<□0.001) in the total population. The improvement was greater in males (C-index increase from 0.684 to 0.771; Δ□=□+0.087) than in females (from 0.720 to 0.769; Δ□=□+0.049). The NRI was 19.9% for the total population, 36.3% for males, and 18.2% for females. The WAP four-disulfide core domain protein (WFDC2), which modulates extracellular matrix degradation, impacting fibrosis and plaque stability, and the growth/differentiation factor 15 (GDF15), reflecting increased inflammatory activity, were the proteins contributing the strongest C-index increase in both sexes; even more than the N-terminal prohormone of brain natriuretic peptide (NTproBNP), which was also selected. Conclusions and Relevance The derived s ex-specific 10-year MACE risk prediction models, combining 12 protein concentrations among men and 11 protein concentrations among women with the SCORE2 model, significantly improved the discriminative abilities of the SCORE2 model. This study shows the potential of sex-specific proteomic profiles for enhanced cardiovascular risk stratification and personalized prevention strategies.