Proteomic Profiles of Obesity-Related Phenotypes and Incident Cardiovascular Events

BackgroundObesity is a modifiable cardiovascular risk factor. Proteomic profiling may improve the understanding of obesity and cardiovascular risk prediction. This study explores the use of protein-predicted scores for body mass index (PPSBMI), body fat percentage (PPSBFP), and waist-hip ratio (PPSWHR) in predicting major adverse cardiovascular events (MACE). MethodsWe used data from the UK Biobank with proteome profiling. PPSBMI, PPSBFP, and PPSWHR were derived using LASSO algorithm. The association between these protein scores and incident MACE was evaluated using competing risk model. MACE prediction using the protein scores was compared with the PREVENT equation. ResultsStrong correlations were observed between protein-predicted obesity phenotypes and their measured counterparts (R2: BMI = 0.78, BFP = 0.85, WHR = 0.63). Per standard deviation of elevated PPSBFP and PPSWHR, but not PPSBMI, was associated with increased risk of MACE (Hazard Ratio [HR] 1.25, 95% CI 1.14 - 1.38; HR 1.15, 95% CI 1.06 - 1.24, respectively), independent of traditional risk factors and outperformed the measured obesity-related phenotypes. For predicting MACE, compared with the PREVENT equation (C-statistic 0.694), the models adjusted for only age, sex, current smoking, and protein scores of obesity showed comparable performance (C-statistics 0.685 - 0.688). ConclusionProtein-predicted scores are better predictors of MACE than measured obesity phenotypes. Integration of protein-predicted scores provides a biologically relevant risk assessment using a single biochemical assay to potentially simplify and standardize the measure of obesity risk in clinical practice.