A novel plasma proteomic signature predicts long-term incident heart failure risk among persons living with HIV

Abstract Background: Persons with HIV (PWH) have an increased risk of heart failure compared with the general population despite effective viral suppression. No methods currently exist to predict heart failure risk in PWH. We sought to develop a proteomics-based risk model of incident heart failure in a longitudinal cohort of PWH in the Veterans Aging Cohort Study Biomarker Cohort. Methods: We characterized the baseline plasma proteome of 1,398 PWH using the SomaScan 4.0 aptamer-based platform. We incorporated traditional risk factors, HIV-specific variables, and inflammatory/coagulation markers into Cox proportional hazards models of incident heart failure. For protein risk models, we used least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation to identify the optimal set of predictors. Model performance was evaluated using Harrell’s Concordance Index (C-index). Results: Among 1,398 PWH, 240 (17.1%) developed incident heart failure over 13.1 years of median follow-up. The sample was randomly divided 2:1 into training (n = 932) and test (n = 466) sets. In the training set, 4,248 proteins were screened, identifying 227 significant markers (Bonferroni-corrected p-value