Epigenetic signatures of biological vulnerability and residual risk in heart failure
Background: DNA methylation (DNAm) signatures capture cumulative lifestyle exposures and biological aging. This prospective study evaluated whether DNAm-based scores and epigenetic aging clocks are associated with clinical outcomes and mortality in a multinational cohort of patients with heart failure (HF). Methods: We studied 2,594 patients with HF from 40 countries in the Global Congestive Heart Failure (G-CHF) registry with whole-blood DNAm data. Fifteen published DNAm-based scores and epigenetic aging clocks reflecting lifestyle, environmental and physiological exposures, inflammation, frailty, mortality risk, and biological aging were derived. Associations with HF hospitalization, cardiovascular death, and all-cause death were assessed using multivariable Cox regression adjusted for age, sex, ancestry, the MAGGIC risk score, and NT-proBNP. Incremental prognostic value was compared to MAGGIC score and NT-proBNP. Extreme DNAm profiles were defined as scores or clocks exceeding {+/-}1.5 standard deviations (s.d.) from the population mean. Results: Mean age was 62.7{+/-}14.0 years, 66.2% were male, and mean left ventricular ejection fraction was 40.1{+/-}14.1%. During a median follow-up of 3.0 years, 338 patients were hospitalized for HF, 349 died from cardiovascular causes, and 565 died from any cause. Higher epigenetic age and DNAm scores for CRP, frailty, and mortality were associated with increased risk, whereas higher diet-related DNAm scores were inversely associated. For all-cause death, adjusted hazard ratios per 1-s.d. were 1.36 (95% CI, 1.24-1.50) for GrimAge, 1.27 (95% CI, 1.17-1.38) for the DNAm score for CRP, 1.44 (95% CI, 1.28-1.63) for the DNAm score for frailty, and 1.48 (95% CI, 1.32-1.65) for the DNAm score for mortality, compared with 0.81 (95% CI, 0.75-0.88) and 0.86 (95% CI, 0.79-0.94) for the DNAm scores for Alternative Healthy Eating Index and Mediterranean Diet Score. These patterns were directionally consistent for cardiovascular death and weaker for HF hospitalization and were more pronounced among patients with lower clinical risk (MAGGIC<17, Pinteraction<0.05), particularly for all-cause death. Patients with 4-5 extreme-high DNAm scores or clocks had more than twice the risk of death (HR, 2.27; 95% CI, 1.65-3.13). Conclusions: DNAm-based scores and epigenetic aging clocks reflect multiple dimensions of biological vulnerability in HF and are associated with clinical outcomes and mortality beyond clinical risk factors.