Epigenetic vs chronologic age for cardiovascular risk assessment

Background/Introduction Chronological age is a dominant determinant of cardiovascular risk prediction. However, DNA methylation-derived epigenetic clocks capture biological aging processes that may reflect cumulative cardiometabolic and inflammatory exposures. Whether biological age improves contemporary cardiovascular risk estimation, including the PREVENT equations, remains uncertain. Purpose To evaluate whether substituting chronological age with epigenetic age within PREVENT risk equations improves discrimination for long-term all-cause and cardiovascular mortality, and to determine whether epigenetic age provides independent prognostic information beyond chronological PREVENT risk. Methods We conducted a retrospective cohort study of 1,516 primary prevention participants aged 50–79 years from NHANES 1999–2002 with linked mortality follow-up through 2019. Chronologic PREVENT risk was calculated using published equations. Biologic PREVENT variants were constructed by substituting chronologic age with eight epigenetic clocks (Horvath, Hannum, Skin-Blood, PhenoAge, Lin, Weidner, Vidal-Bralo, Zhang). Survey-weighted Cox models estimated hazard ratios (HRs) per 5-percentage-point higher predicted risk. Discrimination was assessed using Harrell’s C statistics. Additional models evaluated epigenetic age independently adjusted for chronologic PREVENT risk. Results Mean age was 60.3±8.0 years; 507 all-cause and 140 cardiovascular deaths occurred over 17.4 years. Each 5% higher chronologic PREVENT risk was associated with increased all-cause mortality (HR 1.57, 95% CI 1.45–1.70) and cardiovascular mortality (HR 1.66, 95% CI 1.50–1.84). Biologic PREVENT variants showed similar associations (e.g., Hannum HR 1.51 for all-cause mortality). Chronologic PREVENT demonstrated superior discrimination (C=0.744 for all-cause mortality) compared with epigenetic substitutions (ΔC −0.011 to −0.081). After adjustment for chronologic PREVENT risk, multiple epigenetic clocks remained independently associated with mortality (e.g., Zhang HR 1.89 for all-cause mortality per 5-year higher age). Associations persisted within both low (<10%) and high (≥10%) baseline PREVENT risk strata. Conclusions Epigenetic age is strongly associated with long-term mortality but does not improve discrimination when substituted for chronological age in PREVENT models. Biological aging appears to function as a complementary marker of residual cardiovascular risk rather than a replacement for chronological age in established prediction frameworks.Figure 1Figure 2