Can a Combined Epigenetic-Biomarker Score Supplant Chronologic Age as an Independent Determinant of Incident Death in Adults With Dysglycemia

OBJECTIVES: Diabetes and prediabetes are associated with premature death and are recognized as conditions of accelerated biologic aging. To date, the best measurement of biologic age is chronologic age. Measures of biologic age that can replace chronologic age as a predictor of death can better approximate risk in affected individuals. METHODS: The relationship between 238 biomarkers, epigenetic age scores, and incident death was analyzed in 2,755 participants (mean age 63.7±8 years) in the Outcomes Reduction with an Initial Glargine Intervention (ORIGIN) trial. Independent biomarkers for death identified using Cox models with forward selection were used to derive a biomarker risk score, which, after validation, was combined with epigenetic scores. Hazards for death per standard deviation higher epigenetic-biomarker score, chronologic age, or the age after adjustment for the score were estimated, and the respective β coefficients were compared. RESULTS: Four hundred eighty-one participants died during a median follow-up of 6.2 years. Each standard deviation higher age increased the hazard of death 1.69-fold (95% confidence interval [CI] 1.58 to 1.81, β=0.53). When 11 independent death biomarkers were combined with 3 epigenetic risk scores to yield an epigenetic-biomarker score, each standard deviation higher score increased the hazard of death 3.27-fold (95% CI 2.90 to 3.68). Adding standardized age to this model yielded a β coefficient for age of 0.00 (p=0.93). C statistics for the epigenetic-biomarker score alone and age alone were 0.77 (95% CI 0.74 to 0.78) and 0.66 (95% CI 0.63 to 0.68), respectively (p<0.001 for the difference). CONCLUSION: An epigenetic-biomarker risk score is a better predictor of death than chronologic age.