Alcohol and aging: Next-generation epigenetic clocks predict biological age acceleration in individuals with alcohol use disorder

Abstract

Background: Chronic heavy alcohol use is a major risk factor for premature aging and age-related diseases. DNA methylation (DNAm)-based epigenetic clocks are novel tools for predicting biological age. However, the newest configurations, causality-enriched epigenetic clocks, have not been assessed in the context of alcohol consumption and alcohol use disorder (AUD).

Methods: Epigenetic aging was evaluated in a sample of 615 individuals (372 AUD patients and 243 healthy controls) by applying the GrimAge Version 1 (V1) and Version 2 (V2) clocks alongside three causality-enriched clocks (CausAge, DamAge, and AdaptAge). A linear model controlling for AUD diagnosis, sex, race, BMI, smoking status, and five blood cell types was leveraged to test associations between alcohol-related metrics and age-adjusted epigenetic clocks.

Results: GrimAge V1 and V2 maintained significant associations with AUD and drinking behavior measures within the total sample and both the young (<40 years old) and old (≥40 years old) subgroups. Generally, GrimAge V2 slightly outperformed GrimAge V1, while none of the causality-enriched epigenetic clocks demonstrated significant associations with AUD. However, in the young subgroup, DamAge had a significant association with the total number of drinks. Across the total sample and age subgroups, with liver function enzymes, GrimAge V2 consistently sustained stronger associations compared with GrimAge V1. Among fourth-generation clocks, DamAge exhibited significant associations with gamma-glutamyl transferase (GGT) and aspartate aminotransferase in the total sample and young subgroup; CausAge displayed a significant association with GGT in the total sample. Examining clinical biomarkers, GrimAge V2 showed improved associations with C-reactive protein compared to GrimAge V1 in the total sample and age subgroups.

Conclusions: Overall, we observed moderately improved performance of GrimAge V2 compared with GrimAge V1 with the majority of the parameters tested. The causality-enriched epigenetic clocks lacked significant associations but demonstrate the complexities of aging and inspire further research of AUD and drinking dynamics.

Keywords: DNA methylation clock; alcohol consumption; alcohol use disorder; epigenetic aging.

FIGURE 1

Boxplots illustrating epigenetic age acceleration (EAA) in alcohol use disorder (AUD) groups across different epigenetic clock models without adjustment for covariates. (A) EAA of DNAm GrimAge Version 1 (V1); (B) EAA of DNAm GrimAge Version 2 (V2); (C) EAA of CausAge; (D) EAA of DamAge; (E) EAA of AdaptAge. AUD, alcohol use disorder; EAA, epigenetic age acceleration; HC, healthy control. Significance codes are as follows: “***” when p‐value (p) < 0.0001, “**” when p < 0.001, “*” when p < 0.01, “·” when p < 0.05, ‘NS’ (not significant) otherwise. p‐values were obtained through two sample t‐tests.