Study reveals DNA methylation score that better detects alcohol intake than standard biomarkers

By Tarun Sai LomteReviewed by Susha Cheriyedath, M.Sc.Feb 9 2026

Scientists identify a DNA methylation alcohol biomarker that tracks exposure and aging signals, offering new insight into how drinking patterns may shape long-term health.

Study: Alcohol Consumption and DNA Methylation in a Mediterranean Cohort: A Focus on Oxidative Stress and Aging Biomarkers. Image Credit: PanuShot / Shutterstock

In a recent study published in the journal Antioxidants, researchers characterized the performance of a DNA methylation-based epigenomic biomarker of alcohol intake.

The relationship between alcohol intake and aging has received substantial interest. Genomic instability, epigenetic alterations, telomere attrition, chronic inflammation, cellular senescence, mitochondrial dysfunction, and altered intercellular communication, among others, are hallmarks of aging. Alcohol intake has been implicated in all of these hallmarks through increased oxidative stress and other mechanisms.

Since self-reported alcohol intake is prone to recall bias, it is essential to have objective biomarkers. Conventional biomarkers, such as gamma-glutamyl transferase (GGT), are used as indirect indicators and have limited specificity. Furthermore, while emergent biomarkers, including phosphatidylethanol and ethyl glucuronide, have improved specificity, they differ in their detection windows. As such, methylation-based biomarkers have been posited to address current limitations.

About the study

In the present study, researchers characterized a DNA methylation-based epigenomic biomarker of alcohol intake. Participants were adults aged 55 to 75 years from Valencia, Spain, referred to as the main cohort, recruited for the PREDIMED Plus Valencia study. A replication cohort of adults aged 55 to 80 years with high cardiovascular risk from the same region was also included. Participants in both cohorts were free of cardiovascular disease at baseline and represented predominantly older Mediterranean adults with relatively low to moderate alcohol intake, which may influence biomarker performance and generalizability.

Fasting blood samples were collected for biochemical analyses. Alcohol intake was self-reported using a food frequency questionnaire. Adherence to the Mediterranean diet was evaluated using a validated adherence score. Physical activity, education, and tobacco smoking were assessed using questionnaires. Genomic DNA was extracted from blood, followed by quantification and epigenome-wide methylation profiling.

A DNA methylation-based score for alcohol intake was estimated using 450 cytosine phosphate guanine (CpG) sites. General linear models were used to analyze associations between self-reported alcohol consumption and biomarkers of alcohol intake, including the epigenomic score and conventional biomarkers. The ability of the biomarkers to predict high alcohol intake was assessed using receiver operating characteristic curves, with performance dependent on how drinking categories were defined and distributed in each cohort.

The researchers also conducted epigenome-wide association studies (EWAS) to investigate the effects of self-reported alcohol intake and biomarkers of alcohol consumption on DNA methylation patterns in the main cohort. Associations with aging biomarkers were examined, including telomere length, PhenoAge, GrimAge, and CausalityAgeYing clocks, which are epigenetic aging indices estimating biological rather than chronological age. Interactions between alcohol intake and adherence to the Mediterranean diet were also evaluated.

Findings

The main cohort included 414 individuals with a mean age of 65.08 years. Average alcohol intake was 8.16 grams per day, with males consuming significantly more alcohol than females. Among beverage types, beer was consumed more frequently than wine and spirits. Plasma GGT was significantly associated with alcohol intake. The DNA methylation-derived epigenomic score was also significantly associated with alcohol consumption, showing a modest correlation with self-reported intake.

The epigenomic biomarker performed better at predicting high alcohol intake, with an area under the curve of 0.76 compared with 0.66 for GGT. However, the epigenomic score performed worse when distinguishing alcohol consumers from non-consumers, whereas GGT retained significant predictive performance. The replication cohort comprised 150 individuals with a mean age of 67.59 years; GGT data were unavailable for this cohort.

Alcohol intake in the replication cohort was marginally lower than in the main cohort, with persistent sex differences. The epigenomic biomarker again showed a significant association with alcohol intake and moderate discrimination of drinking categories, yielding an area under the curve of approximately 0.70. In this cohort, where heavy drinking was rare, performance reflected category definitions rather than a strong classification of heavy drinkers. The EWAS for self-reported alcohol intake identified cg06690548 as the most significant differentially methylated site, although most signals were suggestive rather than consistently reaching stringent epigenome-wide significance.

The cg06690548 site was annotated to the solute carrier family 7 member 11 (SLC7A11) gene, which plays a role in defense against oxidative stress. Alcohol consumption was inversely associated with methylation at this site, consistent with prior studies linking hypomethylation at cg06690548 to elevated alcohol intake. EWAS analyses of plasma GGT and the epigenomic biomarker identified the same site as the most significant, supporting its role as an exposure-related signature rather than a direct indicator of biological damage.

Higher alcohol intake was associated with shorter telomere length but not with other aging biomarkers. In contrast, plasma GGT and the epigenomic biomarker were significantly associated with all aging biomarkers in the main cohort. These associations were positive for GrimAge, PhenoAge, and CausalityAgeYing and inverse for telomere length. Higher Mediterranean diet adherence showed an exploratory interaction with alcohol intake and telomere length, but this association emerged only after collapsing alcohol intake into broader categories and was not consistent across models.

Conclusions

Overall, the study characterized a DNA methylation-based epigenomic biomarker of alcohol intake that was significantly associated with self-reported consumption and demonstrated predictive ability for distinguishing heavier drinkers within cohort-specific drinking patterns. The biomarker was also associated with multiple epigenetic aging indices and inversely with telomere length. These associations likely reflect cumulative alcohol exposure rather than definitive biological aging acceleration. The cross-sectional design, reliance on self-reported intake, relatively low prevalence of heavy drinking, and cohort-specific characteristics limit causal inference and broader applicability.