Introduction

Despite increasing attention to the rising incidence of Barrett’s esophagus (BE) and its progression to esophageal adenocarcinoma (EAC) in Western populations, the interplay between host factors and epigenetic alterations, including DNA methylation, remains poorly characterized (1). DNA methylation changes are among the earliest and most consistent molecular events observed in BE and EAC, but it is not yet well understood whether these changes are systematically correlated with demographic variables such as age and gender (2).

Aims & Methods

In this study, we explore whether methylation-based molecular subtypes exist within BE and EAC and examine how they are distributed across gender and age groups, leveraging integrated analysis of publicly available datasets. This approach may reveal novel insights into inter-individual variability in the epigenetic landscape of esophageal disease.
DNA methylation data from three GEO datasets (GSE104707, GSE81334, GSE89181) comprising 153 BE and 60 EAC samples were used. Samples with complete diagnostic, age, and gender annotations were retained, and the corresponding metadata was harmonized and used to build a cross-cohort to ensure robustness. Our previously defined Barrett’s progression-associated signature (3) was used to construct a unified methylation matrix across datasets. Batch effects were evaluated and corrected using ComBat, adjusting for diagnosis, age, and gender. Consensus clustering was used to identify molecular subtypes, and associations between clusters and clinical covariates were assessed using Wilcoxon rank-sum tests with FDR correction. Methylation patterns and BE status were visualized to explore subtype-specific features.

Results

Cross-cohort individual and integrated analyses identified two distinct molecular clusters, defined by broad differences in mean DNA methylation across regions selected by the progression-associated signature. Statistically significant differences in mean methylation were observed in both female (FDR = 1.92 × 10⁻⁸) and male (FDR = 1.03 × 10⁻⁶) groups, indicating gender-specific methylation variation stratified by cluster. Age-stratified analyses further revealed significant inter-cluster methylation differences within the 50–60 (FDR = 1.9 × 10⁻⁶) and 60–70 (FDR = 5.2 × 10⁻⁴) age groups, with a modest effect also observed in the 70–80 group (FDR = 1.26 × 10⁻²). These findings suggest the presence of age-associated methylation drift and gender-related differences; however, neither appears sufficient alone to fully account for the observed methylation patterns across BE and EAC patients.

Conclusion

Aging and male gender are established risk factors for BE and EAC, yet their impact on the epigenetic landscape remains poorly defined. Our findings reveal consistent methylation-based clustering across gender and age groups, suggesting that DNA methylation profiles capture biologically relevant variation linked to these demographics. This supports the utility of methylation signatures for early detection and risk stratification. However, key limitations such as sample size imbalances and male overrepresentation reflect broader biases in BE/EAC research and underscore the need for more representative study designs. Further confirmation studies using prospectively collected clinical samples are currently ongoing (NCT06523374) to advance the development of robust, demographically inclusive methylation-based biomarkers.

References

1. Kailasam A, Mittal SK, Agrawal DK. Epigenetics in the Pathogenesis of Esophageal Adenocarcinoma. Clin Transl Sci. 2015 Aug;8(4):394–402.
2. Kaz AM, Wong CJ, Varadan V, Willis JE, Chak A, Grady WM. Global DNA methylation patterns in Barrett’s esophagus, dysplastic Barrett’s, and esophageal adenocarcinoma are associated with BMI, gender, and tobacco use. Clin Epigenetics. 2016;8:111.
3. Dehkordi NR, Higareda-Almaraz JC, Mancuso FM, Kruusmaa K, Bertossi A. 365. METHYLATION BIOMARKERS TO STRATIFY BARRETT’S ESOPHAGUS PATIENTS PROGRESSING AND NON-PROGRESSING TO ESOPHAGEAL ADENOCARCINOMA. Diseases of the Esophagus. 2024 Sep; 37(Supplement_1)

Disclosure

N. Riahi Dehkordi and K. Kruusmaa are employees of Universal Diagnostics d.o.o., Ljubljana, Slovenia
J.C. Higareda-Almaraz and F.M. Mancuso are employees of Universal Diagnostics S.A., Seville, Spain