Abstract B017: Detection of aberrant DNA methylation profiles in saliva samples as a potential screening method for Barrett's Esophagus and esophageal adenocarcinoma

Abstract Background: Barret’s esophagus (BE) is a well-defined premalignant lesion of esophageal adenocarcinoma (EAC), that if identified early can be treated to reduce the risk of progression to cancer. Endoscopy is the gold standard in screening and diagnosis, but challenges of underutilization exist due to cost, resource limitations, and invasiveness. This represents a substantial missed opportunity for early detection and intervention, underscoring the need for an accessible, affordable, and less invasive screening test. BE and EAC have been shown to share a common molecular feature of aberrant DNA methylation, which is readily detectable in a variety of sample types, making it a favorable target in the development of a minimally invasive screening test. Aim: This is a pilot study to evaluate the suitability of DNA methylation analysis in saliva samples to serve as a potential minimally invasive screening method for BE and EAC. Methods: This was a nonrandomized observational study. From a parent study, saliva samples from 9 biopsy-proven healthy controls, 9 participants with treatment-naive BE, and 10 participants with treatment-naive EAC were selected to undergo human DNA analysis based on participant consent. DNA extraction was performed followed by Whole Genome Bisulfate Sequencing. Sequencing reads were trimmed to remove adapters and low-quality regions to yield genome-wide methylation profiles. Statistical analysis was performed between comparison groups of control vs BE, control vs EAC, and BE vs EAC using pairwise Mann-Whitney U test. Results: After data normalization and filtering, there were 676 CpGs retained for further evaluation. A statistically significant difference (defined as p<0.05) in overall methylation percenteges level was found in BE vs EAC (p=0.029). There was no significant difference found in the control vs BE (p=0.219) and control vs EAC (p=0.313) comparison groups. In a subgroup analysis BE was distinguished into BE with dysplasia (BED, n=4) and BE without dysplasia (BEND, n=5). Methylation percentage levels in BEND were found to be significantly different when compared to control (p=0.014) and EAC (p=0.00067). There were no significant differences for BED compared to control and EAC, and BED and BEND did not differ significantly from each other. Similar analysis was performed at the level of individual CpG sites, identifying 28 sites that are significantly differentially methylated across the 4 conditions. UCSC Genome Browser query of these CpG sites revealed associations with the genes AGGF1P9, C2orf69P4, and LINC01566 on chromosome 16. Conclusion: Analysis of DNA methylation levels in saliva samples shows promising prospects of serving as a screening tool for BE and EAC. Next steps in investigation include validation with a larger cohort and validation against methylation levels in corresponding tissue samples. The genes AGGF1P9, C2orf69P4, and LINC01566 may be candidate biomarkers of interest for studies to follow. Citation Format: Ying Yu, Ruslan Strogantsev, Elisabeth Goldman, Hisham Mohammed, Zhenzhen Zhang, Duygu Altinok Dindar. Detection of aberrant DNA methylation profiles in saliva samples as a potential screening method for Barrett's Esophagus and esophageal adenocarcinoma [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: DNA Methylation, Clonal Hematopoiesis, and Cancer; 2025 Feb 1-4; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2025;85(3 Suppl):Abstract nr B017.