Abstract 3496: EpigenPlot: A tool for the gene-level methylation analysis of colorectal tumors

Abstract

Abstract Introduction. Aberrant DNA methylation patterns are frequent in colorectal adenocarcinoma (CRC) tissues. The investigation of this layer of epigenetic regulation holds the promise of a better understanding of tumour progression and the identification of novel biomarker candidates. Purpose. Our goal was to collect and analyse publicly available DNA methylation data of colorectal samples and create a user-friendly tool for exploring and visualizing the CRC methylome. Methods. We used Illumina Infinium Methylation Arrays, which is a popular platform for genome-wide methylation analysis of large cohorts. Gene Expression Omnibus database (GEO) and TCGA-COAD (The Cancer Genome Atlas Colon Adenocarcinoma) datasets including normal colorectal mucosa, adenoma, and adenocarcinoma samples were systematically collected and filtered. Differentially methylated regions were identified using the Kruskal-Wallis test. Genes with significant methylation changes were further analysed using ROC (Receiver Operator Characteristic) analysis. Finally, we established a web application using the shiny R package. Results. We assembled a database including methylation data of 2,586 samples. Adenoma and adenocarcinoma samples were globally hypomethylated. Hypermethylation (p < 0.05, Δβ >= 0.2) in adenocarcinoma samples was predominantly present in regions near proximal promoters. Out of the 74 previously proposed biomarkers examined, 50 had a cross-validated AUC of over 0.8. FDA-approved biomarkers BMP3, NDRG4, and SEPT9 have shown the highest performance in the TSS200 region (cvAUC = 0.8, 0.83, and 0.76, respectively). Genes with the highest performance were ITGA4 (5’UTR, first exon: cvAUC = 0.9), MDFI (5’UTR: cvAUC = 0.9), CNRIP1 (5’UTR, first exon: cvAUC = 0.89). The established web application can visualize methylation at CpG sites and gene regions, and KEGG pathway genes. The platform is available at https://epigenplot.com. Conclusion. We collected and analysed a sizeable database of colorectal tissue data. Gene regions with the best performance include previously proposed and novel diagnostic biomarker candidates. The web platform provides an tool for the efficient exploration of the assembled data. Citation Format: Dalma Müller, Balazs Gyorffy. EpigenPlot: A tool for the gene-level methylation analysis of colorectal tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3496.