Abstract P028: Methylation biomarkers for colorectal cancer early detection and survival prognostics impact gene expression and link to cancer-related biological pathways
Pol Canal Noguer,Alejandro Requena Bermejo,Francesco Mattia Mancuso,Juan Carlos Higareda,Marina Manrique López,Marko Chersicola,Pablo Pérez Martínez,Pablo Antonio Camino,Primoz Knap,Vivian Erklavec Zajec,Kristi Kruusmaa
DOI: https://doi.org/10.1158/1940-6215.precprev22-p028
2023-01-05
Cancer Prevention Research
Abstract:Background: DNA methylation has been previously shown to have diagnostic and predictive potential for colorectal cancer (CRC). Aim of this study was to evaluate putative methylation markers in the context of early cancer development and diagnostics as well as further investigate the biological significance of these regions. Methods: Biomarker discovery was done by whole genome bisulfite sequencing (WGBS) of 88 CRC, 48 advanced adenoma (AA) and corresponding adjacent normal tissue (NAT) samples. Short-list of significantly hypermethylated regions (DMRs) was correlated to transcriptomics data from 512 CRC patients in The Cancer Genome Atlas (TCGA) cohort. Pathway enrichment for biological pathway analysis of the DMRs was done by using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database. Survival analysis was performed using Kaplan–Meier method on sub-groups of patients divided by the methylation status of individual markers. Finally, individual marker significance of selected regions was evaluated by analyzing 26 plasma samples from early stage (stage I-IIA) CRC samples and 42 colonoscopy verified controls (CNT) with targeted methylation sequencing assay. Results: 4167 putative marker regions were identified from biomarker discovery with WGBS. Differential signal could be observed both between AA and NAT and CRC and NAT, while several of these regions were differentially methylated also between AA and CRC samples, indicating biological signal change with adenoma progression to cancer. 84 hypermethylated DMRs from several verification studies were further evaluated against transcriptome data from TCGA, where overlap for 69 genes was found. 19 of these genes showed a significant down- regulation (p< 0.05), indicating a link between hypermethylation and gene expression. 2 genes showed significant up-regulation (p< 0.05), which could indicate other epigenic processes to be in place. KEGG pathway analysis revealed that the top pathways involved were axonal guidance, ephrin receptor signaling, epithelial-mesenchymal transition and FGF signaling, which all play significant role in the context of cancer development and progression. Kaplan-Meier analysis showed significant correlation to patients 5- year survival prediction linked to 3 genes: FGF14 (p=0.025, HR = 1.75) DPY19L2P1 (p=0.012, HR = 1.86), PTPRO (p=0.046, HR = 1.63). Targeted sequencing analysis on plasma samples of patients with early stage (I-IIA) colorectal cancer and age and gender matching colonoscopy-verified controls, showed high individual marker accuracy with AUC= 0.78 for FGF14, AUC= 0.81 for DPY19L2P1 and AUC= 0.73 for PTPRO. Conclusions: Methylation markers have distinct signals in early development of CRC, with high individual accuracy for separating early-stage cancers from matching controls. These regions have impact on gene expression and can be linked to relevant biological pathways. Extending early detection potential of the markers to further prognostics and stratification, could lead to better outcomes and improved survival of the patients. Citation Format: Pol Canal Noguer, Alejandro Requena Bermejo, Francesco Mattia Mancuso, Juan Carlos Higareda, Marina Manrique López, Marko Chersicola, Pablo Pérez Martínez, Pablo Antonio Camino, Primoz Knap, Vivian Erklavec Zajec, Kristi Kruusmaa. Methylation biomarkers for colorectal cancer early detection and survival prognostics impact gene expression and link to cancer-related biological pathways. [abstract]. In: Proceedings of the AACR Special Conference: Precision Prevention, Early Detection, and Interception of Cancer; 2022 Nov 17-19; Austin, TX. Philadelphia (PA): AACR; Can Prev Res 2023;16(1 Suppl): nr P028.
oncology