DNA methylation data‐based prognosis‐subtype distinctions in patients with esophageal carcinoma by bioinformatic studies. Issue 3 (23rd August 2020)

• Record Type: Journal Article
• Authors: Chen, Hui; Qin, Qin; Xu, Zhipeng; Chen, Tingting; Yao, Xijuan; Xu, Bing; Sun, Xinchen
• Is Part Of: Journal of cellular physiology. Volume 236:Issue 3(2021)
• Page Start: 2126
• Page End: 2138
• Publication Date: 2020-08-23
• Subjects: CpGs -- DNA methylation -- esophageal carcinoma -- TCGA -- WGCNA; Physiology -- Periodicals; Cell physiology -- Periodicals; 571.6
• DOI: 10.1002/jcp.29999
• Languages: English
• ISSNs: 0021-9541

Abstract

Abstract: Esophageal carcinoma (ESCA) is caused by the accumulation of genetic and epigenetic alterations in esophageal mucosa. Of note, the earliest and the most frequent molecular behavior in the complicated pathogenesis of ESCA is DNA methylation. In the present study, we downloaded data of 178 samples from The Cancer Genome Atlas (TCGA) database to explore specific DNA methylation sites that affect prognosis in ESCA patients. Consequently, we identified 1, 098 CpGs that were significantly associated with patient prognosis. Hence, these CpGs were used for consensus clustering of the 178 samples into seven clusters. Specifically, the samples in each group were different in terms of age, gender, tumor stage, histological type, metastatic status, and patient prognosis. We further analyzed 1, 224 genes in the corresponding promoter regions of the 1, 098 methylation sites, and enriched these genes in biological pathways with close correlation to cellular metabolism, enzymatic synthesis, and mitochondrial autophagy. In addition, nine representative specific methylation sites were screened using the weighted gene coexpression network analysis. Finally, a prognostic prediction model for ESCA patients was built in both training and validation cohorts. In summary, our study revealed that classification based on specific DNA methylation sites could reflect ESCA heterogeneity and contribute to the improvement of individualized treatment and precise prognostic prediction. Abstract : We identified specific methylation sites utilizing The Cancer Genome Atlas database, and subsequently performed serial bioinformatics at the prognostic level. We also constructed a prognostic prediction model for esophageal carcinoma (ESCA) patients to identify novel markers for ESCA, determine various targets for ESCA precision medicine, and more accurately classify molecular subtypes in ESCA patients. Of note, this promising model will assist and guide clinicians in clinical diagnosis, therapy, and prognostic assessment of ESCA patients with different epigenetic subtypes.