Dan Luo1, QingLing Yang2, HaiBo Wang1, Mao Tan1, YanLei Zou3, Jian Liu4. 1. Department of General Surgery, Chengdu Fifth People's Hospital, 33 Mashi St, Chengdu, 610000, Sichuan, China. 2. Department of Pulmonary and Critical Care Medicine, Chengdu Fifth People's Hospital, 33 Mashi St, Chengdu, 610000, Sichuan, China. 3. Department of General Surgery, Chengdu Fifth People's Hospital, 33 Mashi St, Chengdu, 610000, Sichuan, China. 501298916@qq.com. 4. Department of General Surgery, Chengdu Fifth People's Hospital, 33 Mashi St, Chengdu, 610000, Sichuan, China. luodan178628131@163.com.
Abstract
BACKGROUND: The role(s) of epigenetic reprogramming in gastric cancer (GC) remain obscure. This study was designed to identify methylated gene markers with prognostic potential for GC. METHODS: Five datasets containing gene expression and methylation profiles from GC samples were collected from the GEO database, and subjected to meta-analysis. All five datasets were subjected to quality control and then differentially expressed genes (DEGs) and differentially expressed methylation genes (DEMGs) were selected using MetaDE. Correlations between gene expression and methylation status were analysed using Pearson coefficient correlation. Then, enrichment analyses were conducted to identify signature genes that were significantly different at both the gene expression and methylation levels. Cox regression analyses were performed to identify clinical factors and these were combined with the signature genes to create a prognosis-related predictive model. This model was then evaluated for predictive accuracy and then validated using a validation dataset. RESULTS: This study identified 1565 DEGs and 3754 DEMGs in total. Of these, 369 were differentially expressed at both the gene and methylation levels. We identified 12 signature genes including VEGFC, FBP1, NR3C1, NFE2L2, and DFNA5 which were combined with the clinical data to produce a novel prognostic model for GC. This model could effectively split GC patients into two groups, high- and low-risk with these observations being confirmed in the validation dataset. CONCLUSION: The differential methylation of the 12 signature genes, including VEGFC, FBP1, NR3C1, NFE2L2, and DFNA5, identified in this study may help to produce a functional predictive model for evaluating GC prognosis in clinical samples.
BACKGROUND: The role(s) of epigenetic reprogramming in gastric cancer (GC) remain obscure. This study was designed to identify methylated gene markers with prognostic potential for GC. METHODS: Five datasets containing gene expression and methylation profiles from GC samples were collected from the GEO database, and subjected to meta-analysis. All five datasets were subjected to quality control and then differentially expressed genes (DEGs) and differentially expressed methylation genes (DEMGs) were selected using MetaDE. Correlations between gene expression and methylation status were analysed using Pearson coefficient correlation. Then, enrichment analyses were conducted to identify signature genes that were significantly different at both the gene expression and methylation levels. Cox regression analyses were performed to identify clinical factors and these were combined with the signature genes to create a prognosis-related predictive model. This model was then evaluated for predictive accuracy and then validated using a validation dataset. RESULTS: This study identified 1565 DEGs and 3754 DEMGs in total. Of these, 369 were differentially expressed at both the gene and methylation levels. We identified 12 signature genes including VEGFC, FBP1, NR3C1, NFE2L2, and DFNA5 which were combined with the clinical data to produce a novel prognostic model for GC. This model could effectively split GC patients into two groups, high- and low-risk with these observations being confirmed in the validation dataset. CONCLUSION: The differential methylation of the 12 signature genes, including VEGFC, FBP1, NR3C1, NFE2L2, and DFNA5, identified in this study may help to produce a functional predictive model for evaluating GC prognosis in clinical samples.
Authors: X Liu; X Wang; J Zhang; E K Y Lam; V Y Shin; A S L Cheng; J Yu; F K L Chan; J J Y Sung; H C Jin Journal: Oncogene Date: 2009-11-02 Impact factor: 9.867
Authors: Wei Dai; Constanze Zeller; Nahal Masrour; Nadeem Siddiqui; James Paul; Robert Brown Journal: Clin Cancer Res Date: 2013-08-21 Impact factor: 12.531