A methylation-based mRNA signature predicts survival in patients with gastric cancer

Abstract

Background: Evidence suggests that altered DNA methylation plays a causative role in the occurrence, progression and prognosis of gastric cancer (GC). Thus, methylated-differentially expressed genes (MDEGs) could potentially serve as biomarkers and therapeutic targets in GC.

Methods: Four genomics profiling datasets were used to identify MDEGs. Gene Ontology enrichment and Kyoto Encyclopaedia of Genes and Genomes pathway enrichment analysis were used to explore the biological roles of MDEGs in GC. Univariate Cox and LASSO analysis were used to identify survival-related MDEGs and to construct a MDEGs-based signature. The prognostic performance was evaluated in two independent cohorts.

Results: We identified a total of 255 MDEGs, including 192 hypermethylation-low expression and 63 Hypomethylation-high expression genes. The univariate Cox regression analysis showed that 83 MDEGs were associated with overall survival. Further we constructed an eight-MDEGs signature that was independent predictive of prognosis in the training cohort. By applying the eight-MDEGs signature, patients in the training cohort could be categorized into high-risk or low-risk subgroup with significantly different overall survival (HR = 2.62, 95% CI 1.71-4.02, P < 0.0001). The prognostic value of the eight-MDEGs signature was confirmed in another independent GEO cohort (HR = 1.35, 95% CI 1.03-1.78, P = 0.0302) and TCGA-GC cohort (HR = 1.85, 95% CI 1.16-2.94, P = 0.0084). Multivariate cox regression analysis proved the eight-MDEGs signature was an independent prognostic factor for GC.

Conclusion: We have thus established an innovative eight-MDEGs signature that is predictive of overall survival and could be a potentially useful guide for personalized treatment of GC patients.

Keywords: DNA methylation; MDEGs; Prognosis; Signature.

Fig. 1

Flowchart of this study

Fig. 2

The methylated‐differentially expressed genes identification and function. a Venn of methylated‐differentially expressed genes in gene expression datasets (GSE13911, GSE79973) and gene methylation datasets (GSE30601, GSE25869). b The volcano plots of GSE13911 and GSE79973 for differentially expressed mRNA. Red and green dots represent significantly up-regulated and down-regulated genes, respectively (FDR < 0.05). c The significant enriched gene ontology of MDEGs. d The significant enriched KEGG pathways of MDEGs

Fig. 3

Construction of the eight-MDEGs signature of GC. The patients were stratified into high-risk group and low-risk group based on median of risk score. a Kaplan‐Meier curve of the overall survival for high-risk and low-risk scores ranking by the eight-MDEGs signature. b The distribution of death in high-risk and low-risk group. c Risk score distribution of GC patients, Survival status of each patient and Expression heatmap of the eight MDEGs corresponding to each sample above

Fig. 4

Validation of the eight-MDEGs signature in two independent datasets. Kaplan‐Meier curve of the overall survival for high-risk and low-risk scores ranking by the eight-MDEGs signature in TCGA-GC dataset (a) and GSE84437 dataset (b). Risk score distribution of GC patients, Survival status of each patient and Expression heatmap of the eight MDEGs corresponding to each sample above in TCGA-GC dataset (c) and GSE84437 dataset (d)