Research Paper|Volume 12, Issue 17|pp 17418—17435

Identification of subtype-specific genes signature by WGCNA for prognostic prediction in diffuse type gastric cancer

Qi Zhou¹, Li-Qiang Zhou¹, Shi-Hao Li¹, Yi-Wu Yuan¹, Li Liu¹, Jin-Liang Wang¹, Deng-Zhong Wu¹, You Wu¹, Lin Xin¹

¹Department of General Surgery, The Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi Province, China

* Equal contribution

Received: April 6, 2020Accepted: July 7, 2020Published: September 11, 2020

Copyright: © 2020 Zhou et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Abstract

Background: Gastric cancer is a common malignancy and had poor response to treatment due to its strong heterogeneity. This study aimed to identify essential genes associated with diffuse type gastric cancer and construct a powerful prognostic model.

Results: We conducted a weighted gene co-expression network analysis (WGCN) using transcripts per million (TPM) expression data from The Cancer Genome Atlas (TCGA) to find out the module related with diffuse type gastric cancer. Combining Least Absolute Shrinkage and Selection Operator (LASSO) with multi-cox regression, the 10 specific genes risk score model of diffuse type gastric cancer was established. The concordance index (0.97), the area under the respective ROC curves (AUCs) (1-years: 0.98; 3-years: 1; 5-years: 1) and survival difference of high- and low risk groups (p=2.84e-10) of this model in TCGA dataset were obtained. The moderate predicting performance was observed in the independent cohort of GSE15459 and GSE62254. The results of the gene set enrichment analysis (GSEA) using high-and low risk group as phenotype indicated differential expression of tumor-related pathways.

Conclusion: Thus, we constructed a reliable prognostic model for diffuse type gastric cancer, which should be beneficial for clinical therapeutic decision-making.