Nine-Gene Prognostic Signature Related to Gut Microflora for Predicting the Survival in Gastric Cancer Patients.

Abstract

The purpose of this study is to screen the feature genes related to gut microflora and explore the role of the genes in predicting the prognosis of patients with gastric cancer. We downloaded the gene profile of gastric cancer from the University of California Santa Cruz, the gut microflora related to gastric cancer from The Cancer Microbiome Atlas. The GSE62254 dataset was downloaded from National Center for Biotechnology Information Gene Expression Omnibus as a validation dataset. A correlation network between differentially expressed genes and gut microflora was constructed using Cytoscape. The optimized prognostic differentially expressed genes were identified through least absolute shrinkage and selection operator (LASSO) algorithm and univariate Cox regression analysis. The risk score model was established and then measured via Kaplan-Meier and area under the curve. Finally, the nomogram model was constructed according to the independent clinical factors, which was evaluated using C-index. A total of 754 differentially expressed genes and 8 gut microflora were screened, based on which we successfully constructed the correlation network. We obtained 9 optimized prognostic differentially expressed genes, including HSD17B3, GNG7, CHAD, ARHGAP8, NOX1, YY2, GOLGA8A, DNASE1L3, and ABCA8. Moreover, Kaplan-Meier curves indicated the risk score model correctly predicted the prognosis of gastric cancer in both University of California Santa Cruz and GSE62254 dataset (area under the curve >0.8; area under the curve >0.7). Finally, we constructed the nomogram, in which the C index of 1, 3, and 5 years was 0.824, 0.772, and 0.735 representing that the nomogram was consistent with the actual situation. These results indicate the 9 differentially expressed genes related to gut microflora might predict the survival time of patients with gastric cancer. Both risk signature and nomogram could effectively predict the prognosis for patients with gastric cancer.