A signature of 14 immune-related gene pairs predicts overall survival in gastric cancer.

Abstract

Increasing evidence demonstrates that immune signature plays an important role in the prognosis of gastric cancer (GC). We aimed to develop and validate a robust immune-related gene pair (IRGP) signature for predicting the prognosis of GC patients. RNA-Seq data and corresponding clinical information of GC cohort were downloaded from the TCGA (The Cancer Genome Atlas Program) data portal. GSE84437 and GSE15459 microarray datasets were included as independent external cohorts. Least absolute shrinkage and selection operator (LASSO) regression analysis was used to build the best prognostic signature. All patients were classified into the high immune-risk and low immune-risk groups via the optimal cut-off of the signature scores determined by time-dependent receiver-operating characteristic (ROC) curve analysis. The prognostic role of the signature was measured by a log-rank test and a Cox proportional hazard regression model. 14 immune gene pairs consisting of 25 unique genes were identified to construct the immune prognostic signature. High immune-risk groups showed poor prognosis in the TCGA datasets and GSE84437 datasets as well as in the GSE15459 datasets (all P < 0.001). The 14-IRGP signature was an independent prognostic factor of GC after adjusting for other clinical factors (P < 0.05). Functional analysis revealed that DNA integrity checkpoint, DNA replication, T-cell receptor signaling pathway, and B-cell receptor signaling pathway were enriched in the low immune-risk groups. B cells naive and Monocytes were significantly higher in the high-risk group, and B-cell memory and T-cell CD4 memory activated were significantly higher in the low-risk group. The prognostic signature based on IRGP reflected infiltration by several types of immune cells. The novel proposed clinical-immune signature is a promising biomarker for prediction overall survival in patients with GC and providing new insights into the treatment strategies.