An immune-relevant signature of nine genes as a prognostic biomarker in patients with gastric carcinoma.

BACKGROUND

As one of the most common malignant tumors worldwide, the morbidity and mortality of gastric carcinoma (GC) are gradually increasing. The aim of this study was to construct a signature according to immune-relevant genes to predict the survival outcome of GC patients using The Cancer Genome Altas (TCGA).

METHODS

Univariate Cox regression analysis was used to assess the relationship between immune-relevant genes regarding the prognosis of patients with GC. The least absolute shrinkage and selection operator (LASSO) Cox regression model was used to select prognostic immune-relevant genes and to establish the signature for the prognostic evaluation of patients with GC. Multivariate Cox regression analysis and Kaplan-Meier survival analysis were used to assess the independent prognostic ability of the immune-relevant gene signature.

RESULTS

A total of 113 prognostic immune-relevant genes were identified using univariate Cox proportional hazards regression analysis. A signature of nine immune-relevant genes was constructed using the LASSO Cox regression. The GC samples were assigned to two groups (low- and high risk) according to the optimal cutoff value of the signature score. Compared with the patients in the high-risk group, patients in the low-risk group had a significantly better prognosis in the TCGA and GSE84437 cohorts (log-rank test < 0.001). Multivariate Cox regression analysis demonstrated that the signature of nine immune-relevant genes might serve as an independent predictor of GC.

CONCLUSIONS

Our results showed that the signature of nine immune-relevant genes may potentially serve as a prognostic prediction for patients with GC, which may contribute to the decision-making of personalized treatment for the patients.