The immune response plays a critical role in gastric cancer (GC) development, metastasis, and treatment. A better understanding of the tumor-immune system interactions in gastric cancer may provide promising diagnostic, prognostic, and therapeutic biomarkers for patients with this disease. In the present study, we aimed to identify a prognostic signature of GC through a comprehensive bioinformatics analysis on the tumor-immune interactions as well as the molecular characteristics. We firstly identified two immunophenotypes and immunological characteristics by employing multiple algorithms, such as the single sample Gene Sets Enrichment Analysis and Cell type Identification By Estimating Relative Subsets of RNA Transcripts. Next, we developed a six-immune-gene signature as a promising independent prognostic biomarker for GC using Lasso Cox regression and verified it via the external validation set and systematically correlated the immune signature with GC clinicopathologic features and genomic characteristics. Finally, a nomogram was successfully constructed based on the immune signature and clinical characteristics and showed a high potential for GC prognosis prediction. This study may shed light on the treatment strategies for GC patients from the perspective of immunology.
Keywords: CIBERSORT; gastric cancer; immune signature; ssGSEA; tumor immunity.
Copyright © 2021 Guan, Xu, Chen, Qin and Cheng.