Adaptive immunity profiling associated with histological subtypes and postoperative survival in gastric cancer

In gastric cancer (GC), abnormal adaptive immunity is correlated with chronic inflammatory disorders and poor prognosis. However, the global study of adaptive immunity involving genes expression is insufficient.

In this study, we investigated the transcriptional profile of adaptive immunity involving genes in GC from TCGA (The Cancer Genome Atlas). The relevance of adaptive immunity and the clinical features of patients with GC were assessed. Differences in gene expression between each feature and the correlation between gene expression and prognosis were elucidated.

According to the expressional profile of adaptive immunity-related genes, 412 patients with GC were grouped into two primary classifications and three secondary classifications. There were no differences in prognosis detected between each subgroup. In the immune subgroups, the distributions of pathological type were obviously different. Additionally, histological types, AJCC (American Joint Committee on Cancer) stage features, grade, tumor stage, aneuploidy score, and fraction genome altered in different subgroups were significantly discrepant. There were 95 differently expressed genes (DEGs) detected between each histological type, which were represented by LAIR1, BTK and LAT2. According to identification of DEGs in the MSTAD (mucinous stomach adenocarcinoma) and SRCC (signet ring cell carcinoma) types, which were relevant to the best and worst prognosis types, respectively, we constructed a model combining seven genes to recognize the MSTAD type (AUC =0.91) and a model combining six genes to recognize the SRCC type (AUC =0.91). Moreover, the expression of FGL1 gene was notably contrasting among the different histological types, and the high-expression of FGL1 was correlated with a poor prognosis.

This study showed that the expressional patterns of adaptive immunity-related genes are closely related to the histological type of GC, and demonstrated that the expression of immune molecules is correlated to the prognosis. Our results are expected to promote immunological therapy for GC.