Immune Infiltration in Gastric Cancer Microenvironment and Its Clinical Significance

Abstract

Immunotherapy has developed rapidly and has gradually become one of the important methods for treatment of gastric cancer (GC). The research on tumor infiltrating immune cells (TIICs) and immune-related genes in the tumor microenvironment (TME) greatly encourages the development of immunotherapy. The devolution algorithm (CIBERSORT) was applied to infer the proportion of 22 TIICs based on gene expression profiles of GC tissues, which were downloaded from TCGA and GEO. TCGA was utilized to analyze the differential expression of immune-related genes, and explore the potential molecular functions of these genes. We have observed the enrichment of multiple TIICs in microenvironment of GC. Some of these cells were closely related to tumor mutational burden (TMB), microsatellite instability (MSI), Fuhrman grade, and TNM staging. Survival analysis showed that the infiltration level of CD8⁺ T cells, activated CD4⁺ memory T cells and M2 macrophages were significantly related to the prognosis of GC patients. The functional enrichment analysis of immune-related genes revealed that these genes were mainly associated with cytokine activation and response. Four significant modules were screened by PPI network and 20 key genes were screened from the modules. The expression levels of CALCR and PTH1R are strikingly related to the expression of immune checkpoint and the prognosis of GC patients. The type and number of TIICs in microenvironment of GC, as well as immune-related genes are closely related to tumor progression, and can be used as important indicators for patient prognosis assessment.

Keywords: GC; TIICs; immune genes; prognosis; tumor microenvironment.

FIGURE 1

The landscape of immune infiltration in GC and difference of immune infiltration between normal tissues and tumor tissues in GSE66229 dataset. (A) Normal tissues (n = 50) and GC tissues (n = 150). (B) Normal tissues (n = 50) and GC tissues (n = 150).

FIGURE 2

The distribution and correlation analysis of 22 TIICs in GC tissues and normal tissues in GSE66229 dataset. (A) Heat map of the 22 immune cell proportions. Green represents low infiltration, black represents medium infiltration and red represents high infiltration. (B) Violin plot shows the difference in the proportion of immune cell infiltration. Blue is normal tissues and red is tumor tissues. (C) The correlation heat map describes the correlation between TIICs. Red representing positive correlation and blue representing negative correlation.

FIGURE 3

The landscape of immune infiltration in GC and difference of immune infiltration between normal tissues and tumor tissues in TCGA. (A) Normal tissues (n = 15) and GC tissues (n = 221). (B) Violin plot shows the difference in the proportion of immune cell infiltration. Blue is normal tissues and red is tumor tissues.

FIGURE 4

The difference and correlation analysis of immune cell infiltration between GC tissues and normal tissues in TCGA. (A) Heat map of the 22 immune cell proportions. Green represents low infiltration, black represents medium infiltration and red represents high infiltration. (B) The correlation heat map describes the correlation between TIICs. Red representing positive correlation and blue representing negative correlation.

FIGURE 5

The correlation of TIICs with microsatellite instability (MSI) and tumor mutational burden (TMB). (A) Radar map of correlation between TIICs and TMB. (B) Radar map of correlation between TIICs and MSI. Spearman’s coefficient was used to calculate the correlation between immune cell infiltration and TMB and MSI. *p < 0.05, **p < 0.01, and ***p < 0.001.

FIGURE 6

Correlation analysis between TIICs and clinical characteristics and prognosis of patients with GC. (A–G) are respectively resting dendritic cells, M1 Macrophages, resting mast cells, CD8⁺ T cells, M0 macrophages, activated mast cells, M2 macrophages and the correlation analysis of Fuhrman grade. G1 represents high differentiation, G2 represents moderate differentiation and G3 represents poor differentiation. (H,I) are the correlation analysis between eosinophils and M2 macrophages with TNM staging, respectively. (J) CD8⁺ T cells, and (K) activated CD4⁺ memory T cells is closely related to the better prognosis of the patient. (L) M2 macrophages is closely related to the poor prognosis of patients.

FIGURE 7

Enrichment analysis of immune-related differential genes. (A) The volcano map shows immune-related differential genes. Red is the up-regulated gene and the green is the down-regulated gene. (B) KEGG enrichment analysis. (C) GO enrichment analysis.

FIGURE 8

Screening of key genes and correlation analysis with Fuhrman grade and TNM staging. (A–D) are the four main modules in the PPI network (Score>10, number of nodes>20). (E) (ADRB2), (F) (AGTR1), (G) (AGTR2), (H) (C3), (I) (C3AR1), (J) (CCL21), (K) (CXCL2), (L) (CXCL12), (M) (F2R), (N) (HGF), (O) (PENK), and (P) (PTAFR) correlation analysis with Fuhrman grade. (Q) (C3), (R) (F2R), (S) (HGF), (T) (KRAS), and (U) (MET) correlation analysis with TNM staging.

FIGURE 9

Enrichment analysis of four main modules. (A–D) are module (A), module (B), module (C), and module (D), respectively.

FIGURE 10

Analysis of the correlation between key genes and immune checkpoints and their prognosis. (A) CALCR was significantly positively correlated with the expression of immune checkpoints (PD-1, PD-L1, CTLA4, LAG3, and VSTM3). (B) PTH1R is negatively correlated with the expression of immune checkpoints (PD-1, PD-L1, CTLA4, LAG3, and VSTM3). Correlation analysis between (C) (PTH1R) and (D) (CALCR) and the prognosis of patients with GC. TIMER was used to analyze the expression of (E) (PTH1R) and (F) (CALCR) in tumors. STAD is gastric adenocarcinoma. Red is tumor tissues and blue is normal tissues. Spearman’s rank test was used to analyze the correlation between immune-related genes and immune checkpoints. Statistical significance: *p < 0.05, **p < 0.01, and ***p < 0.001.