Predictive value of PIMREG in the prognosis and response to immune checkpoint blockade of glioma patients

Abstract

Glioma is the most common primary brain tumor in the human brain. The present study was designed to explore the expression of PIMREG in glioma and its relevance to the clinicopathological features and prognosis of glioma patients. The correlations of PIMREG with the infiltrating levels of immune cells and its relevance to the response to immunotherapy were also investigated. PIMREG expression in glioma was analyzed based on the GEO, TCGA, and HPA databases. Kaplan-Meier survival analysis was used to examine the predictive value of PIMREG for the prognosis of patients with glioma. The correlation between the infiltrating levels of immune cells in glioma and PIMREG was analyzed using the CIBERSORT algorithm and TIMRE database. The correlation between PIMREG and immune checkpoints and its correlation with the patients' responses to immunotherapy were analyzed using R software and the GEPIA dataset. Cell experiments were conducted to verify the action of PIMREG in glioma cell migration and invasion. We found that PIMREG expression was upregulated in gliomas and positively associated with WHO grade. High PIMREG expression was correlated with poor prognosis of LGG, prognosis of all WHO grade gliomas, and prognosis of recurrent gliomas. PIMREG was related to the infiltration of several immune cell types, such as M1 and M2 macrophages, monocytes and CD8+ T cells. Moreover, PIMREG was correlated with immune checkpoints in glioma and correlated with patients' responses to immunotherapy. KEGG pathway enrichment and GO functional analysis illustrated that PIMREG was related to multiple tumor- and immune-related pathways. In conclusion, PIMREG overexpression in gliomas is associated with poor prognosis of patients with glioma and is related to immune cell infiltrates and the responses to immunotherapy.

Keywords: PIMREG; glioma; immune cell infiltrates; immune checkpoint; immune checkpoint blockade (ICB) therapy.

Figure 1

Differential expression of PIMREG in glioma and normal tissues. PIMREG expression was higher in glioma than in normal brain tissue in the GSE16011 (A), GSE14805 (B), and GSE19728 (C) datasets. (D) PIMREG expression was higher in both LGG and GBM than in normal brain tissue in the GEPIA 2.0 database. PIMREG expression analysis results from data downloaded from TGCA (E) and CGGA (F). (G) Correlation between PIMREG expression in gliomas and the histology of gliomas. O: oligodendroglioma; A, astrocytoma; rO, recurrent oligodendroglioma; rA, recurrent astrocytoma; AO, anaplastic oligodendroglioma; AA, anaplastic astrocytoma; rAA, recurrent anaplastic astrocytoma; rGBM, recurrent glioblastoma. (H) PIMREG expression in gliomas with 1p/19q codeletion and non1p/19q codeletion. (I) Expression of PIMREG in IDH-mutant gliomas and wild-type gliomas. *P < 0.05; ***P < 0.001; ****P < 0.0001.

Figure 2

Immunohistochemical images of PIMREG in glioma. Immunohistochemical images of PIMREG in normal brain tissue (A), LGG (B), and GBM (C).

Figure 3

The prognostic value of PIMREG in glioma. (A, B) Univariate and multivariate Cox regression visualized in the forest plot. (C) Nomogram and calibration plots (D–F) predicting the 1-year, 3-year and 5-year OS of glioma patients. (G–I) Predictive ability for 1-, 3-, and 5-year prognosis with PIMREG expression by time-dependent ROC curve analysis.

Figure 4

Predictive value of PIMREG (FAM64A) in glioma prognosis. Correlation of PIMREG expression with OS of glioma (A), LGG (B), and GBM (C). Correlation of PIMREG with the DFS of glioma (D), LGG (E) and GBM (F). Correlation between PIMREG and survival probability of all WHO-grade primary (G–I) and recurrent (J–L) glioma in CGGA datasets.

Figure 5

PIMREG enhances the migration and invasion of glioma cells in vitro. qRT–PCR results showing PIMREG expression in U251 (A) and U87MG (B) glioma cells after transfection. CCK-8 assays showing the viability of U251 (C) and U87 (D) cells. Wound healing was performed to determine the migration ability of U251 cells (E, F) and U87MG cells (G, H). Transwell assays were performed to determine their invasion ability (I–K). **P < 0.01; ***P < 0.001; $$$P < 0.001; &&&; P< 0.0001.

Figure 6

Correlation between PIMREG expression and tumor immunity of gliomas. Correlation of PIMREG with immune score (A), stromal score (B) and ESTIMATE score (C) in LGG+GBM. Correlation of PIMREG with the infiltrating levels of M1 macrophages (D), M2 macrophages (E), monocytes (F), neutrophils (G), activated NK cells (H), resting memory CD4 T cells (I), and follicular helper T cells (J) in GBM. Correlation of PIMREG with the infiltrating level of M0 macrophages (K), M1 macrophages (L), activated mast cells (M), monocytes (N), resting memory CD4 T cells (O), and follicular helper T cells (P) in LGG. Correlation between PIMREG and immune cells in LGG (Q) and GBM (R) in the TIMER database.

Figure 7

Correlation between PMREG and immune checkpoints and response to immunotherapy. Correlation of PIMREG with the immune checkpoints CD274 (A), HAVCR2 (B), CTLA4 (C), LAG3 (D), PDCD1 (E), PDCD1LG2 (F), TIGIT (G), and SIGLEC15 (H) in gliomas (LGG+GBM). (I) Heatmap of immune checkpoint expression in gliomas with high and low PIMREG expression (LGG+GBM). (J) Different TIDE scores in gliomas (LGG + GBM) with high and low expression of PIMREG. (K) Different TIDE scores in LGG with high and low expression of PIMREG. (L) Different TIDE scores in GBM with high and low expression of PIMREG. (M) PIMREG expression in the response and nonresponse groups to anti-PD-1 therapy in the IMvigor210 cohort (anti-PD-L1, urological). (N) The ratio of patients who responded to anti-PD-1 therapy in the low- and high-PIMREG groups of the IMvigor210 cohort. (O) Kaplan–Meier curves for the low- and high-PIMREG patient groups in the IMvigor210 cohort. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 8

Genes coexpressed with PIMREG. (A) Coexpression of PIMREG with immune-related genes. (B) Coexpression of PIMREG with inhibitory and stimulatory immune checkpoints. Significantly positively (C) and negatively (D) coexpressed genes with PIMREG. *P < 0.05.

Figure 9

Protein–protein interaction network, GO and KEGG pathway enrichment analysis of PIMREG in glioma. PPI network of PIMREG in the GeneMINIA (A) and SRTING (B) tools. GO analysis (biological process) of PIMREG (C) and KEGG pathway enrichment analysis (D) in LGG in LinkedOmics. GO analysis (biological process) of PIMREG (E) and KEGG pathway enrichment analysis (F) in GBM in LinkedOmics.