Increased IGFBP7 Expression Correlates with Poor Prognosis and Immune Infiltration in Gastric Cancer

Abstract

Background: Insulin-like growth factor binding protein-7 (IGFBP7) contributes to multiple biological processes in various tumors. However, the role of IGFBP7 in gastric cancer (GC) is still undetermined. The study aims to explore the role of IGFBP7 in GC via an integrated bioinformatics analysis. Methods: IGFBP7 expression levels in GC and its normal gastric tissues were analyzed using multiple databases, including the Tumor Immune Estimation Resource (TIMER), Oncomine, The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases, as well as by our clinical gastric specimens. The methylation analysis was conducted with MEXPRESS, UALCAN and Xena online tools. The survival analysis was conducted using the Kaplan-Meier Plotter and Gene Expression Profiling Interactive Analysis (GEPIA) databases. Coexpressed genes of IGFBP7 were selected with the cBioPortal tool and enrichment analysis was conducted with the clusterProfiler package in R software. Gene set enrichment analysis (GSEA) was performed to explore the IGFBP7-related biological processes involved in GC. Correlations between IGFBP7 and immune cell infiltrates were analyzed using the TIMER database. Results: IGFBP7 expression was significantly upregulated in GC and correlated with stage, grade, tumor status and Helicobacter pylori infection. High IGFBP7 expression and low IGFBP7 methylation levels were significantly associated with short survival of patients with GC. Univariate and multivariate analyses revealed that IGFBP7 was an independent risk factor for GC. The coexpressed genes LHFPL6, SEPTIN4, HSPB2, LAYN and GGT5 predicted unfavorable outcomes of GC. Enrichment analysis showed that the coexpressed genes were involved in extracellular matrix (ECM)-related processes. GSEA indicated that IGFBP7 was positively related to ECM and inflammation-related pathways. TIMER analysis indicated that the mRNA level of IGFBP7 was strongly correlated with genes related to various infiltrating immune cells in GC, especially with gene markers of tumor associated macrophages (TAMs). Conclusions: Increased IGFBP7 expression correlates with poor prognosis and immune cell infiltration in GC, which might be a potential biomarker for the diagnosis of GC.

Keywords: IGFBP7; bioinformatics analysis; gastric cancer; immune infiltration; prognosis.

Figure 1

The transcription level of IGFBP7 was increased in gastric cancer. A, The expression of IGFBP7 in different types of cancers as assessed with the TIMER database. *, P<0.05; **, P<0.01; ***, P<0.001. B, The expression of IGFBP7 in different types of cancers as assessed with the Oncomine database. Colors: red, high expression; blue, low expression. C-F, The expression of IGFBP7 in gastric cancer based on TCGA (C), GSE54129 (D), GSE118916 (E), and GSE79973 (F) datasets. G, the mRNA levels in 16 paired GC and normal gastric tissues collected from our hospital.

Figure 2

Expression levels of IGFBP7 in paraffin-embedded GC and normal gastric tissues.

Figure 3

Association between IGFBP7 expression and clinicpathological characteristic of GC patients. The mRNA expression level of IGFBP7 is shown with scatter plots using the TCGA-STAD dataset according to age (A), sex (B), grade (C),tumor status (D), node status (E), metastasis (F), stage (G) and H. pylori infection (I), as well as according to stage in GSE15459 dataset (H).

Figure 4

High IGFBP7 expression was associated with poor prognosis in GC. A-C, The relationship of IGFBP7 mRNA expression with overall survival (OS) (A), first progression survival (FPS) (B) and post progression survival (PPS) (C) was assessed with probe 213910_at in the Kaplan-Meier Plotter database. D-E, The association of IGFBP7 mRNA expression and overall survival (D) and disease-free survival (E) was assessed in the GEPIA database. The median value was used as the cutoff for dividing the high and low groups.

Figure 5

Analyses of genes coexpressed with IGFBP7. A-H, The top eight coexpressed genes of IGFBP7 were selected and correlation analysis was performed in the cBioPortal online database. These genes included LHFPL6 (A), MGP (B), SEPT4 (C), HSPB2 (D), ACTA2 (E), LAYN (F), NDN (G) and GGT5 (H). I, Survival map of the 8 coexpressed genes based on TCGA-STAD project data was generated with the GEPIA online tool. A red box indicates those genes with statistical significance. J-N, Overall survival analysis was performed for LHFPL6 (J), SEPT4 (K), HSPB2 (L), LAYN (M) and GGT5 (N). Briefly, GC samples were divided into high and low expression groups by the median expression value of the gene. The log-rank method was used for the hypothesis test.

Figure 6

GSEA analysis of IGFBP7 based on expression in the TCGA-STAD dataset. NES: normalized enrichment score; NOM p-value: nominal p value; FDR q-val: false discovery rate.

Figure 7

Correlation of IGFBP7 expression with immune cell infiltration in GC. The study was performed in the TIMER database. Spearman's correlation was used to evaluate the correlation between IGFBP7 expression and immune cells.