The immune-related gene CD5 is a prognostic biomarker associated with the tumor microenvironment of breast cancer

Abstract

The occurrence and progression of breast cancer (BCa) are complex processes involving multiple factors and multiple steps. The tumor microenvironment (TME) plays an important role in this process, but the functions of immune components and stromal components in the TME require further elucidation. In this study, we obtained the RNA-seq data of 1086 patients from The Cancer Genome Atlas (TCGA) database. We calculated the proportions of tumor-infiltrating immune cells (TICs) and immune and stromal components using the CIBERSORT and ESTIMATE methods, and we screened differentially expressed genes (DEGs). Univariate Cox regression analysis of overall survival was performed on the DEGs, and a protein-protein interaction network of their protein products was generated. Finally, the hub gene CD5 was obtained. High CD5 expression was found to be associated with longer survival than low expression. Gene set enrichment analysis showed that DEGs upregulated in the high-CD5 expression group were mainly enriched in tumor- and immune-related pathways, while those upregulated in the low-expression group were enriched in protein export and lipid synthesis. TIC analysis showed that CD5 expression was positively correlated with the infiltration of CD8⁺ T cells, activated memory CD4⁺ T cells, gamma delta T cells, and M1 macrophages and negatively correlated with the infiltration of M2 macrophages. CD5 can increase anticancer immune cell infiltration and reduce M2 macrophage infiltration. These results suggest that CD5 is likely a potential prognostic biomarker and therapeutic target, providing novel insights into the treatment and prognostic assessment of BCa.

Keywords: Breast cancer; CD5; Tumor immune infiltrates; Tumor microenvironment.

Fig. 1

Analytical flowchart of this study

Fig. 2

Relationships of the immune score and stromal score with the prognosis of BCa patients. A Immune score. A high immune score indicates a good prognosis. B Stromal score

Fig. 3

Heat map, Venn diagrams, and GO and KEGG for differentially expressed genes. A, B Heat maps of DEGs obtained for the immune score and stromal score, respectively. C Venn diagram of upregulated DEGs according to the immune score and the stromal score. D Venn diagram of downregulated DEGs according to the immune score and the stromal score. E, F GO and KEGG analyses of DEGs shared by the immune score and stromal score

Fig. 4

Construction of the PPI network of DEGs and univariate Cox regression analysis. A Construction of the PPI network of DEGs. B Top 20 interacting genes in terms of the number of connection nodes in the PPI network. C Results of Cox regression analysis. D The intersection of the hub genes screened by the PPI network and the Cox analysis results. Finally, four hub genes, including CD5, were identified

Fig. 5

Pan-cancer analysis of CD5 expression. A Comparative analysis of CD5 gene expression levels in different cancer types. B Analysis of the prognostic value of CD5 gene in different cancer types

Fig. 6

Correlation between overall survival and CD5 expression in cancer patients. A The survival differed significantly between the BRCA patients with high or low CD5 expression (P < 0.05). B–F Differential CD5 expression in patients with CHOL, ESCA, KRIC, LUSC and THCA was not significantly correlated with patient OS

Fig. 7

Analysis of an independent data set (GSE1456) supports the correlation between CD5 expression and survival and its association with clinicopathological characteristics. A An independent data set from the GEO database (GSE1456) was used for validating the results of survival analysis. We observed a correlation between high CD5 expression and improved OS in patients (P < 0.05). B–D Analysis of CD5 expression and clinicopathological features showed that CD5 expression is significantly correlated with cancer stage and lymph node metastasis status (P < 0.05)

Fig. 8

MethSurv graphical output generated by CpG cg24674703 in BRCA samples, and the association of CD5 expression with TMB. A KM plot of cg24674703-CD5 using BRCA samples dichotomized using the MaxStat method. Blue represents CD5 low expression group, and red represents CD5 high expression group (P < 0.05). B CD5 expression was significantly and positively correlated with TMB (P < 0.05)

Fig. 9

Immunohistochemical staining and western blot analysis confirmed higher CD5 expression in cancer tissue compared to normal breast tissue. Fresh tissue specimens from either breast cancer tissue (A–C) or normal breast tissue (D–F) were subjected to immunohistochemical staining and western blot analysis (G, H). In A–C CD5 was highly expressed in breast cancer cells, and a few lymphocytes also had CD5 expression. Gray-scale value analysis of western blot analysis results

Fig. 10

The expression distribution of 22 TICs in BCa samples and the correlation analysis between each of the 22 TICs, and GSEA correlation analysis of CD5 expression level. A The top 10 most significantly enriched pathways in the samples from the high-CD5 expression group and the only three significantly enriched pathways in the samples from the low-CD5 expression group. B The proportion of distribution of 22 TICs in the BCa sample. C Correlation analysis between TICs. Heat map showing the pairwise correlations of all 22 types of immune cells. Red represents a positive correlation, and blue represents a negative correlation

Fig. 11

Correlation analysis between TICs and CD5 expression. A Violin plot showing the results of the correlation analysis between CD5 expression and the 22 types of TICs in BCa samples. B Scatter plot showing the immune cells whose proportions were most significantly correlated with CD5 expression. Among them, CD8⁺ T cells, activated memory CD4⁺ T cells, gamma delta T cells, and M1 macrophages were positively correlated with CD5 expression, while M0 and M2 macrophages were negatively correlated with CD5 expression. Red color represents high CD5 expression