Background: CLDN5 protein is essential for the formation of tight junctions in epithelial cells, and has been associated with epithelial-mesenchymal transition. Research has indicated that CLDN5 is associated with tumor metastasis, the tumor microenvironment, and immunotherapy in multiple types of cancer. Also, no comprehensive evaluation of the expression of CLDN5 and immunotherapy signatures through a pan-cancer analysis or immunoassay has been performed.
Methods: We explored CLDN5's differential expression, survival analysis and clinicopathological staging through the TCGA database, and then corroborated the expression of CLDN5 by utilizing the GEO (Gene expression omnibus) database. To analyze CLDN5 KEGG, GO, and Hallmark mutations, as well as TIMER for immune infiltration, GSEA was utilized with ROC curve, mutation, and other factors such as survival, pathological stage, TME, MSI, TMB, immune cell infiltration, and DNA methylation. Immunohistochemistry was used to assess CLDN5 staining in gastric cancer tissues and paracancerous tissues. Visualization was done with R version 4.2.0 (http://www.rproject.org/).
Results: According to TCGA database, CLDN5 expression levels differed significantly between cancer and normal tissues, and the GEO database (GSE49051 and GSE 64951) and tissue microarrays confirmed this result. Infiltrating cluster of differentiation 8+ (CD8+) T cells, CD4+ cells, neutrophils, dendritic cells, and macrophages revealed a correlation with CLDN5 expression. DNA methylation, TMB, and MSI are related to CLDN5 expression. Based on the ROC curve analysis, CLDN5 demonstrates outstanding diagnostic effectiveness for gastric cancer and is comparable to CA-199.
Conclusions: The findings suggest that CLDN5 is implicated in the oncogenesis of diverse cancer types, underscoring its potential significance in cancer biology. Notably, CLDN5 could have implications in immune filtration and immune checkpoint inhibitor therapies, however, further research is needed to confirm this.
Keywords: CLDN5; EMT; bioinformatics; immune infiltrating; pan-cancer analysis.