Screening potential biomarkers of cholangiocarcinoma based on gene chip meta-analysis and small-sample experimental research.

Cholangiocarcinoma (CCA) is a rare malignant tumor associated with poor prognosis. This study aimed to identify CCA biomarkers by investigating differentially expressed genes (DEGs) between CCA patients and healthy subjects obtained from the Gene Expression Omnibus database. Bioinformatics tools, including the Illumina BaseSpace Correlation Engine (BSCE) and Gene Expression Profiling Interactive Analysis (GEPIA), were used. The initial DEGs from GSE26566, GSE31370, and GSE77984 were analyzed using GEO2R and Venn, and protein-protein interaction networks were constructed using STRING. The BSCE was applied to assess curated CCA studies to select additional DEGs and them DEGs across the 10 biosets, which was supported by findings in the literature. The final 18 DEGs with clinical significance for CCA were further verified using GEPIA. These included , , , , , , , , and , which were upregulated, and , , , , , , , , and , which were downregulated in CCA patients. Among these 18 genes, 56 groups of genes (two in each group) were significantly related, and none were independently and differentially expressed. The hub genes , , , and , which were most correlated with the 18 DEGs, were screened using STRING. The significantly low expression of , , and and significantly high expression of were verified by immunohistochemical analysis. Overall, four CCA biomarkers were identified that might regulate the occurrence and development of this disease and affect the patient survival rate, and they have the potential to become diagnostic and therapeutic targets for patients with CCA.