[Auxiliary diagnosis and prognosis evaluation of KIF4A, RAD51AP1 and CDKN3 in esophageal cancer]
S Qiu,C X Guo,L Li,J F Yuan,J Liu,Y Y Peng,Q Qiu
DOI: https://doi.org/10.3760/cma.j.cn112150-20230717-00008
2024-05-06
Abstract:To investigate the expression of mRNA in esophageal cancer (ESCA) tissues and its potential and diagnostic and prognostic value by high-throughput sequencing data. Using the Cancer Genome Atlas Program (TCGA) database in USA by integrative bioinformatics analysis methods, the gene expression profiles and clinical data of 173 patients with ECSA were collected. The mRNA expression levels in ESCA tissue and para-cancerous tissue samples were analyzed using DESeq2, edgeR and limma to screen the differentially expressed genes (DEGs). DEGs-related protein network diagrams were drawn. GO and KEGG function enrichment analysis were performed and the hub genes were screened and the survival analysis of hub genes was analyzed. Genes related to the prognosis of ESCA were selected and their prognostic value in ESCA was analyzed. Finally, the receiver operating characteristic curve was drawn to evaluate its diagnostic value. The results showed that using TCGA cancer data, a total of 620 up-regulated DEGs and 668 down-regulated DEGs with significant differential expression between ESCA and para-cancerous tissues were screened. DEGs were mainly involved in receptor complexes, ubiquitin ligase complexes, etc., playing GTPase activity, phospholipid binding, and other molecular functions, and participating in the regulation of intracellular substance transport, small molecule metabolism, and other biological processes. Protein functional enrichment analysis showed that these proteins were mainly enriched in the IL-17 signaling pathway, TNF signaling pathway, Toll-like receptor signaling pathway, Epstein-Barr virus infection, neutrophil extracellular trap formation, and other pathways involved in the formation and development process of ESCA. Survival analysis showed that the overall survival rate of ESCA patients with high expression of KIF4A, RAD51AP1, and CDKN3 was significantly shortened, and the difference was statistically significant (P<0.05). Furthermore, the areas under the curve (AUC) of KIF4A, RAD51AP1, and CDKN3 for diagnosing esophageal cancer were 0.956, 0.951 and 0.979, respectively, with sensitivities and specificities both exceeding 80%. Additionally, ROC results of the combined diagnostic model of these three genes showed an AUC of 0.979, with sensitivities and specificities of 0.914 and 1, respectively. This indicates that KIF4A, RAD51AP1 and CDKN3 have individual or combined auxiliary diagnostic value for ESCA. In conclusion, KIF4A, RAD51AP1 and CDKN3 have high diagnostic efficiency for ESCA, and their increased expression is closely related to the prognosis, suggesting that these three genes could be used as auxiliary diagnostic and prognostic factors for ESCA.