CDCA4 as a novel molecular biomarker of poor prognosis in patients with lung adenocarcinoma

BACKGROUND: Because of the high incidence and poor prognoses of lung adenocarcinoma (LUAD), it is essential to identify cost-effective treatment options and accurate and reliable prognostic biomarkers. CDCA4 upregulation has been identified in many cancers. However, the prognostic importance of CDCA4 and its role in LUAD remain unknown. METHODS: CDCA4 expression was assessed through IHC, Western blotting (WB) and RT-PCR. The Cancer Genome Atlas (TCGA) provided data from 513 patients to study the expression and prognostic relevance of CDCA4 in LUAD. This study used gene set enrichment analyses (GSEA), gene ontology and KEGG pathway analyses for elucidating potential mechanisms underpinning the function of CDCA4 in LUAD. We also investigated correlations between immune infiltration and CDCA4 expression with single specimen GSEA (ssGSEA). RESULTS: According to database analysis and identification of patient tissue samples, CDCA4 expression in tumour tissues surpassed that in normal tissues (P< 0.001). Increased CDCA4 expression was positively correlated with a higher T, N, pathologic stage and poor primary therapy outcome. In addition, the Kaplan–Meier plotter exhibited that an elevated CDCA4 expression was related to worse disease-specific survival(DSS) and overall survival (OS) (DSS HR= 5.145, 95% CI=3.413-7.758, P<0.001; OS HR=3.570, 95% CI=2.472-5.155, P<0.001). Then multivariate COX regression analyses indicated that the CDCA4 gene was an independent risk consideration for prognoses. GO and KEGG results showed that CDCA4 and its neighbouring genes were enriched in the cell cycle and DNA replication. As determined by GSEA, CDCA4 was related to various immune-related signalling pathways (SPs), Homologous recombination, DNA replication and the cell cycle. SsGSEA analysis showed a significant association between CDCA4 expression and Th2 cells, mast cells, eosinophils and Th17 cells. CONCLUSIONS: CDCA4 expression is increased in LUAD and is a potential predictive biomarker and therapeutic target.