Purpose: The present study was designed to explore the significant biomarkers and pathway-related modules for predicting the effects of eribulin relative to paclitaxel in ovarian cancer.
Methods: The gene expression data E-GEOD-50831 were downloaded from the European Bioinformatics Institute (EBI) database. Differentially expressed genes (DEGs) were screened. Subsequently, differential coexpression network was constructed. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis and pathway-related modules mining were conducted. Topological centralities (degree, betweenness, closeness and stress) analyses for coexpression network and pathway-related modules were performed to explore hub genes and the most significant pathways. Then, we verified our findings in an independent sample set via RT-PCR and Western blotting.
Results: Centralities results of ESCO1, CDC27and MCM4 ranked the top five. Moreover, among the top 10% hub genes, CDC27, MCM4 and SOS1 were pathway-enriched genes in two networks. A total of 5 and 6 pathway-related modules were obtained under two drugs treatment. Based analyses of degree, betweenness and other centralities, DNA replication pathway-related module was the most significant under paclitaxel treatment, while cell cycle pathway-related module was the most significant under eribulin treatment. RT-PCR and Western blotting results were consistent with the bioinformatics results. The expression level of MCM4 was remarkably decreased under eribulin treatment relative to paclitaxel.
Conclusions: The inhibition of ovarian cancer growth by paclitaxel and eribulin might be connected with downregulation of cell cycle and DNA replication pathway. Moreover, MCM4 signature might be a potential biomarker to predict the effect of eribulin in ovarian cancer.