Early detection and prediction of prognosis and treatment responses are all the keys in improving survival of ovarian cancer patients. This study profiled an ovarian cancer progression model to identify prognostic biomarkers for ovarian cancer patients. Mouse ovarian surface epithelial cells (MOSECs) can undergo spontaneous malignant transformation in vitro cell culture. These were used as a model of ovarian cancer progression for alterations in gene expression and signaling detected using the Illumina HiSeq2000 Next-Generation Sequencing platform and bioinformatical analyses. The differential expression of four selected genes was identified using the gene expression profiling interaction analysis (http://gepia.cancer-pku.cn/) and then associated with survival in ovarian cancer patients using the Cancer Genome Atlas dataset and the online Kaplan–Meier Plotter (http://www.kmplot.com) data. The data showed 263 aberrantly expressed genes, including 182 up-regulated and 81 down-regulated genes between the early and late stages of tumor progression in MOSECs. The bioinformatic data revealed four genes (i.e., guanosine 5′-monophosphate synthase (GMPS), progesterone receptor (PR), CD40, and p21 (cyclin-dependent kinase inhibitor 1A)) to play an important role in ovarian cancer progression. Furthermore, the Cancer Genome Atlas dataset validated the differential expression of these four genes, which were associated with prognosis in ovarian cancer patients. In conclusion, this study profiled differentially expressed genes using the ovarian cancer progression model and identified four (i.e., GMPS, PR, CD40, and p21) as prognostic markers for ovarian cancer patients. Future studies of prospective patients could further verify the clinical usefulness of this four-gene signature.
On the Reproducibility of TCGA Ovarian Cancer MicroRNA Profiles