Background: Platinum-based chemotherapy plays a crucial role in pre- and post-operative therapy in advanced stage ovarian cancer (OC). The objective of this study was to explore differentially expression genes (DEGs) and their survival impact after exposing to platinum-based chemotherapy in OC patient via integrated bioinformatics analysis.
Methods: Gene expression profiles of RNA-seq data in OC were extracted from the GEO and TCGA databases respectively. DEGs were sent to perform functional Gene Ontology and KEGG pathway enrichment analyses. Survival analysis was processed to identify significant prognostic genes. After overlapping between DEGs and prognostic genes, univariate and multivariate Cox proportion hazards models were utilized to estimate the hazard ratio of the potential genes with a 95% confidence interval. Finally, the Kaplan-Meier log rank test and the time-dependent receiver operating characteristic (ROC) curve were performed to evaluate the potential prognostic prediction in platinum-based chemotherapy OC patients.
Results: A total of 484 up-regulated and 495 down-regulated DEGs were identified. Down-regulated DEGs remarkedly enriched in the cell cycle, oocyte meiosis, progesterone-mediated oocyte maturation, homologous recombination in KEGG pathway enrichment analyses. After overlapping with survival genes in the TCGA cohort, 64 DEGs were demonstrated prognostic potential. Then 29 genes were further identified by univariate analyses. Multivariate analyses indicated eight of 29 genes (DHRS9, OVOS2, STAC2, TCF15, AADAC, LOC730183, LOC440910, ARHGDIG) demonstrated survival prediction potential in platinum-based chemotherapy OC patients. The area under the curve of the time-dependent ROC curve was 0.725 for 5-year survival prediction based on those eight genes.
Conclusion: These prognostic genes identified in this study indicate some significance for prognosis prediction in platinum-based chemotherapy OC patients.