Identification of pathological grade and prognosis‐associated lncRNA for ovarian cancer

Ying Chen,Fangfang Bi,Yuanyuan An,Qing Yang
DOI: https://doi.org/10.1002/jcb.28704
2019-04-29
Journal of Cellular Biochemistry
Abstract:Ovarian carcinoma (OC) is one of the most common malignant tumors in female genitals. In recent years, the therapeutic effect of OC has been significantly improved through the application of effective chemotherapy regimen. However, the 5‐year survival rate is also lower than 30% due to high rate of relapse. So, it is needed to screen reliable predictive and prognostic markers of OC. Ovarian cancer gene expression data and corresponding clinical data used were downloaded from Gene Expression Omnibus database. Weighted gene expression network analysis (WGCNA) and Cox proportional hazards regression (PHR) were used to screen Pathological Grade and Prognosis‐associated long noncoding RNA (lncRNA). Kaplan‐Meier analysis and receiver operating characteristic curves analysis were performed to evaluate the predictive ability of the selected lncRNA. Gene Ontology (GO) enrichment and Gene Set Enrichment Analysis (GSEA) enrichment analysis methods were used to explore the possible mechanisms of the selected lncRNA affecting the development of OC. Five reliably lncRNAs (LINC00664, LINC00667, LINC01139, LINC01419, and LOC286437) was identified through a series of bioinformatics methods. In testing cohorts, we found that the five lncRNAs in predicting the risk of OC recurrence is robustness, and multivariate Cox PHR analysis indicate that the five lncRNAs is an independent risk factor for OC recurrence. Moreover, GO and GSEA enrichment analysis showed that the five lncRNAs are involved in multiple ovarian cancer occurrence mechanism. In summary, all these findings indicated that the five lncRNAs can effectively predict the risk of recurrence of ovarian cancer.
cell biology,biochemistry & molecular biology
What problem does this paper attempt to address?