A novel risk score system for assessment of ovarian cancer based on co-expression network analysis and expression level of five lncRNAs

Abstract

Background: Ovarian cancer (OC) is the most deadly gynaecological cancer, contributing significantly to female cancer-related deaths worldwide. Improving the outlook for OC patients depends on the identification of more reliable prognostic biomarkers for early diagnosis and survival prediction. The various roles of long non-coding RNAs (lncRNAs) in OC have attracted increasing attention. This study aimed to identify a lncRNA-based signature for survival prediction in OC patients. Methods: RNA expression data and clinical information from a large number of OC patients were downloaded from a public database. These data were regarded as a training set to construct a weighed gene co-expression network analysis (WGCNA) network, mine stable modules, and screen differentially expressed lncRNAs. The prognostic lncRNAs were screened using univariate Cox regression analysis and the optimal prognosis lncRNA combination was screened using a Cox-PH model. The finalised lncRNA combination was used to construct the risk score system, which was validated and assessed for effectiveness using other independent datasets. Further functional pathway enrichment was performed using gene set enrichment analysis (GSEA). Results: A co-expression network was constructed and four stable modules with OC-related biological functions were obtained. A total of 19 lncRNAs significantly related to prognosis of ovarian cancer were obtained using univariate Cox regression analysis, and the 5 prognostic signature lncRNAs GAS5, HCP5, PART1, SNHG11, and SNHG5 were used to establish a risk assessment system. The reliability of the prognostic scoring system was further confirmed using validation sets, which indicated that the risk assessment system could be used as an independent prognostic factor. Pathway enrichment analysis revealed that the network modules related to the above five prognostic genes were significantly associated with cell local adhesion, cancer signaling pathways, JAK-STAT signalling, and endogenous cell receptor interaction. Conclusions: The risk score system established in this study could provide a novel reliable method to identify individuals at high risk of OC. In addition, the five prognostic lncRNAs identified here are promising potential prognostic biomarkers that could help to elucidate the pathogenesis of OC.

The Non-Coding RNA GAS5 and Its Role in Tumor Therapy-Induced Resistance

“…Ovarian cancer is a grave gynecological tumor and there are not many studies concerning the role of GAS5 in it or its association with chemoresistance in this tumor. However, similar to other gynecological tumors, it seems that all studies converge to the conclusion that GAS5 acts as a tumor suppressor in ovarian cancer, as well [44][45][46][47]. In a recent study, GAS5 was found to be down-regulated in this disease [41].…”

LINC01094/miR-577 axis regulates the progression of ovarian cancer

“…Exploring the abnormal expression, function and mechanism of lncRNA in the development of OC can provide biomarkers and therapy targets [22]. For example, a recent study reports that a predictive model, based on the expressions of lncRNAs, GAS5, HCP5, PART1, SNHG11 and SNHG5 included, can identify OC patients at high risk of poor prognosis [23]; the up-regulation of LINC00319 expression accelerates the proliferation, migration and invasion of OC cells [24]; LncRNA PTAR expression is up-regulated in OC and it promotes the EMT and metastasis of OC cells [25]. In this work, we focused on the role of LINC01094 in the progression of OC and found that LINC01094 was highly expressed in OC tissues and cell lines.…”

Identification and Validation of Hub Genes Associated With Hepatocellular Carcinoma Via Integrated Bioinformatics Analysis

“…In recent years, bioinformatics analysis of expression profile has been widely and rapidly used to identify novel and more effective potential biomarkers for cancer treatment and patient prognosis (3)(4)(5)(6)(7). However, how to transform the microarray information into a better understanding of biology through traditional differential expression analysis is still a major challenge.…”