The Midkine-related ceRNA Network is a Prognostic Marker of Colon Cancer With Perineural Invasion

Background: Perineural invasion (PNI) is a prominent characteristic of multiple solid tumors and indicates poor prognosis[1, 2]. A growing body of evidence emphasizes the critical role of competitive endogenous RNA(ceRNA) regulatory networks in various human cancers. However, the complexity and behavior characteristics of the ceRNA network in colon cancer with perineural invasion were still unclear. In this study, we aimed to clarify a midkine(MDK)-related ceRNA regulatory network and identify potential prognostic markers associated with colon cancer with perineural invasion.Material and MethodsWe extract the expression profiles of three RNAs (long non-coding RNAs [lncRNAs], microRNAs [miRNAs], and mRNAs) from The Cancer Genome Atlas (TCGA) database. In order to construct the lncRNA-miRNA-mRNA triple regulatory network in colon cancer with perineural invasion, we performed joint analysis in the MDK high and MDK low expression groups, as well as the colon cancer and paracancerous tissue groups. Kaplan-Meier method was used to plot the Overall survival (OS) curves, and the log-rank test was used for testing. Univariate and multivariate Cox regression analysis were used to determine OS-related characteristics.ResultsKCNQ1OT1-miR-454-3p/miR-204-5p-MAP1B ceRNA network related to the prognosis of colon cancer was obtained through bioinformatics analysis. Importantly, we determined the KCNQ1OT1/MAP1B axis in the ceRNA through correlation analysis, and through Cox regression analysis it seems to be a clinical prognostic model. ConclusionIn conclusion, the current study constructing a ceRNA based KCNQ1OT1/MAP1B axis may be a novel important prognostic factor for the diagnosis and prognosis of colon cancer with perineural invasion.