An Integrated mRNA-lncRNA Signature for Overall Survival Prediction in Cholangiocarcinoma

Abstract

BackgroundThe incidence and mortality rate of cholangiocarcinoma (CCA) have been rising globally. Patients with CCA have extremely poor prognosis, partly due to the silent clinical character and hence diagnosed at advantage stage without effective treatments. There is growing evidence showing that aberrant expression of messenger RNAs (mRNAs) and long non-coding RNAs (lncRNAs) are involved in tumorigenesis and development of CCA. It is essential to establish an integrated mRNA-lncRNA signature to improve the ability of prognostic prediction in CCA patients.MethodsWe collected a training dataset of 45 patients from The Cancer Genome Atlas dataset and a validation cohort (GSE107943) of 57 patients from Gene Expression Omnibus. An integrated mRNA-lncRNA risk score was established by a univariate and a multivariate Cox regression analyses. Time-dependent receiver operating characteristic (ROC) analysis was used to evaluate prognostic performance. Moreover, we conducted a correlation analysis between the signature and different clinical characteristics, and preformed weighted gene co-expression network analysis (WGCNA) and functional enrichment analysis to investigate functional roles of the integrated signature.ResultsA total of two mRNAs (CFHR3 and PIWIL4) and two lncRNAs (AC007285.1 and AC134682.1) were identified to construct the integrated signature through a univariate Cox regression (P-value = 1.35E-02) and a multivariable Cox analysis (P-value = 1.12E-02). The ROC curve suggested the integrated mRNA-lncRNA signature possessed a high specificity and sensitivity of prognostic prediction with an area under the curve (AUC) of 0.872 and 0.790 at 1-year and 3-years, respectively. Subsequently, the signature was validated in GSE107943 cohort and combined dataset, and an area under the ROC curve reached up to 0.750 and 0.819 at 1-year. The signature was not only independent from different clinical features (P-value= 1.12E-02), but also outperformed other clinical characteristics as prognostic biomarkers with AUC of 0.781 at 3 years. These molecules in the integrated signature may associated with metabolic-related biological process and lipid metabolism pathway, which was highly involved in CCA carcinogenesis. ConclusionThese results showed that the integrated mRNA-lncRNA signature had an independent prognostic value for risk stratification, and further facilitated personalized treatment for CCA patients.