Identification and Validation of Three Autophagy-Related Long Noncoding RNAs as Prognostic Signature in Cholangiocarcinoma

Abstract Cholangiocarcinoma (CCA) is featured by common occurrence and poor prognosis. Autophagy is a biological process that has been extensively involved in the progression of tumors. Long non-coding RNAs (lncRNAs) have been discovered to be critical for diagnosing and predicting various tumors. It may be valuable to elaborate autophagy-related lncRNAs (ARlncRNAs) in CCA, and indeed, there is still few study concerning the role of ARlncRNAs in CCA. Here, a prognostic ARlncRNA signature was constructed to predict the survival outcome of CCA patients. Through identification, three differentially expressed ARlncRNAs (DEARlncRNAs), including CHRM3.AS2, MIR205HG and LINC00661, were screened and were considered as predictive signatures. Furthermore, the overall survival (OS) of patients with high risk scores was significantly lower than that of patients with low scores. Interestingly, the risk score was an independent factor for the OS of patients with CCA. Moreover, receiver operating characteristic (ROC) curve analysis showed that the screened and constructed prognosis signature for 1-year (AUC= 0.884), 3-year (AUC=0.759), and 5-year (AUC=0.788) presented a high score of accuracy in predicting OS of CCA patients. Gene set enrichment analysis (GSEA) revealed that the three DEARlncRNAs were significantly enriched in CCA-related signaling pathways, including "pathways of basal cell carcinoma", and "glycerolipid metabolism", etc. Quantitative Real-time PCR (qRT-PCR) showed that expressions of CHRM3.AS2, MIR205HG and LINC00661 were higher in CCA tissues than those in normal tissues, similar to the trends detected in the CCA dataset. Further Pearson’s analysis reported an intimate correlation of the risk score with immune cells infiltration, indicating a predictive value of the signature for the efficacy of immunotherapy. In addition, the screened lncRNAs were found to have the ability to modulate the expression of mRNAs by interacting with miRNAs based on the established lncRNA-miRNA-mRNA network. In conclusion, our study develops a novel nomogram with good reliability and accuracy to predict the OS of CCA patients, providing significant guiding value for developing tailored therapy for CCA patients. Keywords: Autophagy, Long non-coding RNAs, Cholangiocarcinoma, The Cancer Genome Atlas, Prognostic Signature, Gene Expression Omnibus