Abstract
BACKGROUND: The epithelial-mesenchymal transition (EMT) plays a vital role in the progression of lung adenocarcinoma (LUAD). Long non-coding RNAs (lncRNAs) participate in the EMT process as an important regulatory factor and have the potential to serve as prognostic biomarkers. We aimed to construct a novel lncRNA prognostic signature for LUAD based on EMT-related lncRNAs, identify EMT-related hub lncRNA, and investigate its biological functions. METHODS: RNA-seq data, clinical and survival information were obtained from The Cancer Genome Atlas database. The EMT-related lncRNA prognostic signature (EMTscore) was constructed using the Least Absolute Shrinkage and Selection Operator Cox regression analysis. The efficiency of EMTscore in predicting the prognosis of LUAD was evaluated through the area under the time-dependent receiver operating characteristic (ROC) curves. The hub lncRNA of the prognostic signature was selected using a co-expression network map, and its effects on cell proliferation and metastasis were explored by in vitro experiments. RESULTS: We constructed a prognostic signature (EMTscore) containing 8 tumor-high expressed lncRNAs. The EMTscore performed well in predicting overall survival rates with AUC values of 0.708 at 5 years in the training set. EMTscore could independently predict the survival of LUAD, with HR = 4.011 (95% CI 2.430-6.622) in the multivariate Cox regression. Importantly, we identified LINC01615 as the hub lncRNA in the EMTscore and revealed that LINC01615 enhanced the proliferation, migration, and EMT of lung cancer cells. CONCLUSIONS: A new EMT-related lncRNA prognostic signature named EMTscore was developed, and LINC01615 was identified as the hub lncRNA of EMTscore. The hub lncRNA LINC01615 had an oncogenic biological function in LUAD.