An autophagy-associated lncRNAs model for predicting the survival in non-small cell lung cancer patients.

Long non-coding RNAs (lncRNAs) can influence the proliferation, autophagy, and apoptosis of non-small cell lung cancer (NSCLC). LncRNAs also emerge as valuable prognostic factors for NSCLC patients. Consequently, we set out to discover more autophagy-associated lncRNAs. We acquired autophagy-associated genes and information on lncRNAs from The Cancer Genome Atlas database (TCGA), and the Human Autophagy Database (HADb). Then, the prognostic prediction signature was constructed through using co-expression and Cox regression analysis. The signature was constructed including 7 autophagy-associated lncRNAs (ABALON, NKILA, LINC00941, AL161431.1, AL691432.2, AC020765.2, MMP2-AS1). After that, we used univariate and multivariate Cox regression analysis to calculate the risk score. The survival analysis and ROC curve analysis confirmed good performances of the signature. GSEA indicated that the high-risk group was principally enriched in the adherens junction pathway. In addition, biological experiments showed that ABALON promoted the proliferation, metastasis and autophagy levels of NSCLC cells. These findings demonstrate that the risk signature consisting of 7 autophagy-associated lncRNAs accurately predicts the prognosis of NSCLC patients and should be investigated for potential therapeutic targets in clinic.