Identification of Six Autophagy-Related-lncRNA Prognostic Biomarkers in Uveal Melanoma

Abstract

Currently, no autophagy-related long noncoding RNA (lncRNA) has been reported to predict the prognosis of uveal melanoma patients. Our study screened for autophagy-related lncRNAs in 80 samples downloaded from The Cancer Genome Atlas (TCGA) database through lncRNA-mRNA coexpression. We used univariate Cox to further filter the lncRNAs. Multivariate Cox regression and LASSO regression were applied to construct an autophagy-associated lncRNA predictive model and calculate the risk score. Clinical risk factors were validated using Cox regression to determine whether they were independent prognostic indicators. Functional enrichment was performed using Gene Ontology and Kyoto Encyclopedia of Genes and Genomes. The model was built with six predictive autophagy-associated lncRNAs and clustered uveal melanoma patients into high- and low-risk groups. The risk score of our model was a significant independent prognostic factor (hazard ratio=1.0; p<0.001). Moreover, these six lncRNAs were significantly concentrated in the biological pathways of cytoplasmic component recycling, energy metabolism, and apoptosis. Thus, the six autophagy-associated lncRNAs are potential molecular biomarkers and treatment targets for uveal melanoma patients.