Background: Lung cancer is the most prevalent and deadliest cancer worldwide. The present study aims to determine the prognosis value of low expression long non-coding RNAs (lncRNAs) in LUAD.
Methods: RNA-seq data and clinical information were downloaded from The Cancer Genome Atlas (TCGA) data-base. Dysregulated genes between LUAD and paracancerous tissue were screened by GeneSpringGX. Prognostic lncRNAs which were low expressed in LUAD were filtrated by Ualcan, then further verified through the TCGA database. The association between clinicopathological features and the expression level of these lncRNAs was tested by chi-square test. Cox regression analysis was performed to test independent prognosis risk factors. Diagnostic efficiency was predicted by receiver operating characteristic (ROC) analysis. Gene Ontology (GO) functional and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to explore potential functions of these prognostic signatures.
Results: Nine prognostic lncRNAs (LINC00092, LINC00908, WWC2-AS2, RPL13AP17, CHIAP2, SFTA1P, SIGLEC17P, CYP2B7P1, CYP4Z2P) were screened out through Ualcan and further verified by TCGA. Among them, six lncRNAs (RPL13AP17, CHIAP2, SFTA1P, SIGLEC17P, CYP2B7P1, CYP4Z2P) were pseudogene transcripts. Multivariate Cox regression analysis showed that three lnRNAs (LINC00908, WWC2-AS2 CYP2B7P) were independent prognostic risk factors for OS and two lncRNAs (WWC2-AS2, SIGLEC17P) were independent prognostic risk factors for RFS in LUAD patients. Meanwhile, they showed powerful diagnostic value by ROC curve analysis. GO analysis revealed correlation genes of prognostic signatures were mainly enriched in plasma membrane, plasma membrane part, purine nucleotide binding, cytoskeleton and ribonucleotide binding and KEGG pathway analysis showed mainly enriched in cell adhesion molecules.
Conclusions: The results illuminated that four lncRNAs (LINC00908, WWC2-AS2, CYP2B7P, SIGLEC17P) may be a powerful diagnostic and prognostic assessment tool for human LUAD.