Long noncoding RNA signatures involved in the genomic instability of papillary thyroid carcinoma

Abstract

We aimed to identify long noncoding RNAs involved in the genomic instability of papillary thyroid carcinoma (PTC). Expression profiles of RNA-seq and gene mutation profiles were downloaded from the Cancer Genome Atlas (TCGA) database, and differentially expressed lncRNAs (DElncRNAs) and DE messenger RNAs (DEmRNAs) were determined. We constructed an lncRNA-mRNA network, analyzed mRNA enrichment, and compared the immune cell proportions and tumor mutation burdens between the low- and high-risk groups using a prognostic model. We identified 95 DElncRNAs and 421 DEmRNAs and constructed a network comprising 33 lncRNAs and 201 mRNAs. The mRNAs in the network were enriched in 36 Gene Ontology biological processes and 7 Kyoto Encyclopedia of Genes and Genomes pathways. A five-lncRNA prognostic model was constructed; the AUCs of the training, validation, and entire sets were 0.955, 0.805, and 0.901, respectively. The proportions of six types of tumor-infiltrating immune cells and neuroblastoma RAS viral (V-ras) oncogene homolog expression differed significantly between the low- and high-risk groups. LncRNAs may be involved in the genomic instability of PTC via cytokine-cytokine receptor interactions, cell adhesion molecules, and chemokine signaling pathways. Our five-lncRNA prognostic model may enable the prognostic evaluation of PTC patients. Highlights The 5-lncRNA prognostic model may be an independent prognostic factor for PTC. Six TIICs are associated with the prognosis of PTC. NRAS gene mutation plays a vital role in the progression of PTC. The model included the lncRNAs WARS2-IT1, LINC00536, ATP13A4-AS1, LINC01561, and FENDRR.

Keywords

• papillary thyroid carcinoma
• genomic instability
• lncrna
• immune cell
• prognostic model