Identification of a 5-lncRNA signature-based risk scoring system for survival prediction in colorectal cancer

The present study aimed to investigate potential prognostic long noncoding RNAs (lncRNAs) associated with colorectal cancer (CRC). An mRNA-seq dataset obtained from The Cancer Genome Atlas was employed to identify the differentially expressed lncRNAs (DELs) between CRC patients with good and poor prognoses. Subsequently, univariate and multivariate Cox regression analyses were conducted to analyze the prognosis-associated lncRNAs among all DELs. In addition, a risk scoring system was developed according to the expression levels of the prognostic lncRNAs, which was then applied to a training set and an independent testing set. Furthermore, the co-expressed genes of prognostic lncRNAs were screened using a Multi-Experiment Matrix online tool for construction of lncRNA-gene networks. Finally, Kyoto Encyclopedia of Genes and Genomes pathway and Gene Ontology (GO) function enrichment analyses were performed on genes in the lncRNA-gene networks using KOBAS, GOATOOLS and ClusterProfiler. The present study identified 82 DELs, of which long intergenic nonprotein coding RNA 2159, RP11-452L6.6, RP11-894P9.1 and RP11-69M1.6, and whey acidic protein four-disulfide core domain 21 (WFDC21P) were reported to be independently associated with the prognosis of patients with CRC. A 5-lncRNA signature-based risk scoring system was developed, which may be used to classify patients into low- and high-risk groups with significantly different recurrence-free survival times in the training and testing sets (P<0.05). Co-expressed genes of WFDC21P or RP11-69M1.6 were utilized to construct the lncRNA-gene networks. Genes in the networks were significantly enriched in ‘tight junction’, ‘focal adhesion’ and ‘regulation of actin cytoskeleton’ pathways, and numerous GO terms associated with ‘reactive oxygen species metabolism’ and ‘nitric oxide metabolism’. The present study proposed a 5-lncRNA signature-based risk scoring system for predicting the prognosis of patients with CRC, and revealed the associated signaling pathways and biological processes. The results of the present study may help improve prognostic evaluation in clinical practice.