Abstract
Background: Colorectal cancer (CRC) is prevalent worldwide, and improvements in timely and effective diagnosis are imperatively needed. We aimed to identify potential long noncoding RNA (lncRNA) biomarkers for CRC diagnosis along with prognosis prediction. Methods: LncRNA expression profiles were studied by microarray in paired tumor and normal tissues from six patients with CRC. The expression levels of candidate lncRNAs were analyzed in 80 pairs of tissues and two independent cohorts of serum samples. Receiver-operating characteristic (ROC) curves were employed to evaluate the performance of the lncRNAs identified. The correlation between lncRNAs and disease-specific survival rate of CRC patients was assessed to explore prognostic potential. Results: Four lncRNAs (BANCR, NR_026817, NR_029373, and NR_034119) were identified to be significantly dysregulated in both tissue and serum samples with consistent pattern, and a panel was established based on this result. The performance of the 4-lncRNA panel was measured with an area under the ROC curve (AUC) of 0.881. The corresponding AUCs of the panel for patients with TNM stageI, II and III were 0.774, 0.844 and 0.949, respectively, significantly higher than that of CEA. Kaplan-Meier analysis showed that patients with low levels of NR_029373 and NR_034119 had significantly lower disease-specific survival rate (p = 0.013 and 0.044, respectively). Multivariate Cox analysis demonstrated that NR_029373 and NR_034119 were both independently associated with disease-specific survival rate (p = 0.013 and 0.038, respectively). Conclusions: Our study established a distinctive 4-lncRNA panel with considerable diagnostic value and identified NR_029373 and NR_034119 as potential biomarkers for CRC prognosis prediction.