Abstract
Esophageal carcinoma (ESCA) is one of the most common malignancies in the world, and has high morbidity and mortality rates. Necrosis and long noncoding RNAs (lncRNAs) are involved in the progression of ESCA; however, the specific mechanism has not been clarified. The aim of the present study was to investigate the role of necrosis-related lncRNAs (nrlncRNAs) in patients with ESCA by bioinformatics analysis, and to establish a nrlncRNA model to predict ESCA immune infiltration and prognosis. To form synthetic matrices, ESCA transcriptome data and related information were obtained from The Cancer Genome Atlas. A nrlncRNA model was established by coexpression, univariate Cox (Uni-Cox), and least absolute shrinkage and selection operator analyses. The predictive ability of this model was evaluated by Kaplan-Meier, receiver operating characteristic (ROC) curve, Uni-Cox, multivariate Cox regression, nomogram and calibration curve analyses. A model containing eight nrlncRNAs was generated. The areas under the ROC curves for 1-, 3- and 5-year overall survival were 0.746, 0.671 and 0.812, respectively. A high-risk score according to this model could be used as an indicator for systemic therapy use, since the half-maximum inhibitory concentration values varied significantly between the high-risk and low-risk groups. Based on the expression of eight prognosis-related nrlncRNAs, the patients with ESCA were regrouped using the 'ConsensusClusterPlus' package to explore potential molecular subgroups responding to immunotherapy. The patients with ESCA were divided into three clusters based on the eight nrlncRNAs that constituted the risk model: The most low-risk group patients were classified into cluster 1, and the high-risk group patients were mainly concentrated in clusters 2 and 3. Survival analysis showed that Cluster 1 had a better survival than the other groups (P=0.016). This classification system could contribute to precision treatment. Furthermore, two nrlncRNAs (LINC02811 and LINC00299) were assessed in the esophageal epithelial cell line HET-1A, and in the human esophageal cancer cell lines KYSE150 and TE1. There were significant differences in the expression levels of these lncRNAs between tumor and normal cells. In conclusion, the present study suggested that nrlncRNA models may predict the prognosis of patients with ESCA, and provide guidance for immunotherapy and chemotherapy decision making. Furthermore, the present study provided strategies to promote the development of individualized and precise treatment for patients with ESCA.