A neutrophil extracellular traps-associated lncRNA signature predicts the clinical outcomes in patients with lung adenocarcinoma.

Abstract

Backgrounds: Neutrophil extracellular traps (NETs) play an important role in the occurrence, metastasis, and immune escape of cancers. We aim to investigate Long non-coding RNAs (lncRNAs) that are correlated to NETs to find some potentially useful biomarkers for lung adenocarcinoma (LUAD), and to explore their correlations with immunotherapy and chemotherapy, as well as the tumor microenvironment. Methods: Based on the The Cancer Genome Atlas (TCGA) database, we identified the prognosis-related lncRNAs which are associated with NETs using cox regression. The patients were then separated into two clusters based on the expression of NETs-associated lncRNAs to perform tumor microenvironment analysis and immune-checkpoint analysis. Least absolute shrinkage and selection operator (LASSO) regression was then performed to establish a prognostic signature. Furthermore, nomogram analysis, tumor mutation burden analysis, immune infiltration analysis, as well as drug sensitivity analysis were performed to test the signature. Results: Using univariate cox regression, we found 10 NETs-associated lncRNAs that are associated with the outcomes of LUAD patients. Also, further analysis which separated the patients into 2 clusters showed that the 10 lncRNAs had significant correlations with the tumor microenvironment. Using LASSO regression, we finally constructed a signature to predict the outcomes of the patients based on 4 NETs-associated lncRNAs. The 4 NETs-associated lncRNAs were namely SIRLNT, AL365181.3, FAM83A-AS1, and AJ003147.2. Using Kaplan-Meier (K-M) analysis, we found that the risk model was strongly associated with the survival outcomes of the patients both in the training group and in the validation group 1 and 2 (p < 0.001, p = 0.026, and p < 0.01). Using receiver operating characteristic (ROC) curve, we tested the sensitivity combined with the specificity of the model and found that the risk model had a satisfactory level of 1-year, 3-year, and 5-year concordance index (C-index) (C = 0.661 in the training group, C = 0.679 in validation group 1, C = 0.692 in validation group 2). We also explored the immune microenvironment and immune checkpoint correlation of the risk model and found some significant results. Conclusion: We constructed a NETs-associated lncRNA signature to predict the outcome of patients with LUAD, which is associated with immunephenoscores and immune checkpoint-gene expression.