Identification and validation of a novel redox-related lncRNA prognostic signature in lung adenocarcinoma

Lung adenocarcinoma (LUAD) is one of the main causes of cancer deaths globally. Redox is emerging as a crucial contributor to the pathophysiology of LUAD, which can be regulated by long non-coding RNAs (lncRNAs). The aim of our research is to identify a novel redox-related lncRNA prognostic signature (redox-LPS) for better prediction of LUAD prognosis. 535 LUAD samples from The Cancer Genome Atlas (TCGA) database and 226 LUAD samples from the Gene Expression Omnibus (GEO) database were included in our study. 67 redox genes and 313 redox-related lncRNAs were identified. After performing LASSO-Cox regression analysis, a redox-LPS consisting of four lncRNAs (i.e., CRNDE, CASC15, LINC01137, and CYP1B1-AS1) was developed and validated. Our redox-LPS was superior to another three established models in predicting survival probability of LUAD patients. Univariate and multivariate Cox regression analysis revealed that risk score and stage were independent prognostic indicators. A nomogram plot including risk score and stage was constructed to predict survival probability of LUAD patients; this was further verified by calibration curves. Functional enrichment analysis and gene set enrichment analysis, were performed to determine the differences in cellular processes and signaling pathways between the high - and low-risk subgroups. A variety of algorithms (such as single-sample gene set enrichment analysis and CIBERSOFT) were conducted to uncover the landscape of tumor immune microenvironment in the high- and low-risk subgroups. In conclusion, a novel independent redox-LPS was constructed and validated for LUAD patients, which might provide new insights for clinical decision-making and precision medicine.

Automated summary

A novel redox-related lncRNA prognostic signature was identified and validated for predicting LUAD patient prognosis, outperforming existing models. Risk score and stage were independent prognostic indicators, and a nomogram plot was constructed to predict survival probability. Functional enrichment analysis and tumor immune microenvironment landscape analysis were performed to uncover differences between high- and low-risk subgroups.