A novel ferroptosis-related lncRNA pair prognostic signature and nomogram for predicting the survival of patients with lung adenocarcinoma

Abstract

Abstract Purpose Ferroptosis is a novel programmed cell death process and is considered a feasible cancer treatment strategy. Long non-coding RNAs (lncRNAs) are key mediators of ferroptosis and iron metabolism in cancer. However, the prognosis of ferroptosis-related lncRNAs (FRLs) in lung adenocarcinoma (LUAD) remains unclear. Method RNA-seq data and clinical information of patients with LUAD were retrieved from GSE81089. Differentially expressed lncRNAs (DELs) and ferroptosis-related genes (DE-FRGs) were identified in LUAD and normaltissues, followed by the construction of a DEL and DE-FRG network to obtain FRLs. Univariate Cox regression and LASSO regression analyses were performed to select lncRNA pairs to construct a prognostic model. Receiver operating characteristic (ROC) curve analysis was used to assess the performance of the model. Independent prognostic factors were screened to establish a nomogram model. Based on the optimal cut-off value, patients were classified into low-risk (LR) and high-risk (HR) groups. Differences in pathways, immune cell infiltration, immune checkpoint genes, and drug susceptibility between the two risk groups were evaluated. Results Eighteen FRL pairs were identified to construct a prognostic model, and patients were divided into LR and HR groups based on the median risk score of this model. Patients in the HR group had worse prognosis than those in the LR group. The area under the curve of the ROC curve verified that this model had good predictive ability. The pTNM stage and risk scores were independent prognostic factors that were used to develop the nomogram, which accurately predicted the survival of patients with LUAD. Functional analysis showed that starch and sucrose metabolism pathways were enriched in the LR group. Compared with the LR group, the HR group was associated with higher immune infiltration status, increased expression levels of immune checkpoints (BTLA, CD274, and HAVCR2), and higher drug sensitivity. Conclusion In conclusion, our prognostic model containing 18 FRL pairs showed excellent predictive ability for LUAD prognosis. This may guide personalized immunotherapy strategies for patients with LUAD.