A scoring model based on bacterial lipopolysaccharide-related genes to predict prognosis in NSCLC.

Background: Non-small cell lung cancer (NSCLC) has high incidence and mortality rates. The discovery of an effective biomarker for predicting prognosis and treatment response in patients with NSCLC is of great significance. Bacterial lipopolysaccharide-related genes (LRGs) play a critical role in tumor development and the formation of an immunosuppressive microenvironment; however, their relevance in NSCLC prognosis and immune features is yet to be discovered. Methods: Differentially expressed LRGs associated with NSCLC prognosis were identified in the TCGA dataset. Prognostic LRG scoring and nomogram models were established using single-variable Cox regression, Least Absolute Shrinkage, and Selection Operator (LASSO) regression. The prognostic value of the scoring and nomogram models was evaluated using Kaplan-Meier (KM) analysis and further validated using an external dataset. Patients were stratified into high- and low-risk groups based on the nomogram score, and drug sensitivity analysis was performed. Additionally, clinical characteristics, mutation features, immune infiltration characteristics, and responses to immunotherapy were compared between the two groups. Results: We identified 15 differentially expressed LRGs associated with NSCLC prognosis. A prognostic prediction model consisting of 6 genes (VIPR1, NEK2, HMGA1, FERMT1, SLC7A, and TNS4) was established. Higher LRG scores were associated with worse clinical prognosis and were independent prognostic factors for NSCLC. Subsequently, a clinical risk prediction nomogram model for NSCLC was constructed, incorporating the status of patients with tumor burden, tumor T-stage, and LRG scores. The nomogram model demonstrated good predictive performance upon validation. Additionally, NSCLC patients classified as high risk based on the model's predictions exhibited not only a poorer prognosis but also a more pronounced inflammatory immune microenvironment phenotype than low-risk patients. Furthermore, high-risk patients showed disparate predicted responses to various drugs and immunotherapies compared with low-risk patients. Conclusion: The LRGs scoring model can serve as a biomarker that contributes to the establishment of a reliable prognostic risk-prediction model, potentially facilitating the development of personalized treatment strategies for patients with NSCLC.