Identification of an Exosome-Related Signature for Predicting Prognosis, Immunotherapy Efficacy, and Tumor Microenvironment in Lung Adenocarcinoma

Abstract

Accumulating evidence suggests that exosomes can affect lung adenocarcinoma (LUAD) progression. However, there is still a lack of understanding of the global influence of exosome-related genes (ERGs) on prognostic relevance, tumor microenvironment features, and immunotherapy responsiveness in patients with LUAD. In the TCGA dataset, differential analysis of 490 LUAD samples and 59 normal samples yielded 30 ERGs with differential expression. We have created a predictive signature based on 10 overall survival (OS)-related ERGs and confirmed it in two external cohorts (GSE72094 and GSE68465) via the least absolute shrinkage and selection operator (LASSO) and Cox regression analysis in the TCGA dataset. The new signature revealed superior robustness and prognostic capacity for overall patient survival. Univariate and multivariate Cox regression analyses indicated that this signature was an independent risk factor for survival in patients with LUAD. In addition, for predicting the 1-year, 3-year, and 5-year OS of LUAD patients, we developed a nomogram and confirmed its predictive ability via the C-index and calibration curve. In addition, patients categorized by risk score exhibited distinct immunological states, stemness index, immune subtypes, and immunotherapy response. In conclusion, we created a risk signature for LUAD that was tightly associated with the immune landscape and therapeutic response. Also, such a risk signature effectively promotes the ability of the clinicians in making more precise and individualized treatment recommendations for patients with LUAD.