Identification of a Three-Gene Signature Based on Epithelial-Mesenchymal Transition of Lung Adenocarcinoma Through Construction and Validation of a Risk-Prediction Model

• Published in: Frontiers in oncology 2021-10, Vol.11, p.726834-726834
• Main Authors: Shi, Jianguang, Wang, Zishan, Guo, Jing, Chen, Yingqi, Tong, Changyong, Tong, Jingjie, Hu, Wentao, Li, Chenwei, Li, Xinjian
• Format: Article
• Language: English
• Subjects: epithelial-mesenchymal transition; LASSO; lung adenocarcinoma; Oncology; prognosis; risk score
• ISSN: 2234-943X; 2234-943X
• DOI: 10.3389/fonc.2021.726834

Epithelial-mesenchymal transition (EMT) process, which is regulated by genes of inducible factors and transcription factor family of signaling pathways, transforms epithelial cells into mesenchymal cells and is involved in tumor invasion and progression and increases tumor tolerance to clinical interventions. This study constructed a multigene marker for lung predicting the prognosis of lung adenocarcinoma (LUAD) patients by bioinformatic analysis based on EMT-related genes. Gene sets associated with EMT were downloaded from the EMT-gene database, and RNA-seq of LUAD and clinical information of patients were downloaded from the TCGA database. Differentially expressed genes were screened by difference analysis. Survival analysis was performed to identify genes associated with LUAD prognosis, and overlapping genes were taken for all the three. Prognosis-related genes were further determined by combining LASSO regression analysis for establishing a prediction signature, and the risk score equation for the prognostic model was established using multifactorial COX regression analysis to construct a survival prognostic model. The model accuracy was evaluated using subject working characteristic curves. According to the median value of risk score, samples were divided into a high-risk group and low-risk group to observe the correlation with the clinicopathological characteristics of patients. Combined with the results of one-way COX regression analysis, HGF, PTX3, and S100P were considered as independent predictors of LUAD prognosis. In lung cancer tissues, HGF and PTX3 expression was downregulated and S100P expression was upregulated. Kaplan-Meier, COX regression analysis showed that HGF, PTX3, and S100P were prognostic independent predictors of LUAD, and high expressions of all the three were all significantly associated with immune cell infiltration. The present study provided potential prognostic predictive biological markers for LUAD patients, and confirmed EMT as a key mechanism in LUAD progression.