Prognosis Prediction of Lung Adenocarcinoma Patients Based on Molecular Subgroups of DNA Methylation

• Published in: Applied immunohistochemistry & molecular morphology Vol. 31; no. 4; p. 255
• Main Authors: Rao, Xiao, Xue, Jianbo, Du, Yinggan, Zhou, Zhiyou, Lu, Yunping
• Format: Journal Article
• Language: English
• Published: United States 01.04.2023
• Subjects: Adenocarcinoma of Lung; Cell Cycle; Cell Division; DNA Methylation; Humans; Lung Neoplasms
• ISSN: 1533-4058
• DOI: 10.1097/PAI.0000000000001114

Lung adenocarcinoma (LUAD) is a malignant tumor with high mortality. At present, the clinicopathologic feature is the main breakthrough to assess the prognosis of LUAD patients. However, in most cases, the results are less than satisfactory. Cox regression analysis was conducted in this study to obtain methylation sites with significant prognostic relevance based on mRNA expression, DNA methylation data, and clinical data of LUAD from The Cancer Genome Atlas Program database. LUAD patients were grouped into 4 subtypes according to different methylation levels using K-means consensus cluster analysis. By survival analysis, patients were grouped into high-methylation and low-methylation groups. Later, 895 differentially expressed genes (DEGs) were obtained. Eight optimal methylation signature genes associated with prognosis were screened by Cox regression analysis, and a risk assessment model was constructed based on these genes. Samples were then classified into high-risk and low-risk groups depending on the risk assessment model, and prognostic, predictive ability was assessed using survival and receiver operating characteristic (ROC) curves. The results showed that this risk model had a great efficacy in predicting the prognosis of patients, and it was, therefore, able to be an independent prognostic factor. At last, the enrichment analysis demonstrated that the signaling pathways, including cell cycle, homologous recombination, P53 signaling pathway, DNA replication, pentose phosphate pathway, and glycolysis gluconeogenesis were remarkably activated in the high-risk group. In general, we construct an 8-gene model based on DNA methylation molecular subtypes by a series of bioinformatics methods, which can provide new insights for predicting the prognosis of patients with LUAD.