Comprehensive analysis of m6A modification patterns and m6A-related lncRNAs as potential biomarkers in lung adenocarcinoma

Background: N6-methyladenosine (m6A) methylation is considered to induce tumor cell proliferation, migration, and apoptosis. Understanding the mechanism of m6A-related lncRNAs in the development of lung adenocarcinoma (LUAD) may help predict prognosis. Methods: m6A-related lncRNAs related to lung cancer were identified and combined with the MeRIP-Seq dataset. The consensus clustering method was utilized to divide LUAD patients, and prognostic model was constructed using the Lasso Cox algorithm. The cluster profiler package was used for gene ontology and KEGG enrichment. The proportion of immune infiltration was estimated using the CIBERSORT algorithm. The decision tree was constructed by the rpart package, and nomograms were built by the rms package. The Connectivity Map database was analyzed for the therapeutic effects of small molecule drugs for LUAD. In addition, qPCR, colony formation and transwell assays were performed to validate functions of m6A-associated lncRNAs. Results: Nineteen m6A-modified lncRNAs in LUAD were identified. LUAD patients were divided into two categories based on the expression of 19 m6A-related lncRNAs. Cluster 2 patients had better antigen production and expression, while naive B cells, plasma cells, and activated NK cells were lower in cluster 1. Nine m6A-related lncRNAs were selected to establish a risk model for evaluating the prognosis of LUAD patients. The high-risk group had higher tumor mutational burden and lower TIDE scores with more gamma delta T cells and neutrophils. Nomograms showed that the prognostic model had predominant predictive ability for LUAD patients based on the risk score analyzed by the decision tree model. Benzo(a)pyrene and neurodazine might improve the prognosis of LUAD patients. The qRT-PCR results confirmed the reliability of the analytical results. Conclusion: The establishment of a prognostic model of m6A-related lncRNAs can independently predict overall survival in LUAD and may help to develop personalized immunotherapy strategies.