Comprehensive Analysis of Immune Cell Infiltration of m6a-Related lncRNA in Lung Squamous Cell Carcinoma and Construction of Relevant Prognostic Models

Abstract

Lung squamous cell carcinoma (LUSC) is the main cause of cancer-related mortality. Some studies demonstrate that m6a and long noncoding RNA (lncRNA) are vital in the pathogenesis of LUSC. In this study, we aimed to further understand the prognostic value of m6a-related lncRNAs in LUSC and their role in the immune microenvironment. For this, we obtained LUSC transcriptome and clinical data from the TCGA database. Further, the identified m6a-related and prognostically relevant lncRNAs were clustered into groups based on prognostic lncRNA expression. Further analysis of the differences between clusters was performed. Five m6A-related lncRNAs were used for model construction using the LASSO regression. The receiver-operating characteristic curve (ROC curves) and decision curve analysis (DCA) were used to assess the model accuracy. Finally, the model was validated using polymerase chain reaction (PCR). We identified 12 m6a-related lncRNAs that were associated with prognosis and were lowly expressed in tumors. Cytotoxic T-lymphocyte-associated protein 4 (CTLA4) highly correlated with prognostic genes, and differential analysis indicated that it was highly expressed in the tumor group and cluster 1. In cluster 2 TIME, tumor cells were less pure and more immune, and stromal-associated cells were present. A prognostic model was constructed based on five m6a-lncRNAs. The area under the curve (AUC) was >0.5 in test group and train group. The PCR results showed that the genes in the prognostic model were lowly expressed in the tumor and were statistically significant ( $p<0.05$ ). We noted that m6a-lncRNAs were strongly associated with LUSC prognosis and the immune microenvironment. Thus, PRC1-AS1, AL132780.2, AC013731.1, SNHG30, and AL358472.2 can be considered as new targets for the treatment of patients with LUSC.