Abstract
Pulmonary lymphoepithelioma-like carcinoma (pLELC) is a rare malignancy that lacks specific biomarkers. N6-methyladenosine (m6 A) is the most widespread internal modification of messenger RNA (mRNA), and its dysregulation is involved in the development of many cancers. However, the expression of m6 A genes in pLELC and their roles are unknown. We obtained an exosomal transcriptome data set of patients diagnosed with pLELC and healthy controls using RNA sequencing and identified differentially expressed genes (DEGs) in the two groups using R software. The differential expression of the 37 m6 A genes in the two sets of samples was further analyzed, and receiver operating characteristic (ROC) curves were plotted for each gene to identify their grouping ability. The STRING database was used to construct a protein-protein interaction network for m6 A genes. An mRNA-miRNA regulatory network of m6 A-related DEGs was constructed using the miRNet database, and a prediction score formula was established. A nomogram was constructed based on the candidate m6 A genes and prediction scores. The expression of key genes was determined through the immunohistochemical (IHC) staining of clinical tissue sections. Using ROC curves, nine m6 A genes were revealed to have classification efficacy in both groups of samples. We screened seven m6 A-related DEGs (MAN2C1, HNRNPCL1, FUS, EIF6, DIP2A, COA3, and BUD13) that were beneficial for grouping and constructed nomogram models. Through IHC, we identified FUS and EIF6 as being possibly involved in the occurrence and development of pLELC. The m6 A gene expression patterns in pLELC-derived exosomes were significantly different from those in healthy controls. We screened several key genes to facilitate the development of diagnostic markers for pulmonary lymphoepithelioma.