Screening biomarkers of oral leukoplakia canceration based on GEO database and in vitro experiments

Abstract Background: Oral leukoplakia (OLK) is the most common precancerous lesion in the oral cavity. This study aimed to screen the key genes of OLK canceration using the Gene Expression Omnibus (GEO) database. Methods: GSE52088 dataset was downloaded from GEO database to screen differentially expressed genes (DEGs) in precancerous cells and head and neck squamous cell carcinoma (HNSCC) cell samples. The Comparative Toxicogenomics Database (CTD) was employed to screen OLK canceration related genes, which were subsequently conducted with a series of bioinformatic analyses. The GSE26549 dataset was then used as an external validation and the immune cell infiltration was assessed by ssGSEA. Finally, real-time PCR was exploited to verify the database results. Results: 439 DEGs were selected from GSE52088 data set (| log2 (Fold change) | > 2.0 and P < 0.001). On this basis, 12 DEGs were selected by CTD database, among them, LAPTM4B, NR3C1 and COX6A1 are finally chosen as three key genes of OLK canceration through external validation by GSE26549. Receiver operating characteristic curve (ROC) analysis showed that the model constructed based on LAPTM4B, NR3C1 and COX6A1 had high accuracy in diagnosing OLK canceration. The area under the curve (AUC) value was 0.753. Moreover, three potential key genes had certain correlation with immune cell infiltration. In addition, real-time PCR results were consistent with the results of three potential key genes in the dataset. Conclusions: In this study, three key genes (LAPTM4B, NR3C1 and COX6A1) were screened as potential biomarkers for the diagnosis and treatment of OLK canceration, laying a foundation for clinical research on OLK canceration.