Transcriptomic analysis predicts the risk of progression of premalignant lesions in in human tongue.

ABSTRACT: The 5-year survival rate for patients with oral squamous cell carcinomas (SCC), including tongue SCC, has not significantly improved over the last several decades. Oral potentially malignant disorders (OPMD), including oral dysplasias, are oral epithelial disorders that can develop into oral SCCs. To identify molecular characteristics that might predict conversion of OPMDs to SCCs and guide treatment plans, we performed global transcriptomic analysis of human tongue OPMD (n=9) and tongue SCC (n=11) samples with paired normal margin tissue from patients treated at Weill Cornell Medicine. Compared to margin tissue, SCCs showed more transcript changes than OPMDs. OPMDs and SCCs shared some altered transcripts, but these changes were generally greater in SCCs than OPMDs. Both OPMDs and SCCs showed altered signaling pathways related to cell migration, basement membrane disruption, and metastasis. We suggest that OPMDs were on the path towards malignant transformation. Based in patterns of gene expression, both OPMD samples and SCC samples can be categorized into subclasses (mesenchymal, classical, basal, and atypical) similar to those seen in human head and neck SCC (HNSCC). These subclasses of OPMDs the potential to be used to stratify patient prognosis and therapeutic options for tongue OPMDs. Lastly, we developed Firth logistic regression models to classify OPMDs and SCCs using a signature gene set ELF5, RPTN, IGSF10, HTR3A, and CRMP1, and this predictive model is amenable to testing as data sets become available. We suggest that changes in these five transcripts relative to paired normal tissue could be used to predict of the likelihood of an OPMD developing into a SCC.

ORGANISM(S): Homo sapiens

PROVIDER: GSE202048 | GEO | 2023/10/08

REPOSITORIES: GEO