ObjectivesTo determine whether m6A/m1A/m5C/m7G/m6Am/Ψ‐related genes influence the prognosis of a patient with oral squamous cell carcinoma.Materials and MethodsWe investigated the changes in regulatory genes using publicly available data from The Cancer Genome Atlas. Consensus clustering by RNA methylation‐related regulators was used to describe oral squamous cell carcinomas (OSCCs). Then, we developed the prediction model. The tumor microenvironment was investigated using ESTIMATE. Gene set enrichment analysis was used to determine whether pathways or cell types were enriched in different groups. The association between the model and immune‐related risk scores was investigated using correlation analysis.ResultsWe found 22 gene signatures in this analysis and then developed a predictive model that reveals the genes that are highly connected to the overall survival of OSCC patients. The survival and death rates were substantially different in the two groups (high and low risk) classified by the risk scores. The validation cohort verified the phenotypic diversity and prognostic effects of these genes.ConclusionOur data reveal that immune cell infiltration, genetic mutation, and survival potential in OSCC patients are linked to m6A/m1A/m5C/m7G/m6Am/Ψ‐related genes, and we constructed a dependable prognostic model for OSCC patients.

Elevated polyamine levels are required for tumor transformation and development; however, expression patterns of polyamines and their diagnostic potential have not been investigated in oral squamous cell carcinoma (OSCC), and its impact on prognosis has yet to be determined. A total of 440 OSCC samples and clinical data were obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). Consensus clustering was conducted to classify OSCC patients into two subgroups based on the expression of the 17 polyamine regulators. Polyamine-related differentially expressed genes (PARDEGs) among distinct polyamine clusters were determined. To create a prognostic model, PARDEGs were examined in the training cohorts using univariate-Lasso-multivariate Cox regression analyses. Six prognostic genes, namely, “CKS2,” “RIMS3,” “TRAC,” “FMOD,” CALML5,” and “SPINK7,” were identified and applied to develop a predictive model for OSCC. According to the median risk score, the patients were split into high-risk and low-risk groups. The predictive performance of the six gene models was proven by the ROC curve analysis of the training and validation cohorts. Kaplan–Meier curves revealed that the high-risk group had poorer prognosis. Furthermore, the low-risk group was more susceptible to four chemotherapy drugs according to the IC50 of the samples computed by the “pRRophetic” package. The correlation between the risk scores and the proportion of immune cells was calculated. Meanwhile, the tumor mutational burden (TMB) value of the high-risk group was higher. Real-time quantitative polymerase chain reaction was applied to verify the genes constructing the model. The possible connections of the six genes with various immune cell infiltration and therapeutic markers were anticipated. In conclusion, we identified a polyamine-related prognostic signature, and six novel biomarkers in OSCC, which may provide insights to identify new treatment targets for OSCC.