Screening Prognostic Risk Markers For Head And Neck Squamous Cell Carcinoma Based Expression Profile G2M Checkpoint Genes

Objective:Head and neck squamous cell carcinoma（HNSCC）is a common cancer worldwide.Since most patients have no preclinical history,they are diagnosed as advanced HNSCC at the time of diagnosis,and the prognosis is poor.Therefore,the use of bioinformatics to study its biomarkers is of great significance to HNSCC.Most of the previous studies used single genes as tumor markers,which have great limitations in application.This article mainly excavated the prognostic risk combined marker model with high sensitivity and specificity about HNSCC.Methods:We first obtain HNSCC gene expression profile and clinicopathological parameters from the human tumor genome（The Cancer Genome Atlas,TCGA）,and use the R language DECENTER package to obtain differential genes.Then take the intersection with the GSE6631 chip differential gene in GEO（The Gene Expression Omnibus）database.Then,the obtained differential genes were enriched with GSEA,and the relevant genes of the most enriched G2M checkpoint were selected for single-factor and Cox multivariate regression analysis to construct a seven-gene risk regression model,and use the KM curve to perform the model Prognostic analysis while using ROC curve to test the sensitivity and specificity of the molecular model.Results:1.This study downloads 500 cases of head and neck squamous cell carcinoma tissues and 44 cases of adjacent tissues from TCGA.4,892 differential genes were obtained,of which 2,525 were up-regulated genes and 2,367 were down-regulated genes.2.The difference genes obtained by TCGA and the difference genes obtained by GSE6631 were crossed,and 861 difference genes were obtained.3.GSEA enrichment analysis has obtained 5 gene sets that are significantly enriched,and the G2M checkpoint pathway is selected,which contains a total of 65genes.4.The single-factor COX risk model analysis of the above 65 genes yielded 35prognostic significant genes,and the multi-factor COX analysis of 35 genes yielded a seven-gene risk regression model.Survival risk score=0.2598*AURKA_EXP+0.2984*BIRC5_EXP+0.28*CKS1B_EXP+0.4865*CDC20_EXP+0.2255*HMMR_EXP-0.2996*SLC12A2_EXP-0.2148*MEIS1_EXP5.According to the median（0.98）,patients were divided into high-risk groups and low-risk groups.The Kaplan-Meier survival prognosis analysis of the two groups indicated that the high-risk prognosis was poor（P=5e-05）.At the same time,the area under the ROC curve AUC=0.763 is between 0.7 and 0.9.Shows that the model has good sensitivity and specificity.Conclusion:In this study,we conducted in-depth mining of HNSC through bio-information analysis and obtained a seven-gene combined prognostic marker,namely protective factors SLC12A2,MEIS1,risk factors AURKA,BIRC5,CDC20,CKS1B,HMMR.The high-risk and low-risk scores of this model are significantly related to the prognosis of HNSCC.With good sensitivity and specificity at the same time,indicating that this model can be used as a good combination marker for HNSCC.