Screening And Analysis Of Biomarkers For HPV-related Head And Neck Squamous Cell Carcinoma

Head and neck squamous cell carcinoma(HNSCC)is the most common type of head and neck tumor.With the rapid development of high-throughput sequencing technology,a large number of cancer-related studies have been carried out at the genetic level,and it has been found that the differential expression of genes and the changes of DNA methylation modification play a very important role in the occurrence and development of tumors.Studies have shown that among HNSCC patients,the overall survival time and tumor-free survival of HPV-positive patients are generally longer than those of HPV-negative patients.This article is based on the RNA-Seq data,450 k methylation data,patient clinical information of HNSCC in the TCGA database,combined with the HNSCC-related research results of Lawrence,Michael S,then the information of HNSCC patients with confirmed HPV status in the TCGA were obtained.According to these patients' IDs,the data of HPV-related patients needed for this study were extracted from the total data.Using transcriptome data to screen the differentially expressed genes,then Cox regression analysis,intersection with oncogenes and tumor suppressor genes,Kaplan-Meier analysis were used to screened out 7 genes that are significantly related to patient survival.Using DNA methylation data to screen the differential methylation sites,Cox regression analysis,Lasso regression analysis and Kaplan-Meier analysis were used to constructed a prognostic model consisting of15 differentially methylated sites.The performance evaluation results of this model are good(AUC=0.787,0.79,0.796).The correlation between 7 prognostic-related genes and 15 prognostic-related sites were studied.It was found that they have similar characteristic change patterns.According to the analysis of characteristic patterns,it is believed that they may be one of the factors leading to the excellent prognosis of HPV-positive HNSCC patients.Finally,the weighted gene co-expression networkanalysis was used to cluster genes into 19 modules according to their expression patterns.Three of the modules that are significantly related to the HPV state were grouped according to the HPV state,selected the DEGs or genes that are highly related to DEGs in each module,bulid the protein protein interaction network,and analysed the connection and function between genes in HPV-positive and HPV-negative network in each module.The results show that the HPV-positive network of each module contains more moudlegenes,genes have closer connections and more similar functions,and the key difference genes between the HPV-positive and the HPV-negative network were obtained.These researches in this article may have potential significance for the understanding of expression patterns of gene in different HPV states,and the research of targeted therapy related of the survival of HPV-negative HNSCC patients.