Screened Biomarkers Related To LncRNA And Establishment Of A Methylated Prognostic Model In Head And Neck Squamous Cell Carcinoma Based On TCGA Database

Head and neck squamous cell carcinoma(HNSCC)is a group of heterogenous solid tumors originating from squamous epithelial cells of the upper respiratory tract.It has a strong tendency of metastasis and recurrence,and therefore has a high morbidity and mortality.Finding effective transcriptomic and epigenetic biomarkers closely associated with HNSCC is critical for the early diagnosis,treatment and prognosis monitoring of patients.The important role of long non-coding RNA(lncRNA)in the regulation of tumorigenesis and metastasis has become a research focus in recent years.Studies have indicated that lncRNAs can serve as potential biomarkers for the early diagnosis and prognosis of multiple cancers,including HNSCC.DNA methylation is an epigenetic change,playing a critical role in HNSCC etiology and pathogenesis.Abnormal DNA methylation can effectively predict overall survival(OS)in patients with HNSCC.The main research contents of this study are as follows:1.Based on TCGA database,lncRNA related to HNSCC were screened for biomarkers,in order to provide reference for the precise diagnosis and treatment of clinical HNSCC.RNA sequencing data including long non-coding RNAs(lncRNAs),messenger RNA(mRNAs)and microRNAs(miRNAs)of 141 HNSCC and 44 adjacent normal tissues were obtained from the TCGA database.Differentially expressed genes were analyzed using the R package DESeq.GO terms and Kyoto Encyclopedia of Genes and Genomes(KEGG)pathway enrichment analyses were conducted.A competing endogenous RNAs(ceRNA)network was constructed.The most differentially expressed genes in the main ceRNA network were chosen for nasopharyngeal carcinoma(NPC)cell lines and NPEC2 Bmi-1 cell line verification.A receiver operating characteristic(ROC)curve was constructed for 141 specimens of HNSCC tissues from 44 control samples.In our study,HNSCC-associated abnormally expressed lncRNAs(79),miRNAs(86)and mRNAs(324)were identified.The public microarray results showed that LINC00958 and HOXC13-AS expression levels were upregulated in HNSCC tissues compared with the matched adjacent normal tissues in this study(P<0.0001).LINC00958 and HOXC13-AS expression levels in NPC cell lines were higher than those in the NPEC2 Bmi-1 cell line(P<0.05).The results showed that the area under the ROC curve(AUC)of LINC00958 reached up to 0.906 at a cutoff value of 7.96,with a sensitivity and specificity of 80.85%and 90.91%,respectively.The AUC of HOXC13-AS reached up to 0.898 at a cutoff value of 0.695,with sensitivity and specificity values of 86.23%and 83.78%,respectively.The study indicates that LINC00958 and HOXC13-AS are new candidate diagnostic biomarkers for HNSCC patients.2.Develop a predictive methylation signature based on bioinformatics analysis to improve HNSCC prognosis efficiency and optimize the therapeutic effect.In this study,we identified 22 DMGs corresponding to 246 methylated sites from 529 HNSCC tissue samples and 50 adjacent normal tissue samples,by setting P values and false discovery rates(FDRs)less than 0.05.About 512 HNSCC samples with intact medical records and methylated information were randomized into a training group(n=341)and a test group(n=171).A univariate Cox proportional hazards regression analysis was applied to explore the relationship between OS and 246 differentially methylated sites in training group.A total of 19 methylated sites were found to be significant correlation with OS(P<0.05).The most predictive differentially expressed methylated sites as strongly correlated to patient survival were then screened out from the 19 methylation sites by Cox multivariate analysis.An optimal predictive model for HNSCC prognostic,composed of cg26428455,cg13754259,cg17421709,cg19229344 and cg11668749 was constructed.The patients can be separated into a high-risk and a low-risk group using the median RS from the training group as the cutoff point.The Kaplan-Meier survival curves showed that the OS were significantly different between the high-and low-risk groups sorted by the signature in the training group(median:1.38 vs.1.57 years,log-rank test,P<0.001).The predictive power was then validated in the test groups(median:1.34 vs.1.75 years,log-rank test,P<0.001).The area under the receiver operating characteristic curve(AUC)based on the signature for predicting the five-year survival rates,was 0.7 in the training and 0.73 in test groups,respectively.The result of multivariate Cox regression analysis showed that the RS signature based on five-mathylation site was an independent prognostic tool for OS prediction in patients.In addition,a predictive nomogram model which incorporated RS signature and patient clinical information was then developed.The innovative model could be a robust prognostic tool in guiding clinical therapy and predicting the probability of OS for patients with HNSCC.We also performed GO and KEGG analyses on 22 DMGs to explore the potential biological functions of DMGs.In addition,five methylation sites were also validated in clinical samples.In this study,bioinformatics and experimental methods were used to screen and verify the diagnostic and prognostic biomarkers related to HNSCC transcriptomics and epigenetics.LINC00958 and HOXC13-AS were screened out as HNSCC diagnostic markers.A Nomogram was established that can intuitively predict the prognosis of HNSCC patients.The results of this research provide new targets and monitoring tools for precision treatment of HNSCC.