Identification Of LncRNAs For Predicting Prognosis Of Head And Neck Squamous Cell Carcinoma Based On Bioinformatics Analysis

BackgroundHead and neck cancer is the sixth most common malignancy around the world,90%of which are squamous cell carcinoma.Most of patients are diagnosed at advanced stages with high rate of lymphatic metastasis and poor prognosis due to the fact that the symptoms of head and neck squamous cell carcinoma are not specific at early stages.If there were some markers that could predict the therapeutic effects and prognosis,it would be beneficial to treatment and the most appropriate therapeutic strategy could be chosen,which may improve outcomes.Long noncoding RNA is a big class of noncoding RNAs.Although IncRNAs don't encode proteins,they have so many biological functions and can affect the expression of genes.IncRNAs play an extremely important role in tumorigeneis and tumor growth and are associated with the tumor cell proliferation,apoptosis,cell cycle,migration,invasion,tumor immunity and cancer stem cell feature.lncRNAs could be promising therapeutic targets and biomarkers.ObjectiveThere are no effective independent prediction factors for prognosis of head and neck squamous cell carcinoma.lncRNAs widely exist in body fluids and tissues with the time-,space-,cell-and tissue-specific features,which may be excellent molecular biomarkers.However,there are a great many of IncRNAs and the functions and mechanisms of these are still unknown.It is necessary to screen out the prognosis related IncRNAs through bioinformatics methods,which may provide new ideas to further studies.Materials and MethodsRNA sequencing data and according clinical information of head and neck squamous cell carcinoma were obtained from TCGA database.Then IncRNAs and survival data were selected from the primary data.All the data processing,analyzing and visualization were performed through R.The method of Cox proportional hazard model combining with lasso regression was used to screen out IncRNAs for predicting prognosis of head and neck squamous cell carcinoma.Furthermore,the receiver operating characteristic curve and C-index were used to evaluate the prediction accuracy of the method.Finally,the selected lncRNAs were testified by survival analysis.ResultsData from TCGA contained 501 tumor samples and 44 adjacent normal tissue samples.There are 1446 differentially expressed lncRNAs,of which 1110 are upregulated and 336 are downregulated.Then univariate Cox regression analysis was performed and 222 lncRNAs were selected for further lasso regression analysis.According to the minimum lasso coefficient,13 IncRNAs were chosen.Multivariate Cox regression analysis showed the risk score and only 8 lncRNAs were selected.Finally,Kaplan-Meier curves of survival analysis demonstrated four lncRNAs.ConclusionlncRNA AC020637.1,AC092100.1,AP001042.1 and LINC01305 can be used as independent biomarkers for predicting prognosis of patients with head and neck squamous cell carcinoma.