AUTHOR=Yin Ji , Li Xiaohui , Lv Caifeng , He Xian , Luo Xiaoqin , Li Sen , Hu Wenjian 

TITLE=Immune-Related lncRNA Signature for Predicting the Immune Landscape of Head and Neck Squamous Cell Carcinoma

JOURNAL=Frontiers in Molecular Biosciences

VOLUME=Volume 8 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/molecular-biosciences/articles/10.3389/fmolb.2021.689224

DOI=10.3389/fmolb.2021.689224

ISSN=2296-889X

ABSTRACT=Methods: We obtained clinical data and corresponding full transcriptome expression of HNSCC patients from TCGA, downloaded GTF files to distinguish lncRNAs from Ensembl, discerned irlncRNAs based on a co-expression analysis, distinguished differentially expressed irlncRNAs (DEirlncRNAs), and paired these DEirlncRNAs. Univariate Cox regression analysis, LASSO regression analysis, and stepwise multivariate Cox regression analysis were then performed to screen lncRNA pairs, calculate the risk coefficient, and establish a prognosis model. Finally, the predictive power of this model was validated through AUC and the ROC curves, and the AIC values of each point of the 5-year ROC curve were calculated to select the maximum inflection point, which was applied as a cut-off point to divide patients into low or high-risk groups. Based on this methodology, we were able to more effectively differentiate between these groups in terms of survival, clinic-pathological characteristics, tumor immune infiltrating status, chemotherapeutics sensitivity and immunosuppressive molecules. 
Results: A 13 irlncRNA pairs signature was built, and the ROC analysis demonstrated high sensitivity and specificity of this signature for survival prediction. The Kaplan-Meier analysis indicated that the high-risk group had a significantly shorter survival than the low-risk group, and chi square test certified that the signature was highly related to survival status, clinical stage, T stage, and N stage. Additionally, the signature was further proven to be an independent prognostic risk factors via the Cox regression analyses, and immune infiltrating analyses showed that the high-risk group had significant negative relationships with various immune infiltrations. Finally, the chemotherapeutics sensitivity and the expression level of molecular markers were also significantly different between high and low-risk groups. 
Conclusion: The signature established by paring irlncRNA, regard to specific expression levels, can be utilized for survival prediction and to guide clinical therapy in HNSCC.