Development and validation of a novel survival model for head and neck squamous cell carcinoma based on autophagy-related genes.

BACKGROUND: In view of the critical role of autophagy-related genes (ARGs) in the pathogenesis of various diseases including cancer, this study aims to identify and evaluate the potential value of ARGs in head and neck squamous cell carcinoma (HNSCC). METHODS: RNA sequencing and clinical data in The Cancer Genome Atlas (TCGA) were analyzed by univariate Cox regression analysis and Lasso Cox regression analysis model established a novel 13- autophagy related prognostic genes, which were used to build a prognostic risk model. A multivariate Cox proportional regression model and the survival analysis were used to evaluate the prognostic risk model. Moreover, the efficiency of prognostic risk model was tested by receiver operating characteristic (ROC) curve analysis based on data from TCGA database and Gene Expression Omnibus (GEO). Besides, the other independent datasets from Human Protein Atlas dataset (HPA) also applied. RESULTS: 13 ARGs (GABARAPL1, ITGA3, USP10, ST13, MAPK9, PRKN, FADD, IKBKB, ITPR1, TP73, MAP2K7, CDKN2A, and EEF2K) with prognostic value were identified in HNSCC patients. Subsequently, a prognostic risk model was established based on 13 ARGs, and significantly stratified HNSCC patients into high- and low-risk groups in terms of overall survival (OS) (HR = 0.379，95% CI: 0.289-0.495, p < 0.0001). The multivariate Cox analysis revealed that this model was an independent prognostic factor (HR = 1.506, 95% CI = 1.330-1.706, P < 0.001). The areas under the ROC curves (AUC) were significant for both the TCGA and GEO, with AUC of 0.685 and 0.928 respectively. Functional annotation revealed that model significantly enriched in many critical pathways correlated with tumorigenesis, including the p53 pathway, IL2 STAT5 signaling, TGF beta signaling, PI3K Ak mTOR signaling by gene set variation analysis (GSVA) and gene set enrichment analysis (GSEA). In addition, we developed a nomogram shown some clinical net could be used as a reference for clinical decision-making. CONCLUSIONS: Collectively, we developed and validated a novel robust 13-gene signatures for HNSCC prognosis prediction. The 13 ARGs could serve as an independent and reliable prognostic biomarkers and therapeutic targets for the HNSCC patients.

Potential Novel Modules and Hub Genes as Prognostic Candidates of Thyroid Cancer by Weighted Gene Co-Expression Network Analysis.

BACKGROUND: Although thyroid cancer (THCA) is one of the most common type of endocrine malignancy, its highly complex molecular mechanisms of carcinogenesis are not completely known. MATERIALS AND METHODS: In this study, weighted gene co-expression network analysis (WGCNA) was utilized to construct gene co-expression networks and evaluate the relations between modules and clinical traits to identify potential prognostic biomarkers for THCA patients. RNA-seq data and clinical data were downloaded from The Cancer Genome Atlas (TCGA). Other independent datasets from the Gene Expression Omnibus (GEO) database and the Human Protein Atlas database were performed to validate findings. RESULTS: Finally, 11 co-expression modules were constructed and four hub genes, CCDC146, SLC4A4, TDRD9 and MUM1L1, were identified and validated statistically, which were considerably interrelated to worse survival of THCA patients. CONCLUSION: This research study revealed four hub genes may be considered candidate prognostic biomarkers and potential therapeutic targets for THCA patients in the future.