A novel metabolic-related gene signature for predicting clinical prognosis and immune microenvironment in head and neck squamous cell carcinoma

Objectives: Metabolic reprogramming is not only an essential hallmark in the progression of head and neck squamous cell carcinoma (HNSCC), but also an important regulator of cancer cell adaptation to tumor microenvironment (TME). However, the potential mechanism of metabolic reprogramming in TME of HNSCC is still unknown. Methods: The head and neck squamous cell carcinoma with survival information were obtained the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. The metabolic-related genes were identified by differential analysis and survival analysis. Univariate and multivariate Cox regression analyses were applied to determine an overall estimate of metabolic-related risk signature and related clinical parameters. The sensitivity and specificity of the risk signature were evaluated by time-dependent receiver operation characteristic (ROC) curves. TME immune cell infiltration mediated by metabolic-related genes was explored by gene set enrichment analysis (GSEA) and correlation analysis. Results: Seven metabolic-related genes (SMS, MTHFD2, HPRT1, DNMT1, PYGL, ADA, and P4HA1) were identified to develop a metabolic-related risk signature. The low-risk group had a better overall survival compared to that of the high-risk group in the TCGA and GSE65858 cohorts. The AUCs for 1-, 3-, and 5-year overall survival were 0.646 vs. 0.673, 0.694 vs. 0.639, and 0.673 vs. 0.573, respectively. The AUC vale of risk score was 0.727 vs. 0.673. The low-risk group was associated with immune cell infiltration in the TME. Conclusions: The metabolic-related risk signature were constructed and validated, which could involve in regulating the immune cell infiltration in the TME and act as an independent biomarker that predicted the prognosis of HNSCC.