摘要Objective:To explore the correlation between mitophagy and the tumor microenvironment(TME)in patients with head and neck squamous cell carcinoma(HNSCC),with an aim to enhance therapeutic efficacy for HNSCC.Methods:A machine learning-based multigene prognostic signature was developed based on mitophagy-related differentially expressed genes(MRGs)identified in The Cancer Genome Atlas cohort.This signature was corre-lated with the TME using gene set enrichment analysis.The association between this prognostic signature and various immunological features of the TME was explored,including status of tumor-infiltrating immune cells,expression of immune checkpoint molecules,and the immunoscore.Immunohistochemistry validated the expression of hub gene CSNK2A2 and assessed its relationship with immunomarker expression.Quantitative PCR validated CSNK2A2 knockdown in HNSCC cell lines.Functional experiments including Transwell assays to determine cell migration and invasion,Cell Counting Kit 8 assay,and 5-ethynyl-2-deoxyuridine assay were performed to confirm the role of CSNK2A2 in HNSCC.Finally,a subcutaneous xenograft model was generated in C3H mice to validate our findings.Results:The MRG-based prognostic signature showed excellent predictive performance.High-risk patients had significantly shorter progression-free and overall survival(P＜0.0001)than low-risk patients.CD8+T cell infiltration was lower in high-risk groups,whereas low-risk groups showed higher immunological marker expression.Thus,the low-risk HNSCC subtype may benefit from immune therapy,while high-risk subtypes may benefit from chemotherapy(P＜0.001).CSNK2A2 was highly expressed and strongly correlated with CD8 and PD-L1 based on immunohistochemistry of the HNSCC tissue microarray.CSNK2A2 knockdown reduced cell migration,invasion,and proliferation,and arrested cells in G1 phase.In vivo,it led to slower tumor growth and smaller tumor volumes.Conclusion:We established a potential prognostic signature that could improve HNSCC management in the future.CSNK2A2 may be a new biomarker to predict immunotherapy efficacy in HNSCC.