Abstract
Immunotherapy has become a promising therapeutic option for Head and neck squamous cell carcinoma (HNSC). However, only a small percentage of patients could benefit from it, and the overall prognosis was far from satisfactory. In this study, by comprehensively computational analyses of hundreds of HNSC samples, a prognostic signature composed of 13 immune-related genes (IRGs) was constructed. The results of the analyses in multiple datasets indicated that our signature had high predictive accuracy and could serve as an independent prognostic predictor. Based on this signature and multiple clinical variables, we also established a prognostic nomogram to quantitatively predict the survival risk of individual patients. Moreover, this signature could accurately predict survival, reflect the immune microenvironment, and predict immunotherapy efficacy among HNSC patients. Two potential drugs (doxorubicin and daunorubicin) were also identified via Connectivity Map and molecular docking, which could be used for HNSC combination therapy. Taken together, we developed and validated a robust IRG-based prognostic signature to monitor the prognosis of HNSC, which could provide a solid foundation for individualized cancer immunotherapy.