Gene

Survival analysis

Machine learning

Cancer

Internal medicine

Oncology

Gene expression

Multivariate statistics

Univariate

Immunology

Head and neck cancer

Head and neck squamous-cell carcinoma

Immunotherapy

Gene signature

Nomogram

Computer science

Proportional hazards model

Biochemistry

Medicine

Biology

Immune system

Frontiers in Immunology

Background Head and neck squamous cell carcinoma (HNSCC), the most common head and neck cancer, is highly aggressive and heterogeneous, resulting in variable prognoses and immunotherapeutic outcomes. Natural killer (NK) cells play essential roles in malignancies’ development, diagnosis, and prognosis. The purpose of this study was to establish a reliable signature based on genes related to NK cells (NRGs), thus providing a new perspective for assessing immunotherapy response and prognosis of HNSCC patients. Methods In this study, NRGs were used to classify HNSCC from the TCGA-HNSCC and GEO cohorts. The genes were evaluated using univariate cox regression analysis based on the differential analysis of normal and tumor samples in TCGA-HNSCC conducted using the “limma” R package. Thereafter, we built prognostic gene signatures using LASSO-COX analysis. External validation was carried out in the GSE41613 cohort. Immunity analysis based on NRGs was performed via several methods, such as CIBERSORT, and immunotherapy response was evaluated by TIP portal website. Results With the TCGA-HNSCC data, we established a nomogram based on the 17-NRGs signature and a variety of clinicopathological characteristics. The low-risk group exhibited a better effect when it came to immunotherapy. Conclusions 17-NRGs signature and nomograms demonstrate excellent predictive performance and offer new perspectives for assessing pre-immune efficacy, which will facilitate future precision immuno-oncology research.

Natural killer cell-related prognosis signature characterizes immune landscape and predicts prognosis of HNSCC