Abstract
BACKGROUND: Globally, head and neck squamous cell carcinoma (HNSCC) is a common malignant tumor with high morbidity and mortality. Hence, it is important to find effective biomarkers for the diagnosis and prediction of the prognosis of patients with HNSCC. FAM3D had been proven to be vital in other cancers. However, its predictive and therapeutic value in HNSCC is unclear. Therefore, it is valuable to explore the association between the expression level of FAM3D and its impacts on the prognosis and tumor microenvironment in HNSCC. METHODS: The Cancer Genome Atlas (TCGA) dataset, Genotype-Tissue Expression (GTEx) dataset, the Clinical Proteomic Tumor Analysis Consortium (CPTAC) dataset, and The Human Protein Atlas (THPA) website were used to assess HNSCC expressions in tumor and nontumor tissues. Then, we further conducted immunohistochemistry experiment as internal cohort to validate the same results. The Cox regression analysis, Kaplan-Meier analysis, and nomograms were performed to find the predictive prognostic value of FAM3D in HNSCC patients and its relationship with the clinicopathological features in HNSCC. The Gene Expression Omnibus (GEO) dataset was utilized to externally verify the prognosis value of FAM3D in HNSCC. Gene Set Enrichment Analysis (GESA) was applied to search the molecular and biological functions of FAM3D. The association between FAM3D and immune cell infiltration was investigated with the Tumor Immune Estimating Resource, version 2 (TIMER2). The relationships between FAM3D expression and tumor microenvironment (TME) scores, immune checkpoints, and antitumor compound half-maximal inhibitory concentration predictions were also explored. RESULTS: In different datasets, FAM3D mRNA and protein levels were all significantly lower in HNSCC tissues than in normal tissues, and they were strongly inversely associated with tumor grade, stage, lymph node metastasis, and T stage. Patients with high-FAM3D-expression displayed better prognosis than those with low-FAM3D-expression. FAM3D was also determined to be a suitable biomarker for predicting the prognosis of patients with HNSCC. This was externally validated in the GEO dataset. As for gene and protein level, the functional and pathway research results of FAM3D indicated that it was enriched in alteration of immune-related pathways in HNSCC. The low-expression group had higher stromal and ESTIMATE scores by convention than the high-expression group. FAM3D expression were found to be positively correlated with immune infiltrating cells, such as cancer-associated fibroblasts, myeloid-derived suppressor cells, macrophage cells, T cell CD8+ cells, regulatory T cells, and T cell follicular helper cells. FAM3D's relationships with immune checkpoints and sensitivity to antitumor drugs were also investigated. CONCLUSION: Our study explored the impact of FAM3D as a favorable prognostic marker for HNSCC on the tumor immune microenvironment from multiple perspectives. The results may provide new insights into HNSCC-targeted immunotherapy.