Abstract 1065: Exploring differential gene expression in head and neck squamous cell carcinoma using TCGA data

Abstract Introduction: The prognosis for head and neck squamous cell carcinoma (HNSCC) has shown limited advancements, and our understanding of how gene expression relates to patient outcomes remains unclear. With advancements in bioinformatics, the characterization of molecular abnormalities and the identification of prognostic signatures have become increasingly important for risk stratification, survival prediction, and treatment management of HNSCC patients. However, due to its lower prevalence, challenges remain in utilizing genetic markers to better understand prognosis. In this study, we performed a differential gene expression analysis (DEGs) to identify genes associated with overall survival. Methods: RNA sequencing data from TCGA, including gene expression profiles and clinical information (tumor stage, overall survival), were analyzed to identify DEGs in HNSCC. DEGs associated with overall survival were identified using the R package DESeq2, and an integrated prognostic signature was developed. Kaplan-Meier survival analysis and log-rank tests were used to evaluate the prognostic value of DEGs, stratifying patients into high- and low-risk groups based on expression levels. A multivariate Cox regression analysis was performed to calculate the hazard ratios (HRs) for each gene, which were then used to construct a risk score. Results: RNA expression data from 521 HNSCC patients were obtained from the TCGA database, comprising 111 early-stage (stage I, II) and 410 advanced-stage (stage III, IV) patients. DEG analysis comparing early-stage and advanced-stage tumors identified 51 genes, including 45 upregulated and 6 downregulated genes. Among these 51 genes, patients were stratified into high- and low-expression groups for each of the genes based on median expression levels. Kaplan-Meier survival analysis revealed that three genes—BRDT (p = 0.0870), ZFP42 (p = 0.0213), and FDCSP (p = 0.0343)—were significantly associated with overall survival. The HRs obtained from the multivariate Cox regression model are as follows: BRDT (HR = 1.51, 95% CI: 1.11-2.06, p = 0.009), ZFP42 (HR = 1.43, 95% CI: 1.05-1.94, p = 0.022), and FDCSP (HR = 0.72, 95% CI: 0.53-0.98, p = 0.035). Conclusions: The study identified a novel 3-gene signature that can stratify HNSCC patients into distinct risk groups based on overall survival. Through DEG analysis, key genes linked to the prognosis of HNSCC were identified, providing valuable insights into the prognostic factors of the disease. Citation Format: Donghoon Shin, Ji Hyung Hong. Exploring differential gene expression in head and neck squamous cell carcinoma using TCGA data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1065.