Integrating bulk and single cell RNA sequencing to predict the potential therapeutic efficacy of DLX5 in hypopharyngeal squamous cell carcinoma.

Abstract

Background: Biological targeted therapies show promise for personalized treatment of head and neck squamous cell carcinoma (HNSCC), but the intrinsic heterogeneity of hypopharyngeal squamous cell carcinoma (HPSCC) presents significant treatment challenges.

Methods: CIBERSORT quantified immune cell composition, and weighted gene co-expression network analysis (WGCNA) identified gene modules linked to immune infiltration. Protein-protein interaction (PPI) networks were constructed using Search Tool for the Retrieval of Interacting Genes/Proteins (STRING), followed by the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis. InferCNV analyzed copy number variations (CNV) in malignant epithelial cells, and pySCENIC was used to build the DLX5 transcription factor network. The spatial localization of DLX5 in HPSCC and its correlation with the immune microenvironment were evaluated. In FaDu cells, DLX5 knockdown and overexpression were tested using CCK-8, wound healing, and Transwell assays. OncoPredict screened small molecules affecting DLX5 expression.

Results: The transcription factor DLX5 is significantly upregulated in hypopharyngeal squamous cell carcinoma (HPSCC) and is closely associated with poor prognosis (log-rank p = 0.037). Tumor evolutionary tree analysis reveals a co-amplification of DLX5 with TP63, which is enriched in the MAPK signaling pathway. Functional experiments further confirm that DLX5 promotes the proliferation, migration, and invasion of HPSCC cells. Spatial transcriptomic analysis shows high expression of DLX5 in both tumor and immune regions. Immune cell composition analysis indicates a positive correlation with CD4+ T cells and M2 macrophages, and a negative correlation with CD8+ T cells. Additionally, three small molecule compounds targeting DLX5 (BI 2536, MN-64, and Ulistatin) were identified, which may serve as potential therapeutic agents.

Conclusion: Through bioinformatics analysis, it was discovered that DLX5 exerts an oncogenic role in HPSCC through co-amplification with TP63 and activation of the MAPK signaling pathway, with functional assays further confirming its promotion of malignant phenotypes. High expression of DLX5 correlates positively with immunosuppressive cells, such as M2 macrophages, and negatively with antitumor CD8+ T cells, indicating an association with an immunosuppressive microenvironment. These findings highlight DLX5 as a key determinant of poor prognosis in HPSCC.