Single-Cell Deconvolution of Head and Neck Squamous Cell Carcinoma

Area of Science:

• Immunogenomics
• Computational Biology
• Cancer Research

Background:

• Interpreting bulk transcriptomic data is challenging due to complex cell-type compositions.
• Cell deconvolution algorithms estimate cell proportions from bulk samples, but their efficacy in tumor tissues like head and neck squamous cell carcinoma (HNSCC) is unclear.
• Understanding immune cell infiltration is crucial for HNSCC prognosis.

Purpose of the Study:

• To evaluate the performance of deconvolution algorithms (CIBERSORTx and MuSiC) for estimating cell-type proportions in HNSCC.
• To investigate the correlation between immune cell composition and patient survival in HNSCC.
• To identify specific immune cell populations and molecular markers associated with improved HNSCC outcomes.

Main Methods:

• Generated cell-type-specific gene expression signatures using single-cell RNA sequencing data from 21 HNSCC samples.
• Applied CIBERSORTx and MuSiC deconvolution algorithms to bulk RNA sequencing data from over 500 TCGA HNSCC samples.
• Correlated estimated cell-type proportions with patient survival data and analyzed gene expression within specific T-cell subpopulations.

Main Results:

• CIBERSORTx and MuSiC yielded comparable estimates of major cell-type proportions in HNSCC.
• A higher fraction of T-cells was significantly associated with improved patient survival.
• Regulatory T-cells (Tregs) were identified as the primary T-cell subpopulation contributing to better survival, with higher TNFRSF4 expression linked to enhanced outcomes.

Conclusions:

• Cell deconvolution is a viable method for analyzing immune cell composition in HNSCC.
• T-cell infiltration, particularly Tregs expressing TNFRSF4, is a positive prognostic indicator in HNSCC.
• TNFRSF4 may play a critical role in the anti-tumor immune response and survival benefits mediated by Tregs in HNSCC.