Machine learning-based prognostic modeling integrating PANoptosis in head and neck squamous cell carcinoma

Area of Science:

• Oncology
• Immunology
• Computational Biology

Background:

• Head and neck squamous cell carcinoma (HNSCC) is a complex cancer with unclear prognostic factors.
• The role of PANoptosis, a cell death pathway, in HNSCC and its impact on the tumor microenvironment require further investigation.

Purpose of the Study:

• To develop a robust prognostic model for HNSCC using PANoptosis-related genes.
• To identify key molecular targets for improved HNSCC diagnosis and treatment.

Main Methods:

• Utilized single-cell RNA sequencing data (GSE181919) to analyze PANoptosis.
• Constructed a prognostic model with 101 machine learning algorithms using TCGA-HNSCC data and external validation sets (GSE41613, GSE65858).
• Identified Glutathione S-transferase omega 1 (GSTO1) as a key prognostic gene.

Main Results:

• Developed a StepCox + RSF prognostic model that effectively stratified HNSCC patients into high-risk and low-risk groups.
• The high-risk group demonstrated significantly poorer prognosis, confirmed by ROC and PCA analyses.
• GSTO1 was found to be upregulated in HNSCC and associated with unfavorable outcomes.

Conclusions:

• The PANoptosis-based prognostic model shows significant predictive power for HNSCC patient outcomes.
• GSTO1 emerges as a potential biomarker and therapeutic target for HNSCC.
• Further research into modulating PANoptosis could lead to novel therapeutic strategies for HNSCC.