Machine learning-based identification of a consensus immune-derived gene signature to improve head and neck squamous...

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Head and neck squamous cell carcinoma (HNSCC) is an aggressive malignancy with poor patient outcomes.
• Current staging systems lack the precision needed for effective treatment selection.
• There is an urgent need for a reliable biomarker for early HNSCC detection and tailored therapy.

Purpose of the Study:

• To develop and validate a novel gene signature for predicting HNSCC prognosis.
• To identify a biomarker that can guide treatment decisions and improve patient stratification.
• To explore the relationship between the gene signature, genomic alterations, immune cell infiltration, and treatment response.

Main Methods:

• Utilized ten distinct machine learning algorithms and 85 combinations to construct a consensus immune-derived gene signature (CIDGS).
• Applied the CIDGS to analyze gene expression profiles from three independent cohorts (719 HNSCC patients).
• Validated the predictive performance of CIDGS against existing signatures and clinical features.

Main Results:

• The developed CIDGS accurately predicted prognoses across training, testing, and meta-cohorts for HNSCC.
• CIDGS demonstrated superior predictive capacity compared to other signatures when integrated with clinical data.
• Patients with low CIDGS exhibited fewer genomic alterations, higher immune infiltration, and better response to immunotherapy, alongside improved prognoses.
• High CIDGS scores were associated with shorter survival times in HNSCC patients.
• CIDGS also showed efficacy in stratifying immunotherapy response and prognoses for bladder cancer.
• Niclosamide and ruxolitinib emerged as potential therapeutic agents for HNSCC patients with high CIDGS.

Conclusions:

• The consensus immune-derived gene signature (CIDGS) shows significant potential for clinical application in risk stratification and outcome prediction for HNSCC.
• CIDGS can aid in identifying patients who may benefit from specific therapies, including immunotherapy.
• Further research may validate CIDGS for broader applications in other cancer types, such as bladder cancer.