Models for predicting stage in head and neck squamous cell carcinoma using proteomic data

Area of Science:

• Oncology
• Biomarker Discovery
• Proteomics

Background:

• Head and neck squamous cell carcinoma (HNSCC) presents a significant clinical challenge, with advanced stages correlating to poorer patient prognoses compared to early-stage diagnoses.
• Accurate staging is crucial for effective treatment planning and improving survival rates in HNSCC patients.
• Identifying molecular differences between early and advanced HNSCC could lead to improved diagnostic and prognostic tools.

Purpose of the Study:

• To develop predictive models capable of discriminating between early and advanced stages of head and neck squamous cell carcinoma (HNSCC).
• To leverage reverse phase protein array (RPPA) data for molecular profiling of HNSCC.
• To identify informative protein signatures associated with HNSCC stage.

Main Methods:

• Utilized reverse phase protein array (RPPA) data to analyze protein expression profiles in HNSCC.
• Employed individual and ensemble binary classifiers for predictive modeling.
• Applied filter-based and wrapper-based feature selection techniques to identify relevant protein markers.

Main Results:

• Developed several predictive models that demonstrated moderate performance, indicated by balanced accuracy (MCC) and area under the curve (AUC) values.
• Successfully identified specific sets of proteins that are informative in distinguishing between early and advanced HNSCC stages.
• The selected protein features provide insights into the molecular heterogeneity of HNSCC progression.

Conclusions:

• The study successfully identified protein expression patterns that can help differentiate between early and advanced HNSCC.
• The identified informative protein feature sets may enhance our understanding of the molecular mechanisms driving HNSCC progression.
• These findings could potentially contribute to the development of novel biomarkers for improved HNSCC staging and patient management.