Molecular Decision Tree Algorithms Predict Individual Recurrence Pattern for Locally Advanced Nasopharyngeal...

Area of Science:

• Oncology
• Biostatistics
• Molecular Biology

Background:

• Recurrence after radical radiation is a primary challenge in treating locally advanced nasopharyngeal carcinoma (NPC).
• Predicting individual recurrence patterns is crucial for effective patient management and treatment planning.

Purpose of the Study:

• To develop and validate decision tree algorithms for predicting individual recurrence patterns in locally advanced NPC.
• To integrate multiple molecular and clinical variables for enhanced predictive accuracy.

Main Methods:

• Utilized data from 136 locally advanced NPC patients from a phase III trial.
• Detected expression levels of 33 biomarkers, 3 Epstein-Barr virus serological antibodies, and 5 clinicopathological variables.
• Constructed decision tree classifiers using adaptive boosting for variable selection and 10-fold cross-validation.

Main Results:

• Selected 13 molecular biomarkers (e.g., AKT1, Bcl-2, HIF-1α) and N stage for classifier construction.
• Achieved high predictive sensitivity (87.2-93.3%), specificity (69.0-100.0%), and accuracy (84.5-95.2%) for recurrence prediction.
• Confirmed decision tree classifiers as independent prognostic factors for individual recurrence.

Conclusions:

• Integrated decision tree algorithms show potential for predicting recurrence patterns in locally advanced NPC.
• This approach can identify high-risk patients for intensified follow-up and early salvage treatment.
• The algorithms offer a valuable tool for personalized medicine in NPC management.