MRI texture analysis predicts p53 status in head and neck squamous cell carcinoma

Area of Science:

• Oncology
• Radiology
• Biomarkers

Background:

• Head and neck cancers, including oropharyngeal squamous cell carcinoma, require accurate prognostic tools.
• The Tumor, Node, Metastasis staging system is standard, but tumor biomarkers offer additional prognostic value.
• p53 protein status is an important biomarker in cancer prognosis and treatment.

Purpose of the Study:

• To evaluate the efficacy of MR imaging texture analysis in predicting p53 protein status in oropharyngeal squamous cell carcinoma.
• To determine if texture features from MR images can classify tumors based on p53 status.

Main Methods:

• Prospective evaluation of 16 patients with oropharyngeal squamous cell carcinoma using MR imaging, histopathology, and p53 staining.
• MR imaging regions of interest were analyzed using a 2D fast time-frequency transform tool to quantify texture features.
• A statistical model and Bayesian network classifier with 10-fold cross-validation were used to predict p53 status based on texture variables.

Main Results:

• Seven significant texture variables were identified and used to build a predictive model.
• The MR imaging texture analysis model accurately predicted p53 status with 81.3% accuracy (P < .05).
• Cross-validation indicated a moderate agreement (κ = 0.625).

Conclusions:

• MR imaging texture analysis demonstrates significant potential for predicting p53 status in oropharyngeal squamous cell carcinoma.
• This non-invasive imaging technique offers a promising approach for enhancing prognostic assessment in head and neck cancers.
• Further research into MR imaging texture analysis for oropharyngeal squamous cell carcinoma and other head and neck tumors is warranted.