A Radiomics-Based Classifier for the Progression of Oropharyngeal Cancer Treated with Definitive Radiotherapy

Simple Summary Oropharyngeal cancer is the most common type of head-and-neck squamous cell carcinoma. Although patients with HPV-associated cancers have a better prognosis than patients with HPV-negative cancers, there is a lack of robust biomarkers that describe the relative risk for disease progression. To investigate this problem, we extracted quantitative descriptors from medical images (known as radiomics features) that could not otherwise be assessed by the naked eye. We built a machine learning model based on clinical and radiomics features to predict whether a patient will exhibit disease progression at 2 years post-treatment. These findings are important for identifying patients treated with definitive radiotherapy with low and high risk of disease progression and formulating patient-specific treatment strategies. In this study, we investigated whether radiomics features from pre-treatment positron emission tomography (PET) images could be used to predict disease progression in patients with HPV-positive oropharyngeal cancer treated with definitive proton or x-ray radiotherapy. Machine learning models were built using a dataset from Mayo Clinic, Rochester, Minnesota (n = 72) and tested on a dataset from Mayo Clinic, Phoenix, Arizona (n = 22). A total of 71 clinical and radiomics features were considered. The Mann-Whitney U test was used to identify the top 2 clinical and top 20 radiomics features that were significantly different between progression and progression-free patients. Two dimensionality reduction methods were used to define two feature sets (manually filtered or machine-driven). A forward feature selection scheme was conducted on each feature set to build models of increased complexity (number of input features from 1 to 6) and evaluate model robustness and overfitting. The machine-driven features had superior performance and were less prone to overfitting compared to the manually filtered features. The four-variable Gaussian Naive Bayes model using the 'Radiation Type' clinical feature and three machine-driven features achieved a training accuracy of 79% and testing accuracy of 77%. These results demonstrate that radiomics features can provide risk stratification beyond HPV-status to formulate individualized treatment and follow-up strategies.