Head & nEck canceR treatment through AI-based magnEtic reSonance-guided Photon/Protontherapy – HERESP

Cancer is the leading cause of mortality and morbidity globally. Radiation therapy is used in more than 50% of patients treated for cancer. Head and neck tumors have the sixth most frequent cancer prevalence and are often associated with poor outcomes and lifelong sequels. Radiation therapy is particularly challenging due to anatomical complexity and physiological functions associated with the organs at risk. Current treatment implementation relies mostly on CT imaging that fails to provide functional information regarding tumor infiltration. MR imaging (MRI) has the potential to improve efficiency of treatment due to its unique ability for better tissue visualization of organs at risk and tumor physiological characterization. Multiple parametric MRI could further improve the lack of reproducibility that can be encountered using contrast MR images only to translate to corresponding synthetic CT. Multiple parametric MRI and has the additional potential to improve radiation dose planning (local dose control) based on sub-regional radiomic analysis. Adoption of proton therapy is another potential revolution for head and neck radiotherapy in the context of preservation of multiple vital organs at risk during treatment. Artificial intelligence has all the potentials to revolutionize cancer care by harvesting existing treatment plans and incorporating domain knowledge. Overall, this proposal has the unique ambition to introduce a fully MR-based proton therapy treatment solution that is able to grasp and encode tumor biological infiltration, and convert to dose painting constraints that are effectively, efficiently and optimal implemented in treatment thanks to the use of the latest advances of artificial intelligence.