Abstract:
Today’s technology allows for a highly flexible and precise treatment of cancer with radiation therapy (RT). However, RT in advanced head-and-neck cancer, among other cancer types, is still associated with a high rate of tumor recurrence. One reason is that in general, patients with similar cancer type and stage receive similar treatment even though they may respond differently. The concept of precision medicine in cancer treatment is to take individual factors for different treatment response into account and to tailor the treatment accordingly to improve the outcome. In this context, modern imaging modalities like positron emission tomography (PET) and magnetic resonance imaging (MRI) have great potential to help identify cancer patients with recurrence or poor response to RT as well as to serve as a basis for RT treatment adaptation. This potential is even further expanded with the availability of combined PET/MRI. However, this hybrid imaging technology also comes with new challenges in terms of adapting the imaging to the specific needs of RT. In this thesis, three different imaging strategies were presented to address these challenges and to leverage precision RT.
First, a novel hardware setup was developed which allows to examine head and neck cancer patients with hybrid PET/MRI in RT specific treatment position. Here, PET photon attenuation correction of the hardware setup was implemented and image quality was assessed. Second, a correction method for image distortions was implemented for diffusion-weighted MRI such that according image information can be more accurately integrated into radiation treatment planning. And third, a novel strategy was developed based on supervised Machine Learning of multiparametric MRI data. This strategy allowed to predict PET information on hypoxic tumor regions, i.e., tumor subregions deprived of adequate oxygen supply and indicating poor response to RT.
In future, the presented PET/MRI based imaging strategies may help adapt RT in cancer patients on an individual patient level and thus help to improve treatment outcome.