Magnetic Resonance-Guided Radiotherapy

The just recently introduced magnetic resonance (MR)-guided radiotherapy (MRgRT) is a very promising treatment modality allowing for an optimized dose coverage of the tumor while sparing the surrounding normal tissue. This is realized by online MR imaging showing superior soft tissue contrast as compared to x-rays normally used for imaging in radiotherapy. However, MRgRT poses several challenges for clinical implementation originating from distortions of the dose distribution and the detector reading due to the Lorentz force acting on secondary electrons, "MR-only"-based treatment planning up to the development and verification of truly adaptive treatment workflows.

To solve these issues, our group contributes to the metrological foundations of reference and small field dosimetry in presence of magnetic fields for photon (MRgRT-DOS project) and ion beams.

Further activities involve the development of dosimetry protocols and their application in phantom measurements. For this, dosimeters ranging from 1D to 3D are combined to validate the geometric and dosimetric accuracy of new MRgRT-treatment workflows. This includes so-called end-to-end tests where the entire patient workflow is validated using in-house developed patient-equivalent phantoms with anthropomorphic imaging contrast and attenuation properties.

Further activities address optimization of imaging protocols for MRI, image registration, generation of pseudo-CTs solely based on MR data, fractionated irradiation and verification of adaptive treatment strategies such as online plan adaption while the patient is on the table as well as gated treatments to account for organ motion. In many of these projects, gel dosimetry plays an important role.

The research activities in photon MRgRT are performed at the MR-Linac of the University Hospital Heidelberg in close cooperation with the responsible clinical medical physicists (Clinical Research Group Medical Physics). Research activities in ion beam MRgRT are performed together with medical physicists of the university hospital and the Heidelberg Ion Beam Therapy Center (HIT).

Within the ARTEMIS Project funded by the Federal Ministry of Education and Research (FMER), technical feasibility of MR-guided ion radiotherapy (MRgIT) was Investigated (grant no. 13GW0436B). Within this project, phantom materials with tissue-specific adjustable CT values as well as T1 and T2 relaxation times in MR imaging were developed for three different magnetic field strengths. Since the requirements for the use of MRI in MRgIT go far beyond those of diagnostic applications, sequence parameter were optimized with respect to spatial fidelity and the generation of synthetic CTs. To handle the large number of sequence parameters and setting options, an automated sequence optimization tool was developed, implemented and tested for two clinical scenarios: (i) achieving the same signal as on a given target image and (ii) maximizing the signal difference between given tissue materials. The tool iteratively changes the sequence parameter settings, applies them on the MRI scanner and automatically evaluates the resulting images. In addition, a quality assurance protocol for an off-line MRI scanner was developed and the accuracy and long-term stability of an optimized sequence was repeatedly examined over a period of 20 months and tested for stability. Furthermore, the impact of the scanner magnets of the beam line on the image quality of an MRI scanner, integrated at the beamline of the Heidelberg Ion-Beam Therapy Center (HIT), was investigated.