Aswin Hoffmann: Integration of Magnetic Resonance Imaging and Proton Therapy

The integration of magnetic resonance imaging (MRI) into anti-cancer radiation therapy has become a topic of interest in the field of medical physics research over the past years. Several prototypes for real-time MRI-guided photon therapy have been developed since, some of which have already found their way into clinical practice. For proton therapy, the integration with MRI (MRiPT) is considered to be even more important than for photon therapy, because protons are more sensitive to anatomical variations and patient set-up inaccuracies than photons. This is due to the fact that the range of the proton beam strongly depends on the material composition in the beam path. These uncertainties currently translate into relatively large treatment margins, thus compromising the potential dosimetric benefit of proton therapy to better spare healthy tissue surrounding the tumor than photon therapy. So far, MRiPT has only been studied at a conceptual level, because it requires fundamental technical and physical challenges to be overcome. Among these are the mutual electromagnetic interactions between the imaging and therapy systems and the fact that charged particle beams will be deflected in magnetic field of the MRI scanner due to the Lorentz force.
After a brief overview of the current state of the art in MRiPT, in particular the effects of magnetic field-induced distortions in dose distributions deposited in tissue-equivalent material will be highlighted. Although theoretical methods have recently been proposed for this purpose, experimental data is lacking so far. Therefore, an experimental setup will be presented enabling first in-magnet measurement of magnetic field induced dose distortions of a slowing-down proton beam within a tissue-like medium. A 0.95 T permanent dipole magnet having a transverse magnetic field was used to design and realize an experimental setup for simulation and measurement of the magnetically deflected proton pencil beam’s trajectory inside a tissue-equivalent slab phantom incorporating a centrally placed film dosimeter. A finite element model of the magnet was used to generate 3D magnetic field data. Hall-probe magnetometry was used for experimental validation. The 3D magnetic field data was used as input for Monte Carlo simulations to track the transport of proton pencil beams through the magnetic field to the phantom, and to predict radioactivation and demagnetization effects of the magnet. The predicted beam trajectory and lateral deflection were extracted from the film’s planar dose distribution of the Monte Carlo calculations and validated against the measured dose distributions.
Quantitative evaluation of the measured and calculated 3D magnetic field data showed excellent agreement. The dimensions of the phantom were optimized for beam energies of 70-180 MeV. Both predicted and measured planar dose distributions showed comparable beam trajectories and percentage depth-dose graphs. Repeated Hall probe measurements showed excellent reproducibility and no signs of radiation-induced degradation of the magnetic field over time.
An experimental framework for simulation and measurement of proton beam transport through, and dose deposition in a realistic magnetic field and dosimetry phantom has been realized. For the first time, magnetic and Monte Carlo simulations of magnetic deflected proton beams inside tissue-equivalent material have been experimentally validated with dose measurements. Effects of radiation-induced magnet damage have been shown to be negligible.
The results of this study justify further research on the physical and technical challenges of MRiPT.