Researchers at Sunnybrook Health Sciences Centre in Toronto are investigating the feasibility of using an MRI-Linac to treat breast cancer patients, using hypofractionated partial breast irradiation (HPBI). Sunnybrook's Odette Cancer Centre is evaluating a clinical prototype of Elekta's MRI-Linac. Its clinical staff believe that the system's online visualization and tumour contouring capabilities, combined with the ability to reduce internal motion margins using multileaf collimator tracking or exception gating from real-time MR images, will be advantageous for treating intact breast tumours.

However, one concern is that treatment with an MRI-Linac can cause elevated radiation doses to the skin. The ever-present magnetic field can create electron return effects (ERE), in which electrons liberated at tissue–air and tissue–lung interfaces curl back on themselves and deposit larger radiation doses in tissue at these interfaces.

Medical physicist Anthony Kim and colleagues conducted a simulation study to determine the impact of the magnetic field on HPBI dose distributions. After evaluating a tangential beam arrangement (TAN), 5-beam intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT), the researchers confirmed their hypothesis that the magnetic field increases the skin dose. The magnetic field had clinically negligible effects on radiation dose to the heart and the lung (J. Appl. Clin. Med. Phys. doi: 10.1002/acm2.12182).

Impact of the magnetic field

The researchers developed treatment plans for five patients who did not have surgery due to metastatic disease or severe medical comorbidities. They analysed a total of seven tumours close to the skin, with planning target volumes (PTV) of between 37.3 cc and 341.1 cc. For each tumour, treatment plans for the three beam geometries were optimized with and without a 1.5 T magnetic field, using the same PTV isocoverage for all six plans.

The authors used the same patient image data and target contours as used clinically. They evaluated two skin depths, 3 and 5 mm, to determine whether magnetic dose effects were more prevalent closer to the patient's external surface.

All plans had acceptable PTV coverage. The authors reported that with the magnetic field on, the skin dose was considerably higher for the TAN plan compared with the IMRT plan, which in turn delivered a higher dose to the skin than the VMAT plan.

The skin dose correlated with the number of beam angles used. Specifically, skin dose can be reduced by increasing the number of beam entry angles. Hence, in the presence of a magnetic field, VMAT spared the skin more than IMRT, which in turn spared more than the TAN beam arrangement. Also, the researchers found that the ERE due to the magnetic field was greatest very near the surface of the skin.

The authors explain these phenomena by the fact that the ERE tends to have much less impact at the entry points compared with the beam exit points. Only the beam angles near the tumour created a higher magnetic field dose. The ERE had a much larger impact with TAN radiation delivery compared with IMRT or VMAT delivery when the beam angles were spread far more apart.

Based on their analyses, the authors stated, "The number of beam angles matter, and it is likely that beam arrangement also matters… Skin dose is significantly impacted not only by the magnetic field, but also varies with depth and when increasing the number of beam angles."

The MRI-Linac is currently being installed at Odette Cancer Centre. "The installation will be completed soon, and in the early part of 2018 we will be conducting basic physics research and volunteer MR imaging, to evolve our understanding of this device ahead of treating patients," Kim told medicalphysicsweb. "We have been able to do some basic groundwork with respect to how to best optimize these plans without adverse effects. This groundwork has been possible because of our access to the radiation treatment planning system (Elekta's Monaco) that can simulate the MRI-Linac beam on actual patient data."