Purpose: The advent of MR-LINAC makes it feasible to adapt treatment plans based on the anatomy-of-the-day, especially for low x-ray contrast tumor sites including abdomen and breast. However, under the presence of a magnetic field (especially for a 1.5T model), the known ‘electron return effect’ (ERE) at high-gradient tissue boundaries may make the planning more challenging. In this study we investigated such an effect on the dosimetric endpoints of breast planning.
Methods: A pilot study was performed on pre-operative breast SBRT cases, with a prescription of 40Gy in 5 fractions. We planned and compared the dosimetric results based on four techniques: 1. A non-parallel-opposed beam configuration featuring 6 beams from 45⁰ to 330⁰ (counter-clockwise), with the magnetic field turned ‘on’; 2. The same beam configuration as 1, but with the magnetic field turned ‘off’; 3. A parallel-opposed beam configuration featuring 12 beams ranging from 45⁰ to 150⁰ (counter-clockwise), with the magnetic field turned ‘on’; and 4. The same beam configuration as 3, but with the magnetic field turned ‘off’; The plans were optimized under the same objectives and constraints.
Results: Results showed a clear presence of ERE, in the form of high dose regions extending into the chest wall and lungs for plans with the magnetic field ‘on’. However, with the introduction of proper tuning structures, all four planning techniques achieved adequate target coverage and minimized chest wall and lung doses. Although non-parallel-opposed beam plans show similar dosimetric endpoints to parallel-opposed ones, they need to be further compared regarding to their relative robustness against intra-fractional motion/deformation under ERE.
Conclusion: ERE in the presence of magnetic fields is non-negligible in MR-LINAC plans. Inverse treatment planning with curated tuning structures can mitigate its dosimetric impact, while potential intra-fractional patient motion should be taken into account in the selection of a robust plan.
Funding Support, Disclosures, and Conflict of Interest: The study was supported by funding from the National Institutes of Health (R01CA240808) and from the University of Texas Southwestern Medical Center.
Not Applicable / None Entered.