FLASH Dose Estimation for Whole Breast Irradiation with Proton Transmission Beams

Abstract

<h3>Purpose/Objective(s)</h3> The sparing effect of ultrahigh dose rate (UHDR; FLASH) radiotherapy (RT) is not only interesting for organs at risk, but also for healthy tissue within the PTV. This makes whole breast irradiation (WBI) an attractive option for FLASH RT, as the entire (mostly healthy) breast is given the prescribed dose. We investigated whether UHDR proton transmission beams (TBs) could achieve clinical plan quality for hypofractionated WBI and determined the amount of FLASH dose for different FLASH thresholds and machine settings. <h3>Materials/Methods</h3> Three planning scenarios, each using a single 250MeV TB delivering 5 × 5.7Gy or 2 × 9.7Gy, were considered: (1) spots with equal monitor units (MUs) were placed in a uniform square grid of various spacing sizes, (2) spot MUs were optimized with a set minimum MU threshold and (3) the TB of the second scenario was split into two beams based on spot MUs: one beam delivered all spots above a certain MU threshold, while another delivered the remaining spots to improve plan quality. All TB plans were compared for a test case and finally the third scenario was planned for three breast cancer patients. The FLASH dose was calculated using two methods and various machine parameters were considered: a minimum spot irradiation time (minST) of 2 ms/1 ms/0.5 ms, a maximum nozzle current (maxN) of 200 nA/400 nA/800 nA and two gantry current (GC) techniques: energy layer based and spot based. <h3>Results</h3> For the test case we found: (1) a spot spacing distance of 7 mm achieved the best balance between plan quality and FLASH dose for equal MU spots; (2) near the target boundary lower MU spots are necessary for better homogeneity, but these spots lower the amount of FLASH dose; (3) for a 40 Gy/s FLASH threshold, the nonsplit beam achieved up to 100% FLASH dose for optimal machine parameters (SB GC, low minST, high maxN), but <5% for clinical settings (EB GC, minST=2 ms, maxN=200 nA) and for 100 Gy/s the FLASH dose dropped to <5% for almost all scenarios; (4) beam splitting gave better plan quality and increased FLASH dose for a 40 Gy/s FLASH threshold to ∼40% for clinical settings and 90-100% for almost all other settings, for 100Gy/s a minST=0.5 ms and maxN=800 nA were necessary for a reasonable FLASH dose and (5) different dose rate calculations influence the amount of FLASH dose significantly. In all scenarios we ensured that plan quality was sufficient and comparable to that of the clinical plan. <h3>Conclusion</h3> We demonstrated that using a single UHDR TB for WBI can achieve acceptable plan quality and a high amount of FLASH dose. Hypofractionation in 5 fractions is clinically common; a potential FLASH effect might lead to shorter fractionation schemes in the future. WBI is given to a large population and even if a small part of patients benefits from UHDR irradiation, it involves many patients. Also, the single TB setup is currently used in a clinical trial, meaning that a clinical framework is already present. Based on our results we conclude that WBI makes a very suitable candidate for FLASH RT.