ESTRO 2022 - Abstract Book

S794

Abstract book

ESTRO 2022

volume (CTV) was either overlapping with or adjacent to the brain stem the CTV dose coverage was compromised in the clinical treatment plans. The FLASH painting plans had a FLASH painting field with a single energy layer (~234MeV) that delivered most of the dose in a FLASH painting volume in or near the brain stem (Figs 1a,2a) and three non-FLASH proton fields that ensured a uniform CTV dose (Figs 1b,2b). The FLASH painting fields had ~26cm water equivalent beam degrader material in the entrance path to adjust the Bragg peak to the FLASH painting volume. Delivery of the FLASH painting fields at a clinical proton facility (ProBeam, Varian Medical Systems) was simulated by assuming 800nA cyclotron current and the same beam transmission efficiency as measured for the FLASH painting beam energies at the facility (~5%). The simulated spot durations were combined with spot specific dose distributions to determine the dose rate as function of time in each voxel. The FLASH fraction was then calculated as the fraction of the dose in each voxel that was delivered continuously with at least 40Gy/s mean dose rate (Figs 1c,2c). Finally, the FLASH weighted dose to normal tissue was calculated as the dose delivered with FLASH weighted by 80% plus the dose delivered without FLASH weighted by 100% (i.e by assuming a FLASH sparing effect of 20%) (Figs 1d,2d). The brain stem dose with and without FLASH weighting was reported to illustrate the potential brain stem sparing effect of FLASH painting. Plan 1 was made without robust optimization while Plan 2 was optimized and evaluated robustly (±2mm setup, ±3.5% range uncertainties).

Results Both FLASH painting plans had full CTV dose coverage (Figs 1b,2b). FLASH painting reduced the brain stem V54Gy to clinically acceptable levels: from 1.54cm 3 to 0.0 cm 3 for Plan 1 and from 0.33cm 3 to 0.02cm 3 for Plan 2 (worst case robustness scenario) (Figs 1d-e,2d-e).