A. Bookbinder1, M. Krieger2, P. Lansonneur2, A. Magliari2, X. Zhao1, I. J. Choi1,3, C. B. Simone II1,3, H. Lin1,3, M. M. Folkerts2, and M. Kang1; 1New York Proton Center, New York, NY, 2Varian Medical Systems, Palo Alto, CA, 3Memorial Sloan Kettering Cancer Center, New York, NY
Purpose/Objective(s): To improve ease of Pencil Beam Scanning (PBS) Bragg Peak (BP) FLASH planning using universal range shifters and patient-specific range compensators. We hypothesize that BP FLASH can achieve comparable plan quality to conventional PBS intensity modulated proton therapy (IMPT) techniques, and the wide availability of commercial treatment planning systems (TPS) can progress this novel method towards clinical application. Materials/ Methods: IMPT plans were created with a research build of a commercial TPS. A script converted these plans into Bragg Peak FLASH plans using the same target and beam angles. This script created one range compensator for each beam angle based on the target geometry from the beam’s-eye-view and replaced each multi-layered field with a pristine 250 MeV beam with a range shifter. This approach does not spread out the Bragg Peak, requiring only a universal range shifter and a target-specific range-compensator, without range modulators or ridge filters. Spot positions were optimized to evenly distribute sparse spots in the target. Inverse planning was done using a scorecard-based optimizer that adjusted spot weights with a soft minimum monitor unit constraint to achieve FLASH dose rates. A phantom case with a C-shaped target and a core (AAPM TG-119) was showcased to provide details of the implementation of this BP FLASH technique. Results: Plans were normalized to 100% D95%. As comparing IMPT to BP FLASH, target Dmax went from 11.2Gy to 11.5Gy, and core OAR Dmax went from 8.2Gy to 9.9Gy. 40 Gy/s FLASH coverage at 2Gy and 5Gy dose thresholds were 100% and 100% for the target and 80.4% and 89.5% for the core OAR. Conclusion: By introducing range compensators to PBS plans, single-energy FLASH planning can be optimized efficiently using an existing optimizer resulting in a plan quality comparable to conventional IMPT.
