Mabroor Ahmed ^1,2 Elke Beyreuther ^3,4 Sebastian Gantz ^4,5 Felix Horst ^4,5 Juergen Meyer ⁶ Jörg Pawelke ^4,5 Thomas E Schmid ^2,3 Jessica Stolz ^2,3 Jan Jakob Wilkens ² Stefan Bartzsch ^1,2*

• ¹ Institute of Radiation Medicine, Helmholtz Center München, Helmholtz Association of German Research Centres (HZ), Neuherberg, Germany
• ² Technical University of Munich, TUM School of Medicine and Health, Klinikum rechts der Isar, Department of Radiation Oncology, Munich, Bavaria, Germany
• ³ Institute of Radiation Physics, Helmholtz Center Dresden-Rossendorf, Helmholtz Association of German Research Centres (HZ), Dresden, Lower Saxony, Germany
• ⁴ OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universit ̈at Dresden, Helmholtz-Zentrum Dresden – Rossendorf, Dresden, Lower Saxony, Germany
• ⁵ Institute of Radiooncology – OncoRay, Helmholtz Center Dresden-Rossendorf, Helmholtz Association of German Research Centres (HZ), Dresden, Lower Saxony, Germany
• ⁶ Fred Hutchinson Cancer Center, Seattle, Washington, United States

Background: Proton Minibeam Radiation Therapy has shown to widen the therapeutic window compared to conventional radiation treatment in pre-clinical studies. The underlying biological mechanisms, however, require more research.The purpose of this study was to develop and characterize a mechanical collimation setup capable of producing 250 µm wide proton minibeams with a center-to-center distance of 1000 µm.Methods: To find the optimal arrangement Monte Carlo simulations were employed using the Geant4 toolkit TOPAS to maximize key parameters such as the peak-to-valley dose ratio (PVDR) and the valley dose rate. The experimental characterization of the optimized setup was carried out with film dosimetry at the University Proton Therapy beamline in Dresden and the proton beamline of the University of Washington Medical Center in Seattle with 150 MeV and 50.5 MeV, respectively. A microDiamond detector (PTW, Freiburg, Germany) was utilized at both beamlines for online proton minibeam dosimetry.Results: A PVDR of 10 was achieved in Dresden and a PVDR of 14 in Seattle. Dosimetry measurements were carried out with EBT3 films at a depth of 5 mm in a polymethylmethacrylate (PMMA) phantom. When comparing film dosimetry with the microDiamond, excellent agreement 1 Ahmed et al.was observed in the valleys. However, the peak dose showed a discrepancy of approximately 10 % in the 150 MeV beam and 20 % in the 50.5 MeV beam between film and microDiamond.Discussion: The characteristics of the minibeams generated with our system compares well with those of other collimated minibeams despite being smaller. The deviations of microDiamond measurements from film readings might be subject to the diamond detector responding differently in the peak and valley regions. Applying previously reported correction factors aligns the dose profile measured by the microDiamond with the profile acquired with EBT3 films in Dresden.The novel proton minibeam system can be operated independently of specific beamlines. It can be transported easily and hence used for inter-institutional comparative studies.The quality of the minibeams allows us to perform in vitro and in vivo experiments in the future.The microDiamond was demonstrated to have great potential for online dosimetry for proton minibeams, yet requires more research to explain the observed discrepancies.