Monte Carlo simulations of microdosimetry and radiolytic species production at long time post proton irradiation using GATE and Geant4-DNA.

BACKGROUND: Radiobiological effectiveness of radiation in cancer treatment can be studied at different scales (molecular till organ scale) and different time post irradiation. The production of free radicals and reactive oxygen species during water radiolysis is particularly relevant to understand the fundamental mechanisms playing a role in observed biological outcomes. The development and validation of Monte Carlo tools integrating the simulation of physical, physico-chemical and chemical stages after radiation is very important to maintain with experiments. PURPOSE: Therefore, in this study, we propose to validate a new Geant4-DNA chemistry module through the simulation of water radiolysis and Fricke dosimetry experiments on a proton preclinical beam line. MATERIAL AND METHODS: In this study, we used the GATE Monte Carlo simulation platform (version 9.3) to simulate a 67.5 MeV proton beam produced with the ARRONAX isochronous cyclotron (IBA Cyclone 70XP) at conventional dose rate (0.2 Gy/s) to simulate the irradiation of ultra-pure liquid water samples and Fricke dosimeter. We compared the depth dose profile with measurements performed with a plane parallel Advanced PTW 34045 Markus ionization chamber. Then, a new Geant4-DNA chemistry application proposed from Geant4 version 11.2 has been used to assess the evolution of HO • ${\mathrm{HO}}^ \bullet $ , e aq - ${\mathrm{e}}_{{\mathrm{aq}}}^ - $ , H 3 O + ${{\mathrm{H}}}_3{{\mathrm{O}}}^ + $ , H 2 O 2 ${{\mathrm{H}}}_2{{\mathrm{O}}}_2$ , H 2 ${{\mathrm{H}}}_2$ , HO 2 • ${\mathrm{HO}}_2^ \bullet $ , HO 2 - , O 2 • - ${\mathrm{HO}}_2^ - ,{\mathrm{\ O}}_2^{ \bullet - }$ and HO - ${\mathrm{HO}}^ - $ reactive species along time until 1-h post-irradiation. In particular, the effect of oxygen and pH has been investigated through comparisons with experimental measurements of radiolytic yields for H 2 O 2 ${{\mathrm{H}}}_2{{\mathrm{O}}}_2$ and Fe3+. RESULTS: GATE simulations reproduced, within 4%, the depth dose profile in liquid water. With Geant4-DNA, we were able to reproduce experimental H 2 O 2 ${{\mathrm{H}}}_2{{\mathrm{O}}}_2$ radiolytic yields 1-h post-irradiation in aerated and deaerated conditions, showing the impact of small changes in oxygen concentrations on species evolution along time. For the Fricke dosimeter, simulated G(Fe3+) is 15.97 ± 0.2 molecules/100 eV which is 11% higher than the measured value (14.4 ± 04 molecules/100 eV). CONCLUSIONS: These results aim to be consolidated by new comparisons involving other radiolytic species, such as e aq - ${\mathrm{e}}_{{\mathrm{aq}}}^ - $ or , O 2 • - $,{\mathrm{\ O}}_2^{ \bullet - }$ to further study the mechanisms underlying the FLASH effect observed at ultra-high dose rates (UHDR).

Methodology for small animals targeted irradiations at conventional and ultra-high dose rates 65 MeV proton beam.

As part of translational research projects, mice may be irradiated on radiobiology platforms such as the one at the ARRONAX cyclotron. Generally, these platforms do not feature an integrated imaging system. Moreover, in the context of ultra-high dose-rate radiotherapy (FLASH-RT), treatment planning should consider potential changes in the beam characteristics and internal movements in the animal. A patient-like set-up and methodology has been implemented to ensure target coverage during conformal irradiations of the brain, lungs and intestines. In addition, respiratory cycle amplitudes were quantified by fluoroscopic acquisitions on a mouse, to ensure organ coverage and to assess the impact of respiration during FLASH-RT using the 4D digital phantom MOBY. Furthermore, beam incidence direction was studied from mice µCBCT and Monte Carlo simulations. Finally,in vivodosimetry with dose-rate independent radiochromic films (OC-1) and their LET dependency were investigated. The immobilization system ensures that the animal is held in a safe and suitable position. The geometrical evaluation of organ coverage, after the addition of the margins around the organs, was satisfactory. Moreover, no measured differences were found between CONV and FLASH beams enabling a single model of the beamline for all planning studies. Finally, the LET-dependency of the OC-1 film was determined and experimentally verified with phantoms, as well as the feasibility of using these filmsin vivoto validate the targeting. The methodology developed ensures accurate and reproducible preclinical irradiations in CONV and FLASH-RT without in-room image guidance in terms of positioning, dose calculation andin vivodosimetry.