Analytical dose modeling for preclinical proton irradiation of millimetric targets

Abstract

Purpose: Due to the considerable development of proton radiotherapy, several proton platforms have emerged to irradiate small animals in order to study the biological effectiveness of proton radiation. A dedicated analytical treatment planning tool was developed in this study to accurately calculate the delivered dose given the specific constraints imposed by the small dimensions of the irradiated areas. Methods: The treatment planning system (TPS) developed in this study is based on an analytical formulation of the Bragg peak and uses experimental range values of protons. The method was validated after comparison with experimental data from the literature and then compared to Monte Carlo simulations conducted using GEANT4. Three examples of treatment planning, performed with phantoms made of water targets and bone-slab insert, were generated with the analytical formulation and GEANT4. Each treatment planning was evaluated using dose-volume histograms and gamma index maps. Results: We demonstrate the value of the analytical function for mouse irradiation, which requires a targeting accuracy of 0.1 mm. Using the appropriate database, the analytical modeling limits the errors caused by misestimating the stopping power. For example, 99% of a 1-mm tumor irradiated with a 24-MeV beam receives the prescribed dose. The analytical dose deviations from the prescribed dose remain within the dose tolerances stated by report 62 of the International Commission on Radiation Units and Measurements for all tested configurations. In addition, the gamma index maps show that the highly constrained targeting accuracy of 0.1 mm for mouse irradiation leads to a significant disagreement between GEANT4 and the reference. This simulated treatment planning is nevertheless compatible with a targeting accuracy exceeding 0.2 mm, corresponding to rat and rabbit irradiations. Conclusion: Good dose accuracy for millimetric tumors is achieved with the analytical calculation used in this work. These volume sizes are typical in mouse models for radiation studies. Our results demonstrate that the choice of analytical rather than simulated treatment planning depends on the animal model under consideration.

Dosimetry and characterization of a 25‐MeV proton beam line for preclinical radiobiology research

“…We measured the lateral fluence of the beam with the CMOS sensor 26 compared to Geant4 and PSD function calculation, demonstrating an excellent homogeneity and flatness, however, the integral depth dose measurement in water of the 10-mm collimated proton beam at 23.5 MeV would have been required to fully characterize the proton beam and the validity of the in-house treatment plan. 25 The percentage depth dose will be measured in a future study. Nevertheless, the nominal proton beam energy was appropriately assessed as well as each energy downstream Al thicknesses attenuator.…”

“…The energy uncertainty was obtained by means of residuals on the calibration curve, and the energy straggling uncertainty was obtained by unfolding the silicon detector resolution and accounting for the impact of the proton beam scattered by the two 200‐μm‐thick and 50‐μm‐thick Al foils (most significant contributions). The energy straggling assessment was previously described by Vanstalle et al …”

“…The energy uncertainty was obtained by means of residuals on the calibration curve, and the energy straggling uncertainty was obtained by unfolding the silicon detector resolution and accounting for the impact of the proton beam scattered by the two 200-lm-thick and 50-lm-thick Al foils (most significant contributions). The energy straggling assessment was previously described by Vanstalle et al 25 Finally, the platform offers both in vitro and in vivo irradiations. For example, tumor xenograft located in mice hind paws can be irradiated directly in contact with the collimator, using an in-house immobilization bed [ Fig.…”

“…An analytical algorithm, referred to as PStar Databased (PSD) function, was developed to define the weight of each Bragg peak composing the SOBP delivery in small animal tumors. 25 It is noteworthy that the PSD function is valid for low-energy protons (below 50 MeV) and applicable for various geometry and size of tumors, but in this study, an example of a 6-mm-thick xenograft tumor will be considered. It consists in a calculation based on the study by Bortfeld et al 37 using the proton range extrapolated from the PSTAR database.…”

“…[22][23][24] This article presents the development of a proton beam platform dedicated to in vivo and in vitro radiobiological studies, using the 25-MeV cyclotron CYRC e (IPHC) as a proton source. A dedicated treatment planning system (analytical calculation) is used to deliver the absorbed dose to small animals and cells, 25 and the beam energy and fluence are characterized. Based on the energy measurement and calculation, and a beam energy degrader wheel, leading to a broad range of LET d , this platform is demonstrating the feasibility of radiobiological studies of clinical interest.…”