Author: Niels Bassler, Jonathan Bortfeldt, Francesco Evangelista, Giulio Lovatti, Jasper Nijkamp, Rasmus Nilsson, Munetaka Nitta, Katia Parodi, Marco Pinto, Per R. Poulsen, Brita Singers Sørensen, Peter Thirolf, Erik Traneus, Taiga Yamaya 👨🔬
Affiliation: RaySearch Laboratories AB, National Institutes for Quantum Science and Technology, Department of Medical Physics, Ludwig-Maximilians-Universität München (LMU Munich), Ludwig-Maximilians Universität, Danish Centre for Particle Therapy, Aarhus University Hospital, Department of Medical Physics, Ludwig-Maximilians-Universität (LMU) München, Department of Experimental Clinical Oncology, Aarhus University, Department of Oncology, Aarhus University Hospital 🌍
Purpose: Within the SIRMIO project (Small Animal Proton Irradiator for Research in Molecular Image-Guided Radiation Oncology), we aimed to perform the first in vivo proton beam treatment with a dedicated beamline (Parodi et al., Acta Oncologica, 2019) and online range control using a novel Positron-Emission-Tomography (PET) scanner for quasi-real-time range verification (Lovatti et al., PMB, 2023). A Monte Carlo simulation code was developed to utilize RayStation treatment plans information to predict PET image responses.
Methods: An experimental campaign was conducted at the Danish Centre for Particle Therapy in Aarhus (Denmark), where 8 mice (C57BL) received proton beam irradiation to lung tissue with a median dose (D50%) ranging from 30 to 50 Gy (RBE). The PET scanner, developed by LMU (Germany) in collaboration with QST (Japan), consists of 56 scintillator blocks in a spherical configuration, each incorporating three layers of LYSO crystals (0.9 mm x 0.9 mm x 6.3 mm) for depth-of-interaction information (Nitta et al., IEEE, 2022). A Monte Carlo simulation method was developed to account for intra-spill motion of the mice while the beam remained fixed (Evangelista, PTCOG, 2023). This method utilized RayStation β+ emitter maps data and a novel online image reconstruction algorithm for co-registering moving mice to a fixed CT reference frame.
Results: Preliminary findings showed promising agreement between the predicted activation map and quasi-real-time PET imaging acquired and reconstructed during treatment (60 seconds delay). The comparison with the simulated expectations after sensitivity correction highlighted the potential of this technology to ensure precision delivery. The washout effect was also observed during irradiation, providing additional biological insights.
Conclusion: This study demonstrates the feasibility of online range verification with an in-beam PET scanner for precision proton therapy in preclinical settings. These advancements improve treatment accuracy and pave the way for enhanced molecular imaging and radiation oncology research.