S1898
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
96.1%; P=0.221), while V105% was significantly reduced with EZFluence (0.4% vs 2.5%; P=0.002). For both sites, no relevant OAR and hotspot differences were observed, likely reflecting the identical field geometry. Average planning time decreased from approximately 15 minutes to 3 minutes per plan with the automatic tool. In blind evaluation, automated plans were rated superior in 48% of cases and equivalent in 35%. The remaining 17% corresponded to slightly underdosed (“cold”) breast plans, later corrected by normalization. Conclusion: EZFluence can reproduce clinical plan quality while substantially reducing planning time. The tool maintained acceptable target coverage and OAR sparing, producing more homogeneous dose distributions. These preliminary findings support the integration of EZFluence into clinical workflow for efficient, standardized forward-planned breast and WB treatments. Keywords: Forward planning, automatism, 3DCRT A Proof-of-Concept Study of Internal Radiotherapy Using Ultracompact Laser-Accelerated Electrons Jennifer J Hardt 1,2 , Joseph Natal 3 , Oliver Jäkel 1,2 , Matthias Fuchs 3 , Niklas Wahl 1,2 1 Department of Medical Physics in Radiation Oncology,, German Cancer Research Center – DKFZ, Heidelberg, Germany. 2 Heidelberg Institute for Radiation Oncology and National Center for Radiation Research in Oncology, HIRO and NCRO, Heidelberg, Germany. 3 Institute for Beam Physics and Technology, Karlsruhe Institute of Technology, Karlsruhe, Germany Purpose/Objective: In the past several decades, ultracompact electron accelerators based on laser-plasma acceleration have been developed. More recently, fundamentally new accelerating regimes show promise for generating high-charge MeV electron beams [1]. Their compact size enables a novel approach to radiation therapy, in which the accelerator is placed close to the tumor site through a minimally invasive endoscope to generate highly localized electron beams (Figure 1). This technique has the potential to significantly spare surrounding healthy tissue. Digital Poster 2762
Material/Methods: The accelerator was modeled using the Particle-in-Cell (PIC) simulation code Smilei [2], and phase-space datasets of the accelerated electrons were generated for various combinations of laser powers and gas densities within the accelerator. The energies of the electrons reach up to 10 MeV. The phase-space files were imported into the Monte Carlo simulation tool TOPAS [3] to calculate the corresponding dose distributions. Simulated dose distributions in water were analyzed to identify the optimal beam parameters and evaluate different applicator configurations, including the use of a potential scattering foil with varying materials and thicknesses. Furthermore, this novel radiation modality was integrated into the open-source treatment planning framework pyRadPlan [4]. The integration enables the automatic selection of positions and orientations of the accelerator within the patient’s airway. The dose distribution within the patient can be calculated analytically or with the help of Monte Carlo (TOPAS) simulations. Results: Figure 2 shows the integrated and lateral depth dose distributions for an initially narrow beam. The introduction of a scattering foil can be used to broaden the beam, enabling a more uniform irradiation of the tumor while also reducing the entrance dose. Figure 2 presents the dose distribution for a NSCLC patient case using a 1 mm scattering foil. The slice shows the dose resulting from two different accelerator positions inside the airway.
Made with FlippingBook - Share PDF online