ESTRO 2024 - Abstract Book

S4696

Physics - Optimisation, algorithms and applications for ion beam treatment planning

ESTR0 2024

1 Department of Physics and Technology, University of Bergen, Bergen, Norway. 2 Department of Oncology and Medical Physics, Haukeland University Hospital, Bergen, Norway

Purpose/Objective:

Proton arc therapy (PAT) is a new and promising delivery technique with the potential to reduce doses to normal tissues compared to conventional intensity-modulated proton therapy (IMPT). More dosimetric studies comparing IMPT and PAT are however needed to determine which cases where PAT would provide a lower healthy tissue dose. This is an especially important factor in treatment of pediatric patients which are more susceptible to radiation induced side effects compared to adults. In this study, we therefore compared the two modalities for a cohort of pediatric ependymoma patients.

Material/Methods:

IMPT and PAT plans were made in Eclipse Treatment Planning System (Varian Medical System, Palo Alto, CA, USA) for 10 pediatric ependymoma patients, with a prescribed dose of 54 Gy(RBE). For the IMPT plans, three fields were used, with two opposing lateral fields and one posterior field (Figure 1a and c). For the PAT plan, an analysis of the optimal number of beams was performed, where plans with 3,6,8,12,16 and 20 beams between 95º and 265º degrees were created and compared for a brain tumor patient, avoiding fields anterior of the patient. Further, the final PAT plans were created using a pseudo-arc with the optimal number of beams. The plans were robustly optimized with a ± 0.2 cm isocenter shift and a ± 3% error in the CT calibration curve, with clinical goals from EPTNs and the Danish National Guidelines for proton treatment (DNOG) recommendations. Additionally, a statistical analysis was performed, using the non-parametric Wilcoxon signed-rank test to determine any statistical significance in the differences between the modalities. The optimal number of beams in the PAT plans were found to be 8, as this plan provided the optimal dose distribution based on our criteria, and thus the ependymoma PAT plans were created with 8 beams (Figure 1b and d). In terms of robustness, the PAT plans showed better V 95% and D 98% compared to IMPT for all but one patient. The signed-rank test showed a statistical significance for the dose difference between PAT and IMPT for D 98% and all OARs except the hippocampus and spinal cord (Table 1). The maximum dose to both the brainstem core and surface were reduced slightly, however, the first and third quartile of the surface dose was significantly lower for PAT compared to IMPT (Figure 2a). The D 98% were increased for the target volume, with a median increase of 0.5% with PAT compared to IMPT, showing a slight improvement in tumor coverage (Figure 2b). Further the median dose were reduced for the PAT plans compared to IMPT for both cochleae, with a median reduction in mean dose of 8.0 Gy(RBE) and 6.5 Gy(RBE) for the right and left cochlea, respectively (Figure 2c). Another aspect is the patient variation, as the dose ranges varied significantly between the different patients. This can be seen for the left Cochlea, where the max dose from IMPT varied between 4.28 and 31.99 Gy(RBE), while the PAT plans varied between 2.79 and 30.76 Gy(RBE). The integral dose of the different patients were however similar for both modalities (Figure 2d). Results:

Table 1 Median dose parameters and the ranges for the CTV and the OARs for IMPT and PAT plans. The bold p-values indicate statistical significant differences according to the Wilcoxon signed rank test.