ESTRO 2024 - Abstract Book

S3997

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2024

motion mitigation techniques and, at the same time, an evaluation method to assess the target coverage periodically during treatment to determine the impact of the interplay effects and the effectiveness of motion mitigation strategies. For this purpose, 4D phantoms were used in this study to experimentally validate our clinical control infrastructure for moving targets.

Material/Methods:

This study utilized a 4D-moving phantom to simulate controlled tissue motion amplitudes. A custom-designed movable platform was connected to the CIRS Dynamic Thorax Phantom, enabling controlled motion. The CIRS phantom was positioned on the platform, with a 4 cm water solid slab beneath it, as shown in Figure 1. EBT3 radiochromic films were placed between the CIRS phantom and the solid water slabs. This setup facilitated the simulation of moving tissue and the assessment of accumulated dose, accounting for interplay effects resulting from the reciprocal movement of the detector relative to the pencil beam. The study employed three configurations: static (no motion), 10-, and 20-mm I-S motion amplitudes.

A pressure belt system was employed to collect motion surrogate data. 3D- and 4D-CT images were used to generate treatment plans with Raystation 11B. For the static configuration, a CTV was delineated around three tissue inserts (Bone 200 (1.16 g/cc), Breast 50/50 (0.9 g/cc), and Liver (1.07 g/cc), by doing so, the region of effective irradiation also encompassed the Lung Inhale (0.2 g/cc) and Lung Exhale (0.5 g/cc) inserts. These conditions were similar also for the motion configurations. Specifically for 10- and 20-mm motion configurations, a 4DCT was acquired and reconstructed into 10 phases. CTV delineation was conducted individually for each phase, and the union of all delineated contours defined the ITV on the 4D-average CT. Treatment plans in Raystation were optimized based on this ITV. The plans were designed for robustness against setup variations (6 mm) and range uncertainty (3%), according with our clinical practice for lung tumor treatment. The EBT3 films were irradiated using an IBA Proteus Plus delivering one fraction for the static and eight fractions for the two moving configurations. Dose reconstruction employed the method currently used in our clinic as described by Mejers et al. [1] and Visser et al [2]. It incorporated 4DCT, respiratory signals from the Anzai belt system, and treatment delivery machine log files. Treatment spots were divided into DICOM sub-plans, each containing spots associated with a specific respiratory phase. These sub-plans were imported into our treatment planning system, where dose calculations were performed. Deformable image registration (DIR) was generated using the ANACONDA algorithm within Raystation between a reference phase (50%, end of exhale) and other phases of the 4DCT dataset to warp dose contributions. To ensure high-quality vector fields, multiple CIRS inserts were used as controlling ROIs. Reconstructed fraction doses and accumulated course doses were computed for two motion amplitudes (10 mm and 20 mm) across eight fractions.

Results: