ESTRO 2024 - Abstract Book

S4346

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2024

[4] Raoul R.F. Stevens, Colien Hazelaar, Marta Bogowicz et al (2023), A Framework for Assessing the Effect of Cardiac and Respiratory Motion for Stereotactic Arrhythmia Radioablation Using a Digital Phantom With a 17-Segment Model: A STOPSTORM.eu Consortium Study, Int. J Rad. Oncol. Biol. Phys. doi: 10.1016/j.ijrobp.2023.08.059.

2460

Poster Discussion

Online inter-beam treatment adaptation for structures with differential motion

Björn Eiben 1 , Edmund Goodwin 1 , Emilia Persson 1,2,3 , Tarun Gangil 1 , Simeon Nill 1 , Uwe Oelfke 1

1 The Institute of Cancer Research, Joint Department of Physics, Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London, United Kingdom. 2 Radiation Physics, Department of Hematology, Oncology and Radiation Physics, Skåne, Sweden. 3 Lund University, 3) Department of Translational Medicine, Malmö, Sweden

Purpose/Objective:

Target and OAR motion during radiotherapy can cause differences between the delivered and the originally planned dose. This can lead to suboptimal treatment reflected by not fulfilling clinical DVH criteria. The motion in abdominal treatment sites, such as pancreas, is primarily caused by respiration. However, due to the soft-tissue nature and other effects complex differential motion between the target and adjacent OARs may be present. MR-linac technology enables online measurement of the organ motion during beam-on and thus subsequent action can be taken using this information. We have developed an online-capable inter-beam treatment adaptation method that recalculates the segment weights of each next beam with the objective to minimise deviations from the planned dose to the target whilst adjusting for anatomical changes. Our inter-beam segment weight optimisation (IBSWO) utilizes the reconstructed motion-including delivered dose to various structures while accounting for differential motion. We simulated treatment delivery for one fraction of three pancreas cancer patients (35Gy/5#) with target tracking as a motion management strategy, either with or without IBSWO. Patient-specific motion of the target and duodenum for each treatment were calculated from the patients’ cine images using an optical flow algorithm. In addition, a zero motion trace was used for comparison. Our in-house treatment adaptation framework integrates the Research Monaco TPS (v.6.09.00, Elekta AB, Stockholm, Sweden) with our treatment delivery simulator, our fast online dose reconstruction, and the IBSWO. The treatment delivery simulation emulates the linac, most notably the multi-leaf collimator (MLC) motion according to the target position. The simulator sends MLC and position data to the iterative online dose reconstruction which utilises the GPUMCD library (Elekta AB, Stockholm, Sweden). For each iteration step, a dose cube comprising the full anatomy of interest is calculated using the MLC information. Then, to account for differential organ motion, this cube is subsequently sampled by dose point clouds. Each point cloud represents either a target or an OAR structure. For every dose calculation step, each point cloud is shifted according to the reported position and each point accumulates Material/Methods: