ESTRO 2025 - Abstract Book

S3268

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

ESTRO 2025

2592

Poster Discussion Balancing treatment efficiency and accuracy: the role of gating and drift correction in MRI- guided liver SBRT Janita E van Timmeren 1 , Astrid van der Horst 2 , Marlies E Nowee 2 , Jan-Jakob Sonke 2 , Erik van der Bijl 1 , Tomas M Janssen 2 1 Radiation Oncology, Radboud University Medical Center, Nijmegen, Netherlands. 2 Radiation Oncology, Netherlands Cancer Institute, Amsterdam, Netherlands Purpose/Objective: Intra-fraction tumor motion caused by drifts and breathing motion can be managed using a combination of online drift correction and gating. The choice of gating window and number of drift corrections implies making a trade-off between treatment efficiency and accuracy. The aim of this study was to quantify the combined impact of both gating and online drift corrections on PTV margins for liver metastases treated with MRI guided radiotherapy (MRIgRT). Material/Methods: Prior to and during MRI-guided liver SBRT, 4D-MRI scans were acquired every 4 minutes, based on continuously acquired 2D multi-slice cine images. From these, the cranio-caudal position of the diaphragm was extracted with a time resolution of ~3Hz, serving as a validated surrogate for cranio-caudal tumor motion [1,2]. The position-trace derived from the initial 4D-MRI (MRI1), acquired prior to irradiation, was used to set a patient specific gating window threshold at end-expiration to achieve a predefined duty cycle. By subdividing the signal acquired during irradiation into intervals of predefined length, drift corrections were simulated for various drift correction frequencies. In each interval, the drift was calculated based on the position of the average tumor position in the previous interval. This drift was then applied to all subsequent intervals. All per-fraction signals acquired during irradiation for each patient were combined to calculate the average and standard deviation of (motion-compensated) tumor positions. These were used to calculate the systematic and random errors due to intra-fraction motion, including both breathing and drift motion. Population-based PTV margins for different duty cycle and drift correction intervals were calculated using Van Herk’s non-linear margin recipe, with β=0.64 (i.e. for dose prescription at 75%) and a penumbra of 3.2 mm (1SD). Results: Thirty-eight 4D-MRI-derived tumor motion signals from 11 patients treated with 3–5 fractions of SBRT were evaluated. Compared to no-intervention, applying a gating window at a 70%-duty cycle reduced the required PTV margin from 6.0 to 4.2 mm, while drift corrections every 10 minutes reduced the margin from 6.0 to 4.9 mm. Combining both techniques reduced the margin to 3.5 mm. Further reduction is possible using different parameters (Figure 1).