ESTRO 2022 - Abstract Book

S187

Abstract book

ESTRO 2022

Sagittal 2D cine MRI videos were acquired at 4 Hz during each fraction of MR-guided online adaptive radiotherapy for 10 prostate cancer patients at a 0.35 T MR-linac (ViewRay-MRIdian). The CTV (prostate) was chosen as the tracking contour and the gating window was defined as the CTV with an isotropic expansion in the [3, 5] mm range. During treatment, the target was continuously tracked by the vendor’s optical flow algorithm. Using the videos and in-house software, the centroid coordinates of the target were calculated relative to the static gating window in anterior-posterior (AP) and superior- inferior (SI) direction. Using the static dose cloud approximation, the planned fraction dose was shifted according to the extracted motion during the beam delivery (in the gating window) to reconstruct the delivered dose by superimposing and averaging the shifted doses. For the CTV, the rectum and the bladder, DVH parameters derived from the planned and the reconstructed delivered dose distributions were compared on a fraction per fraction basis. Results Prostate motion was evaluated for 174 fractions totaling 15.7 hours of cine MRI videos. Averaged over all patients, the average (± 1s) target motion was (-0.6 ± 1.0) mm in the AP and (0.0 ± 0.6) mm in the SI direction. On average, 95% of the motions were within [-3.5 mm, 2.7 mm] in AP and [-2.9 mm, 3.2 mm] in SI direction. In few single fractions, with pronounced intra-fractional motion, substantial deviations of reconstructed and planned fraction dose were observed (Fig. 1). CTV D 98% decreased by up to 7%, rectum and bladder D 2% increased by up to 25% and 3%. However, averaged over all treatment fractions, CTV D 98% showed a decrease within 2% in all patients. The rectum and the bladder increase was on average below 3% and 0.5%, respectively (Fig. 2).

Conclusion A workflow for extraction of the prostate motion during MR-guided radiotherapy based on 2D cine-MRI has been implemented. The obtained motion data can support the estimation of treatment efficiency in different margin/gating window scenarios and enable reconstruction of the delivered dose using a dose cloud approximation in future studies. On average, only minor deviations in target and organs-at-risk dose parameters were observed, indicating safety of the currently adopted MR-guided treatment workflow.

PD-0228 Time-dependent non-linear respiratory tracking models to cope with complex breathing motion patterns

M. Gi ż y ń ska 1 , Y. Seppenwoolde 1 , B. Heijmen 1

1 Erasmus MC Cancer Institute, University Medical Center Rotterdam, Department of Radiotherapy, Rotterdam, The Netherlands Purpose or Objective Correlations between external marker positions and internal fiducial positions are used by the robotic CyberKnife (CK) for tracking of tumors with respiratory motion. Applied correlation models are established pre-treatment and updated during treatment. Apart from linear models, the CK can also use non-linear models, e.g. for handling phase shifts between internal and external motion. Currently, none of the implemented models has an option to explicitly include time dependency to account for baseline drifts. In this study, we extended the currently implemented non-linear models with an explicit time dependency, and we compared accuracy of predicted internal fiducial positions with and without this option for respiratory motion tracks including baseline drifts and phase shifts between external and internal motion.

Materials and Methods