ESTRO 2021 Abstract Book

S1483

ESTRO 2021

Conclusion We have demonstrated an example of real time dose reconstruction including soft tissue motion derived from high-field MR images, and quantified the effect of PTV motion on key clinical DVHc during treatment delivery.

PO-1757 Monitoring the intra-fracion motion with optical surface scanners during palliative radiotherapy N.A. Hoffmans-Holtzer 1 , M. Olofsen-van Acht 1 , M. Hoogeman 1 , S. Petit 1 1 Erasmus MC, Radiotherapy, Rotterdam, The Netherlands Purpose or Objective Patients receiving palliative RT are often frail and in pain. It is therefore expected that for these patients it may be challenging to lie still on the treatment couch. The purpose of this study was to evaluate the intra- fraction motion of patients treated with palliative intent using optical surface scanners (AlignRT), and to investigate the impact of pain on the motion. Materials and Methods Patients were included who were treated with palliative intent between October 2019 to October 2021. Treatment sites included the thoracic, abdominal and pelvic area, sacrum and spine, and extremities. Patients were asked during intake to score the maximum pain experienced (scale 0 to 10) on the day of the first fraction. During the first treatment fraction AlignRT was used to monitor the patients. The 3D translation of the surface of interest was extracted with a resolution of ±6Hz. Only the signal acquired during beam on was considered. Breathing motion was filtered from the data using a fast Fourier transform for thoracic, abdominal, and pelvic sites. The baseline position was defined as the average position during the first 20 s of the treatment. Patients were categorized as a Mover if more than 20% of the time their position (i.e. the 3D translation) deviated by more than 3 mm from the baseline. To evaluate the impact of pain on motion, a linear regression analysis was applied between the standard deviation of the 3D translation vector (as measure of motion) and the maximum pain scores. Results The motion data of 120 patients were recorded and corrected (typical examples in Figure 1). 19% of the patients moved more than 3mm for at least 20% of the time and were classified as Mover with large variations between treatment sites (0% (abdomen) to 41.7% (extremities)) (Table 1). For 52 patients the pain score was available, of which only 5 patients were classified as Mover. The average maximum pain scored of 6.1 for Non- Movers was comparable to 6.2 scored for Movers . No statistically significant relation was observed between pain score and motion (r 2 << 0.1).