ESTRO 2025 - Abstract Book

S3240

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

ESTRO 2025

6) D. Grama et al., Deep learning-based markerless lung tumor tracking in stereotactic radiotherapy using Siamese networks 1924 Digital Poster The impact of adaptation time on treatment quality in CBCT-based ultra-hypofractionated adaptive radiation therapy for prostate cancer Miriam Eckl, Nour Alfakhori, Kerstin Siebenlist, Tonja Zakrzewski, Judit Boda-Heggemann, Constantin Dreher, Frank A. Giordano, Jens Fleckenstein Radiotherapy and Radiation Oncology, University Medical Center Mannheim, University of Heidelberg, Mannheim, Germany Purpose/Objective: The Ethos therapy system (Siemens Healthineers) allows for CBCT-based (HyperSight) daily adaptation of treatment doses in ultra-hypofractionated adaptive radiation therapy for prostate cancer. During the adaptation process, two CBCTs are acquired: CBCT1 for adaptive planning and CBCT2 for repositioning prior to beam-on. This work aims to assess the dosimetric benefits gained through adaptation and its robustness against organ changes during the adaptation process. Material/Methods: Clinical target volume (CTV), planning target volume (PTV), bladder and rectum were retrospectively, manually contoured on every CBCT for 5 patients treated with an SBRT regimen (PACE-B trial, 40/36.25Gy to the CTV/PTV in 5 fractions). Dice-similarity-coefficients (DSC) were determined between corresponding structures in CBCT1 and CBCT2. Based on a CBCT number-to-electron-density calibration and the applied couch-shift, the adapted daily plans (ADP) optimized on CBCT1 were recalculated on CBCT2. Dose distributions of reference scheduled plans (SCP) were compared to both ADP1 on CBCT1 and ADP2 on CBCT2. Pearson's correlation coefficient (PCC) was used to identify potential correlations between the adaptation time and relevant bladder metrics. Results: The mean CTV V40Gy coverage of 25 adaptive fractions was 80.8±5.7 %, 92.0±1.0 % and 87.7±4.0 % for SCP on CBCT1, ADP1 on CBCT1 and ADP2 on CBCT2, respectively. PTV V36.25Gy were 85.2±5.3 %, 96.6±1.4 % and 95.0±2.5 %. The bladder V37Gy were 5.4±6.1 cc, 8.1±2.2 cc and 13.7±6.5 cc and rectum V36Gy were 0.8±0.8 cc, 0.4±0.6 cc and 0.7±1.2 cc for the SCP, ADP1 and ADP2 plans, respectively. This coincided with the DSC of 0.94±0.02 (CTV), 0.93±0.02 (PTV), 0.90±0.03 (rectum) and 0.73±0.18 (bladder). The PACE-B dosimetric criteria for PTV (V36.25Gy≥90%), bladder (V37Gy<10cc) and rectum (V36Gy<2cc) were fulfilled in (16, 92, 96)% (SCP), (100, 80, 92)% (ADP1) and (96, 28, 88)% (ADP2) of all fractions. Time for adaptation between CBCT1 and CBCT2 was 35.1±10.9 min. PCC between adaptation time and bladder volume or bladder V37Gy constraint violation, respectively, were +0.46 or +0.41. Conclusion: CBCT1-based ADP1 substantially improved the daily treatment plan quality compared to the SCP. However, prolonged treatment adaptation times lead to a degradation of treatment plan quality. Particularly varying bladder filling affects the treatment quality for longer adaptation times, markedly impairing dosimetric benefits through adaptation. Target coverage and rectum sparing can be preserved throughout the adaptation process. Future adaptive SBRT treatments should thus focus on patient preparation and faster adaptation workflows.

Keywords: prostate SBRT, CBCT, dosimetric benefit