S2154
Physics - Inter-fraction motion management and daily adaptive radiotherapy
ESTRO 2026
and requires improved bladder and rectum preparation procedures. This study compares two preparation procedures in terms of prostate stability and evaluates reproducibility of bladder and rectum volumes and their overlap with the high-dose region. Material/Methods: Eighty consecutive patients with intermediate-risk prostate cancer were included: 40 before (Group-A) and 40 after (Group-B) procedure change. All patients were instructed to empty their bladder and drink two cups (Group-A) or one cup (Group-B) of water 30 minutes before simulation and treatment. Enema was administered before simulation for Group-A, and before both simulation and treatment for Group-B.A total of 240 T2-weighted 3D MRIs were analyzed;the reference planning scan and the first two scans acquired during the initial treatment fraction on a 1.5 T MRI-linac for each patient. Image sets were contoured using a validated in-house AI model[2]. Evaluation included the following metrics: prostate displacement rate (mm/min) between scans at the first fraction, volume of the bladder and rectum, and their wall areas within the high-dose region (defined as prostate+2cm). Wilcoxon and Leven’s tests assessed differences in median and variance, respectively. Spearman’s correlation coefficient evaluated associations between metrics at simulation and treatment. Results: Group-B showed significantly reduced median absolute displacement rate of 0.11 mm/min compared to 0.16 mm/min for Group-A (Figure-1), and 50% lower interquartile ranges in superior–inferior and anterior– posterior directions (Table- 1).
a five-voxel control point grid spacing, and three multi- level resolutions were selected.The PSM resulted in a smooth and continuous bladder emptying for all patients. For smaller bladder volume changes ( ∆ VB < 100 mL), the PSMs captured the full emptying range. However, for patients with larger differences ( ∆ VB > 100 mL), the full range was not captured (Figure 1).
DICE scores between the σ = 1 PSM prediction and empty-bladder CT were calculated. For small ∆ VB, mean DICE = 0.8 ± 0.1, compared to 0.5 ± 0.1 for large ∆ VB. Conclusion: The PSMs successfully modelled smooth bladder emptying for small bladder volume changes. For larger changes, stepwise registrations may be needed to capture full bladder deformation. References: [1] Chen, V. E., et.al, (2019). Medical Dosimetry, 44(2), 102-106. [2] Ma, S., et.al, (2019). Cancer Management and Research, 11, 7171-7181. [3] Rozilawati Ahmad, et.al, (2008). Radiotherapy and Oncology, 89(2), 172- 179.[4] Rozilawati Ahmad, et.al, (2011). Radiotherapy and Oncology, 98(3), 340-346.[5] Jiang, P., et.al, (2024). Bioengineering, 11(12), 1304. [6] Modat, M., et.al, (2010). Computer Methods and Programs in Biomedicine, 98(3), 278–284. Keywords: Cervix,Plan of the day,Patient-specific Models Digital Poster 3319 Optimizing enema and bladder filling procedures for ultra-hypofractionated online adaptive prostate radiotherapy Anders Smedegaard Bertelsen 1,2 , Rasmus Lübeck Christiansena 1,3 , Bahar Celika 1,2 , Louise Lærke Nielsenc 2 , Søren Nielsen Agergaard 1,2 , Christina Junker Nyborg 2 , Annie Kleis Skriver 2 , Carsten Brink 1,2 , Tine Schytte 2,4 , Uffe Bernchou 1,4 1 Laboratory of Radiation Physics, Odense University hospital, Odense, Denmark. 2 Department of Oncology, Odense University Hospital, Odense, Denmark. 3 Department of Oncology, Odense University hospital, Odense, Denmark. 4 Department of Clinical Research, University of Southern Denmark, Odense, Denmark Purpose/Objective: Shifting from mild (3Gy × 20) to ultra-hypofractionated (8Gy × 5) online adaptive prostate radiotherapy increases treatment sessions to more than 30 min[1]
Made with FlippingBook - Share PDF online