S2235
Physics - Intra-fraction motion management and real-time adaptive radiotherapy
ESTRO 2026
Poster Discussion 2707
Patient-Specific CTV-PTV Margins in Online Adaptive Radiotherapy for Bladder Cancer Frank Brewster 1 , Michael Wake 1 , Hannah Carter 1 , Alberto Salas Mellado 2 , Marcel van Herk 3 , Joe Wood 1 , Peter Hoskin 4,3 , Ananya Choudhury 4,3 , Robert Chuter 1,3 1 Christie Medical Physics & Engineering, The Christie NHS Foundation Trust, Manchester, United Kingdom. 2 Hospital Clínico Universitario de Santiago, CHUS, Santiago de Compostela, Spain. 3 Division of Clinical Cancer Science, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom. 4 Clinical Oncology, The Christie NHS Foundation Trust, Manchester, United Kingdom Purpose/Objective: Online-adaptive radiotherapy (oART) in bladder cancer allows for adaptation of treatment plans for daily target volume. As a result of the extended treatment sessions, PTV margins are primarily required to account for intra-fraction bladder filling and inter- observer variability. However, bladder filling is highly variable between patients [1]. There may be opportunities to adapt margins based on intra-fraction motion information. Material/Methods: Ten patients with bladder cancer treated to 55Gy/20# with oART using an Elekta Unity MR-Linac were included in this study. Intra-fraction bladder filling was quantified using bladder CTVs contoured on daily pre- treatment (CTVpre) and post-treatment images (CTVpost) using a bidirectional local distance measure (BLD) [2]. To decompose distances into the cardinal directions, regions were made covering the most distal 20 mm in each direction. The distance for a direction was taken as the maximum component of the BLD vector in that region. The patient-specific PTV margin for each direction was taken as the 95th percentile BLD across the first 5 fractions with an additional 2 mm for inter-observer variability [3,4].The patient- specific PTV (PTVPS) was then grown from the CTVpre for a sample of 10/15 subsequent fractions due to data completeness and compared to the CTVpost. The volume of CTVpost not covered by PTVPS was extracted. Results: There were large amounts of variation in the derived patient-specific margins with the largest degree of variation in the superior direction, as shown in Figure 1. In the anterior, posterior, left and right directions, patient-specific margins were smaller than population margins. However, most (8/10) patients had a patient- specific margin that was larger than the oART population margin. This is likely due to the nature of population margins, that are calculated to cover a percentile of the population, and the oART margins
Figure 1: Beam-On ratio and Maximum 3D displacement per patient and fraction.
Figure 2: Cumulative VOICE distribution across all fractions Conclusion: Pancreatic MRIgSBRT performed with abdominal compression and CMM-based gating achieved high beam-on efficiency and low baseline shift frequency despite measurable intrafraction motion and a minimal 3-mm PTV expansion. Audit-log analysis combined with VOICE evaluation confirmed robust real-time motion management, supporting the clinical feasibility of precise and efficient MRIgRT for pancreatic cancer. Keywords: MRIgRT, Intrafraction Motion Management, Pancreas
Made with FlippingBook - Share PDF online