ESTRO 2025 - Abstract Book

S3308

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

ESTRO 2025

3375

Poster Discussion Geometric accuracy of ultra-hypofractionated prostate treatment using fiducial markers and kV intra fraction motion monitoring Emilia Persson 1,2 , Per R Poulsen 3 , Tobias Pommer 1 , Mustafa Kadhim 1,4 , Adalsteinn Gunnlaugsson 1 , Per Munck af Rosenschold 1,4 , André Haraldsson 1,4 1 Radiation Physics, Department of Hematology, Oncology, Radiation Physics, Skåne University Hospital, Lund, Sweden. 2 Department of Translational Medicine, Lund University, Malmö, Sweden. 3 Danish Centre for Particle Therapy, Aarhus University Hospital, Aarhus, Denmark. 4 Medical Radiation Physics, Department of Clinical Sciences, Lund University, Lund, Sweden Purpose/Objective: The prospective REMIND trial (NCT05844761) aims to reduce planning margins for localized prostate cancer based on dosimetric and geometric accuracy. The two prostate cohorts in the trial involve use of (1) image-guided MLC tracking, and (2) gating using intra-fraction images. In this preliminary interim analysis, we report on geometric accuracy of gated radiotherapy. Material/Methods: A total of 17 patients were investigated in this interim analysis, prescribed ultra-hypofractionated treatments of the prostate (42.7Gy/7 fractions), using 1-2 arc VMAT-treatment on TrueBeam (Varian Medical systems, Palo Alto, CA, USA) linear accelerators. Gating was used as intra-fraction motion management with images every 15° and a gating threshold of 5mm radius around three (1x5mm) intra-prostatic markers. The geometrical impact of inter-fraction prostate pitch was investigated by calculating the pitch between planning CT and setup kV-image pairs. Intra-fraction marker motion was estimated with in-house software that segmented the markers in the setup kV-images and intra-fraction triggered images and determined their 3D-traces by a probability-based method [1]. Intra-fraction marker motion calculations used the setup kV-images as baseline corrected for initial patient setup. Results: In total 108 fractions out of 119 fractions were analyzed. Eleven fractions were excluded due to incomplete data sets and further eleven fractions were excluded from the intra-fraction motion analysis due to disruptions or segmentation problems. The inter-fraction prostate pitch ranged between -9.2-10.4° with an absolute median pitch of 2.9 (95%CI 2.9-3.9). The DICE-index ranged between 0.67-1.00, with the lowest value corresponding to the highest pitch (Figure 1). The intra-fraction motion ranged between -22-16mm with a median value of -0.13 (Figure 2). The mean and 95%CI in the different anatomical directions were -0.42mm (-0.44, -0.40), -0.29mm (-0.31, -0.27) and 0.05mm (0.04, 0.06) in AP, CC and LR respectively. A statistically significant difference in intra-fraction motion was seen between all anatomical directions (p<0.05, unpaired two-sample Wilcoxon test).