ESTRO 2025 - Abstract Book

S3117

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2025

Conclusion: Soft-check warning/failure thresholds for pre-treatment sanity checks were systematically defined and validated for two target entities, proving the approach’s feasibility. The proposed commissioning method can be extended to other entities and adaptation techniques. The developed sanity checks will be applied in a multi-institutional OAPT implementation project.

Keywords: Online-adaptive PT, Patient-specific QA

1805

Proffered Paper First-in-human 4D treatment verification with a prompt-gamma imaging slit-camera for a pancreatic-cancer patient in proton therapy Jonathan Berthold 1,2,3 , Lena Nenoff 1,4,5 , Stefanie Bertschi 1 , Julia Thiele 6 , Fabian Lohaus 6 , Guillaume Janssens 7 , Julien Smeets 7 , Kristin Stützer 1,4 , Christian Richter 1,4,6 1 OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine, Dresden, Germany. 2 Center for Advanced Systems Understanding, CASUS, Görlitz, Germany. 3 Helmholtz-Zentrum Dresden-Rossendorf, CASUS - Center for Advanced Systems Understanding, Dresden, Germany. 4 Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiooncology – OncoRay, Dresden, Germany. 5 Faculty of Medicine Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany. 6 Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany. 7 Ion Beam Applications, SA, Louvain-la-Neuve, Belgium Purpose/Objective: Treatment verification with prompt-gamma imaging (PGI) provides an important safety net in proton therapy. At our institute, this is already applied clinically for static tumors using a non-commercial PGI system whose capability to detect anatomical changes was shown for prostate treatments [1]. In this proof-of-concept investigation, we show the feasibility of time-resolved 4D-PGI treatment verification for patients with moving tumors based on the simultaneous measurement of the breathing and PGI signals. Material/Methods: During two fractions (Fx 15&25) of a pancreatic-cancer treatment, we monitored the delivery of one proton beam with the PGI slit camera in parallel to the breathing motion (pressure-belt-system AZ-733V, Anzai Medical), the latter of which enables sorting delivered spots to breathing phases. Additionally, a 4D-control CT (4D-cCT) was acquired directly after each of these fractions. For two additional fractions with cCTs (Fx 11&23), PGI measurements were acquired without respiration monitoring. First, for the two fractions with breathing monitoring, PGI reference simulations were generated on the average CT (3D-PGI) and on the respective 4D-CT phases (4D-PGI), and deviations between these references were evaluated (Fig. 1a). In a second step, 3D-PGI reference simulations on the planning CT were compared to each of the four PGI measurements (Fig. 2a). This represents the standard application of PGI treatment verification for static tumors, for example to trigger a cCT. Results: First, the comparison between 3D- and 4D-PGI reference data showed no systematic deviations, mean differences were <1mm (Fig. 1b). This was expected, since the patient’s breathing motion was limited by a breathing suppression belt. Second, the comparison of the PGI measurements with the 3D reference data on the planning CT showed a good agreement at the beginning of the treatment (Fx 11&15), but larger deviations (>6mm) towards the end of treatment (Fx 23&25). This could be visually verified on the respective cCTs, showing a weight loss leading to a shift