S899
Clinical - Mixed sites & palliation
ESTRO 2026
prior simulated plans, served as a pre-delivery plan complexity QA tool; Mobius provided independent pre-RT dose verification; and ArcCheck, a diode-based 3D array, performed a post-delivery accuracy check.
re-RT remains limited but encouraging. Identifying predictors of toxicity and treatment response is essential to guide future research and to establish this approach as a viable option for selected patients. Keywords: re-irradiation; hypofractionation; proton therapy Digital Poster 3569 Simulation-free, same-day, palliative online adaptive radiotherapy: Clinical implementation of a phantom-based workflow, compressing days into hours Monica Serban 1,2 , Sam Appiah 3 , Michael Velec 1,2 , Nauman Malik 1,4 , Philip Wong 1,2 , Laura A. Dawson 1,2 , David G. Kirsch 1,2 , Adriel Ngo 2 , Edward Taylor 1 , Tony Tadic 1,2 , Alan Wong 1 , Rebecca Wong 1,2 , Jan Seuntjens 1,2 , C. Jillian Tsai 1,2 , Teuta Zoto Mustafayev 1,2 , Dana Keilty 1,2 1 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Canada. 2 Radiation Oncology, University of Toronto, Toronto, Canada. 3 Radiation Medicine Program, University of Toronto, Toronto, Canada. 4 Radiation Oncology, Princess Margaret Cancer Centre, Toronto, Canada Purpose/Objective: To demonstrate proof-of-concept and evaluate the feasibility, dosimetric performance, and clinical implementation of a phantom-based, simulation-free (sim-free), online adaptive radiotherapy (oART) workflow for same-day palliative treatment, designed to enable rapid plan generation and RT delivery without prior simulation or diagnostic CT. Material/Methods: We implemented a clinical, CT-independent, phantom- based, sim-free oART workflow on the Ethos platform (Varian) for abdominopelvic bone and soft tissue metastases requiring palliative treatment with stereotactic RT, 8–20 Gy in 1 fraction, within two hours of consultation (Figure 1a–d). Prior to clinical treatment, radiation oncologists completed a standardized form specifying the number of targets, location, and approximate tumor dimensions; prescription; and organ-at-risk (OAR) goals. Prior to the on-couch session, a generic phantom-based reference plan with dummy contours was created in Ethos. A single Ethos template was designed to perform robustly across the range of target numbers and anatomical variations. At the on-couch session, HyperSight CBCT was acquired and used for contouring, plan optimization, and dose calculation; OARs were AI-generated, and targets were manually contoured by the radiation oncologist. A monitor unit (MU)-predictive regression model, developed from 190
Results: To date, three patients with multi-target metastases in the pelvis requiring palliation have been treated using this workflow. Pre-RT reference planning and review required 15–20 minutes; pre-RT physics checks, 10 minutes; and total on-couch time – including setup, imaging, contouring, plan generation, and delivery – took 40–50 minutes. All plans met clinical goals, demonstrating excellent OAR sparing and conformal target coverage across complex geometries of up to three distinct targets, with conformity indices ranging from 1.02 to 1.06 (Figure 1e–f, Table 1). Mean ArcCheck gamma pass rate was 96.1% (3%/2 mm). Calculated and predicted MUs agreed within ±20%, well below the ±30% tolerance.
Conclusion: This study demonstrates clinical implementation of a phantom-based, sim-free, oART workflow, enabling planning and delivery within hours of consultation and decision to treat. The workflow uses HyperSight CBCT imaging for on-couch planning, maintaining dose distribution quality through a robust Ethos template that performs effectively the “first time, on-couch”.
Made with FlippingBook - Share PDF online