S2783
RTT - RTT contouring, target definition, and treatment planning
ESTRO 2026
treatments on Elekta Agility linacs to Ethos template- based automated planning (Varian). To ensure consistent plan quality during this transition, we investigated whether a RapidPlan (RP) model trained on Pinnacle VMAT plans could be effectively integrated into the Ethos IMRT planning to enable a fast high- quality transition with minimal extra burden for the planners. Material/Methods: This study focusses on treatment planning of hypofractioned localized prostate cancer patients, which are treated in 28 fractions with a total dose up to 86.8Gy to the GTV, 70Gy to the prostate and 64.4Gy to the proximal seminal vesicles. A RP model was created, using a cohort (N=45) of high-quality Pinnacle Auto-Planning prostate cancer treatment plans, and incorporated into an Ethos planning template to guide fully automated plan generation.For evaluation of the plan quality, data of five prostate cancer patients, not from the training set, were used for replanning. For each patient an additional RP plan was generated and compared with the original non-RP clinical Ethos plan. The dosimetric comparisons between Ethos plans were performed using dose-volume histogram (DVH) metrics for the anal wall, rectal wall and bladder. Results: All RP-based Ethos plans met institutional clinical objectives. Compared to the corresponding original clinical Ethos plans, the RP Ethos IMRT plans demonstrated improved organ-at-risk (OAR) sparing. The mean dose to the anal wall was reduced by 5.4 Gy, to the bladder by 2.3 Gy, and to the rectal wall by 1.5 Gy. Table 1 summarizes the mean dose differences (Non-RP – RP) per patient. Figure 1 illustrates representative DVHs comparing with and without RP- based Ethos plans for patient 3, showing overall improved OAR sparing without compromising target coverage.
Conclusion: Integrating a RP model trained on Pinnacle plans into the Ethos planning workflow improved OAR sparing in automated prostate IMRT planning. These results demonstrate the feasibility of transfer learning between treatment planning systems and support consistent, high-quality automated planning during clinical system transitions, limiting the burden of transitioning to a new treatment planning system in times of severe strains on the department. Keywords: Prostate, Ethos Rapid Plan, Plan quality Efficacy and efficiency of the online adaptive workflow in Raystation for bladder patients Rianne de Jong, Ernst Harderwijk, Koen Crama, Mila Donker, Eleftheria Astreinidou, Steven Habraken Radiation Oncology, Leiden University Medical Center, Leiden, Netherlands Purpose/Objective: Successful online adaptive radiotherapy (ART) requires both high-quality treatment plans and streamlined (fast) workflows. This study evaluates both the efficacy and efficiency of the newly implemented ‘automated replanning’-module in Raystation v2025 (RS), focusing on the trade-off between plan quality and workflow duration - from CBCT acquisition to plan approval. Material/Methods: Digital Poster Highlight 2682 Ten patients, previously treated in the pelvic region, were included for simulation. For each patient two CT scans showing significant anatomical differences (body contour, bladder, rectum) were used. One scan served as the reference CT, the other as CBCT. A bladder reference treatment plan (20 × 275cGy, 1 dose level) was created (dosimetrist) using clinical objectives, a 5 mm PTV margin and DLS-based structures. For efficacy, six plans were assessed for PTV coverage and maximum dose (Dmax) to both PTV and bowel:1. Reference treatment plan2. Scheduled plan (reference plan recalculated on new anatomy)3. RS-
Made with FlippingBook - Share PDF online