ESTRO 2025 - Abstract Book

S3142

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2025

The deviations between finADP and initADP for TVs and OARs are displayed in Fig. 1. In most cases (Fig. 1(a)), finADP provided superior dose coverage (D98%) of the PTV (0.7 ± 1.5)%. However, coverage and homogeneity for CTV and uterus showed no differences between finADP and initADP. Fig. 1(b) highlights the results for the OARs, showing higher mean dose values for rectum when using the finADP, while only minor differences were observed for bladder and bowel. Conclusion: Retrospective analysis of dosimetric parameters indicates that initADP offers comparable TV coverage and similar OAR dose values compared to finADP treatment plans. These findings suggest that Ethos-propagated contours, combined with Boolean operations, offer an accurate foundation for adaptive treatment in the pelvic area and can help reduce overall treatment time.

Keywords: Adaptive workflow, Dose distribution, CBCT based

2446

Proffered Paper Development of a quality assurance method for dose accumulation in proton therapy of patients with lung and esophageal cancer Richard Canters 1 , Djoya Hattu 1 , Maaike Berbee 1 , Dirk De Ruysscher 1 , Vicki Trier Taasti 1,2 1 Department of Radiation Oncology (MAASTRO), GROW-School for Oncology, Maastricht University Medical Center, Maastricht, Netherlands. 2 Danish Particle Therapy Centre, Arhus University Hospital, Arhus, Denmark Purpose/Objective: Patients undergoing proton therapy for lung or esophageal cancer often experience anatomical changes during treatment. Reliable dose accumulation is essential for accurately assessing the delivered dose throughout the treatment course. This study developed and evaluated a quality-assurance (QA) method to address uncertainties in dose accumulation Material/Methods: We included 20 lung cancer patients (30x2Gy or 20x2.75Gy) and 20 esophageal cancer patients (23x1.8Gy or 28x1.8Gy) treated at our proton center, each receiving weekly repeat CTs (reCTs). Using a reference deformable image registration (DIR) method (Mirada Medical, Oxford, UK), we generated deformed planning CTs (dCTs) by deforming the planning CT (pCT) to the reCTs, creating known deformations. We performed dose deformation and accumulation (DDA) from all dCTs to the pCT using both known deformation (DDA ref ) and DIRs created in Raystation (RaySearch Laboratories, Stockholm, Sweden) (DDA test ). Inspired by the distance discordance metric (DDM) method, doses recalculated on the dCTs were deformed to all other dCTs and then to the pCT (Figure 1). The deformed doses were combined into voxelwise minimum and maximum dose distributions to create DDM intervals. We evaluated dose differences between DDA ref and DDA test in the target and surrounding ring structures, assessing whether these differences fell within the DDM interval. Results: For esophageal cancer patients, the median absolute difference between DDA ref and DDA test was 0.31 (IQR 0.13–0.52) pp for ITV_3mm D98%, 0.62 (IQR 0.47-0.91) Gy for mean heart dose (MHD), and 0.09 (IQR 0.07-0.13) Gy for mean lung dose (MLD). In lung cancer patients, differences were 0.13 (IQR 0.04-0.40) pp for ITV_2mm D98%, 0.13 (IQR 0.04-0.22) Gy for MHD, and 0.13 (IQR 0.04-0.23) Gy for MLD. Dose differences around the target in lung cancer patients were generally small (<2%), with 85% of rings showing >90% of accumulated doses within the DDM interval. For esophageal cancer patients, all rings cranial to the diaphragm had differences <2%, while 23% of rings below the diaphragm exceeded this threshold. Cranially to the diaphragm, 92% of the rings showed >90% of the accumulated doses within the DDM interval, compared to 53% caudally to the diaphragm (Figure 2).