ESTRO 2025 - Abstract Book

S3190

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2025

1. Reference Plans: For the 86 patients, the 151 reference plans created were reviewed for technique, number of revisions, modulation factor (MU/Gy), plan optimization time, and dose calculation time. 2. Adaptive Plans: Of the 1842 adaptive sessions delivered, 429 adaptive sessions were reviewed for the selection of adaptive vs. scheduled plans, changes in modulation factor, optimization and calculation times, treatment time, intra-fractional movements, and volume changes in target and organs at risk. Results: Of the 151 reference plans, 150 used IMRT (Intensity Modulated Radiation Therapy) technique and one used VMAT (Volumetric Modulated Arc Therapy) technique. Three plans were created in the Varian Eclipse system and imported into Ethos for reoptimization, while 148 were created in Ethos using 7-field (13%), 9-field (40%), or 12-field (47%) IMRT. The modulation factor varied depending on the number of fields and plan complexity, ranging from 475 MU/Gy for simple 7-field plans to more than 1000 MU/Gy for complex curative plans with multiple dose levels. During treatment, the adaptive plan was selected 98% of the time on average, with selection rates ranging from 96% to 100% depending on the treatment site. Conclusion: During oART, the adaptive plan was chosen over the scheduled plan in 98% of cases. Most reference plans were created using IMRT, with 9-field and 12-field configurations being the most common, which also caused higher modulation factors due to their increased complexity. Digital Poster High quality CBCT images for online adaptive radiotherapy for c-arm linacs Eenas A Omari 1 , Silambarasan Anbumani 1 , Jonathan Mason 2 , Thomas Joyce 2 , Jiaofeng Xu 2 , Nicole O'Connell 2 , Dan Thill 3 , Monica Shukla 1 , Eric Paulson 1 1 Radiation Oncology, Medical College of Wisconsin, Milwaukee, USA. 2 Elekta, AB, Stockholm, Sweden. 3 Elekta, AB, stockholm, Sweden Purpose/Objective: MRI guided online adaptive radiotherapy (oART) caused a paradigm shift in the field. However, the implementation of this advanced technology in RT departments is costly. To increase accessibility to oART, CBCT based techniques are being developed for c-arm linacs. Elekta’s Iris contains an underlying algorithm combining AI-based scatter correction with polyquant reconstruction 1 designed to improve CBCT image quality. In this work, we evaluate Iris CBCT images (CBCT Iris ) for feasibility in oART. Material/Methods: A retrospective analysis was performed for 15 pelvic cancer patients. CBCT Iris images were reconstructed using a research system from raw data collected during IGRT. The Dice similarity coefficient (DSC) and mean distance to agreement (MDA) were calculated to evaluate the auto-segmented contours (DL- and thresholding- based) against ground truth contours segmented by an experienced Radiation Oncologist. The mean HU differences between simulation CT (simCT) and CBCT Iris for the corrected pelvic contours (bone, femurs, bladder, rectum, colon, small bowel, lymph nodes, prostate (male), seminal vesicles (male), penile bulb (male)) were compared. Dose was calculated on the CBCT Iris images using a Monte Carlo based algorithm and compared with dose calculated on simCT. Gamma passing criteria of 95% with 10% threshold at 2%/3% dose difference, and 2mm/3mm distance to agreement (i.e 2%/2mm and 3%/3mm) was used. In addition, results from a previously evaluated DL-based synthetic CT (sCT) model 2 were compared to CBCT Iris using similar metrics. Keywords: online adaptive radiotherapy, Ethos Therapy 4440