S2397
Physics - Quality assurance and auditing
ESTRO 2026
pharyngeal constrictor muscles (inferior/middle/superior). Mean OAR doses and NTCP for physician-rated dysphagia [3] were calculated and visualised annually using boxplots. For the NTCP calculation, baseline dysphagia grade was assumed to be 0-1. Results: Of the 8,224 patients with pharyngeal and laryngeal cancer who were curatively treated, full DICOM data and AI segmentation were obtained for 5,770. No dosemetrically relevant outliers were identified on the AI-based segmentations.Mean doses to OARs showed temporal reductions. Particular reductions occurred after 2018, following two DAHANCA treatment- planning audits related to the new proton treatment modality. Additional improvements were seen after 2013, following the introduction of the DAHANCA radiotherapy guidelines, which standardised the margin to a 5 mm GTV-to-high-dose CTV1, replacing previous variable practices [4]. The mean dose to the inferior PCM illustrates these dose-reduction time points, with median doses of 61.4 Gy in 2010 and 32.3 Gy in 2023 (Figure 1). Similarly, trends were observed in other OAR: the mean doses to the superior and middle PCM decreased from medians of 55.2 Gy and 61.2 Gy in 2010 to 49.8 Gy and 50.2 Gy in 2023, respectively. The oral cavity dose decreased from 40.2 Gy in 2010 to 35.8 Gy in 2023. Median NTCP for physician-rated dysphagia showed a 16.3%- point improvement, with the biggest decrease after 2018 (Figure 2).
Proffered Paper 4425
National guidelines and audits drive improved organ sparing and dysphagia risk reduction in 5,770 larynx and pharynx DAHANCA patients Sarah W Stougaard 1,2 , Ruta Zukauskaite 3 , Richard Röttger 4 , Jeanette FA Sommer 1 , Maximilian L Konrad 1,2 , Camilla P Nielsen 1,2 , Mogens Bernsdorf 5 , Nikolai H Tarnavski 5 , Camilla K Lonkvist 6 , Susanne Bekke 6 , Anne IS Holm 7 , Cai Grau 7,8 , Jesper G Eriksen 7,9 , Maria Andersen 10 , Kasper Laursen 10 , Mohammad Farhadi 11 , Laura P Kaplan 11 , Thomas Ravkilde 7 , Maiken H Guldberg 8 , Carsten Brink 1 , Jens Overgaard 9 , Christian R Hansen 1,2 1 Department of Oncology – Laboratory of Radiation Physics, Odense University Hospital, Odense, Denmark. 2 Department of Clinical Research, University of Southern Denmark, Odense, Denmark. 3 3) Department of Oncology of Clinical Research, Odense University Hospital, Odense, Denmark. 4 Department of Mathematics and Computer Science, University of Southern Denmark, Odense, Denmark. 5 Department of Oncology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark. 6 Department of Oncology, Copenhagen University Hospital Herlev, Herlev, Denmark. 7 Department of Oncology, Aarhus University Hospital, Aarhus, Denmark. 8 Danish Centre for Particle Therapy, Aarhus University Hospital, Aarhus, Denmark. 9 Department of Experimental Clinical Oncology, Aarhus University Hospital, Aarhus, Denmark. 10 Department of Oncology, Aalborg University Hospital, Aalborg, Denmark. 11 Department of Oncology and Palliative Care, Zealand University Hospital, Roskilde, Denmark
Purpose/Objective: Over the last two decades, technological advancements have significantly influenced
radiotherapy, improving plan quality and presumably patient outcomes. The aim of the current study is to quantify the dosemetric effect of technological advancements on organs at risk (OAR) for head and neck cancer. Material/Methods: Nationwide DAHANCA DICOM data for 8,224 patients with pharyngeal and laryngeal cancer treated curatively with radiotherapy (2010-2023) were collected using automated data-harvesting software. One solution was developed for centres using the Aria database, while a separate system was applied for non-Varian centres, combining the Pinnacle treatment planning system with the Mosaiq record-and-verify system. AI-based segmentation provided robust OAR delineation [1], validated using an outlier detection tool [2]. Included OARs were: buccal mucosa, esophagus, larynx (glottic/supraglottic), mandible, oral cavity, parotid and submandibular glands, and
Made with FlippingBook - Share PDF online