S2443
Physics - Radiomics, functional and biological imaging, and outcome prediction
ESTRO 2026
dose parameters improve the prediction of late toxicity of RT of HNC patients compared to pure clinical markers. Also, the found univariate associations were in accordance with previous studies (see e.g. [1-3]). Further investigations are needed to construct predictive models for late toxicities with respect to the specific OAR dose levels and clinical markers aiding the individually tailored treatment planning in HN-RT. References: [1] Ranta P. et al. Dysphagia, hypothyroidism, and osteoradionecrosis after radiation therapy for head and neck cancer. Laryngoscope Investig Otolaryngology, 2021 Dec 10;7(1):108-116. 10.1002/lio2.711[2] Huynh T. et al. Radiation- induced long-term dysphagia in survivors of head and neck cancer and association with dose-volume parameters. Radiother Oncol. 2024 Jan;190:110044[3] Van den Bosch L. et al. Comprehensive toxicity risk profiling in radiation therapy for head and neck cancer: A new concept for individually optimised treatment. Radiother Oncol. 2021 Apr:157:147-154. 10.1016/j.radonc.2021.01.024 Keywords: head&neckRT, late toxicity Poster Discussion 2681 External validation of an FDG-PET-radiomic model for distant-relapse-free-survival after radio- chemotherapy for pancreatic adenocarcinoma Monica Maria Vincenzi 1 , Luca Boldrini 2,3 , Pasqualina D'Urso 4 , Paolo Passoni 5 , Martina Mori 1 , Roberta Tummineri 5 , Lorenzo Placidi 2 , Marco Fois 1 , Laura Giannini 5 , Emiliano Spezi 6 , Maria Picchio 7 , Antonella del Vecchio 1 , Arturo Chiti 7,8 , Giuseppe Sanguineti 4 , Maria Antonietta Gambacorta 2,3 , Claudio Fiorino 1 , Nadia Gisella Di Muzio 5,9 1 Medical Physics Department, IRCCS San Raffaele Scientific Institute, Milan, Italy. 2 Radiotherapy, Fondazione Policlinico Universitario “A. Gemelli” IRCCS, Rome, Italy. 3 Institute of Radiology, Università Cattolica del Sacro Cuore, Rome, Italy. 4 Department of Radiation Oncology, IRCCS Regina Elena National Cancer Institute, Rome, Italy. 5 Radiotherapy Department, IRCCS San Raffaele Scientific Institute, Milano, Italy. 6 School of Engineering, Cardiff University, Cardiff, United Kingdom. 7 Nuclear Medicine, IRCCS San Raffaele Scientific Institute, Milan, Italy. 8 Medicine, Vita- Salute University, milan, Italy. 9 Medicine, Vita-Salute University, Milan, Italy Purpose/Objective: The “RadiomiPa” study aimed to externally validate previously trained radiomic-based [18F]FDG-PET models for predicting distant relapse-free survival (DRFS) in patients with unresectable locally advanced
Mind Center, Turku University, Turku, Finland. 5 Department of Radiology, Turku University and Turku University Hospital, Turku, Finland. 6 Varian Medical Systems Finland Oy, Varian, Helsinki, Finland. 7 Department of Radiation Biology, Institute for Cancer Research, Norwegian Radium Hospital and Oslo University Hospital, Oslo, Norway. 8 Department of Physics, University of Oslo, Oslo, Norway. 9 Department of Radiation Biology, Oslo University Hospital, Oslo, Norway Purpose/Objective: Despite advances in treatment planning and delivery techniques, predicting which patients will experience severe late toxicity remains a clinical challenge. We studied late toxicities of head and neck cancer radiation therapy (HN-RT) with respect to the clinical patient characteristics and OAR doses. Material/Methods: The data consists of 199 HN-RT patients treated with IMRT in 2010-15. Of the patients, 48% received pre/postoperative, 83% concurrent chemo- and 89% bilateral neck RT, respectively [1]. Reported toxicities were 42% dysphagia, 28% hypothyroidism and 12% osteoradionecrosis among the population. These late toxicity occurrences were compared to doses of OARs defined with MVision AI based autosegmentation. The associations were studied using logistic regression models, using both univariate and multivariate
analysis. Results:
In the univariate setting dysphagia was associated with radiotherapy dose in oral area, pharyngeal constrictor muscle (PCM) and specifically its upper part V20Gy (p<0.05). Among clinical patient characteristics strongest associations were T classification (p=0.04), neck RT (p=0.03) and the primary tumor in the oral cavity (p=0.07). In the multivariate analysis, neck RT and T classification were most important factors associated with dysphagia among the clinical markers (p<0.07). This association was even strengthened when dose parameters of oral area OARs, as buccal mucosa V50Gy, were included in the model.Based on univariate analysis hypothyroidism was strongest associated with thyroid mean dose (p=0.002). Among the clinical markers sex (p=0.0001), T classification (p=0.01), age (p=0.02) and chemotherapy (p=0.02) gave strongest associations for hypothyroidism. In multivariate analysis hypothyroidism was associated with thyroid mean dose, female sex and chemoRT (p<0.05).For osteoradionecrosis, the strongest univariate associations were the mean dose and the V50Gy-bin of mandible and especially molar and frontal areas (p<0.05). The strongest clinical marker was primary tumor in the oral cavity (p=0.003). Conclusion: The multivariate analysis suggests that certain OAR
Made with FlippingBook - Share PDF online