ESTRO 2025 - Abstract Book

S1094

Clinical – Head & neck

ESTRO 2025

3691

Digital Poster Early Results Of RadiomicART Trial: A Prospective Monocenter Study Of Multi-Imaging Adaptive Radiotherapy For Locally Advanced Head-Neck Cancer Mariya Boyanova Ilieva 1,2 , Ciro Franzese 1,2 , Maria Ausilia Teriaca 1 , Manuele Roghi 3 , Marco Badalamenti 1 , Luciana Di Cristina 1,2 , Damiano Dei 1 , Anna Bertolini 1,2 , Beatrice Marini 1,2 , Giacomo Reggiori 1 , Pietro Mancosu 1 , Stefano Tomatis 1 , Marta Scorsetti 1,2 1 Radiotherapy and Radiosurgery, Humanitas Clinical and reasearch center-IRCCS, Rozzano (MI), Italy. 2 Biomedical Sciences, Humanitas University, Pieve Emanuele (MI), Italy. 3 Radioterapia Oncologica, Centro di Riferimento Oncologico (CRO), Aviano (PN), Italy Purpose/Objective: Patients with head and neck cancer (HNC) may develop significant anatomical changes during radiotherapy (RT). Adequate dose coverage and sparing of organs at risk can be guaranteed by re-planning with adaptive radiotherapy (ART). Advanced imaging modalities play a role in the customization of the RT as shown the use of ART and radiomics. The RadiomicART trial evaluates a machine learning-based radiomic approach to predict outcome and toxicity of HNC patients treated with ART by CT, MRI and PET-scan. Material/Methods: Patients with locally advanced HNC treated with radical RT from October 2021 to September 2023 were analyzed. Target volumes were delineated on CT and adjusted on MRI and FDG-PET-scan. The total dose was 66/60/54 Gy in 30 fractions with VMAT-SIB. For the whole sample, 95% of the prescribed dose covered at least 95% of the PTV. At week 3 after the start of RT, CT simulation, MRI and FDG-PET-scan were repeated for re-planning. In week 4, the new plan began. We performed the comparison between the original plan and the re-simulation (adaptive plan) through CT, MRI and FDG-PET showing the variation of PTVs and collected early data on toxicity during follow-up. Results: 47 patients were enrolled and completed the scheduled radiotherapy treatment. Median GTV-T and GTV-N at baseline and replanning were 22.85 cc (2.6-118.7) vs 20.45 cc (2.2-87.5), and 6.3 cc (0-215.9) vs 4.35 cc (0-174.5), respectively. Mean GTV-P were 29.61 cc vs 26.48 cc, and mean GTV-N 19.78 cc vs 14.11 cc, at baseline and re simulation. Median right parotid volumes at baseline and replanning were 29.85 cc (11.9-70.3) and 27.2 cc (9.7-68.2) and left parotid were 31.15 cc (14.7-71.8) and 28.55 cc (12.9-68.8), respectively. Besides, median right submandibular glands volumes at baseline and replanning were 9.3 cc (5.1-29.5) vs 8.45 cc (4.3-12.4), and left gland were 9.1 cc (0.9-15.1) vs 7.5 cc (4.1-13.3). At the time of analysis, the most common grade 3 adverse events during follow up were xerostomia (3%) and weight loss (7%). Grade 2 xerostomia at 3, 6 and 12 months follow up was respectively in 20%, 3%, and 0% of treated patients.