S627
Clinical – Head & neck
ESTRO 2026
learning-derived online prediction models of outcomes for patients with cholelithiasis-induced acute cholangitis: development and validation in two retrospective cohorts. EClinicalMedicine 2024;76:102820. https://doi.org/10.1016/j.eclinm.2024.102820. Keywords: NPC, Distant metastasis, Risk stratification Digital Poster 3580 Patient-reported fatigue in head and neck cancer survivors after radiotherapy Valentin Magnus 1 , Sebastian Schäfer 1 , David Unger 1 , Anna-Maria Tews 1 , Anna Boide 1 , Christos Moustakis 1 , Clemens Seidel 1 , Nils H Nicolay 1 , Alexander Rühle 1,2 1 Department of Radiation Oncology, University Medical Center Leipzig, Leipzig, Germany. 2 Department of Radiation Oncology, Princess Margaret Cancer Centre, University of Toronto, Toronto, Canada Purpose/Objective: To determine the prevalence and severity of fatigue in head and neck cancer survivors and to identify factors associated with its occurrence, including potential associations with radiation dose in cerebral organs at risk (OAR). Material/Methods: This cross-sectional observational study was conducted at a tertiary cancer center. Eligible participants were survivors who had received curative (chemo)radiotherapy for head and neck cancer between January 1, 2007 and December 31, 2023. Fatigue and quality of life (QoL) were assessed using the EORTC QLQ-C30 questionnaire, while psychosocial distress was measured with the NCCN Distress Thermometer. Anxiety and depression were quantified by the GAD-7 and PHQ-9 scales. Dosimetric analyses were conducted to explore associations between fatigue and radiation dose to cerebral OAR (medulla oblongata, brainstem, cerebellum, posterior fossa, pituitary gland, basal ganglia, hippocampi), including voxel-based mapping and multivariable modeling. Correlations between fatigue and sarcopenia were examined using deep-learning-based segmentation of skeletal muscle at the C3 level to derive the Skeletal Muscle Index (SMI). Results: A total of 158 patients were eligible for the analysis. The median age was 60 years (IQR, 50.5–67), and the most common tumor sites were the oropharynx (n=61; 38.6%), oral cavity (n=45; 28.5%), and larynx (n=22; 13.9%). Eighty-eight patients (55.7%) received concurrent systemic therapy. The median interval between completion of radiotherapy and study participation was 41 months (IQR, 22–61). The mean fatigue score was 44.7 points, compared to 29.7 points
primary endpoint was distant metastasis-free survival (DMFS). Five machine learning survival models were trained to estimate the 5-year risk of DM, and patients were stratified into high- and low-risk groups. A nomogram was then constructed and interpretability was achieved by the Shapley Additive exPlanations (SHAP) method. Results:
The median age of the cohort was 48 years 48 years (IQR, 40–56), with 1,930 (27.4%) female and 5,115 (72.6%) male patients. DM occurred in 568 patients (8%). The model achieved a C-index of 0.810 on the test set, effectively distinguishing high- and low-risk patients (log-rank test; P < 0.0001). A nomogram was developed to assess DM risk and SHAP analysis indicated that main risk factors included T stage, N stage, TNM stage, sex, prior to initiation of radiotherapy (pre-RT) and within one week after the completion of RT (post-RT) EBV-DNA levels, induction chemotherapy, adjuvant chemotherapy, pre-RT platelet-to-lymphocyte ratio, pre-RT neutrophil-to- lymphocyte ratio, and pre-RT neutrophil. Conclusion: This study presents a novel risk stratification model to accurately predicts the risk of DM in NPC patients. The interpretable nomogram is instrumental in facilitating the clinical application and paving the way for more personalized treatment strategies for NPC patients. References: [1] Chen Y-P, Chan ATC, Le Q-T, Blanchard P, Sun Y, Ma J. Nasopharyngeal carcinoma. Lancet 2019;394:64– 80. https://doi.org/10.1016/S0140-6736(19)30956- 0.[2] Yesilyaprak A, Kumar AK, Agrawal A, Furqan MM, Verma BR, Syed AB, et al. Predicting Long-Term Clinical Outcomes of Patients With Recurrent Pericarditis. J Am Coll Cardiol 2024;84:1193–204. https://doi.org/10.1016/j.jacc.2024.05.072.[3] Huang S, Zhou Y, Liang Y, Ye S, Zhu A, Li J, et al. Machine-
Made with FlippingBook - Share PDF online