S2983
Invited Speaker
ESTRO 2026
tumor progression is described via the transition matrix of the HMM, which is in turn parameterized via a directed graph representing the lymphatic drainage. The model parameters learned from the data are the probabilities of the tumor to spread to a LNL and between LNLs. The model was used to design a clinical trial on personalized volume-deescalated elective nodal irradiation in oropharyngeal SCC (DeEscO [2], clinicaltrials.gov). The primary endpoint is the rate of N-site recurrences in unirradiated LNLs. The CTV-N is defined such that the estimated risk of occult metastases in all unirradiated LNLs combined is <10%. Main features are: a) ipsilateral level IV is not irradiated if level III is clinically negative, b) contralateral levels III/IV are not electively irradiated unless the upstream levels II/III are involved, c) unilateral irradiation is performed in patients with lateralized tumors and clinically negative contralateral neck, d) levels I and V are irradiated in fewer patients compared to current guidelines. References: [1] https://doi.org/10.1038/s41598-025-99978-7 [2] https://doi.org/10.1101/2025.06.24.25330129 5264 Clinical rationale and needs Esther Troost Image-Guided High-Precision Radiotherapy’, Dresden University of Technology, Dresden, Germany. Department of Radiotherapy and Radiation Oncology, University Hospital Carl Gustav Carus, Dresden, Germany It is needles to emphasize that the goal of radiotherapy is to achieve uncomplicated local tumor control. For this, a trade-off between tumor control probability (TCP) and normal tissue complication probability (NTCP) has to be made. The current radiation dose prescription is based on results from large randomized controlled trials, treating entire patient cohorts with identical radiation doses, notwithstanding the individual patients’ tumor characteristics. On the one hand, in recent years, it has become apparent that HPV+ oropharyngeal tumors, e.g., may be sufficiently treated with lower radiation doses than tumors in the same anatomical subsite being the consequent of nicotine and alcohol abuse. On the other hand, tumors may be relatively radiation resistant and the current one-size-fits-all approach not sufficient to reach a high TCP in many patients. Therefore, biologically-guided radiotherapy is a means to incorporate the tumor characteristics, such as hypoxia, proliferation, stem cell number, into the definition of the biological target volume [BTV; Ling,
measures of tumour cellularity and imaging approaches to hypoxia, will be discussed. The main translational challenges will be addressed, including the need for robust technical validation, reproducibility across scanners and centres, and biological validation linking imaging metrics to underlying tumour characteristics and clinical outcomes. These limitations contribute to the current gap between biomarker development and clinical implementation. Understanding and addressing these challenges is essential for enabling biomarkers to move from exploratory research tools to actionable components of biologically guided RT. 5263 Personalization of the CTV Jan Unkelbach, Roman Ludwig, Yoel Pérez Haas, Noemi Bührer, Esmée Looman, Panagiotis Balermpas Radiation Oncology, University Hospital Zurich, Zurich, Switzerland Definition of the gross tumor volume (GTV) has improved through advances in biomedical imaging. In contrast, the clinical target volume (CTV) appears normal in medical images. Therefore, CTV definition goes beyond image segmentation and analysis, making personalization of the CTV challenging. Radiation oncology distinguishes two parts of the CTV, the primary tumor CTV-T and the nodal CTV-N. The primary tumor CTV-T is usually a margin expansion around the primary tumor GTV-T. The GTV-to-CTV margin varies between tumor sites but is usually not individualized for a given patient. The nodal CTV-N contains parts of the lymph drainage system at risk of harboring occult lymph node metastases. In current practice, the CTV-N often leads to extensive bilateral irradiation of the regional lymphatics with only limited personalization. We conducted research on personalized estimation of the risk of occult lymph node metastases for Head- and-Neck squamous cell carcinoma (HNSCC). We built a large multi-institutional database of 2885 HNSCC patients to quantify patterns of lymph node involvement per lymph node level (LNL) depending on primary tumor characteristics. The web-based platform LyProX.org was developed to make the data publicly available and allow interactive exploration. Based on the data, a statistical model was developed to estimate the individual patient's probability of occult metastases in clinically negative LNLs [1]. The statistical model is based on Hidden Markov Models (HMM). Each LNL is described by a random variable that indicates the true involvement of the LNL including occult metastases. Lymphatic metastatic
Made with FlippingBook - Share PDF online