ESTRO 2025 - Abstract Book

S1045

Clinical – Head & neck

ESTRO 2025

The cumulative delivered dose was 70 Gy in conventional fractionation. The low sample size prevented any statistical tests, however, the dose received from the contralateral parotid gland and the contralateral SCRR was higher in G2 xerostomia pts than in G0 ones. Conclusion: The present work confirms a trend between the dose received from either the contralateral parotid gland or SCRR and the onset of chronic xerostomia. Therefore, the future full data collection and consequent statistical analysis could better define the role of SCRR in developing long-term xerostomia.

Keywords: xerostomia, stem cells, oropharyngeal cancer

2791

Digital Poster Do pretreatment CT and PET radiomics predict individual lymph node failure after (chemo)radiotherapy in head and neck cancer patients? Yan Li 1 , Inge Wegner 2 , Makbule Atasoyu 1 , Tiantian Zhai 3 , Roel Steenbakkers 1 , Johannes A. Langendijk 1 , Nanna M. Sijtsema 1 , Lisanne V. van Dijk 1 1 Department of Radiation Oncology, University Medical Centre Groningen, University of Groningen, Groningen, Netherlands. 2 Department of Otorhinolaryngology, Head and Neck Surgery, University Medical Center Groningen, University of Groningen, Groningen, Netherlands. 3 Department of Radiation Oncology, Cancer Hospital of Shantou University Medical College, Shantou, China Purpose/Objective: Selective nodal dissection prior to (chemo)radiotherapy may reduce the incidence of failure in high-risk lymph nodes (i.e., persistent or recurrent nodal metastases) in head and neck cancer (HNC) patients. Previously, Zhai et al. developed an externally validated clinical model for predicting individual lymph node (iLN) failure, enhanced by shape and radiomics features derived from the contrast-enhanced planning CT scans [1,2]. PET images contain information about the metabolic activity which is associated with the risk of nodal failure [3]. This study aims to evaluate whether the addition of shape, CT and PET radiomics features, individually or in combination, improve the performance of clinical prediction models for iLN failure in an expanded cohort.