ESTRO 2025 - Abstract Book

S3830

Physics - Radiomics, functional and biological imaging and outcome prediction

ESTRO 2025

Conclusion: In our dataset, elevated perfusion was either absent or confined to areas within the conventional GTV. Integrating ASL-based perfusion MRI into glioblastoma target delineation did not change the GTV volume, indicating that its role in target delineation requires further investigation. Keywords: Glioblastoma, perfusion, MRI References: 1. Price, S.J., et al., Improved Delineation of Glioma Margins and Regions of Infiltration with the Use of Diffusion Tensor Imaging: An Image-Guided Biopsy Study. American Journal of Neuroradiology, 2006. 27 (9): p. 1969. 2. Seker-Polat, F., et al., Tumor Cell Infiltration into the Brain in Glioblastoma: From Mechanisms to Clinical Perspectives. Cancers (Basel), 2022. 14 (2). 3. Niyazi, M., et al., ESTRO-EANO guideline on target delineation and radiotherapy details for glioblastoma. Radiother Oncol, 2023. 184 : p. 109663. Proffered Paper Estimating tumor hypoxia from patient-specific microvascular measures: towards microenvironment digital twins in radiotherapy Luca Possenti 1 , Sophie Materne 1,2 , Francesco Pisani 1 , Piermario Vitullo 2 , Alessandro Cicchetti 1 , Nicola Alessandro Iacovelli 3 , Marzia Franceschini 3 , Anna Cavallo 4 , Paolo Zunino 2 , Tiziana Rancati 1 1 Data Science Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy. 2 MOX, Department of Mathematics, Politecnico di Milano, Milan, Italy. 3 Radiotherapy Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy. 4 Medical Physics Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy Purpose/Objective: The microvascular network is an essential player in the microenvironment, shaping oxygen delivery and impacting radiotherapy (RT) outcomes. This work aims to build a patient-specific mechanistic computational model to simulate microvasculature in head-and-neck cancer (HNC) patients. The model is then used to describe single patients by leveraging sublingual microscopy data, gaining personalized estimation of the hypoxia within the microenvironment. Material/Methods: Data from 63 patients were collected using the non-invasive GlycoCheck™ system to analyze sublingual microvasculature. We built 2D synthetic microvascular networks generated using Voronoi-based models [1] and calibrated against patient data. We simulated blood flow considering the red blood cells transport dynamics and vessel diameter using a modified Poiseuille's equation matching blood velocity data. We further developed personalized anatomically accurate models of the microenvironment to simulate oxygen diffusion and uptake [2]. Such a microenvironment oxygen map is then translated into a surviving fraction map by an oxygen-modified Linear Quadratic (LQ) model to evaluate RT outcomes [3]. Results: We built synthetic microvascular networks tailored to nine HNC patients based on their capillary densities (CD) and diameter distribution (Figure 1). 3508