S2996
Invited Speaker
ESTRO 2026
positioning in radiotherapy, but its integration into the treatment planning system (TPS) for dose calculation is becoming increasingly relevant with the rise of adaptive radiotherapy. This presentation will provide an overview of the challenges associated with CBCT- based dose calculation, as well as recent hardware and software developments enabling its use. Despite its widespread availability, CBCT presents several limitations for accurate dose calculation. These include inaccurate CT numbers due to scatter and beam hardening, limited field-of-view, increased image noise, and motion artifacts, particularly in thoracic and abdominal regions. These factors affect electron density estimation and may lead to clinically relevant dose errors. Recent advances in CBCT technology have introduced new hardware and software solutions to improve image quality and quantitative accuracy. Results from recent studies will be presented, focusing on the impact of these developments on image quality and dose calculation accuracy. This includes the evaluation of software solutions previously only available on CT systems, such as metal artifact reduction and extended field-of-view reconstruction, as well as improvements in CT number accuracy. Initial validation was performed using phantoms with known ground truth. The implications of CBCT calibration strategies for accurate dose calculation will be discussed, alongside results obtained in both phantom and clinical settings. Furthermore, the translation to clinical practice will be addressed, including the impact of motion on CBCT accuracy, supported by motion phantom studies. Practical considerations will be highlighted, including scenarios where caution is required. Finally, future directions will be explored, including broader clinical applications of accurate daily CBCT imaging. 5306 Magnetic resonance imaging for treatment planning, where are we on synthetic CT implementation and what comes next? Fernanda Villegas Radiotherapy Physics and Engineering, Karolinska University Hospital, Stockholm, Sweden Despite the overwhelming amount of magnetic resonance (MR) based synthetic computed tomography (sCT) algorithms found today in the literature, the clinical implementation of MR-only radiotherapy workflows has progressed at a relatively slow pace. The lack of international consensus on quality assurance (QA) for sCT may still be an important reason but it is not the only one. This talk aims to discuss key elements that lead to a successful implementation of MR-only workflows while
remain important for risk assessment, but they do not constitute true predictive biomarkers for perioperative immunotherapy benefit. This question has become particularly timely with the introduction of perioperative anti–PD-1 therapy into the treatment landscape of resectable disease. As immunotherapy moves from experimental investigation into clinical implementation, the central challenge is no longer whether it has a role, but how to identify those patients most likely to benefit from it. In this context, it is essential to distinguish clearly between prognostic biomarkers, which are associated with outcome irrespective of treatment, and predictive biomarkers, which identify differential benefit from a specific therapeutic intervention. This presentation will examine the current state of biomarker development for perioperative immunotherapy in resectable HNSCC. PD-L1 has emerged as the first clinically relevant biomarker in this setting, but its utility remains limited by spatial heterogeneity, temporal variability, assay dependence, and its inability to capture the full complexity of the antitumor immune response. Increasing attention is therefore being directed toward more informative and biologically integrated approaches. These include pathologic response after neoadjuvant immunotherapy, which may provide an early treatment-specific readout, as well as multidimensional characterization of the tumor microenvironment, including T-cell inflamed phenotypes, tumor-infiltrating lymphocytes, tertiary lymphoid structures, myeloid and stromal composition, and spatial immune organization. HPV status may contribute to biologic stratification, but is unlikely to function as a standalone predictive marker. Serial liquid biopsy approaches, particularly ctDNA- based monitoring, may further enhance response prediction and treatment adaptation across the perioperative course. Overall, this presentation will argue that while the one- size-fits-all paradigm is beginning to give way, the future is unlikely to rest on a single biomarker. Instead, progress will depend on composite, dynamic, and clinically actionable models capable of supporting more personalized perioperative treatment strategies. 5304 CBCT in the TPS loop: Can we use it for dose calculation? Didier Lustermans, Lars Daenen, Marta Bogowicz, Frank Verhaegen, Gabriel Paiva Fonseca Department of Radiation Oncology (Maastro), GROW Research Institute for Oncology and Reproduction, Maastricht University, Maastricht, Netherlands
Cone-beam CT (CBCT) is routinely used for patient
Made with FlippingBook - Share PDF online