S1953
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
References: [1] Sher, D.J., et al., Stereotactic Ablative Radiotherapy for T1 to T2 Glottic Larynx Cancer: Mature Results From the Phase 2 GLoTtic Larynx-SABR Trial. Int J Radiat Oncol Biol Phys, 2025. 121(1): p. 137-144. Keywords: SABR, early-stage glottic cancer, MR-Linac
Digital Poster 3809
Accuracy of dose calculation based on Dual-Energy CT images for radiotherapy treatment planning Anne Richter, Chiara Jegelka, Andrea Wittig-Sauerwein, Gary Razinskas Department of Radiotherapy and Radiation Oncology, University Hospital Würzburg, Würzburg, Germany Purpose/Objective: Accurate dose calculation in radiotherapy critically depends on precise mapping between CT numbers and electron densities. With the increasing use of Dual-Energy CT (DECT) in clinical workflows, it is essential to evaluate how its image reconstructions— particularly Virtual Monoenergetic Images (VMI)— affect dose calculation accuracy. This study aims to quantify the impact of different VMI energy levels on the accuracy of calculated dose distributions. Material/Methods: A retrospective analysis was conducted for two patients who underwent stereotactic radiotherapy for brain metastases. Imaging for treatment planning and dose calculation was performed using a DECT CT scanner (Somatom go.OpenPro, Siemens Healthineers, Forchheim, Germany). VMI datasets were reconstructed at energy levels ranging from 50 to 180 keV in 10 keV increments. The reference dose distribution was calculated on a conventional Single- Energy CT (SECT) at 120 kV using the standard clinical calibration curve. To simulate a potential mismatch between calibration and image datasets, this same SECT calibration curve was applied to each VMI dataset. All treatment plans were generated with 6 MV flattening-filter-free photons using a 3D conformal beam arrangement. The resulting dose distributions were compared to the SECT reference using dose volume histogram (DVH) parameters within the planning target volume (PTV). Results: Across the VMI spectrum, the PTV dose metrics showed high stability. The most accurate dose calculation, showing the closest agreement in DVH parameters with the reference distribution, was achieved at a VMI energy near 70 keV. For energy levels between 70 and 90 keV, deviations remained below 0.5%, except for Dmax at 90 keV, which showed a slightly higher deviation of 0.68% (Table 1). Between 60 and 150 keV, the parameters D95%, D98%, D99%
Conclusion: Modest dose differences were seen between the two techniques and between patients, but they are unlikely to be clinically significant. MR-guided SABR for early- stage glottic cancer is promising, since it allows for narrow margins and online motion management. As a result of these outcomes, a clinical trial is started on SABR for early-stage glottic cancer using MR-Linac at UMC Utrecht.
Made with FlippingBook - Share PDF online