S1822
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Digital Poster 1201 Internal and external validation of a RapidPlan– MCO workflow for automatic planning Giulia Rambaldi Guidasci 1 , Federica Murtas 1 , Tiziana Malatesta 1 , Antonella Bufacchi 1 , Roberto Capparella 1 , Annabelle Grieco 1 , Marco Fusella 2 , Antonella Fogliata 3 , Silvia Di Franco 4 , Calogero Casà 4 , Giorgio Caramia 4 , Francesco Miccichè 4 1 Centro di Eccellenza Oncologia Radioterapica, Medica e Diagnostica per Immagini - UOS Fisica Sanitaria, Ospedale Isola Tiberina - Gemelli Isola, Rome, Italy. 2 UOC Radioterapia, Policlinico Abano Terme, Padua, Italy. 3 Radiotherapy and Radiosurgery, Humanitas Research Hospital IRCCS, Milan-Rozzano, Italy. 4 Centro di Eccellenza Oncologia Radioterapica, Medica e Diagnostica per Immagini - UOC Radioterapia, Ospedale Isola Tiberina - Gemelli Isola, Rome, Italy Purpose/Objective: To develop and validate an automatic VMAT planning workflow for prostate cancer that combines commercial knowledge-based planning system (RapidPlan) and multicriteria optimisation (MCO) approach, and assessing plan quality efficiency, inter- planner variability and transferability through internal and external validation Material/Methods: Eighty-nine prostate cancer patients treated with VMAT on Truebeam were included. Their clinical plans were used to train an initial RapidPlan model. This model generated new plans for all patients, which were further optimised with MCO.1 These optimised plans were then used to train a second model: RP_MCO.The final RP_MCO model was tested on ten internal patients and compared with manual plans. Plan quality was evaluated in terms of PTV coverage and homogeneity, dose to organs of interest (rectum, bladder, femoral heads) according to institutional dose guidance, monitor units (MUs) and patient-specific QA (2D gamma passing rate). Planning efficiency was measured as total optimisation time. Inter-planner variability was assesded by comparing automatic plans with manual plans from planners with different experience levels.External validation on plan quality performance of the model included ten patients from another centre, treated with VMAT on TrueBeam and Halcyon linacs. RP_MCO-based plans were generated and compared with local clinical reference plans. Results: In the internal validation all automatic plans were clinically acceptable. PTV coverage and homogeneity were comparable to clinical plans (Fig.1). RP_MCO plans showed a statistically significant reduction in rectum V40 and mean dose (p<0.05), and a slight improvement for the bladder, with a minor increase in femoral heads dose. In a blind evaluation, RP_MCO
Digital Poster 1175
Dosimetric impact of 3D-printed dental stents on salivary gland sparing in head and neck cancer radiotherapy: A comparative planning study Meysam Ghanbari Saray 1 , Kourosh Rezazadeh Eskandani 1 , Sana Abdollahnejad 2 , Sevil Akbarpour 1 , Emad Keyvani 1 , Erfan Salimi 1 , Elmira Aliyeva 3 1 Scientific Research Center, Azerbaijan Medical University, Baku, Azerbaijan. 2 Faculty of Dentistry, Istanbul Medipol University, Istanbul, Turkey. 3 department of orthodontics, Azerbaijan Medical University, Baku, Azerbaijan Purpose/Objective: Radiotherapy for head and neck cancer (HNC) often causes xerostomia due to salivary gland irradiation, severely impacting quality of life. While dental stents can displace tissues, their dosimetric impact hasn't been fully quantified. This study evaluates the potential of patient-specific 3D-printed dental stents to reduce radiation dose to salivary glands in HNC patients. Material/Methods: We conducted a comparative treatment planning study on 15 HNC patients. For each patient, two Volumetric Modulated Arc Therapy (VMAT) plans were created: a standard plan and an optimized plan utilizing a 3D-printed dental stent. Dose-volume histograms were compared for parotid glands, submandibular glands, and oral cavity. Primary endpoints were mean dose (Dmean) to parotids and Dmean to oral cavity. Statistical significance was assessed using a paired t-test (p < 0.05). Results: The use of the 3D-printed stent led to a statistically significant reduction in mean radiation dose. The Dmean for the contralateral parotid gland decreased from 32.4 ± 4.1 Gy to 26.1 ± 3.8 Gy (p = 0.003). The Dmean for the ipsilateral parotid showed a reduction from 40.2 ± 5.6 Gy to 35.9 ± 5.1 Gy (p = 0.012). The Dmean to the oral cavity was reduced from 35.8 ± 6.2 Gy to 28.3 ± 5.5 Gy (p = 0.001). Planning target volume (PTV) coverage was maintained in all plans (V95% > 98%). Conclusion: This planning study demonstrates that patient-specific 3D-printed dental stents can significantly reduce radiation dose to salivary glands and the oral cavity in HNC patients, potentially mitigating xerostomia. These findings support the integration of dental expertise into radiotherapy planning. References: Nutting et al. Clin Oncol (R Coll Radiol). 2020. Van der Laan et al. Radiother Oncol. 2023. Keywords: 3D-Printing, Dosimetry, Head and Neck Cancer
Made with FlippingBook - Share PDF online