S1881
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
configurations. Statistical analysis revealed a significant correlation between the mean dose to certain OOIs and the degree of PTV–OOI overlap (Figure 2). The strongest correlations were found for the combined superior and middle constrictor muscles (CMP Sup+Mid), 0.91 (p < 0.05, CI [0.83–0.98]), and the submandibular gland, 0.77 (p < 0.05, CI [0.52–0.91]). Weak or no correlation was observed between dose and centroid distance.
Digital Poster 2324
Advances in head and neck radiotherapy: updated constraints for automated planning and dose– anatomy correlation analysis Filippo Susini 1 , Livia Marrazzo 1,2 , Chiara Arilli 2 , Pierluigi Bonomo 3 , Margherita Zani 2 , Slivia Calusi 2 , Marta Casati 2 , Ilaria Cupparo 2 , Cinzia Talamonti 1,2 , Lorenzo Livi 1,3 , Stefania Pallotta 1,2 1 Department of Experimental and Clinical Biomedical Sciences “Mario Serio”, University of Florence, Florence, Italy. 2 Medical Physic Unit, Careggi University Hospital, Florence, Italy. 3 Radiation Oncology Unit, Careggi University Hospital, Florence, Italy Purpose/Objective: Radiotherapy planning for head and neck (H&N) tumors is one of the most challenging tasks in clinical practice due to the high density of organs of interest (OOIs) and the large variability of target volumes. Starting from the redefinition of dose constraints for high-dose (PTV-HD), low-dose (PTV-LD) PTVs, and OOIs, the aim of this work was to develop a robust template for automated planning using the Monaco TPS, capable of achieving consistent dosimetric results across heterogeneous anatomical configurations. We also aimed to analyze the correlation between OOI dose and anatomical indicators. Material/Methods: The dosimetric constraints in our internal protocol for PTV-LDs and OOIs were updated according to the latest ASTRO 2024 guidelines [1], transitioning from general to case-specific criteria for unilateral and bilateral treatments and relaxing the PTV-LD requirement (V95% > 90%) to improve OOI sparing. These constraints were implemented in two planning templates, iteratively refined for bilateral and unilateral tumor configurations using five patients for each category. Dose distributions were then generated fully automatically from these patient-derived templates on cohorts of 23 and 12 patients, respectively. Correlations were evaluated between the average OOI dose and both the PTV–OOI overlap volume and the distance between the PTV-HD and OOI centroids. Data distribution was verified using the Wilcoxon test, and correlations were calculated using Pearson or Spearman tests accordingly. Each correlation was associated with a confidence interval (CI) estimated through bootstrapping with 10,000 repetitions. Results: The optimized templates demonstrated adequate PTV coverage in line with requirements while remaining close to the ideal OOI constraints (Figure 1), thus enabling fully automated plan generation. Relaxation of the PTV-LD requirement allowed most constraints to be met, even in challenging anatomical
Conclusion: The developed templates proved robust, laying the groundwork for standardization and automation of H&N planning. Statistical analysis showed that the spatial relationship of PTVs, particularly PTV-HDs, with overlapping OOIs affects optimization performance. The strong correlation between mean dose and overlap volume for CMP Sup+Mid enables prediction of the achievable mean dose for this structure in new cases. References: 1. Margalit DN, Anker CJ, Aristophanous M, Awan M, Bajaj GK, Bradfield L, et al. Radiation Therapy for HPV- Positive Oropharyngeal Squamous Cell Carcinoma: An ASTRO Clinical Practice Guideline. Practical Radiation Oncology. 2024; 14(Suppl):1–28. doi:10.1016/j.prro.2024.05.007 Keywords: head and neck, automated planning Digital Poster 2327 AutoRAD: Fully automated RapidArc Dynamic planning Esben S Worm, Anne IS Holm Oncology, Aarhus University Hospital, Aarhus, Denmark Purpose/Objective: Varian’s RapidArc Dynamic (RAD) technique is a recent VMAT-extension that incorporates variable collimator
Made with FlippingBook - Share PDF online