S2020
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Digital Poster 5029
planning system (TPS) in predicting superficial doses at 0.7 mm depth. Material/Methods: This prospective study combined simulations on anthropomorphic phantoms with in vivo measurements in patients treated with VMAT using Flattening Filter Free (FFF) beams on a HALCYON accelerator. Preliminary tests were performed on phantoms to validate the reliability of the measurement setup. Surface dosimetry was then assessed using Gafchromic EBT3 films placed on head- and-neck cancer patients and in vivo measurements of superficial dose were made. This study was conducted from January 15, 2025, to April 15, 2025. Results: Our study enrolled six patients treated for Head and Neck cancer. Preliminary tests showed excellent agreement between measured and planned doses at the beam exit, with a negligible deviation of less than 3%. However, there was a moderate deviation of 5 to 7% at the center, as well as a slight underestimation at the beam entrance. For the patients included, comparisons between treatment planning system (TPS) calculations and in vivo measurements showed average dose differences ranging from 10% to 69%. Differences in maximum doses were generally smaller, ranging from 3% to 43%. The overall mean deviation was 41% for mean doses and 23% for maximum doses, indicating greater variability in the prediction of mean doses. Some patients exhibited negative deviations (up to − 10% on the right side), suggesting TPS underestimation in these cases. Additionally, notable asymmetries were observed between the left and right sides in some patients (63% on the left versus 56% on the right). Conclusion: Although preliminary phantom tests confirmed the reliability of the measurement method, in vivo results revealed substantial variability, emphasizing the limitations of the TPS. Our findings highlight the need for cautious interpretation of superficial dose predictions and encourage further investigations to optimize surface dosimetry accuracy in clinical practice. References: 1) Ma C, Chen M, Long T, Parsons D, Gu X, Jiang S, et al. Flattening filter free in intensity-modulated radiotherapy (IMRT) - Theoretical modeling with delivery efficiency analysis. Med Phys. janv 2019;46(1):34 - 44. Keywords: flattening filter free, Radiotherapy
Development of a parameter for the preselection of patients with breast cancer for treatment with the FAST-Forward protocol using a hybrid technique. Karolina Sanocka Medical physics, Greater Poland Cancer Centre in Pozna ń , Pozna ń , Poland Purpose/Objective: In recent years, there has been a growing use of shortened breast radiotherapy schedules, such as the FAST-Forward protocol.The aim of this study was to develop a simple method to enable a preliminary assessment of patient eligibility for this schedule, using a hybrid technique (3DCRT+VMAT). Material/Methods: A cohort of postoperative patients with left-sided breast cancer was analyzed. Using computed tomography (CT) images and treatment plans, dose distributions in the left lung and heart were assessed. Particular attention was paid to V8Gy, defined as the percentage of the left-lung volume receiving at least 8 Gy ( ≤ 15% optimal; ≤ 20% maximally acceptable). To simplify assessment of lung exposure, a geometric model based on a hemi-ellipsoid approximation was constructed. On this basis, a proprietary index-LuLI (Lung-Load Index) was developed, defined as the ratio of the hemi-ellipsoid volume to the total volume of the left lung, and its correlation with V8Gy was evaluated. This was a single-institution retrospective analysis; DIBH was delivered according to institutional standards. The primary endpoint was the association between LuLI and V8Gy, and secondary endpoints were simple cut-points directing patients to the FAST- Forward schedule, to conventional 15 fractions, or to full dosimetric review. Linear regression and correlation analyses were applied. Internal validation was performed on a hold-out set to assess generalizability and practical decision thresholds. Results:
The strongest correlation with V8Gy was observed for the 70% isodose line (R ² = 0.58), which was selected for subsequent analyses. Using the developed geometric model, the LuLI parameter was retrospectively derived
Made with FlippingBook - Share PDF online