Main Session
Sep
28
PQA 04 - Breast Cancer, Patient Reported Outcomes/QoL/Survivorship, Functional Radiation Medicine, Hematologic Malignancies, Palliative Care, and International/Global Oncology
2717 - Skin Dose Assessment in Prone Accelerated Partial Breast Irradiation Using a Personalized 3D-Printed Immobilization Device
Presenter(s)
Daniel Zwahlen, MD, MBA - Department of Radiation Oncology, Winterthur, Zurich
S. Chouaf1,2, M. Jaccard1, O. Fargier-Bochaton1, O. Matzinger1,3, A. Clivio4, M. M. Aspradakis4, D. R. Zwahlen4, and G. Dipasquale2; 1Department of Radiation Oncology, Clinique de Genolier, Genolier, Switzerland, 2HeroSupport SA, Geneva, Switzerland, 3ETH Zurich, Zurich, Switzerland, 4Department of Radiation Oncology, Kantonsspital Winterthur, Winterthur, Switzerland
Purpose/Objective(s):
Accurate prediction of skin toxicity remains challenging in prone breast radiotherapy due to uncertainties in the build-up region and potential dose perturbations introduced by immobilization devices. This study, based on the APBI-IMRT-Florence clinical protocol, is designed to evaluate skin dose in prone TomoDirect™ treatments using a 3D-printed personalized immobilization system.Materials/Methods:
A controlled treatment planning study was performed using prone breast CT images of a commercial anthropomorphic phantom (Rando) with and without the shell. Five clinically relevant targets were delineated following APBI-IMRT-Florence criteria (GTV = 2.5 cm) in each breast quadrant on both CT datasets near the skin. A hypo-fractionated schedule of 30 Gy in 5 fractions was prescribed. PTVs were generated using a 5 mm isotropic margin and cropped 5 mm from the external body contour. Skin structures were defined as a 5 mm inner ring extending 5 cm around the breast region. Ten treatment plans were generated for a Radixact system using TomoDirect (6FFF) with three optimization runs on RayStation v3.5 (Collapsed-Cone v5.9) and normalized to prescription dose. DVH parameters including PTV_eval_V95%, Conformity Index (CI), Dmax, Skin_V29Gy, Ipsilateral Lung V10Gy and Heart Dmean were extracted and compared.Results:
Implementation of the shell maintained robust PTV coverage (mean ?PTV_eval_V95% = 0.45±0.59%; mean ?CI = -0.05±0.03). No statistically significant increase in high-dose skin exposure was observed between configurations (mean ?Skin_V29Gy = +1.52±1.64 cc; p = 0.18), although a small-to-moderate effect size was noted (r = 0.28). No clinically relevant impact was observed for ipsilateral lung dose (mean ?V10Gy = 0.00±0.04%; range -0.06% to +0.05%) and cardiac dose variation was minimal (?Heart Dmean = 0.00±0.01 Gy; range 0.000–0.014 Gy). All values remained well below Florence-APBI constraints (Lung V10 = 20%, Heart Dmean = 2.5 Gy).Conclusion:
Across all cases,the shell did not adversely affect skin dose, cardiopulmonary sparing, or PTV coverage. Build-up perturbations remained within clinically acceptable limits, confirming its dosimetric safety. Given its high positioning reproducibility and target stability, PTV margin reduction may be feasible, potentially decreasing OAR dose and treatment-related toxicity.