S1897
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Digital Poster 2755
First clinical evaluation of EZFluence automated forward planning for breast and whole-brain radiotherapy in Monaco TPS Valeria Faccenda 1,2 , Denis Panizza 1,2 , Martina Scotti 3,1 , Valeria Enrica Tremolada 1 , Riccardo Ray Colciago 2 , Stefano Arcangeli 2,4 , Elena De Ponti 1 1 Medical Physics, Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy. 2 School of Medicine and Surgery, University of Milan Bicocca, Milan, Italy. 3 Physics Department, University of Milan, Milan, Italy. 4 Radiation Oncology, Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy Purpose/Objective: Forward planning remains the standard approach for several radiotherapy sites, such as breast and whole brain (WB), due to its simplicity and robustness compared with VMAT. The field-in-field technique is commonly used to modulate fluence and improve dose homogeneity while maintaining simple beam geometry. However, manual creation of such plans is time-consuming and subject to operator variability. This study aimed to evaluate the clinical feasibility and integration of EZFluence, a novel automatic forward- planning tool, into routine clinical practice. Material/Methods: Twenty WB patients and twenty breast patients were retrospectively selected. EZFluence, implemented as a script within Monaco TPS, was executed using the same field geometry as in the clinically approved plans: two opposed lateral fields (90°/270°) for WB and two tangential fields for breast cases. Where MLC shielding of OARs was present in the clinical plan, it was preserved in the automated workflow. The fluence optimized by EZFluence was imported into Monaco and calculated using the same dose algorithm. Quantitative comparisons between the two plans included monitor units (MUs), modulation weights, target coverage, homogeneity, hotspots, OAR metrics, and planning time. Additionally, a blinded qualitative evaluation was independently performed by two medical physicists and one radiation oncologist, who classified each plan as superior, inferior, or equivalent. Results: EZFluence WB plans showed slightly higher MUs (median, 322.5 vs 312.2; P<0.001) and modulation weights (median, 6.3 vs 2.2; P<0.001), achieving a more homogeneous dose distribution with higher V95% (98.2% vs 96.7%; P<0.001) and lower V105% (0.0% vs 1.1%; P<0.001). For breast cases, automated plans reduced MUs (414.4 vs 719.0; P<0.001) and increased modulated segment weights (16.7 vs 4.8; P<0.001) due to the absence of motorized wedges. Target coverage was comparable between approaches (95.3% vs
Results: DC - DVH visualizations demonstrated a clear distance– dose relationship, with high - dose contributions predominantly arising from voxels closest to the PTV for both the rectum and the bladder. (Figure 2). At intermediate doses (e.g., 40–50 Gy), the color gradient indicated that most exposure occurred within 15 mm from the target, thus linking dose gradients with anatomical proximity that conventional DVHs cannot represent. At 70 Gy within 5 mm, mean SVAR was 7.8% ± 1.7% for the rectum and 16.6% ± 7.3% for the bladder, consistent with conventional V70Gy yet explicitly localizing exposure to the anterior rectal wall and bladder neck.
Conclusion: DC - DVH incorporates intuitive distance - to - target information into a familiar DVH, enabling immediate spatial interpretation of OAR dose, and may facilitate spatially informed plan optimization and toxicity risk assessment. This simple extension provides clear indicators of whether OAR dose predominantly arises nearer to or farther from the target. Keywords: Dose-Volume Histogram, Spatial Information
Made with FlippingBook - Share PDF online