ESTRO 2025 - Abstract Book

S2856

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2025

Kelly Kisling 1 , Xenia Ray 1 , Grace Gwe-Ya Kim 1 , Ben Archibald-Heeren 2 1 Radiation Medicine and Applied Sciences, University of California, San Diego, San Diego, USA. 2 Department of Radiation Oncology, ICON Cancer Centres, Sydney, Australia Purpose/Objective: To evaluate plan quality for prostate SBRT using a new commercial solution, RapidArc Dynamic - RAD (Varian Medical Systems, Palo Alto, USA), which combines VMAT with modulated ports and dynamic collimator rotation; and to determine the impact of the number of static angle modulated ports (STAMPs) on plans. Material/Methods: We compared RAD to VMAT for nine patients previously treated with prostate SBRT at two clinics (US and Australia based). Three RAD plans were generated per patient using a single-arc and either 4, 5 or 7 STAMPs (4s, 5s, or 7s). These were compared to VMAT plans using 2- or 3-arcs. Other RAD planning parameters were held constant across all patients and were established during initial plan iterations. For both RAD and VMAT, 5/9 patients used RapidPlan to initialize optimization and then pushed for optimal plan quality. The remainder were initialized using a template. We assessed plan quality using 25 dose metrics harmonized from both clinics and eviQ criteria. Delivery efficiency was assessed by comparing MUs. Results: We identified the following parameters for creating single-arc RAD prostate SBRT plans during initial testing: iterations=1000; STAMP weighting=-1 (arc); and dynamic collimator optimization=full. Of 222 dose metrics evaluated across all patients, all metric goals were met except 1/222, 3/222, and 2/222 for 4s, 5s, and 7s RAD, respectively, compared to 8/222 for VMAT. No RAD plans failed any metrics, while one VMAT plan failed a single dose metric: the ratio of V50% prescription dose to PTV volume (V_50%/V_PTV) . The most common metric scored within variation for RAD was V_50%/V_PTV (n=1 per 4s, 5s and 7s RAD) and for VMAT was PTV Dmax (n=3). For most metrics, VMAT and RAD performed similarly. However, RAD plans were more homogenous: average PTV V105% of 1.9% for VMAT vs 0.8%, 0.9%, and 0.9% for 4s, 5s, and 7s RAD. The 4s RAD plans had the best average V_50%/V_PTV: 3.3 vs. 3.5, 3.5, and 3.5 for 5s RAD, 7s RAD, and VMAT. VMAT had the lowest average skin Dmax: 9.6Gy vs 13.9Gy, 13.6Gy, and 13.4Gy for 4s, 5s, and 7s RAD. VMAT had less MU on average: 2421 vs 3056, 3008, and 2956 MU for 4s, 5s, and 7s RAD.