3181 - Clinical Impact of Dynamic Collimator for Single-Isocenter Multi-Target SRS Evaluated through Physician Rating and Dosimetric Analysis
Presenter(s)
Y. Xie1, Y. Gao2, T. C. Mullikin1, Q. J. J. Wu1, T. Li2,3, X. Wu1, D. Yang1, Q. Wu1, and Y. Sheng1; 1Duke University Medical Center, Durham, NC, 2Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, 3Varian Medical Systems, Palo Alto, CA
Purpose/Objective(s): A recently introduced VMATp technique (commonly referred to as RAD) combines traditional VMAT with static angle modulated ports (STAMPs) to enhance conformity and modulation flexibility. This study isolated the impact from dynamic collimator rotation and assessed its clinical impact compared with standard VMAT for single-isocenter multi-target (SIMT) stereotactic radiosurgery, through subjective physician rating and objective dosimetric analysis.
Materials/Methods: Nineteen SIMT SRS cases (each with 5–17 lesions) were retrospectively replanned using RAD following institutional clinical standards. Every RAD plan included 4–8 arcs, with only one STAMP in an arc-dominant setup to isolate impact from only dynamic collimator rotation during arcs. Dose calculations utilized the Acuros XB algorithm with a 1-mm grid, and plans were normalized so each PTV coverage achieved V100% no less than 99%. To evaluate clinical impact, one radiation oncologist specialized in radiosurgery reviewed both plan groups. Clinical VMAT and RAD plans were scored by the attending on a 5-point scale: (1) VMAT significantly better; (2) VMAT slightly better; (3) comparable; (4) RAD slightly better; (5) RAD significantly better. Dosimetric endpoints included conformity index (CI), homogeneity index (HI), and normal brain dose volumes (V60%, V30%, V25%, V20%). Delivery efficiency was assessed by total monitor units (MU) and simulated beam-on time. Paired t-tests were used for statistical comparisons.
Results: Single physician reviews showed a preference for RAD over clinical VMAT, with an average score of 3.89±0.99 (where 3.0 is neutral). Out of 19 cases, RAD was rated as superior (score 5) in six cases (32%), and as preferred with likely clinical benefit (score 4) in seven cases (37%), meaning that RAD was favored in 68% of the cases. Four cases (21%) were judged as clinically equivalent (score 3). Only two cases (11%) favored clinical VMAT (score 2), both due to marginally lower V60% exposure to normal brain tissue, and no case found clinical VMAT to be clearly superior. In terms of planning metrics, compared to VMAT, RAD provided significant improvement in conformity index (CI: 1.43 vs 1.61, p<0.001) while maintaining similar homogeneity (HI: 0.16 vs 0.15, p=0.2675), and reduced V20% to normal brain by 39.8 ± 83.7cc (p=0.047). Moreover, the use of RAD significantly reduced monitor units by ~38% (7282±2334 vs 11634±5081, p<0.001) and significantly decreased delivery time by ~36% (423±103 s vs 569±201 s, p<0.001).
Conclusion: RAD provided better conformity, reduced low-dose brain exposure, and more efficient delivery, supporting its clinical value for multi-target SRS. Single physician review demonstrated that RAD was preferred in majority(68%) of cases, with no VMAT plan rated superior.