3029 - Dosimetric Comparison of Upfront Single-Optimization, Geometrically Split Radiosurgery Planning vs. Sequential Gap-Based Volumetric Staged Radiosurgery Planning for Arteriovenous Malformations
Presenter(s)
J. W. Jung1, Y. Cao1, W. Belcher1, J. Glass2, S. Sharma1, A. W. Ju1, and M. S. Peach1; 1Department of Radiation Oncology, Brody School of Medicine, East Carolina University, Greenville, NC, 2Brody School of Medicine, East Carolina University, Greenville, NC
Purpose/Objective(s): Volumetric staged stereotactic radiosurgery (SRS) is used for large arteriovenous malformations (AVMs) to reduce normal brain toxicity while maintaining high-dose target coverage. A conventional approach sequentially plans each stage with an intentional geometric gap between volumes to limit excess dose overlap. However, gap creation is imprecise and may compromise coverage. We compared traditional sequential gap-based planning with an upfront, single-optimization strategy in which the entire AVM is optimized first and then geometrically divided into staged volumes without introducing a gap.
Materials/Methods: Five patients with AVMs previously treated using two sequential volumetric stages incorporating an inter-stage geometric gap were retrospectively analyzed. For each case, a corresponding contiguous (gap-free) staged plan was generated in Leksell GammaPlan v11.4. The full AVM was contoured and optimized upfront using inverse planning. The optimized shot distribution was then geometrically separated into two approximately equal treatment volumes and delivered independently without a gap. Dosimetric parameters from original gap-based plans were compared with upfront gap-free plans, including prescription coverage (=20 Gy), conformity, gradient index, normal brain V12 and V20, maximum organ-at-risk dose, number of shots, and beam-on time.
Results: Upfront gap-free planning improved prescription coverage relative to gap-based plans (99.5±0.14% vs 98.9±0.97%). Normal brain dose was reduced, with mean V12 and V20 decreases of 24% and 22%, respectively. Maximum skull and optic doses were lower. Shot number decreased by 19%, and beam-on time was reduced by 28%. Conformity and gradient index were maintained.
Conclusion: Single-optimization upfront volumetric staged planning without an inter-stage gap improved target coverage, reduced normal brain dose, and significantly shortened treatment time compared with sequential gap-based planning. This approach simplifies workflow by eliminating iterative gap design while achieving superior cumulative dosimetric performance. Routine inter-stage gap introduction in staged AVM radiosurgery warrants reconsideration.
Table. Dosimetric analysis for a patient| Metric | Organ | Plan with gap | Plan without gap | % difference |
| Coverage (%) | nidus | 98.4 | 99.4 | 1.0 |
| Dmax (Gy) | skull | 40.6 | 34.3 | -15.5 |
| V12 (cc) | skull | 31.6 | 23.6 | -25.3 |
| V20 (cc) | skull | 15.8 | 11.2 | -29.1 |
| Dmax (Gy) | Optic | 11.4 | 9.9 | -13.2 |
| Number of shots | 72 | 58 | -19.4 | |
| Beam-on time (min) | 123 | 87 | -29.3 | |