2593 - Novel Volume Staged Stereotactic Radiosurgery (VS-SRS) for the Treatment of Large Volume Cerebral Arteriovenous Malformation (AVM)

Purpose/Objective(s): Stereotactic radiosurgery (SRS) has become a cornerstone in the management of inoperable arteriovenous malformations (AVMs) of the brain. However single stage SRS may not be feasible for large AVM's. Volume Stage SRS (VS-SRS) with a phased approach to radiation delivery has been considered for large AVM's. We report our unique strategy of VS-SRS to treat large AVM and the clinical outcomes.

Materials/Methods: This is an IRB approved retrospective review to analyze the treatment outcome in the patients with large AVM’s who underwent VS-SRS in our institution from 2009-2024. AVMs were contoured on fused images of simulation CTA, MRA and angiogram by a neurosurgeon and radiation oncologist. The large volume AVM's (>10 cc) were divided into 2-3 sub targets to be treated sequentially in a VS approach at 2 weeks interval. Prescription doses ranged from 16Gy-24Gy/subunit. VMAT plans using base dose technique were generated and optimized. 1st stage optimization was conducted with a high priority of minimizing dose to 2nd stage and base dose technique was used to optimize the remaining stages, so the target volume received the prescription dose while maintaining a low Gradient Index (G.I). Patients were treated on a Linac having HD MLC and external IGRT verification. The cumulative plans were evaluated for target coverage, Conformity Index (CI), GI, and brain V12. Patients were followed with annual MRA and had repeat arteriogram (I.R.A) if MRA showed no residual AVM nidus. Treatment outcomes of response rate, symptomatic radiation necrosis (S-RN) and neurologic sequelae were analyzed. Statistics analysis was with Fisher’s exact test.

Results: We included 22 patients with median age 38.5 yrs and median follow up (m.f.u) 40 months (mo). The median AVM target volume was 9.19 ± 8.59 cc. In VS, 16 patients were treated in 2 subunits stages and 6 patients were treated in 3 subunits stages with median subunit volume of 4.79 ±3.56cc. The cumulative plan parameters achieved include median treatment dose of 2000 ± 160 cGy (Max dose 2660.3 ± 76 cGy). Median D95 was 101% prescription dose (R 92.4-115), median CI of 1.275 (R 0.11-1.93), GI of 3.664 (R 2.75-16.4) and V12 of 21.26 cc (R 5.35- 59.47) respectively. 9 patients (41%) achieved completed response (CR), of whom 6 (27%) patients had CR confirmed by I.R.A ( m.f.u 56 mo), 3 patients (13.6%) had CR on MRA (m.f.u 20.5 mo). 12 (54%) patients had partial response by MRA (m.f.u of 24 mo). 3 patients developed S-RN (13.6%), 2 were treated with steroids with resolution of NS. 1 patient with preexisting neurological deficit required bevacizumab. No statistical difference was observed in patients without S-RN and with S-RN in median V12 or total AVM volumes

Conclusion: This study demonstrated that our unique VS-SRS to large AVM with treatment interval of 2-3 weeks provide comparable treatment outcome with possible lower S-RN rate compared to the reported VS-SRS with treatment interval of 3-6 months. Our lower rate of S-RN maybe associated with the lower hot spots in the cumulative plans