2230 - Efficacy and Safety of Fractionated Stereotactic Radiosurgery for the Treatment of Large Intracranial Arteriovenous Malformations

03:00pm - 04:00pm ET

Poster Hall - Exhibit Hall A

Screen: 7

POSTER

Presenter(s)

Eugene Vaios, MD, MBA - Duke Cancer Institute, Durham, NC

S. E. Spellicy¹, A. Rustad², M. S. Chen³, A. Suarez-Jew⁴, J. Herndon⁵, E. Hauck¹, D. Hasan¹, J. P. Kirkpatrick³, and E. J. Vaios³; ¹Department of Neurosurgery, Duke University Medical Center, Durham, NC, ²Duke University School of Medicine, Durham, NC, NC, ³Department of Radiation Oncology, Duke University School of Medicine, Durham, NC, ⁴Department of Neurosurgery, University of California San Francisco, San Francisco, CA, ⁵Department of Biostatistics, Duke University Medical Center, Durham, NC

Purpose/Objective(s):

Large arteriovenous malformations (AVMs) are difficult to treat safely and effectively with single fraction radiosurgery (SRS) due to reduced obliteration rates and increased risk for adverse sequelae, including radionecrosis. For patients ineligible or refractory to surgery or embolization, there is an urgent need for more effective radiotherapeutic approaches. We hypothesized that fractionated SRS has comparable efficacy and improved safety for large AVMs when compared to single fraction SRS.

Materials/Methods:

Patients with AVMs treated with LINAC-based SRS from October 2002 to October 2024 were identified from a prospective institutional database. Patients with multiple AVMs were excluded. Patients underwent computed tomography (CT) simulation in a thermoplastic facemask, as well as fine-cut (2mm) T2-weighted and contrast enhanced 3D-axial SPGR magnetic resonance (MR) imaging for SRS planning. A median marginal dose of 18 Gy (range: 12-25) was prescribed for single-fraction SRS. Staged dosing (24 Gy (24-30)) typically consisted of two 12-15Gy fractions separated by 2-6 months, and fractionated SRS (30 Gy) consisted of five consecutive daily fractions of 6Gy. AVMs = 3cm were defined as large. Obliteration was determined by MRI or angiography, and the cumulative incidence of obliteration and hemorrhage was determined, with death and salvage SRS as competing risks.

Results:

A total of 129 patients (Single n=99, Staged n=15, and Fractionated n=15) met inclusion criteria (mean age: 43 years; 52% male). Median follow up was 54 months [95%CI: 45-73]. There was no significant difference between SRS cohorts with respect to sex, performance status, smoking history, hypertension, prior hemorrhage, prior AVM resection, or prior AVM embolization. Diabetes and cardiovascular disease were increased in the staged SRS (p=0.04) and fractionated SRS cohorts (p= 0.01). A subset of large AVMs within each cohort was reviewed (Single: median 17Gy (range:12-20), n=21 (21%); Staged: 24Gy (24-30), n=9 (60%); Fractionated: 30Gy (30-30), n=14 (93%)). Among large AVMs, there was no significant difference in adverse events (hospitalization, hemorrhage, neurologic deficit, death) between SRS cohorts (Single: 16.7% [95%CI:3.6-41.9]); Staged:12.5% [0-52.7]; and Fractionated: 0% [0-23.2]). The three-year cumulative incidence of large AVM nidus obliteration was comparable between SRS cohorts (Single: 9.17% [95%CI:1.98-35.66], Staged: 0%, Fractionated: 9.09% [4.38-36.8]), though small sample sizes and short follow up in the fractionated SRS cohort limited comparison.

Conclusion:

In our series, fractionated LINAC-based SRS demonstrated comparable safety and efficacy to single fraction SRS in patients with large AVMs. Fractionated SRS could represent an alternative option for patients with large AVMs, and the present study provides a framework for prospective evaluation.