2026 Poster Presentations
P152: FRAME-BASED VERSUS FRAMELESS GAMMA KNIFE RADIOSURGERY FOR RECURRENT INTRACRANIAL MENINGIOMA
Anthony J Tang, BBA, BSA; Misha Amini, MD; Charles T Borchers, BS; Arjun R Adapa, MD; Eric Giannaris, BS; Michael B Sisti, MD; Jeffrey N Bruce, MD; Brian J Gill, MD; Brett E Youngerman, MD; Guy M McKhann, MD; Tony J Wang, MD; Matthew Gallitto, MD, PhD; Columbia University Irving Medical Center
Background: Gamma knife radiosurgery (GKRS) is a commonly used treatment modality for salvage treatment of recurrent intracranial meningioma. Precise immobilization is critical given high doses of radiation and minimal margins of error with surrounding brain tissue. Historically, immobilization was performed using a stereotactic frame that is placed under local anesthesia and remains in place during planning and treatment. However, frame-based immobilization can be uncomfortable, and complications such as infections and persistent pain at the insertion site have been reported. Beginning in 2016, our institution adopted the frameless technique (The GK Icon) using a lightweight plastic mask for immobilization and allowing for enhanced patient comfort and fractionation of treatments over several days. We report our institution’s long-term outcomes comparing frameless versus frame-based fixation in a recurrent meningioma cohort.
Methods: We conducted a retrospective cohort study of patients with recurrent intracranial meningioma treated with GKRS at our institution from 2014-2021 using frame-based or frameless (mask) head fixation. The primary endpoint was progression-free survival (PFS) from GKRS to radiographic local failure or death. Overall survival (OS) was defined from GKRS to death. Survival was estimated with Kaplan-Meier, reported as median and 1-, 3-, and 5-year rates with 95% confidence intervals, and compared between fixation groups using the log-rank test; median follow-up was estimated using reverse Kaplan-Meier. Secondary outcomes included symptomatic toxicity, use of steroids or Avastin, optic neuropathy, and new tumor and were modeled with logistic regression, estimating odds ratios for the frame versus mask cohorts and adjusting for age at GKRS, follow-up duration, and pathology grade (when available); p-values were corrected using the Benjamini-Hochberg method. Baseline characteristics were tabulated including sex, age at treatment, tumor grade, and GKRS parameters such as total dose and number of fractions.
Results: We analyzed 154 recurrent lesions in 112 patients (Mask n=81, Frame n=73). Most lesions had prior surgery before GKRS (139/154; 90%); among these, grades were WHO I 40% (55/139), WHO II 58% (81/139), and WHO III 2% (3/139). Median follow-up was 68.8 months overall (Mask 59.3; Frame 87.0). Median PFS was 73.2 months (Mask) and 101.3 months (Frame); PFS did not differ by fixation (log-rank p=0.20). One-, three-, and five-year PFS were: Mask 94.9% (90.1–99.9) / 74.9% (65.7–85.4) / 58.5% (47.7–71.8); Frame 91.0% (84.5–98.1) / 71.5% (61.4–83.2) / 64.7% (54.0–77.5). OS did not differ by fixation (log-rank p=0.45). OS at 1/3/5 years was: Mask 100.0% / 97.4% (93.9–100.0) / 95.7% (91.0–100.0); Frame 100.0% / 89.6% (82.6–97.2) / 89.6% (82.6–97.2). In grade-stratified analyses (WHO I vs II/III), fixation remained unassociated with PFS or OS. In adjusted analyses of secondary outcomes, no significant differences were observed between fixation groups.
Conclusion: In this single-institution cohort of 154 recurrent meningioma lesions, framed versus frameless head immobilization during GKRS was not associated with differences in PFS or OS. After adjusting for age, follow-up time, and tumor grade, no significant differences were found in secondary outcomes. These findings support frameless immobilization for GKRS treatment of recurrent meningioma as a viable alternative to frame fixation without compromising survival or measured toxicities.