Outcome evaluation of patients treated with fractionated Gamma Knife radiosurgery for large (> 3 cm) brain metastases: a dose-escalation study

OBJECTIVE Fractionated Gamma Knife radiosurgery (GKS) represents a feasible option for patients with large brain metastases (BM). However, the dose-fractionation scheme balanced between local control and radiation-induced toxicity remains unclear. Therefore, the authors conducted a dose-escalation study using fractionated GKS as the primary treatment for large (> 3 cm) BM. METHODS The exclusion criteria were more than 3 lesions, evidence of leptomeningeal disease, metastatic melanoma, poor general condition, and previously treated lesions. Patients were randomized to receive 24, 27, or 30 Gy in 3 fractions (8, 9, or 10 Gy per fraction, respectively). The primary endpoint was the development of radiation necrosis assessed by a neuroradiologist blinded to the study. The secondary endpoints included the local progression-free survival (PFS) rate, change in tumor volume, development of distant intracranial progression, and overall survival. RESULTS Between September 2016 and April 2018, 60 patients were eligible for the study, with 46 patients (15, 17, and 14 patients in the 8-, 9-, and 10-Gy groups, respectively) available for analysis. The median follow-up duration was 9.6 months (range 2.5–25.1 months). The 6-month estimated cumulative incidence of radiation necrosis was 0% in the 8-Gy group, 13% (95% confidence interval [CI] 0%–29%) in the 9-Gy group, and 37% (95% CI 1%–58%) in the 10-Gy group. Being in the 10-Gy group was a significant risk factor for the development of radiation necrosis (p = 0.047; hazard ratio [HR] 7.2, 95% CI 1.1–51.4). The 12-month local PFS rates were 65%, 80%, and 75% in the 8-, 9-, and 10-Gy groups, respectively. Being in the 8-Gy group was a risk factor for local treatment failure (p = 0.037; HR 2.5, 95% CI 1.1–29.6). The mean volume change from baseline was a 47.5% decrease in this cohort. Distant intracranial progression and overall survival did not differ among the 3 groups. CONCLUSIONS In this dose-escalation study, 27 Gy in 3 fractions appeared to be a relevant regimen of fractionated GKS for large BM because 30 Gy in 3 fractions resulted in unacceptable toxicities and 24 Gy in 3 fractions was associated with local treatment failure.