Re-Irradiation Using Brachytherapy for Recurrent Intracranial Tumors: A Systematic Review and Meta-Analysis of the Literature.

Introduction We aim to compare the efficacy and toxicity of re-irradiation using brachytherapy for patients with locally recurrent brain tumors after previous radiation therapy.  Methods We performed a systematic review of the major biomedical databases from 2005 to 2020 for eligible studies where patients were treated with re-irradiation for recurrent same site tumors using brachytherapy. Tumor types included high-grade gliomas (HGG) (World Health Organization (WHO) Grades 3 and 4), meningiomas, and metastases. The outcomes of interest were median overall survival (OS) and progression-free survival (PFS) after re-irradiation, the incidence of radiation necrosis (RN), and other relevant radiation-related adverse events (AE). We used a fixed-effect meta-analysis regression moderation model to compared results of interstitial versus intracavitary therapy, treatment with low-dose-rate (LDR) versus high-dose-rate (HDR) techniques, and outcomes by tumor type. Results The search resulted in a total of 194 articles. A total of 16 articles with 695 patients fulfilled the inclusion criteria and were selected for analysis. For high-grade glioma, meningioma, and brain metastasis the pooled meta-analysis showed mean symptomatic RN rates of 3.3% (standard error (SE) = 0.8%), 17.3% (SE = 5.0%), and 22.4% (SE = 7.0%), respectively, and mean rates of RN requiring surgical intervention of 3.0% (SE = 1.0%), 11.9% (SE = 5.3%), and 10.0% (SE = 7.3%), respectively. The mean symptomatic RN rates in the meta-analysis comparing interstitial versus intracavitary therapy were 3.4% and 4.9%, respectively (p = 0.36), and for the comparison of LDR versus HDR, the rates were 2.6% and 5.7%, respectively (p = 0.046). In comparing the symptomatic RN rates in comparison to HGG versus meningioma, the means were 3.3% and 17.3%, respectively (p = 0.006), and in HGG versus metastatic tumors, the means were 3.3% and 22.4%, respectively (p = 0.007). There was no significant difference in rates of RN requiring surgery in any of these groups. Due to the small number of studies and inconsistent recording of OS and PFS, statistical analysis of these parameters could not be performed. Conclusion Published literature on the same site re-irradiation using brachytherapy for recurrent brain tumors is highly limited, with inconsistent reporting of safety and efficacy outcomes. To overcome these shortcomings, we utilized a structured meta-analysis approach to show that re-irradiation with modern brachytherapy is generally safe in terms of the risks of symptomatic RN. We also found that symptomatic RN rates for brachytherapy are significantly lower in recurrent HGG compared to recurrent meningiomas (p = 0.006) and metastatic tumors (p = 0.007). Re-irradiation with brachytherapy is a feasible option for appropriately selected patients. The availability of Cesium-131 (Cs-131) shows promise in reducing toxicity while achieving excellent local control due to its physical properties, and the recent introduction of a novel surgically targeted radiation therapy device, that makes brachytherapy less technically demanding, may allow for more widespread adoption. Prospective trials with consistent reporting of endpoints are needed to explore whether these advances improve safety and efficacy in patients with recurrent, previously irradiated tumors.