Purpose This single-institution prospective study was designed to investigate the feasibility and safety of dose escalation with GliaSite (Proxima Therapeutics Inc., Alpharetta, GA) brachytherapy for the treatment of patients with newly diagnosed and recurrent central nervous system (CNS) tumors after neurosurgical resection. We now report mature results of this trial, its outcomes, and a toxicity profile. Methods and Materials Ten adult consecutive patients with recurrent and newly diagnosed CNS malignancies underwent GliaSite brachytherapy after maximally safe neurosurgical resection between 2004 and 2007. GliaSite balloon was placed intraoperatively, and the size was selected so as to conform to the surgical cavity. Low-dose-rate radiation was delivered with an aqueous solution of organically bound 125I (Iotrex: sodium 3-( 125I)-iodo-4-hydroxybenzenesulfonate; Proxima Therapeutics Inc.), introduced into the balloon portion of the device via a subcutaneous port. Two to 3 weeks later, the device was filled with Iotrex for a median dwell time of 94.3 hours (range, 68.0–120.5 hours), after which the balloon was explanted. A commercial 3-D planning system was used for a detailed analysis of dosimetry. Median dose of 52.0 Gy (range, 45.0–60.0 Gy) was prescribed 0.5–1.0 cm from the balloon surface. Radiation Therapy Oncology Group (RTOG) criteria were used to assess acute and long-term toxicities associated with this technique. Followup was assessed with MRI scans and was available on all enrolled patients. Results Median followup for surviving patients was 38 months (range, 18–57 months). Mean size of GliaSite balloon was 3.4 cm (range, 2.0–4.0 cm). Mean volume of filling was 19.0 cc (range, 4.0–35.0 cc). Median activity of Iotrex was 301.6 mCi (range, 95.0–515.4 mCi). Median survival was 14.0 months for the entire cohort after the treatment with the GliaSite device. Of our cohort, 6/10 (60%) patients sustained recurrence (20% local and 40% distant). Median time to recurrence after treatment with GliaSite was 8.0 months, and median time to death after recurrence was 7.5 months. There were no RTOG Grade 3 or 4 acute or late toxicities. Followup MRI imaging did not identify any evidence of radiation necrosis. Conclusions Our data indicate that treatment with GliaSite balloon brachytherapy is feasible and safe, while rendering acceptable local control and minimal acute and long-term toxicities for newly diagnosed and recurrent CNS malignancies. These encouraging results compel us to embark on testing larger numbers of patients with this treatment modality.