A Phase I study of intraoperative radiation therapy in children with recurrent brain tumors- a preliminary report

Abstract

To describe the feasibility, toxicity and early efficacy of intraoperative radiation therapy (IORT) using the photon radiosurgery system (PRS)in children with recurrent brain tumors. The PRS incorporates a miniature 50 kV x-ray source capable of delivering a prescribed radiation dose directly into the surgical cavity using spherical applicators. Thus this treatment approach combines the advantages of surgery and RT with focused high RBE x-rays delivered directly into the tumor bed at a high dose rate. Between 2001 and 2004, 16 children were treated with IORT using the PRS at our institution on an IRB approved phase-I dose escalation protocol. Patients were classified into two groups. One group (Gp A) comprised of 8 children with recurrent tumors after surgery or chemotherapy (CT) and full dose RT (>50 Gy). In this group, 7 children had a posterior fossa ependymoma and 1 had a glioma. Most of them had multiple surgeries, one each had failed additional therapy with gamma knife radiosurgery (12 Gy) and high dose CT- bone marrow transplantation prior to IORT. The other group (Gp B) comprised of 8 children with recurrent tumors after surgery and / or chemotherapy but not RT. Among them 5 had supratentorial ependymoma, 1 each had glioma, astroblastoma and fibrosarcoma. All patients had pathologically confirmed tumor recurrence at surgery prior to IORT. The IORT dose was delivered using PRS applicators that ranged in size from 1.5–5 cm. The IORT dose was 10 Gy prescribed to a depth of 2mm in ten children, 10 Gy to 5mm in four children and 12 Gy to 2mm in two children on this dose escalation study. All patients were periodically evaluated by clinical and radiologic (MRI) examinations per protocol. The primary study end point was the development of unacceptable acute CNS toxicity (RTOG grade 3 or higher). The median follow up for all patients was 12 months (4–24 m). The median follow up in Gp A patients was 8 months (4–24 m), and four patients have been followed for more than one year after therapy. The median follow up for Gp B patients was 14 months (10–19m), and six have been followed for more than one year. There were no IORT-related complications in the intra or peri-operative period. The IORT procedure prolonged the duration of surgery by approximately 30 – 90 minutes for each patient. The median treatment time was 13.7 minutes (5.7–44). The median dose rate was 77.3 cGy/min (22.5–170). Two of 16 patients (12.5%) developed radiation necrosis approximately 3–4 months after IORT. Both of these patients belonged to GpB and had not received prior RT. Both of these patients received a dose of 10 Gy to a depth of 5mm. At the beginning of the study, the protocol allowed for the IORT dose to be prescribed to a depth of 2–5 mm. Thus four patients in GpB received 10 Gy to a depth of 5mm. Two of these four patients (50%) developed radiation necrosis. One developed symptomatic necrosis requiring surgery 5 months after IORT. Pathology revealed radiation necrosis without tumor. The other patient was asymptomatic and had spontaneous resolution of necrosis 15 months after IORT. None of the GpA patients developed radiation necrosis or any other complication following IORT despite previous full dose RT. Thus far, 11 patients (69%) have achieved tumor control after IORT, 5/8 (63%)in GpA and 6/8 (75%)in Gp B. Five patients in Gp A (63%) and 7 in Gp B (88%) are alive, 4 children died of tumor progression. This report describes the feasibility, safety and early efficacy of IORT (10 Gy to 2mm) delivered using the PRS in children with recurrent brain tumors, who may or may not have previously received full dose external RT. Majority of the previously irradiated children had tumors in the posterior fossa adjacent to the brain stem and fourth ventricle. IORT to a dose of 10 Gy prescribed to 5mm was associated with a high incidence of radiation necrosis and should not be performed especially in previously irradiated children. Patients are currently being enrolled at the next dose level (12 Gy to 2mm) of our dose-escalation study