24 Radiation induced meningioma in adult survivors of childhood leukemia or primary brain tumor treated with cranial radiotherapy: Incidence and screening recommendations

Abstract

Purpose: Cranial radiotherapy (CRT) was commonly given for childhood leukemia and brain tumors. Survivors are at risk of late effects including radiation induced meningioma (RIM). Surveillance for RIM is not standardized . We aimed to determine the incidence, latency, and screening patterns for RIM. Materials and Methods: Retrospective chart review of all patients aged <18 years at the time of radiation (RT), treated with CRT for leukemia or a brain tumor in BC between 1981-2006. Patient, tumor, and treatment characteristics were collected. Actuarial statistics were calculated with Kaplan-Meier Curves. Patients were censored at the date of last normal cranial imaging, or development of a RIM. Results: 392 patients were identified. Median age (range) at CRT was 9.6 years. Median CRT dose was 28Gy. The original diagnosis was leukemia in 50%, glioma in 13%, medulloblastoma in 8%, ependymoma in 7%, neuroectodermal tumor in 7%, germ cell tumor in 5%, craniopharyngioma in 4%, and other pathologies in 6%. Median (range) of clinical follow-up (FU) was 13.2 (0-37.5) years. Median (range) of cranial imaging FU was 15.5 (0-21.2) years. There was no documented cranial imaging FU in 144 patients. Forty-eight patients developed a RIM. The median age (range) at RT for patients with RIM was 6.7 years. Only 8 of these cases presented with associated symptoms. The earliest RIM in our cohort occurred 10.2 years after CRT. On actuarial analysis, the median (95% CI) time to development of a meningioma was 29.8 (28.9-30.7) years. Incidence (95% CI) of meningioma at 10 years was 0%, 15 years was 5 (2-9)%, 20 years was 12 (6-18)%, 25 years was 33 (23-43)% and 30 years was 47 (37-68)%. Amongst patients with a RIM, the median dose of CRT was 45 Gy. The lowest dose of RT in a patient who developed RIM was 12 Gy. RT was delivered to the whole brain in 58% and partial brain in 42% of patients with a RIM. Conclusions: After CRT in pediatric patients, there is a significant risk of developing a RIM and a steady increase in this risk with ongoing follow-up. We recommend standardization of surveillance for these patients with screening beginning 10 years after completion of CRT.