Impact of radiation avoidance on survival and neurocognitive outcome in infant medulloblastoma.

PURPOSE: Concerns about radiotherapy-related neurocognitive sequelae in young children have led to deferral or avoidance of radiation in contemporary treatment for this fragile group of patients. We compared survival and neurocognitive outcome in two groups of infants with medulloblastoma who received adjuvant conventional craniospinal irradiation (CSI) or reduced or no radiotherapy during an era of change in the philosophy of infant medulloblastoma treatment. PATIENTS AND METHODS: From 1985 to 2007, 29 patients 3 years of age or younger were diagnosed and treated with curative intent in our institution. Children treated before 1994 received adjuvant radiation with chemotherapy; subsequently, radiation was prescribed essentially for disease progression or relapse. RESULTS: Median age at diagnosis was 24 months (range: 1-36 months); 15 patients (52%) presented with metastatic disease at diagnosis. As part of initial treatment, 8 children received adjuvant radiotherapy with chemotherapy, and 21 children received postoperative chemotherapy only. Five children treated with chemotherapy alone are in prolonged remission. The 5-year event-free and overall survivals were 35.9% +/- 9.8% and 50.2% +/- 9.6% respectively. Extent of resection, metastatic status, and desmoplastic histology were not found to be significant prognostic factors. On serial neurocognitive evaluations, patients treated with chemotherapy with or without reduced radiotherapy demonstrated improvement of intellectual function over time. Patients treated with conventional csi exhibited significantly lower intelligence quotient scores and academic performance, with the exception of receptive vocabulary. CONCLUSIONS: Avoidance of conventional CSI in treatment of very young children with medulloblastoma appears to be associated with a preserved neurocognitive profile. Neurocognitive evaluation should be integrated into the primary objectives of future infant protocols.