ABSTRACT
Children treated for medulloblastoma (MB) exhibit long-term impairments in declarative memory, but the pathophysiology underlying this is unclear. Previous studies report declines in global white matter volume, but have failed to link this to declines in memory performance. We examined the effects of treatment on measures of global brain structure (i.e., total white and gray matter volume) and specific memory structures (i.e., hippocampus and uncinate fasciculus). We used volumetric MRI and diffusion tensor imaging in pediatric survivors of MB and one survivor of astrocytoma treated with cranial-spinal radiation (n = 20), and healthy controls (n = 13). Compared to controls, the survivor group exhibited reduced white matter volume, damage to the uncinate fasciculus, and a smaller right hippocampus. Critically, reduced hippocampal volume was not related to differences in brain volume, suggesting that the hippocampus may be especially vulnerable to treatment effects. A subset of the survivors (n = 10) also underwent memory testing using the Children's Memory Scale (CMS). Performance on the general index of the CMS was significantly correlated with measures of hippocampal volume and uncinate fasciculus. The examination of treatment effects on specific brain regions provides a better understanding of long-term cognitive outcome in children with brain tumors, particularly medulloblastoma.
Subject(s)
Brain/pathology , Infratentorial Neoplasms/complications , Infratentorial Neoplasms/pathology , Medulloblastoma/complications , Medulloblastoma/pathology , Memory Disorders/etiology , Adolescent , Analysis of Variance , Brain/radiation effects , Case-Control Studies , Child , Diffusion Tensor Imaging , Female , Humans , Infratentorial Neoplasms/radiotherapy , Magnetic Resonance Imaging , Male , Medulloblastoma/radiotherapy , Memory Disorders/diagnosis , Neuropsychological Tests , White Matter/pathology , White Matter/radiation effectsABSTRACT
PURPOSE: The outcome of recurrent ependymoma in children is dismal. Reirradiation has been proposed as an effective modality for ependymoma at relapse. However, the toxicity and outcome benefits of this approach have not been well established. METHODS AND MATERIALS: We conducted a retrospective population-based study of all patients with recurrent ependymoma treated between 1986 and 2010 in our institution. Demographic, treatment, and outcome data were analyzed for the entire cohort. RESULTS: Of 113 patients with intracranial ependymoma, 47 patients relapsed. At the time of relapse, 29 patients were treated with surgical resection and/or chemotherapy, and 18 patients received full-dose (≥ 54 Gy focal and/or craniospinal) reirradiation with or without surgery at recurrence. Reirradiation was tolerated well with no severe acute complications noticed. Three-year overall survival was 7% ± 6% and 81% ± 12% for nonreirradiated and reirradiated patients, respectively (p < 0.0001). Time to second progression after reirradiation was significantly longer than time to first progression. This surprising phenomenon was associated with improved progression-free survival for tumors with evidence of DNA damage (n = 15; p = 0.002). At a mean follow-up of 3.73 years, only 2/18 patients had endocrine dysfunction, and 1 patient required special education support. However, a decline in intellectual function from pre- to postreirradiation assessment was observed. CONCLUSIONS: Reirradiation is an effective treatment that may change the natural history of recurrent ependymoma in children. However, this change may be associated with increased neurocognitive toxicity. Additional follow-up is needed to determine the risk of late recurrence, secondary radiation-induced tumors, and long-term functional outcome of these patients.
