Late magnetic resonance imaging features of leukoencephalopathy in children with central nervous system tumours following high-dose methotrexate and neuraxis radiation therapy

• Published in: European journal of cancer (1990) 2005-07, Vol.41 (11), p.1588-1596
• Main Authors: Kellie, Stewart J., Chaku, Jyoti, Lockwood, Liane R., O’Regan, Peter, Waters, Keith D., Wong, Christopher K.F.
• Format: Article
• Language: English
• Subjects: Adolescent; Antimetabolites, Antineoplastic - administration & dosage; Antineoplastic Combined Chemotherapy Protocols - therapeutic use; Carboplatin - administration & dosage; Cerebellar Neoplasms - drug therapy; Cerebellar Neoplasms - pathology; Cerebellar Neoplasms - radiotherapy; Child; Child, Preschool; Combined Modality Therapy - methods; Dementia, Vascular - pathology; Etoposide - administration & dosage; Humans; Infant; Leukoencephalopathy; Magnetic Resonance Imaging; Medulloblastoma; Medulloblastoma - drug therapy; Medulloblastoma - pathology; Medulloblastoma - radiotherapy; Methotrexate; Methotrexate - administration & dosage; Radiation therapy
• ISSN: 0959-8049; 1879-0852
• DOI: 10.1016/j.ejca.2005.02.024

High-dose methotrexate (HDMTX) is used increasingly to treat children with central nervous system (CNS) tumours. Although the neuro-imaging features of leukoencephalopathy associated with systemic or intrathecal methotrexate administered after cranial radiation have been well described, the extent to which the sequencing of HDMTX prior to cranial radiation in infants and children predisposes to late neuroradiological features of leukoencephalopathy is unknown. This report describes the National Cancer Institute (NCI) toxicity grade of leukoencephalopathy based on magnetic resonance imaging (MRI) findings in all patients who survived 4 or more years after treatment on an earlier phase II study. These patients, with newly diagnosed CNS embryonal tumours, were in the age range 3.5–14.2 years (median 6.9 years) at diagnosis, and received four courses of pre-irradiation combination chemotherapy, including HDMTX 8 g/m 2. Following completion of the ‘up-front’ phase II study, all patients received conventionally fractionated whole brain doses of 36–50.4 Gy. The radiation dose and treatment volumes were determined individually according to the primary tumour location and results of extent of disease evaluations. The most recent MRI brain scans, obtained 4.0–10.5 years (median 6.5 years) after radiation therapy and comprising a minimum of T1, T1 following gadolinium and T2 sequences, were reviewed centrally to assess the neuroradiological grade of leukoencephalopathy, based on the NCI Common Terminology Criteria for Adverse Events, v3.0. Grade I changes (mild increase in subarachnoid space, and/or mild ventriculomegaly, and/or small/focal T2 hyperintensities) were evident in 8 of the 12 patients and grade II changes (moderate increase in subarachnoid space and/or moderate ventriculomegaly, and/or focal T2 hyperintensities extending to the centrum ovale) were found in the remaining 4. In conclusion, treatment with multiple courses of HDMTX prior to 36–50.4 Gy cranial radiation did not result in moderate to severe MRI features of leukoencephalopathy. Future studies in paediatric neuro-oncology patients, involving HDMTX combined with prospective neuropsychological evaluations appear justified.