ABSTRACT
Neurotoxic intrathecal chemotherapy for childhood acute lymphoblastic leukemia (ALL) affects developing structures and functions of memory and learning subsystems selectively. Results show significant reductions in magnetic resonance imaging morphometry of mamillary bodies, components of the corticolimbic-diencephalic subsystem subserving functionally later developing, single-trial memory, nonsignificant changes in bilateral heads of the caudate nuclei, components of the corticostriatal subsystem subserving functionally earlier developing, multitrial learning, significant reductions in prefrontal cortical volume, visual and verbal single-trial memory deficits, and visuospatial, but not verbal, multitrial learning deficits. Multiple regression models provide evidence for partial dissociation and connectivity between the subsystems, and suggest that greater involvement of caudate may compensate for inefficient corticolimbic-diencephalic components.
Subject(s)
Antineoplastic Agents/adverse effects , Brain/drug effects , Brain/pathology , Memory Disorders/diagnosis , Memory Disorders/etiology , Nerve Net/drug effects , Nerve Net/pathology , Precursor Cell Lymphoblastic Leukemia-Lymphoma/drug therapy , Adolescent , Case-Control Studies , Caudate Nucleus/drug effects , Caudate Nucleus/pathology , Child , Female , Humans , Magnetic Resonance Imaging , Male , Mammillary Bodies/drug effects , Mammillary Bodies/pathology , Memory Disorders/pathology , Memory, Short-Term/drug effects , Models, Psychological , Multivariate Analysis , Prefrontal Cortex/drug effects , Prefrontal Cortex/pathology , Verbal Learning/drug effects , Visual Perception/drug effectsABSTRACT
BACKGROUND: Following brain insult in early childhood, the later maturing neocerebellum and frontal lobes frequently show abnormalities. OBJECTIVE: To investigate the morphologic characteristics and function of a proposed cerebellar-frontal subsystem in children treated for acute lymphoblastic leukemia (ALL) with intrathecal methotrexate using quantitative magnetic resonance imaging, neuropsychological measures, nonlinear multiple regression analysis, and a statistical effect size model that augments interpretive validity of nonsignificant statistical findings, particularly from small sample size studies. DESIGN: Comparison and relationship of magnetic resonance imaging morphometry of cerebellar lobuli I-V and VI-VII and prefrontal cortices, and performance on 5 neuropsychological tests assessing visual-spatial attention, short-term memory, and visuomotor organization and coordination between childhood survivors of ALL and a matched control group. PARTICIPANTS: Ten childhood survivors of ALL treated between 1982 and 1989 with standard 3-year intrathecal chemotherapy, and matched control subjects. MAIN OUTCOME MEASURES: Morphometric results of cerebellar lobuli I-V and VI-VII and prefrontal cortices, and results of Trail-Making Tests, Rey-Osterreith Complex Figure Test, WISC-III Coding. RESULTS: Significant effect size model values for outcome measures in the ALL group support deficits in lobuli VI-VII and prefrontal cortices, and neuropsychological performance. Multiple regression analysis results were consistent with hypothesized involvement of a cerebellar-frontal brain subsystem. CONCLUSION: Treatment of children with ALL with intrathecal methotrexate before 5 years of age has structural and functional effects on the developing neocerebellar-frontal subsystem.
