paediatrics Brussels 17

Redmond et al.: Radiation to neural progenitor niches and neurocognitive outcomes

at Universitaet Leipzig, Institut fuer Informatik/URZ, Bibliothek on August 25, 2014 http://neuro-oncology.oxfordjournals.org/ Downloaded from

Fig. 1. Representative contours of the hippocampus on an axial (top image) T1-weighted MRI postgadolinium contrast and on the sagittal (bottom left) and coronal reconstructions (bottom right). The T1 postgadolinium and FLAIR-sequence MRIs were fused to the RT-planning CT scan to allow calculation of the doses to contoured structures.

dependent on dorsolateral prefrontal-striatal circuitry, potentially disrupted following damage to the SVZ. Motor speed was assessed using Purdue Pegboard, 36 which measures the time it takes to place pegs into a board with 25 parallel holes. The 2-Hand trial was ana- lyzed for the present study. Measures of motor speed are considered to be dependent on frontostriatal circuitry in- volving the motor circuit including the putamen, which may be disrupted secondary to damage to the SVZ. Visual perception 37 required the participants to match 2-dimensional line drawings to a model. Performance on the visual perception test has been shown to be associated with temporal lobe volumes. 38 Linear mixed effect (LME) regression analyses were used to examine differences in neuropsychological test scores between patients and controls (“group”), withmoderating variables including age at the time of the baseline visit (“age 0 ”) and time since baseline (“time”). Main effects and their 2- and 3-way interactions were used in the anal- yses of the raw scores of the neuropsychological tests. Two-way interaction terms included Age 0 × Time, which examined the change in test score with age over Data Analysis

time; Group × Age 0 , which examined differences in test scores with age between the groups; and Group × Time, which examined differences in test scores with time between the groups. A 3-way interaction term (Group × Age 0 × Time) was used to examine whether the neuropsy- chological scores changed differently with age and over time between patients and controls. For LME analyses showing a significant term “time,” a general linear model ANOVA with Fisher’s least significant difference was used as a post-hoc test to evaluate the differences in neuropsychological performance between patients and controls at individual time points (visits 1–4). In patients, LME analyses were also used to evaluate the overall relationship between the raw scores of the neuropsychological tests and regional radiation doses (“dose”) to SVZ, hippocampus, temporal lobes, and cerebrum. The analyses controlled for age at the time of irradiation (age RT ) and for the presence or absence of concurrent chemotherapy. For presentation of the results in the Figures, Z -scores were used to account for the effect of age on neuropsychological performance. Because we hypothesized that higher radiation doses to specific brain regions in patients would result in a more pronounced impairment on specific neuropsychological tests, no corrections for multiple comparisons were performed. Statistical significance was set to P , .05.

NEURO-ONCOLOGY † M A R C H 2 0 1 3

362