paediatrics Brussels 17

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

reference (http: // headneckbrainspine.com). To mini- mize interindividual differences in measurement, all con- tours were drawn by a single physician. The initial contours were then reviewed by a second physician to confirm their accuracy. Figure 1 shows representative contours of the hippocampus on an axial T1-weighted MRI postgadolinium contrast that has been fused to the treatment-planning CT scan. The mean radiation dose to each of these structures was calculated from the dose volume histogram of the restored radiation treatment plan using the Pinnacle planning system (Philips). The neuropsychological assessment was designed to provide a delineation of selected neurobehavioral func- tions, using nationally standardized tests with good test-retest reliability that have been validated in the age range of interest. For this prospective study, we empha- sized assessment of memory and motor speed, consid- ered to be most sensitive to radiation effects to NPCs in the hippocampus and SVZ, respectively. Testing was performed by a trained master’s-level psychology associ- ate or postdoctoral neuropsychology fellow under the supervision of a board-certified neuropsychologist in an outpatient clinic. Control participants were screened for psychiatric disorders using the Diagnostic Interview for Children and Adolescents, Fourth Edition. 32 Neuropsychological testing was performed at baseline and at approximately 6, 15, and 27 months following completion of RT in patients and at matched time points for controls. Not all patients were evaluated on all tests at all time points, and missing patients varied among time points. Additionally, some of the youngest patients were not tested until the later time points. The neuropsychological protocol was as follows: Verbal memory was assessed using the Memory for Words Test, 33 a measure of short-term auditory verbal memory / learning. Tests of declarative learning and rec- ognition vocabulary are considered to be dependent on the integrity of subcortical systems, including the thala- mus and hippocampus. Vocabulary was measured via the Peabody Picture Vocabulary Test, third edition, 34 which requires the par- ticipant to identify a picture that best describes a word by pointing to it or verbalizing its number. Tests of vo- cabulary are considered to be dependent on temporal lobe functioning, particularly that of the left temporal lobe. Working memory was assessed using (i) the Bead Memory Test, 35 a measure of visual-spatial working memory in which participants look at a picture of beads in a given pattern and then reproduce the pattern from memory; and (ii) the Auditory Working Memory assessment (Woodcock Johnson version III [WJ-III]), a measure of auditory verbal memory in which patients are asked to repeat lists of unrelated words. Working memory tests are considered to be Neuropsychological Measures

zone (SVZ) of the lateral ventricles and the dentate gyrus of the hippocampus. 15 – 17 NPCs are critical to recovery of the CNS from damage, including RT-associated injury. 18 , 19 However, NPCs and their progenitor stem cell populations are highly radiosensitive. 8 , 20 – 22 Emerging data suggest that the human brain has neuro- genic areas similar to the rodent brain 17 that may be associated with neurocognitive toxicity following RT and chemotherapy. Limited retrospective human studies suggest an asso- ciation between radiation dose to the hippocampus and temporal lobes and neurocognitive deficits following cranial irradiation. 23 – 25 The relationship between radia- tion dose to the SVZ and neurocognitive sequelae remains controversial. Although prophylactic intrathe- cal methotrexate administration in children with acute lymphoblastic leukemia has been associated with neuro- logic toxicity, 26 – 29 suggesting a potential relationship with injury to cells immediately adjacent to the ventri- cle, 30 retrospective data have not demonstrated a corre- lation between RT dose to the SVZ and neurocognitive decline following radiation therapy. 31 We present one of the first prospective studies to examine the relationship between RT dose to NPC niches, temporal lobes, and cerebrum and neurocogni- tive dysfunction in children following cranial irradiation for brain tumors. The a priori hypothesis of this study was that increased radiation dose to the temporal lobes and NPC-containing niches would be associated with decreased performance on follow-up neurocognitive testing, specifically on tests of memory, executive func- tion, and motor dexterity. Children ( n ¼ 19) ages 1–18 years at the time of radia- tion to the brain for tumors of any histology or prophy- lactic cranial irradiation were eligible for enrollment in this prospective study approved by the institutional review board at The Johns Hopkins Hospital. Procedures were followed in accordance with the ethical standards of the Helsinki Declaration of the World Medical Association. Written consent was ob- tained prior to enrollment in the study. The comparison group ( n ¼ 55) were healthy, typically developing chil- dren with no history of psychiatric disorder, neurologic illness, or learning disability. The SVZ, hippocampus, temporal lobes, and cerebrum were manually contoured using the treatment-planning CT scan and co-registered T1-weighted postgadolinium contrast and fluid attenuated inversion recovery (FLAIR) MRIs. The SVZ was defined as a 5-mm region adjacent to the lateral wall of the lateral ventricle. The contours of the hippocampus, temporal lobes, and cerebrum were delineated using an online radiographic atlas as a Materials and Methods Study Population and Eligibility Radiation Dose to Brain Structures

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

NEURO-ONCOLOGY † M A R C H 2 0 1 3

361