paediatrics Brussels 17

979

Volume 92 Number 5 2015

Cognitive performance in the PNET4 study

Summary In the PNET4 randomized controlled treatment trial, cognitive performance of children and young adults with standard risk medullo- blastoma allocated to un- dergo hyperfractionated radiation therapy (HFRT) followed by standard chemotherapy was compared to that of subjects allocated to receive standard radiation therapy (STRT) followed by standard chemotherapy regimen. Treatment with HFRT was associated with a trend toward better verbal outcomes in children younger than 8 years of age at diagnosis, but no signifi- cant differences in other cognitive measurements.

Purpose: In the European HIT-SIOP PNET4 randomized controlled trial, children with standard risk medulloblastoma were allocated to hyperfractionated radiation therapy (HFRT arm, including a partially focused boost) or standard radiation therapy (STRT arm), followed, in both arms, by maintenance chemotherapy. Event- free survival was similar in both arms. Previous work showed that the HFRT arm was associated with worse growth and better questionnaire-based executive function, especially in children < 8 years of age at diagnosis. Therefore, the aim of this study was to compare performance-based cognitive outcomes between treatment arms. Methods and Materials: Neuropsychological data were collected prospectively in 137 patients. Using the Wechsler Intelligence Scales, Kaufman Assessment Battery for Children, and Raven’s Progressive Matrices, we estimated full-scale intelligence quo- tient (FSIQ) and, when available, verbal IQ (VIQ), performance IQ (PIQ), working memory index (WMI), and processing speed index (PSI). Results: Among the 137 participants (HFRT arm n Z 71, STRT arm n Z 66, 63.5% males), mean ( SD) ages at diagnosis and assessment respectively were 9.3 ( 3.2) years of age (40.8% < 8 years of age at diagnosis) and 14.6 ( 4.3) years of age. Mean ( SD) FSIQ was 88 ( 19), and mean intergroup difference was 3.88 (95% confidence interval: 2.66 to 10.42, P Z .24). No significant differences were found in children > 8 years of age at diagnosis. In children < 8 years of age at diagnosis, a marginally significant trend toward higher VIQ was found in those treated in the HFRT arm; a similar trend was found for PSI but not for PIQ, WMI, or FSIQ (mean intergroup dif- ferences were: 12.02 for VIQ [95% CI: 2.37-21.67; P Z .02]; 3.77 for PIQ [95% CI: 5.19 to 12.74; P > .10]; 5.20 for WMI [95% CI: 2.07 to 12.47; P > .10]; 10.90 for PSI [95% CI: 1.54 to 23.36; P Z .08]; and 5.28 for FSIQ [95% CI: 4.23 to 14.79; P > .10]). Conclusions: HFRT was associated with marginally higher VIQ in children < 8 years of age at diagnosis, consistent with a previous report using questionnaire-based data. However, overall cognitive ability was not significantly different. 2015 Elsevier Inc. All rights reserved.

fact that proliferating tumor cells are more sensitive than normal tissue to a given dose of RT if it is administered in a larger number of fractions of smaller size. This enhances the antitumor effects of RT while sparing normal tissues (19-22) . Compared with standard fractionated RT (STRT), HFRT can be used either to maintain a given antitumor effect while decreasing unwanted effects on the CNS or to increase the antitumor effect without increasing unwanted effects on the CNS. Previous uncontrolled studies by Carrie et al (22) and Gupta et al (23) reported higher posttreatment full-scale IQ in patients receiving twice-daily HFRT than that in historical controls receiving once-daily STRT. However, using historical controls instead of a controlled experimental randomized design limits interpretation of these data. Furthermore, we could hypothesize that the lack of a significant IQ decline could be related to improved quality of posterior fossa irradiation, even in STRT, with less ra- diation to the temporal and occipital lobes. The HIT-SIOP PNET4 phase 3 European randomized controlled treatment trial (RCT) for MB was designed to investigate the hypothesized biological advantage of HFRT relative to STRT. Five-year event-free survival was similar between the 2 arms (24) . A subsequent cross-sectional study

Introduction

Extensive research has consistently recognized longitudinal impairments associated with medulloblastoma (MB), the most frequent malignant brain tumor of the central nervous system (CNS) during childhood (1-3) . Standard treatment includes surgical resection, postoperative radiation therapy (RT) and adjuvant chemotherapy. MB survivors experience significant health-related problems, namely endocrine and growth morbidity and reduced fertility (4, 5) , second tu- mors (6) , hearing loss (7) , and long-term neurological deficits (8-10) . Among the major complications arising from the tumor and its treatment, predominantly RT and especially when given with chemotherapy, are the high rate of neurocognitive deficits, possibly attributable to the deleterious effects of radiation on white matter develop- ment (11, 12) . MB survivors typically achieve scores below the mean for age-matched peers in measurements of intel- ligence quotient (IQ), verbal and performance IQ (VIQ, PIQ), processing speed index (PSI), working memory index (WMI), and sustained attention (13-16) . Importantly, defi- cits in these core cognitive domains tend to worsen over time (16-18) . To improve tumor control and quality of survival, hyperfractionated RT (HFRT) capitalizes on the