Serial evaluation of academic and behavioral outcome after treatment with cranial radiation in childhood, DJ Mabbott, BJ Spiegler, ML Greenberg

Tags: diagnosis, Teacher Report, Child Behavior Checklist, medulloblastoma, Social problems, SICK CHILDRENS HOSPITAL, difficulties, standard error, Abbreviations, brain tumors, longitudinal study, teacher ratings, Wide Range Achievement Test, linear model, J Clin Oncol, assessment, standard deviation, radiation dose, Estimate, performance, academic achievement, linear models, mean score, arithmetic, behavioral adjustment, novel findings, Attention Problems, Thought problems, brain tumor, treated, adjuvant chemotherapy, Academic performance, behavioral assessment, Assessments, radiation therapy, functional outcome, intellectual decline, reading achievement, academic scores, demographic variables, young children, JOURNAL OF CLINICAL ONCOLOGY
Content: VOLUME 23 NUMBER 10 APRIL 1 2005 JOURNAL OF CLINICAL ONCOLOGY
ORIGINAL REPORT
From the Department of Psychology and Divisions of Hematology/Oncology and Neurosurgery, The Hospital for Sick Children; and Departments of Pediatrics and Surgery, University of Toronto, Toronto, Ontario, Canada. Submitted January 26, 2004; accepted December 23, 2004. Presented in part at the 11th International Symposium on Pediatric NeuroOncology, International Society for Pediatric Neuro-Oncology, Boston, MA, June 13-16, 2004. Authors' disclosures of potential conflicts of interest are found at the end of this article. Address reprint requests to D. Mabbott, PhD, Paediatric Brain Tumor Program, The Hospital for Sick Children, 555 University Ave, Toronto, Ontario, M5G 1X8, Canada; e-mail: [email protected] sickkids.ca. © 2005 by American Society of Clinical Oncology 0732-183X/05/2310-2256/$20.00 DOI: 10.1200/JCO.2005.01.158
Serial Evaluation of Academic and Behavioral Outcome After Treatment With Cranial Radiation in Childhood Donald J. Mabbott, Brenda J. Spiegler, Mark L. Greenberg, James T. Rutka, Douglas J. Hyder, and Eric Bouffet ABSTRACT Purpose To evaluate academic and behavioral outcome in radiated survivors of posterior fossa (PF) tumors. Patients and Methods Fifty-three patients (36 males) treated for malignant PF tumors were seen for evaluation of academics and/or behavioral functioning. Forty-six patients were treated for medulloblastoma, and seven patients were treated for ependymoma. Fourteen patients were treated with reduced-dose cranial radiation, and 34 patients were treated with standard-dose cranial radiation (dose was not available for four patients). All patients received an additional boost to the PF. One patient was treated with PF radiation only. Standardized achievement tests and behavioral questionnaires were administered at different times after diagnosis for each child. First, the influence of demographic and medical variables on outcome was examined. Second, the rate of change in scores was determined using mixed model regression for patients seen for serial assessment. Results The presence of hydrocephalus was related to poorer academics, but outcome was not related to radiation dose, extent of surgery, or treatment with chemotherapy. Younger age predicted poor reading ability and lower parent rating of academic achievement. Children's performance declined for spelling, mathematics, and reading. Significant declines were also evident in parent and teacher's ratings of academic ability. Behavioral functioning was generally not related to medical and demographic variables, and few clinically significant problems in externalizing behavior were evident. Increases in social and attention problems emerged over time. Conclusion Cranial radiation is associated with declines in academic ability, social skills, and attention. However, neither psychological distress nor behavior problems were a significant concern for this sample. J Clin Oncol 23:2256-2263. © 2005 by American Society of Clinical Oncology
INTRODUCTION Medulloblastoma and ependymoma are the most common malignant tumors of the posterior fossa (PF) in children. Survival rates for these tumors have improved with advances in surgery, radiation therapy, and chemotherapy.1,2 As survival rates improve, understanding the long-term func-
tional impact of therapy, particularly craniospinal radiation, is important. Survivors of pediatric brain tumors have lower rates of high school graduation and employment relative to the overall population.3,4 Although a progressive decline in intellectual functioning associated with craniospinal radiation is now well documented,5-9 changes in academic and
2256 Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Effects of Cranial Radiation in Childhood
behavioral functioning after radiation have received considerably less attention. Intelligence is certainly related to academics and, to a lesser degree, behavioral functioning. However, a unique focus on academics and behavior adds to the previous literature by providing more specific and ecologically valid information regarding psychological outcomes. As knowledge about neurocognitive late effects of cranial radiation accumulates, examining whether neurobehavioral changes are observed in real world settings, such as school and home, is increasingly important. Academic achievement is directly related to the skills and knowledge children acquire at school and, hence, is reflective of their daily functioning. Ratings of children's behavior by parents and teachers are direct measures of the child's functioning within home and school settings. Consequently, examining academic and behavioral outcome is essential for understanding children's functional capabilities, their potential for scholastic attainment and vocational success, the extent to which they can live independently as adults, and, ultimately, their quality of life. Furthermore, this information is particularly useful for developing psychosocial interventions and cognitive rehabilitation programs. For children with brain tumors, cranial radiation has been associated with school problems and poor academic performance 2 to 5 years after diagnosis based on standardized achievement tests,10-12 parent report,13 or utilization of special education services.3,10,12,14,15 A decline in academic functioning over time has also been reported for children treated with cranial radiation.14 Despite overall poor performance, no differences between children treated with or without radiation for brain tumors have been found for parent's rating of school functioning.16 In terms of behavioral adjustment, poor social functioning has been documented for heterogeneous samples of children with brain tumors, some of whom were treated with radiation.13,16 Furthermore, problems in mood and withdrawal have been reported for heterogeneous groups of children treated with and without cranial radiation,17,18 patients treated with surgery only,19 and a homogeneous sample of patients whose treatment included radiation.20 However, other studies have not documented such internalizing problems in heterogeneous samples11,16,21 and samples in which all children were treated with cranial radiation.12 Finally, it has been noted that behavioral outcome after diagnosis of and treatment for a brain tumor is related to family factors, such as family stress, and parental coping.11 In most studies of behavioral functioning, patients with different types and locations of brain tumors and who had different modalities and intensities of treatment have been considered as a single group. The use of such heterogeneous samples may obscure patterns in behavioral outcome related to specific locations or treatment modalities.
Although treatment with cranial radiation for brain tumors is associated with poor academic outcome and some difficulties in behavioral adjustment, little is known regarding the medical and demographic variables that influence that outcome, the course of outcome over time, and how that course may be related to intellectual decline. The limitations imposed by heterogeneity of samples are compounded by lack of longitudinal data. The goal of the present study is to examine patterns of academic and behavioral functioning in a homogeneous group of children treated with craniospinal radiation for PF tumors, including the impact of medical and demographic variables on outcome at 2 to 3 years after diagnosis, and patterns of change and stability over time. PATIENTS AND METHODS Patients Fifty-three patients (36 males) treated for malignant PF tu- mors between 1983 and 2004 at Toronto's Hospital for Sick Children (HSC; Toronto, Ontario, Canada) were seen for evaluation of academics and/or behavioral functioning (Table 1). Longitudinal intellectual and neurocognitive outcome for 34 of these patients (32 medulloblastoma and two ependymoma patients) have been reported elsewhere.9 Forty-six patients were treated for medulloblastoma, and seven patients were treated for ependymoma. Gross total resection was achieved in 58.5% of patients. Data on surgical outcome was not available for one patient. One patient was treated for meningitis. Twenty-one patients had ventriculoperitoneal shunts inserted to treat associated hydrocephalus. Fifty-two patients were treated with craniospinal radiation; 14 were treated with reduced-dose radiation (23.4 to 30.2 Gy), and 34 patients were treated with standard-dose radiation (34.0 to 41.4 Gy) to the whole brain. All of these patients received an additional boost to the PF, so that the total PF radiation dose ranged from 45.0 to 55.8 Gy. Information on craniospinal dose and PF boost was unavailable for four patients. One patient was treated with PF radiation only, although the field was large and included the posterior half of the brain to a dose of 54 Gy. Thirty-nine patients received adjuvant chemotherapy. Most patients were seen for their first assessment some months or years after initial diagnosis and radiation treatment, without intervening recurrence or additional treatment. Four exceptions were seen for assessment after treatment for tumor progression or recurrence. Two of these patients experienced disease recurrence as an infant before any treatment with craniospinal radiation; their first evaluations were conducted a number
Table 1. Number of Assessments Across Time
No. of Assessments (No. of patients)
Measure
1
2
3
45
Wide Range Achievement Test,
13
16
16
3
2
Revised or Third Edition
Child Behavior Checklist
21
16
6
1
0
Teacher Report Form
23
10
4
0
0
www.jco.org
2257
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Mabbott et al
of years after cranial radiation for recurrence. The other two patients, who were originally treated with craniospinal radiation, were seen for their final evaluation within weeks of their recurrence before any further treatment. Therefore, all patients were seen for academic and/or behavioral assessment after a single course of cranial radiation. The mean age at diagnosis for the entire sample was 6.60 years (standard deviation [SD], 3.22 years). The patients in this series were accrued by clinical referral. To examine potential sample biases, we compared the 43 patients (67% males) with medulloblastoma in this series who were diagnosed after 1985 with the overall population of patients diagnosed with medulloblastoma and treated with craniospinal radiation at HSC since 1985 (n 180, 66% males). (Information before 1985 is unavailable, and thus, the three patients in the current sample who were diagnosed before this year are not included in this comparison. Furthermore, the three patients treated for ependymoma are not included because many ependymomas are treated without radiotherapy, and hence, comparisons with the medulloblastoma population are not relevant.) The mean age at diagnosis was similar for the current subsample relative to the overall population (6.84 years [SD, 3.19 years] v 6.51 years [SD, 3.93 years], respectively), as was the percentage of children treated with adjuvant chemotherapy (79% v 74%, respectively). Gross total resection was achieved in 50% of patients treated since 1985 for whom information was available (n 164) and in 61% of patients in the present subsample. Postoperative mutism and speech changes were reported for 8.8% of patients in the overall sample compared with 14% of patients in the current subsample. Finally, of the 114 patients treated since 1985 for whom data was available, the majority received standard-dose radiation (74%), which is similar to the present sample (69%) for whom data was available. Thus, the current sample of children treated for medulloblastoma is similar to the overall population of children treated for medulloblastoma since 1985 for a number of important demographic and medical variables. Materials and Procedures Patients were assessed at different time points after diagnosis. Eighteen patients were seen within 12 months from diagnosis, and the remainder had their first assessment beyond 12 months after diagnosis. Thirteen patients were seen for a single assessment only, and 40 patients were seen for serial assessment. Academic performance was evaluated using the Reading, Spelling, and Arithmetic subtests of either the Wide Range Achievement Test Revised or Third Edition, parent ratings of academic performance from the Child Behavior Checklist, and teacher ratings from the Teacher Report Form. Parent and teacher ratings were based on evaluations of school performance (ie, a composite score including estimated performance of the patient across core subjects and whether the patient experienced school problems, received special education support, or repeated a grade). Behavioral and emotional adjustment was evaluated based on parent and teacher ratings from the Child Behavior Checklist and Teacher Report Form, respectively. Each of these questionnaires yields both domain-specific and general scores for internalizing (withdrawal, somatic complaints, and anxiety and depression) and externalizing (oppositional and aggressive behavior) behavior, as well as scores for social problems, thought problems, attention problems, and an overall behavioral composite. Because not all participants were given the achievement test or the standardized questionnaires at each assessment point (because of constraints of time or cooperation), the number of patients with
multiple assessments varies (Table 1). This study was approved by HSC's Research Ethics Board before the data was extracted from the clinical records. Statistical Analyses Individual achievement subtest and questionnaire scores were converted to standard scores (based on age-appropriate means and SDs from test standardization norms). For the Wide Range Achievement Test, standard scores had a mean of 100 and a SD of 15. For the Child Behavioral Checklist and Teacher Report Form, T scores were calculated with a mean of 50 and a SD of 10. Lower T scores reflect poorer ratings for academic scales on the questionnaires, whereas higher T scores reflect worse ratings on the behavioral scales. First, overall outcome at the first assessment after diagnosis was examined. Mean standard scores for each measure were calculated across 53 patients, and effects of extent of surgical resection (gross total v subtotal), insertion of a shunt (presence v absence of a shunt), chemotherapy (present v absent in treatment), and radiation dose (standard v low dose) were examined using one-way analysis of variance (ANOVA). As well, correlation coefficients were calculated for age at diagnosis and each outcome measure. Finally, there is a potential for referral bias in the patients seen for multiple evaluations; these patients may have been most impaired initially and, therefore, were seen more frequently based on clinical recommendation. Hence, mean scores for each measure were compared for patients seen for a single assessment versus those seen for serial assessment using one-way ANOVA. To correct for multiple comparisons, results were considered significant at P .01 only. Second, to determine change or stability over time, longitudinal analyses were conducted. Specifically, the rate of change in scores over time was determined using mixed model regression, which is consistent with the recent literature related to changes in intellectual functioning.5-9 Because patients were evaluated at various times after diagnosis, yielding unbalanced and missing data, the mixed model approach was used. Mixed model techniques were implemented using PROC MIXED from SAS (SAS Institute, Cary, NC). Because fewer analyses were conducted, relative to the previous ANOVAs, a less stringent criterion was necessary, and results were considered significant at the P .05 level for the longitudinal models. RESULTS Outcome After Radiation Academics. Means for reading, spelling, mathematics, and parent and teacher ratings of school performance were approximately 1 SD below normative means (Table 2). Extent of resection, use of chemotherapy, and radiation dose did not have an impact on achievement or ratings of school performance (P .10). However, patients who required a shunt to treat hydrocephalus had lower arithmetic scores relative to those who did not require a shunt (P .01); a similar trend was also evident on measures of reading and parent and teacher's ratings of school performance (P .04). Older age at diagnosis was associated with better reading (r 0.39) and higher ratings of school performance by parents (r 0.39; P .01). Finally, no
2258
JOURNAL OF CLINICAL ONCOLOGY
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Effects of Cranial Radiation in Childhood
Table 2. Academic Outcome As a Function of Medical Variables
Extent of Resection
Shunt Insertion
Chemotherapy
Radiation Dose
Overall
Subtotal
Complete
Yes
No
Yes
No
Low
Standard
No. of Mean No. of Mean No. of Mean No. of Mean No. of Mean No. of Mean No. of Mean No. of Mean No. of Mean Domain Patients Score Patients Score Patients Score Patients Score Patients Score Patients Score Patients Score Patients Score Patients Score
WRAT
Reading
49 86.41 21 88.86 27 85.37 19 79.89 29 91.00 36 86.86 12 86.00 13 82.77 33 88.55
Spelling
47 85.66 21 89.19 25 83.68 19 81.42 27 89.04 36 86.75 10 83.90 14 81.43 31 88.13
Arithmetic 48 87.75 21 92.19 26 85.77 18 79.39 29 93.14 35 87.66 12 88.33 13 82.54 33 90.15
Achenbach
Parent§
39 38.05 16 35.94 21 39.81 11 33.73 26 39.69 27 38.67 10 36.50 12 38.50 24 37.92
Teacher
38 40.26 16 41.75 21 39.38 10 36.20 28 41.71 27 40.22 9 40.78 12 39.92 24 40.75
NOTE. WRAT scores are standard scores (mean, 100; standard deviation, 15), and Achenbach scale scores are T scores (mean, 50; standard deviation, 10). Abbreviation: WRAT, Wide Range Achievement Test (Revised or Third Edition). Mean time, 2.6 years from diagnosis. Significantly different at P .04. Significantly different at P .01. §Mean time, 3.0 years from diagnosis. Mean time, 3.58 years from diagnosis.
differences were evident on any of the measures for patients seen for a single assessment relative to those patients who were subsequently seen for follow-up assessment (P .10). Behavioral and emotional adjustment. Performance across all patients ranged from 0.5 to 1 SD above normative means for parent and teacher ratings of internalizing, social, thought, and attention problems. Means for externalizing problems were within 0.5 SD from normative means (Table 3). Extent of resection, insertion of a shunt, and radiation dose did not have an impact on parent or teacher ratings of behavioral adjustment (P .10). However, patients who were not treated with chemotherapy were rated as having greater problems in anxiety and depression relative to patients who were treated with chemotherapy (62.36 v 54.39, respectively; P .01). There was no association between age
Table 3. Parent and Teacher Ratings of Behavioral Adjustment
CBCL
TRF
Achenbach Scales
No. of Patients
Mean Score
No. of Patients
Mean Score
Withdrawal
45
57.82
37
56.19
Somatic problems
45
60.18
37
58.00
Anxiety/depression
45
56.62
37
54.81
Social problems
45
58.93
37
57.22
Thought problems
45
56.51
37
52.57
Attention problems
45
56.58
37
55.46
Oppositional
45
53.67
37
53.00
Aggressive
45
53.62
37
53.24
Total
44
54.27
37
52.32
Internalizing
44
56.57
37
54.43
Externalizing
44
48.23
37
48.95
Abbreviations: CBCL, Child Behavior Checklist; TRF, Teacher Report Form. Mean time, 2.9 years from diagnosis. Mean time, 3.6 years from diagnosis.
at diagnosis and ratings of behavior (P .10). Finally, no differences were evident on any of the measures for patients seen for a single assessment compared with patients who were subsequently seen for follow-up assessment (P .10). Longitudinal Analyses Separate growth curve models were generated for the different academic and behavioral subtest scores; given the relatively small sample size, medical and demographic variables were not included as covariates. First, simple linear models were generated for the measures of reading, spelling, and arithmetic. Subsequently, Full Scale Intelligence Quotient was included in these models to examine the moderating influence of changes in intelligence. A third model was generated for the achievement measures, which included both the linear and quadratic terms. Only the simple linear model was used for parent and teacher report measures because there were too few patients with the 3 or more data points required to produce relatively stable estimates of curvilinearity. For all models, intercepts represent the estimated baseline functioning, and slopes characterize change in functioning over time. As has been previously noted when working with unbalanced longitudinal data, the predicted values for times farthest from radiation treatment are less stable because of the smaller number of data points contributing to the model at these times.5,8 Hence, caution must be used in interpreting the estimated values at the end of the time period. Unmodeled means for each academic area for 10 years after diagnosis are found in Table 4. These means are based on a different set of patients each year and differ from the predicted means yielded from the growth curve analyses that are presented in the figures. Academics. Median follow-up time for assessment of academics was 4.84 years from diagnosis (with the
www.jco.org
2259
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Mabbott et al
Table 4. Observed Means Calculated for Each Year After Diagnosis for the WRAT, CBCL Social Problems Scale, and the TRF Social Withdrawal, Social Problems, and Attention Problems Scales Over 10 Years
Year
Reading (mean score)
Spelling (mean score)
Mathematics (mean score)
Social (parent)
No. of Patients
Mean Score
Withdrawal (teacher)
No. of Patients
Mean Score
Social (teacher)
No. of Patients
Mean Score
Attention (teacher)
No. of Patients
Mean Score
1
96.31
88.50
2
86.56
85.00
3
86.00
88.10
4
80.00
82.50
5
80.00
85.09
6
74.00
75.42
7
72.14
75.13
8
81.50
81.37
9
62.00
57.50
10
68.75
77.33
97.31 92.81 88.00 80.18 77.31 71.14 71.14 81.50 67.50 73.67
9
54.11
4
50.25
4
56.50
4
54.50
12
57.83
9
52.89
9
52.44
9
50.89
16
57.06
12
57.83
11
58.36
12
55.58
6
58.83
3
52.66
4
57.75
3
51.33
8
54.75
7
57.71
7
57.57
7
60.42
3
75.33
4
57.25
4
57.00
4
57.00
5
60.00
5
58.40
5
61.20
5
54.40
6
58.50
3
71.30
3
62.00
3
67.00
3
69.67
2
59.50
2
60.00
2
61.00
1
68.00
1
68.00
1
70.00
1
69.00
NOTE. The subjects who contribute to each mean differ across time. Abbreviations: WRAT, Wide Range Achievement Test (Revised or Third Edition); CBCL, Child Behavior Checklist; TRF, Teacher Report Form.
maximum time from diagnosis being 15.25 years). For all the models, there was a significant decline in reading, spelling, and mathematics (Table 5 and Fig 1). Change in spelling can be predicted based on the following linear model: spelling 88.80 (1.37 time since diagnosis) e. Linear declines remained significant for spelling and arithmetic even when the influence of changes in intellectual function was accounted for (P .03), with the predictive model for arithmetic being as follows: arithmetic [27.46 (0.72 time varying full-scale intelligence quotient)] (0.82 time since diagnosis) e. Furthermore, the quadratic terms were significant for reading and arithmetic (P .01; Table 5 and Fig 1). A significant quadratic term reflects curvature in the slope of the function representing change over time. The predictive quadratic model for arithmetic is as follows: arithmetic 96.82 (4.96 time since diagnosis) (0.23 time since diagnosis time since diagnosis) e. Thus, for reading and arithmetic, the rate of decline from year to year decreased as time from
diagnosis increased. These findings are consistent with the view that there is an attenuation of the decline in reading and arithmetic over time. Median follow-up time for parents' and teachers' report of school function was 4.13 and 4.00 years from diagnosis (with the maximum time from diagnosis being 15.25 years), respectively. A significant decline in parents' and teachers' rating of school functioning was evident (Table 5). Behavioral and emotional adjustment. Median follow-up time was 4.17 years from diagnosis (with the maximum time from diagnosis being 15.25 years) for assessment of behavioral functioning as reported by parents. Intercepts for internalizing behaviors (social withdrawal and anxiety and depression) were generally 0.5 to 1 SD above the normative mean (Table 6), implying difficulties in affective functioning at the time of diagnosis for the sample. Few problems were noted in externalizing behavior (ie, aggressiveness and oppositional behavior). Parents' reports were relatively stable over time for most behavioral domains;
Table 5. Estimated Intercepts and Slopes for Academic Measures
Domain
No. of Patients With Multiple Assessments
Median Age at Diagnosis (years)
Time (years)
Intercept Estimate SE
Slope
Estimate
SE
Quadratic
Estimate
SE
WRAT, n 51
Reading
37
Spelling
37
Arithmetic
37
Achenbach scales
Parent, CBCL, n 42
20
Teacher, TRF, n 38
12
6.44
2.54
92.19
3.02 3.57 0.81
0.23
0.06
6.44
2.54
88.80
2.41 1.37 0.26
--
--
6.44
2.54
96.82
2.95 4.96 0.96
0.23
0.07
6.14
3.22
41.35
1.50 1.13 0.25
--
--
6.08
3.58
43.53
1.44 0.82 0.27
--
--
Abbreviations: WRAT, Wide Range Achievement Test (Revised or Third Edition); SE, standard error; CBCL, Child Behavior Checklist; TRF, Teacher Report Form. Median time after diagnosis to assessment when the test was first used. WRAT scores are standard scores (mean, 100; standard deviation, 15), and Achenbach scales scores are T scores (mean, 50, standard deviation, 10). P .01.
2260
JOURNAL OF CLINICAL ONCOLOGY
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Effects of Cranial Radiation in Childhood
Fig 1. Estimated stability or decline in standard scores for spelling, arithmetic, and reading subtests from the Wide Range Achievement Test (Revised or Third Edition) over time.
however, a significant increase in the index related to social problems was evident over the modeled time period (Fig 2 and Table 6). Median follow-up time was 4.17 years from diagnosis (with the maximum time from diagnosis being 12.33 years) for assessment of behavioral functioning as reported by teachers. Significant increases in attention difficulties, social problems, and social withdrawal were evident over the modeled time period (Fig 2 and Table 7). By the end of the modeled time period, estimated scores for these domains were approaching the clinically significant range. DISCUSSION We documented a number of novel findings relevant for understanding academic and behavioral outcome in children treated with cranial radiation for malignant brain tumors. First, consistent with previous research, we documented poor academic scores on Standardized tests and on parent and teacher ratings of school performance. However, we also documented the influence of a number of demographic and medical variables on outcome. Academic outcome was poorer for children who required the inser-
Fig 2. Estimated decline in the Child Behavior Checklist Social Problems scale and the Teacher Report Form Social Withdrawal, Social Problems, and Attention Problems scales over time. tion of a shunt for the control of hydrocephalus. Extent of surgery and treatment with chemotherapy did not influence outcome. Notably, no differences in academic outcome were evident for children treated with standard-dose versus reduced-dose radiation. Finally, younger age at diagnosis was associated with poorer parent and teacher ratings of overall academic functioning and reading achievement. A novel contribution of the present study using growth curve analyses is that children fall progressively behind their peers in reading, spelling, and mathematics achievement. We note that these declines in standard scores do not reflect loss of skills but, rather, a reduced rate of skill acquisition. The declines in spelling and mathematics scores remain evident even when the effects of intellectual decline are considered. Hence, poorer academic outcome does not solely reflect the influence of decline in intelligence. This is notable because it appears that factors other than a decline in intelligence contribute to declining academic scores. Perhaps the secondary effects of being treated for a brain
www.jco.org
2261
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Mabbott et al
Table 6. Estimated Intercepts and Slopes for the CBCL
Domain
No. of Patients (n 45)
Intercept
Estimate
SE
Slope
Estimate
SE
Withdrawal
23
Somatic complaints
23
Anxiety/depression
23
Social problems
23
Thought problems
23
Attention problems
23
Oppositional behavior
23
Aggressive behavior
23
Internalizing
23
Externalizing
23
Total
23
57.63
1.47
0.01
0.26
60.45
1.82
0.30
0.32
56.08
1.23
0.13
0.19
54.54
1.46
1.17
0.25
56.08
1.44
0.004
0.25
56.04
1.48
0.35
0.25
53.69
1.13
0.01
0.20
54.22
1.28
0.24
0.21
56.59
2.16
0.13
0.35
49.15
1.98
0.23
0.30
54.39
1.95
0.02
0.30
NOTE. Median age at diagnosis, 4.83 years; median time to assessment when the test was first used, 2.17 years. Abbreviations: CBCL, Child Behavior Checklist; SE, standard error. P .01.
tumor, including fatigue and time missed from school because of health concerns and medical appointments, have an impact on academic achievement, and such variables should be examined in future studies. Finally, the rates of decline in reading and arithmetic scores are attenuated over time. Because children demonstrated progressive academic problems soon after diagnosis and treatment, our findings highlight the need for intervention programs to be instituted early in the treatment course if academic difficulties are to be ameliorated in this population. Educational intervention has been shown to be effective in addressing academic delays in children treated with cranial radiation for leukemia.22 Future studies should further examine the relations between neurocognitive decline and academic decline,23 as mediated by such factors as time missed from school, which grades were missed, and individual and fam-
ily coping. This will help identify which patients are at greatest risk and which patients may most benefit from early intervention. Second, consistent with previous findings, parents and teachers endorsed more internalizing versus externalizing behavioral symptoms. As a group, however, children treated with radiation for brain tumors demonstrated only moderately elevated scores on internalizing measures reflecting psychological distress. Furthermore, they did not exhibit an increase in internalizing or externalizing problems over time. It would seem that psychological problems are not universally experienced by survivors of pediatric brain tumors16; rather, behavioral outcome may depend on many factors, including family stress and parental coping.11 Such findings are reported after childhood-acquired brain injury where recovery is related to family environment.24,25
Table 7. Estimated Intercepts and Slopes for the TRF
Domain Withdrawal Somatic complaints Anxiety/depression Social problems Thought problems Attention problems Oppositional behavior Aggressive behavior Internalizing Externalizing Total
No. of Patients (n 37) 14 14 14 14 14 14 14 14 14 14 14
Intercept
Estimate
SE
52.87
1.78
57.59
1.71
54.58
1.33
54.75
1.55
52.49
1.28
52.69
1.65
51.92
1.18
52.27
1.37
52.23
2.19
47.64
1.97
49.03
1.99
Slope
Estimate
SE
0.98
0.34
0.16
0.32
0.19
0.25
0.71
0.35
0.08
0.24
0.86
0.31
0.27
0.22
0.32
0.25
0.62
0.41
0.49
0.37
1.08
0.38
NOTE. Median age at diagnosis, 5.67 years; median time to assessment when the test was first used, 2.17 years. Abbreviations: TRF, Teacher Report Form; SE, standard error. P .01. P .03.
2262
JOURNAL OF CLINICAL ONCOLOGY
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.
Effects of Cranial Radiation in Childhood
Considering that the child's family environment may have an impact on outcome, an intervention program that fosters competence, coping, and resilience in at-risk families may be important in ameliorating the long-term challenges faced by survivors of pediatric brain tumors. A unique contribution of the present study is that teachers report increasing difficulties in attention over time. Psychopharmacologic and cognitive rehabilitation interventions seem to be effective for addressing such difficulties in children treated for cancer,26,27 and our findings support the need for these interventions. Furthermore, parents and teachers reported increasing social problems over time. Hence, these findings also support the implementation of programs designed to address social skills. Finally, our results must be interpreted in the context of common limitations of Clinical Research. First, this study is limited by the relatively small sample size for the longitudinal analyses of the behavioral adjustment data (ie, Child Behavior Checklist and Teacher Report Form), and these findings must be interpreted cautiously. Second, because a clinical sample was studied, the issue of ascertainment bias is relevant. Although our sample was similar to the overall
population treated at HSC for a number of medical and demographic variables, and no differences were evident between patients seen for only a single assessment versus multiple assessments, these clinically obtained data may not reflect what might have been found had all children been observed prospectively. Third, a common constraint of clinical studies that is not unique to the present study is the absence of a comparison group. For example, the relative effects of surgery alone versus surgery and radiation cannot be evaluated, and conclusions regarding the impact of radiation must be considered within these constraints. Finally, variables that mediate the changes in outcome after treatment should be examined in future research. For example, the effects of age at diagnosis and sex and a number of family (eg, stress and parental coping) and school (eg, number of days missed and which grades missed) variables will be important for better understanding the risk factors associated with changes in functional outcome in this group of high-risk patients.
Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest.
REFERENCES 1. Strother D, Pollack I, Fisher P, et al: Tumors of the central nervous system, in Pizzo P, Poplack D (eds): Principles and Practice of Pediatric Oncology (ed 4). New York, NY, Lippincott Williams & Wilkins, 2002, pp 751-824 2. Packer RJ, Goldwein J, Nicholson HS, et al: Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: A Children's Cancer Group Study. J Clin Oncol 17:2127-2136, 1999 3. Hoppe-Hirsch E, Brunet L, Laroussinie F, et al: Intellectual outcome in children with malignant tumors of the posterior fossa: Influence of the field of irradiation and quality of surgery. Childs Nerv Syst 11:340-346, 1995 4. Kelaghan J, Myers MH, Mulvihill JJ, et al: Educational achievement of long-term survivors of childhood and adolescent cancer. Med Pediatr Oncol 16:320-326, 1988 5. Copeland DR, deMoor C, Moore BDI, et al: NeuroCognitive Development of children after cerebellar tumor in infancy: A longitudinal study. J Clin Oncol 17:3476-3486, 1999 6. Palmer SL, Goloubeva O, Reddick WE, et al: Patterns of intellectual development among survivors of pediatric medulloblastoma: A longitudinal analysis. J Clin Oncol 19:2302-2308, 2001 7. Palmer S, Gajjar A, Reddick WE, et al: Predicting intellectual outcome among children treated with 35-40 Gy craniospinal irradiation for medulloblastoma. Neuropsychology 17:548-555, 2003 8. Ris D, Packer R, Goldwein J, et al: Intellectual outcome after reduced-dose radiation therapy plus adjuvant chemotherapy for medulloblastoma: A Children's Cancer Group Study. J Clin Oncol 19:3470-3476, 2001
9. Spiegler BJ, Bouffet E, Greenberg ML, et al: Change and stability in neurocognitive functioning following treatment with cranial radiation in childhood. J Clin Oncol 22:706-713, 2004 10. Johnson DL, McCabe MA, Nicholson HS, et al: Quality of long-term survival in young children with medulloblastoma. J Neurosurg 80: 1004-1010, 1994 11. Carlson-Green B, Morris RD, Krawiecki N: Family and illness predictors of outcome in pediatric brain tumors. J Pediatr Psychol 20:769-784, 1995 12. Seaver E, Geyer R, Sulzbacher S, et al: Psychosocial adjustment in long-term survivors of childhood medulloblastoma and ependymoma treated with craniospinal irradiation. Pediatr Neurosurg 20:248-253, 1994 13. Carpentieri SC, Mulhern RK, Douglas S, et al: Behavioral resiliency among children surviving brain tumors: A longitudinal study. J Clin Child Psych 22:236-246, 1993 14. Packer JR, Sutton LN, Atkins TE, et al: A prospective study of cognitive function in children receiving whole-brain radiotherapy and chemotherapy: 2-year results. J Neurosurg 70:707-713, 1989 15. Radcliffe J, Packer JR, Atkins TE, et al: Three- and four-year cognitive outcome in children with noncortical brain tumors treated with wholebrain radiotherapy. Ann Neurol 32:551-554, 1992 16. Radcliffe J, Bennett D, Kazak AE, et al: Adjustment in childhood brain tumor survival: Child, mother, and teacher report. J Pediatr Psychol 21:529-539, 1996 17. LeBaron S, Zeltzer PM, Zeltzer LK, et al: Assessment of quality of survival in children with medulloblastoma and cerebellar astrocytoma. Cancer 62:1215-1222, 1988 18. Mulhern RK, Carpentieri S, Shema S, et al: Factors associated with social and behavioral problems among children recently diagnoses with brain tumor. J Pediatr Psychol 18:339-350, 1993
19. Meyer EA, Kieran MW: Psychological adjustment of `surgery-only' pediatric neurooncology patients: A retrospective analysis. Psychooncology 11:74-79, 2002 20. Taphoorn MJB, Schiphorst AK, Snoek FJ, et al: Cognitive functions and quality of life in patients with low-grade gliomas: The impact of radiotherapy. Ann Neurol 36:48-54, 1994 21. Carpentieri SC, Mulhern RK: Patterns of memory dysfunction among children surviving temporal lobe tumors. Arch Clin Neuropsychol 8:345-357, 1993 22. Anderson VA, Godber T, Smibert E, et al: Cognitive and academic outcome following cranial irradiation and chemotherapy in children: A longitudinal study. Br J Cancer 82:255-262, 2000 23. Reddick WE, White HA, Glass JO, et al: Developmental model relating white matter volume to neurocognitive deficits in pediatric brain tumor survivors. Cancer 97:2512-2519, 2003 24. Taylor HG, Drotar D, Wade S, et al: Recover from traumatic head injury in children: The importance of the family, in Broman SH, Michel ME (eds): Traumatic Head Injury in Children. New York, NY, Oxford University Press, 1995, pp 188-216 25. Yeates KO, Taylor HG, Drotar D, et al: Preinjury family environment as a determinant of recovery from traumatic brain injuries in school-age children. J Int Neuropsychol Soc 3:617-630, 1997 26. Butler RW, Copeland DR: Attentional processes and their remediation in children treated for cancer: A literature review and the development of a therapeutic approach. J Int Neuropsychol Soc 8:115-124, 2002 27. Thompson SJ, Leigh L, Christensen R, et al: Immediate neurocognitive effects of methylphenidate on learning-impaired survivors of childhood cancer. J Clin Oncol 19:1802-1808, 2001
www.jco.org
2263
Information downloaded from jco.ascopubs.org and provided by at SICK CHILDRENS HOSPITAL on August 25, 2015 from Copyright © 2005 American So1c4ie2t.y20o.f2C0l.i1n9ic3al Oncology. All rights reserved.

DJ Mabbott, BJ Spiegler, ML Greenberg

File: serial-evaluation-of-academic-and-behavioral-outcome-after-treatment.pdf
Author: DJ Mabbott, BJ Spiegler, ML Greenberg
Published: Mon Mar 21 15:59:04 2005
Pages: 8
File size: 0.2 Mb


INTRODUCING CHILDREN TO, 3 pages, 0.13 Mb

Folk music, 7 pages, 0.14 Mb

Heart to Heart♥, 42 pages, 1.03 Mb
Copyright © 2018 doc.uments.com