|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Patterns of Intellectual Development Among Survivors of Pediatric Medulloblastoma: A Longitudinal AnalysisFrom the Division of Behavioral Medicine, Department of Biostatistics and Epidemiology, Department of Diagnostic Imaging, Department of Hematology/Oncology, and Department of Radiation Oncology, St. Jude Children’s Research Hospital; Department of Psychology and Departments of Electrical and Biomedical Engineering, University of Memphis; and Department of Pediatrics, University of Tennessee College of Medicine, Memphis, TN. Address reprint requests to Shawna L. Palmer, MS, Division of Behavioral Medicine, St. Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105-2794.
PURPOSE: To examine two competing hypotheses relating to intellectual loss among children treated for medulloblastoma (MB): Children with MB either: (1) lose previously learned skills and information; or (2) acquire new skills and information but at a rate slower than expected compared with healthy same-age peers. PATIENTS AND METHODS: Forty-four pediatric MB patients were evaluated who were treated with postoperative radiation therapy (XRT) with or without chemotherapy. After completion of XRT, a total of 150 examinations were conducted by use of the child version of the Wechsler Intelligence Scales. These evaluations provided a measure of intellectual functioning called the estimated full-scale intelligence quotient (FSIQ). Changes in patient performance corrected for age (scaled scores) as well as the uncorrected performance (raw scores) were analyzed. RESULTS: At the time of the most recent examination, the obtained mean estimated FSIQ of 83.57 was more than one SD below expected population norms. A significant decline in cognitive performance during the time since XRT was demonstrated, with a mean loss of 2.55 estimated FSIQ points per year (P = .0001). An analysis for the basis of the intelligence quotient (IQ) loss revealed that subtest raw score values increased significantly over time since XRT, but the rate of increase was less than normally expected, which resulted in decreased IQ scores. CONCLUSION: These results support the hypothesis that MB patients demonstrate a decline in IQ values because of an inability to acquire new skills and information at a rate comparable to their healthy same-age peers, as opposed to a loss of previously acquired information and skills.
MEDULLOBLASTOMA (MB) accounts for approximately 15% to 20% of all primary CNS neoplasms in children.1 Of the common pediatric CNS tumors, MB has the greatest tendency to disseminate to other areas of the brain. Therefore, standard therapy for MB includes treatment to address potential events in volumes of the entire subarachnoid region. Conventional treatment includes maximum surgical resection of the tumor and postoperative radiation therapy (XRT) that consists of craniospinal irradiation (CSI) followed by a boost to the posterior fossa. Dose and treatment volumes depend on the age of the child and the extent of the disease. The standard XRT dose typically has been 36 Gy to the craniospinal axis and 54 Gy to the posterior fossa delivered at a rate of 1.8 Gy per day. In young children or those with low stage or "average risk" presentations, chemotherapy has been used recently in conjunction with irradiation at a lower dose level (23.4 Gy CSI).2 Patients younger than age 3 years typically receive chemotherapy first in an attempt to delay or eliminate the need for XRT.3 The potential benefits of more recent three-dimensional conformal XRT for MB have not yet been documented.4 Survival rates for CNS tumors have slowly but steadily increased, with up to an 85% 5-year survival rate among selected pediatric MB patients treated with XRT and adjuvant chemotherapy.5,6 As treatment for MB has become more effective, greater attention has been given to examination of the long-term neuropsychologic status of childhood MB survivors. Unfortunately, psychological testing reveals that 40% to 100% of long-term brain tumor survivors have some form of cognitive dysfunction,7 and impaired intelligence has been found in nearly 90% of conventionally treated MB patients.8 It does not seem that deficits in cognitive functioning can be attributed routinely to the adverse effects of tumor or tumor surgery.9 However, higher XRT dose and young patient age at the time of treatment seem to be risk factors for lowered intelligence quotient (IQ). In a recent Pediatric Oncology Group study, survivors who received reduced-dose XRT (23.4 Gy) and those who were older (> 8.8 years) at the time of treatment demonstrated higher cognitive functioning than those who received standard dose XRT (36 Gy) and who were younger (< 8.8 years) at time of treatment.10 Declines in cognitive functioning among patients who survive pediatric MB have been demonstrated consistently in the literature,6-14 and these declines have been associated most often with XRT. Few of these studies have been longitudinal,11,14 and even among longitudinal studies, the natural history of cognitive losses has not been well defined. One of two neuropsychologic processes could account for the robust findings that document cognitive decline over time: (1) children could lose previously acquired information and skills, similar to adult dementia conditions; or (2) children could continue to acquire new information and skills but at a slower rate than healthy age-related peers. The primary objective of this study was to examine the pattern of IQ loss over time among children who survive MB. To understand the processes that underlie the decline in cognitive function, IQ raw score values and their relationship to age-adjusted scaled score values was analyzed over prolonged periods of time since the delivery of XRT. A secondary objective of this study was to examine the impact of two known risk factors for cognitive decline: young age at treatment and higher dose of the CSI component of XRT. In addition, the influence of chemotherapy was evaluated. We hypothesized that patients who were younger at the time of treatment, received higher doses of CSI, and received adjuvant chemotherapy, would demonstrate a greater rate of cognitive decline.
Patient Characteristics Patients were considered for inclusion in the study if they had histologically proven MB diagnosed before the age of 17 years, which was treated consecutively between 1984 and 1996, and they had survived 2 years or longer (n = 71). In accordance with the standard of care at our institution, each patient underwent routine psychological assessment during regularly scheduled outpatient visits. Because this study was longitudinal in design, patients with a single psychological evaluation or who had not been evaluated with the child version of the Wechsler Intelligence Scale15,16 were excluded from analysis. In addition, for patients found to have evidence of progressive disease, subsequent evaluations were excluded from analysis. These criteria resulted in a study population of 44 patients who underwent 150 evaluations. There were 28 males and 16 females. Thirty-nine patients were white, four were black, and one was Hispanic. All patients were screened routinely for hearing loss, and hearing aids were provided when warranted. In addition, all patients underwent regular endocrine function analysis and received appropriate replacement therapy when necessary.
All patients underwent surgical resection of the tumor and XRT. XRT included CSI (23.4 to 39.6 Gy at 1.5 to 1.8 Gy once daily, or 44 to 48.4 Gy at 1.1 Gy twice daily) as well as a posterior fossa boost (to a total dose of 49.2 to 55.8 Gy at 1.5 to 1.8 Gy once daily or a total dose of 66 to 70.4 Gy at 1.1 Gy twice daily). Thirty-three patients were treated with conventionally fractionated standard dose CSI, defined as doses of 35.2 to 38.4 Gy. The remainder of the patients were treated either with low-dose CSI (23.4 to 25 Gy; n = 7) or high-dose hyperfractionated CSI (
Psychometric Testing For all patients, IQ tests included the selected subtests (information, similarities, and block design), which allowed an estimation of the patient’s fund of factual knowledge, verbal abstract thinking, and nonverbal abstract thinking, respectively. Subtest raw scores were converted to age-adjusted scaled scores as indicated in the test manuals. The three subtest scaled scores were combined to calculate an estimated full-scale IQ (FSIQ).18 A subgroup of patients were given an additional seven subtests at least twice within the study period (n = 36 patients; 120 examinations). For this subgroup, measures of verbal IQ, performance IQ, and overall FSIQ were derived to provide measures of verbal comprehension and knowledge, visual-perceptual and nonverbal skills, and overall functioning, respectively.
Statistical Analysis The raw scores obtained in the MB study group were compared with established population test norms provided in the scoring tables of the Wechsler manual. The normal population slopes were calculated using linear regression analysis of the raw score values needed to maintain a mean scaled score of 10 for each subtest over time. This estimate represents precisely "normal" functioning at the 50th percentile for age.
Time since XRT was included as a continuous covariate in all analyses of the IQ test data. Patients were stratified according to median age at XRT (
Patient Characteristics Patients ranged in age from 1.08 to 12.52 years (mean, 7.57 years) at the time of diagnosis and from 1.73 to 12.88 years (mean, 7.84 years; median, 8.02 years) at the time of XRT. At the time of the most recent psychological evaluation, patients ranged in age from 7.3 to 16.9 years (mean, 13.19 years) and were 1.9 to 12.6 years past completion of XRT (mean, 5.2 years) ( Table 1).
Psychological Testing Scaled scores. Change in patient performance corrected for age (scaled scores) over time since XRT was evaluated. At the most recent examination, the mean estimated FSIQ was 83.57, which is greater than one SD below the expected population norm (Table 1). Deficits of similar magnitude were observed on the information, similarities, and block design subtests. An analysis of change over time since XRT revealed clearly that the IQ deficit was a result of an insidious decline in cognitive performance over time ( Table 2). A statistically significant decline since XRT was demonstrated for estimated FSIQ as well as for the age-adjusted information, similarities, and block design subtest scaled scores. Patients lost a mean of 2.55 points per year on the estimated FSIQ (P = .0001), 0.41 points on the information subtest scaled score (P = .0001), 0.53 points on the similarities scaled score (P = .0001), and 0.40 points on the block design scaled score (P = .0001) during the study interval. The net result of these losses was that as time since XRT increased, the mean estimated FSIQ decreased and a greater percentage of examinations had estimated FSIQ scores one SD or more below the population norm ( Table 3).
As confirmation of the pattern of loss, a statistically significant decline over time since XRT also was demonstrated on measures of verbal IQ, performance IQ, and FSIQ for the subgroup of patients who received an additional seven subtests (Table 2). Patients lost a mean of 2.66 points per year on the verbal IQ (P = .0001), a mean of 1.34 points per year on the performance IQ (P = .0005), and a mean of 2.21 points per year on the FSIQ (P = .0001). Raw scores. Changes in scores unadjusted for age (raw scores) over time since XRT also were analyzed. The declining trend in the age-adjusted scaled scores was not duplicated in the analysis of raw score values. Significant increases were demonstrated: 0.95 points per year for the information subtest (P = .001); 0.76 points per year for the similarities subtest (P = .0001); and 2.69 points per year for the block design subtest (P = .0001) (Table 2). The raw score slopes for these subtests obtained from the MB patient group were compared with the raw score slopes that would be expected in the normal population ( Fig 1). The normal population slope for each subtest was plotted by use of an intercept equivalent to the corresponding intercept derived from the MB patient group. For each subtest, the rate of increase in raw scores for the MB patient group progressed at less than normal population expectations. To maintain an average score on the information subtest, the normal population sample gains a mean of 1.53 raw score points per year. In contrast, the MB patients gained only a mean of 0.95 points per year. Likewise, the expected gain on the similarities subtest is 1.52 points per year for the normal population sample, whereas the MB patients gained only 0.76 points per year. Finally, subjects in the normal population sample gain a mean of 4.40 points on the block design subtest, although the MB patients gained only 2.69 points per year. In summary, the MB patient group acquired knowledge at only 62.1%, 49.9%, and 61.2% of the expected rate for the information, similarities, and block design subtests, respectively. As illustrated in Fig 1, the ultimate result of these trends is a widening cognitive deficit between patients treated for MB in childhood and their healthy same-age peers.
Effects of Moderator Variables There were no statistically significant relationships among the moderator variables. Therefore, tests of their effects were conducted independently. The rates of change in the age-adjusted scaled scores and estimated FSIQ were analyzed as a function of age at start of XRT, CSI dose, and administration of chemotherapy. MB patients who were younger (  8.02 years) when they received XRT seemed to be at greater risk for cognitive declines than MB patients who were older ( > 8.02 years) at XRT ( Table 4). Statistically significant greater declines over time since XRT were seen for younger MB patients on the information (P = .028) and block design (P = .046) subtests and estimated FSIQ (P = .005), but not the similarities subtest. Of interest, by visual inspection the intercepts for the MB patients who were younger at XRT were consistently lower than those obtained by patients who were older at XRT for each outcome variable, which suggests greater problems among younger patients. However, differences in the intercepts of the two age groups failed to reach statistical significance.
Patients who received CSI doses of 35.2 Gy or less (n = 30) were compared with those who received 36.0 Gy or more (n = 14). In patients who received higher doses of CSI, statistically significant greater declines over time since XRT were observed on the similarities subtest (P = .039) as well as the estimated FSIQ (P = .049), but not for the remaining outcome variables ( Table 5). Patients who received chemotherapy (n = 27) were compared with those who did not receive chemotherapy (n = 17). No statistically significant effects on cognitive outcome were found for the use of chemotherapy; both groups declined at similar rates on all scaled scores.
Patterns of intellectual development among 44 survivors of pediatric MB were examined by use of longitudinal analyses of 150 IQ examinations. In summary, the age-adjusted scaled scores of the patients confirmed a declining pattern of functioning over time since completion of XRT. The process that underlies the cognitive decline was explored through the examination of raw score patterns. Strong evidence was found that the patients continued to acquire new knowledge. However, they acquired new information and skills at only 49% to 62% of the rate expected from healthy same-age peers. Age at XRT was found to moderate the relationship between intellectual decline and time since XRT, and younger patients demonstrated greater declines in overall IQ, fund of information, and nonverbal, but not verbal, abstract thinking. Moderating effects also were found for CSI dose in overall IQ and verbal abstract thinking; patients who received higher doses of CSI demonstrated poorer performance. However, no moderating effects were found for chemotherapy status. The declining pattern of intellectual performance found in the present study reinforces previous reports of progressive intellectual deterioration over time since XRT.8,25 A recent longitudinal analysis of neurocognitive development in children who had a cerebellar tumor in infancy demonstrated a decline in IQ performance of patients who had received XRT,11 and a study of 19 patients diagnosed with MB in infancy revealed an average loss of 3.9 IQ points per year that had not yet reached a plateau.14 Although previous studies have demonstrated declining cognitive functioning, the present study is unique in its comparison of obtained versus expected performance patterns. The present analysis demonstrates that MB patients acquire new information and skills, but at a slower rate than a normal population sample of healthy same-age peers. This finding has critical implications for academic and vocational potential. Problems in academic performance are a common reason for psychological referral, and the academic failure rate is high for MB survivors. At 5 years posttreatment, it has been demonstrated that 34% of posterior fossa tumor patients were 2 or more years behind their same-age peers academically, and an additional 26% attended special schools because of inability to follow a normal curriculum.12 It also has been reported that 73% of an MB patient sample indicated problems with cognition, which makes it the most common quality of life problem, and 54.5% of the same sample needed special education services.10 A slower rate of knowledge acquisition could help to explain the poor academic performance that MB patients experience. However, further research in this area must be completed to before the mechanism that underlies the slowed rate of knowledge acquisition is fully understood. In the present study, a younger age at XRT was found to be a critical risk factor; young patients demonstrated significantly greater declines in age-adjusted scaled scores over time since XRT than their older counterparts. It has been suggested that age at XRT is a proxy variable for underlying neurodevelopmental maturity.26 Although development of cortical gray matter peaks at approximately 4 years of age, cortical white matter volume increases steadily until approximately 20 years of age.27 Therefore, patients who are younger when they receive XRT generally have less fully developed white matter. However, because both younger and older patients have been demonstrated to lose white matter volume at a similar rate,28 the younger irradiated patients continue to display reduced total white matter volume after XRT. These deficits in white matter volume among younger patients have been associated with increased intellectual morbidity.9,26,28 The present study also confirmed CSI dose effects as reported in previous studies; patients who received higher CSI doses demonstrated poorer performance than those who received lower CSI doses. A Pediatric Oncology Group study demonstrated that for children between 4 and 9 years of age, a reduced CSI dose (23.4 Gy) instead of the conventional dose (36 Gy) resulted in a measurable sparing of IQ.10 The present study has several limitations that restrict interpretation. No significant differential effects were found for the use of chemotherapy. Of the study patient group, 27 patients received chemotherapy and 17 did not. Both groups demonstrated cognitive declines at similar rates. However, the type of chemotherapy received by the study group varied widely, and it was combined with various levels of CSI dose. To examine the effects of adjuvant chemotherapy, future studies are planned in which chemotherapy status differs among patients who receive similar CSI doses. This study also focused on IQ performance and did not examine other potentially valuable cognitive functions such as memory, attention, and processing speed. Monitoring a broader scope of abilities in MB patients, to include more specific rather than global measures, may provide critical insight needed to better understand the psychological sequelae experienced by this population. A question that remains is whether the decline in intellectual ability reaches a plateau. The present study applied random-effects analysis of covariance models, and the study data limited our ability to address this issue. To answer this question, we propose consideration of a three-parameter logistic growth model,29 which would require a minimum of three observations for each patient in the study. Future studies are planned that will use these types of nonlinear mixed-effects models. In conclusion, these results suggest that pediatric MB patients need continued service and vigilant follow-up care beyond completion of medical treatment. The present study demonstrated that these patients continue to acquire new information and skills but at a slower rate than expected. Inability to keep pace with peers has far-reaching detrimental implications that extend into academic performance, future economic potential, and social ability, particularly for patients who are treated with conventional XRT at a young age. Efforts are needed to design effective treatment plans aimed at reduction of neurotoxicity and to explore methods of cognitive rehabilitation for patients who experience learning problems. Continued research in these areas has the potential to ensure a high quality of life for pediatric MB patients.
Supported in part by grant no. P30-CA21765 through a Cancer Center Support (CORE) grant from the National Cancer Institute and by the American Lebanese Syrian Associated Charities.
1. Heideman RL, Packer RJ, Albright L, et al: Tumors of the nervous system, in Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology. Philadelphia, PA, Lippincott, 1997, pp 633-681
2.
Packer RJ, Goldwein JL, 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. Ater JL, van Eys J, Woo SY, et al: MOPP chemotherapy without irradiation as primary postsurgical therapy for brain tumors in infants and young children. J Neurooncol 32: 243-252, 1997[Medline] 4. Merchant TE, Happersett L, Finlay J, et al: Preliminary results of conformal radiation therapy for medulloblastoma. Neuro-oncology 1: 177-187, 1999[Abstract] 5. Gurney JG, Smith MA, Bunin GR: CNS and miscellaneous intracranial and intraspinal neoplasms, in Ries LAG, Smith MA, Gurney JG, et al (eds): Cancer Incidence and Survival Among Children and Adolescents: United States SEER Program, 1975-1995. Bethesda, MD, National Institutes of Health, 1999, pp 51-63 6. Packer RJ, Sutton L, Elterman R, et al: Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU, vincristine chemotherapy. J Neurosurg 81: 690-698, 1994[Medline] 7. Glauser TA, Packer RJ: Cognitive deficits in long-term survivors of childhood brain tumors. Childs Nerv Syst 7: 2-12, 1991[Medline] 8. Dennis M, Spiegler BJ, Hetherington CR, et al: Neuropsychological sequelae of the treatment of children with medulloblastoma. J Neurooncol 29: 91-101, 1996[Medline] 9. Mulhern RK, Reddick WE, Palmer SL, et al: Neurocognitive deficits in medulloblastoma survivors and white matter loss. Ann Neurol 46: 834-841, 1999[Medline] 10. Mulhern RK, Kepner JL, Thomas PR, et al: Neuropsychologic functioning of survivors of childhood medulloblastoma randomized to receive conventional or reduced-dose craniospinal irradiation: A Pediatric Oncology Group study. J Clin Oncol 16: 1723-1728, 1998[Abstract]
11.
Copeland DR, deMoor C, Moore B, et al: Neurocognitive development of children after a cerebellar tumor in infancy: A longitudinal study. J Clin Oncol 17: 3476-3486, 1999 12. 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[Medline] 13. Schatz F, Kramer JH, Ablin A, et al: Processing speed, working memory and IQ: A developmental model of cognitive deficits following cranial radiation therapy. Neuropsychology 14: 189-200, 2000[Medline]
14.
Walter A, Mulhern RK, Gajjar A, et al: Survival and neurodevelopmental outcome of young children with medulloblastoma at St. Jude Children’s Research Hospital. J Clin Oncol 17: 3720-3728, 1999 15. The Psychological Corporation : Wechsler Intelligence Test for Children (revised). New York, NY, Harcourt, Brace, Jovanovich, 1974 16. The Psychological Corporation : Wechsler Intelligence Test for Children ( ed 3 ). New York, NY, Harcourt, Brace, Jovanovich, 1992 17. The Psychological Corporation : Wechsler Individual Achievement Test. New York, NY, Harcourt, Brace, Jovanovich, 1992, pp 365-377 18. Sattler JM: Assessment of Children. San Diego, CA, Jerome M. Sattler, Publisher, 1992, p 1172 19. Chambers JM, Hastie TJ: Statistical Models in Southern California. Pacific Grove, CA, Wadsworth & Brooks, 1993 20. Jones RH: Longitudinal Data with Serial Correlation: A State-Space Approach. London, England, Chapman and Hall, 1993 21. Little RC, Milliken GA, Stroup WW: SAS System for Mixed Models. Cary, NC, SAS Institute, 1996 22. Rutter CM: Analysis of longitudinal data: Random coefficient regression modeling. Stat Med 13: 1211-1231, 1994[Medline] 23. Searle SR: Linear Models for Unbalanced Data. New York, NY, Wiley, 1987 24. Venables WN, Ripley BD: Modern Applied Statistics. New York, NY, Springer, 1997 25. Roman DD, Sperduto PW: Neuropsychological effects of cranial radiation: Current knowledge and future directions. Int J Radiat Oncol Biol Phys 31: 983-998, 1995[Medline]
26.
Mulhern RK, Palmer SL, Reddick WE, et al: Risks of young age for selected neurocognitive deficits in medulloblastoma are associated with white matter loss. J Clin Oncol 19: 472-479, 2001
27.
Pfefferbaum A, Mathalon DH, Sullivan EV, et al: A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood. Arch Neurol 51: 874-887, 1994 28. Reddick WE, Russell JM, Glass JO, et al: Subtle white matter volume differences in children treated for medulloblastoma with conventional or reduced dose craniospinal irradiation. Mag Reson Imaging 18: 787-793, 2000[Medline] 29. Pinheiro JC, Bates DM: Mixed-Effects Models in S and S-PLUS. New York, NY, Springer-Verlag, 2000 Submitted September 20, 2000; accepted January 17, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
40 Gy; n = 4). Twenty-seven patients also received pre- and/or post-XRT chemotherapy (carboplatin/vincristine, prednisone [VP]-16 + cyclophosphamide/vincristine, n = 7; cisplatin/VP-16 + cyclophosphamide/vincristine, n = 7; cisplatin/VP-16, n = 3; lomustine/cisplatin/vincristine, n = 3; cyclophosphamide/vincristine/cisplatin/vincristine, n = 2; carboplatin/cyclophosphamide/VP-16, n = 1; cyclophosphamide/vincristine/cisplatin/etoposide/thiotepa, n = 2; mechlorethamine/vincristine/procarbazine/prednisone, n = 1; cyclophosphamide/vincristine/cisplatin/etoposide, n = 1). 
