|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Cognitive Outcome of Long-Term Survivors of Multisystem Langerhans Cell Histiocytosis: A Single-Institution, Cross-Sectional Study
From the Department of Haematology/Oncology and Developmental Cognitive Neuroscience Unit, The Institute of Child Health and Great Ormond Street Hospital, London, United Kingdom; and Department of Oncology and Haematology, Royal Hospital for Sick Children, Edinburgh, Scotland, United Kingdom. Address reprint requests to Vasanta R Nanduri, MRCP, MRCPCH, Department of Paediatrics, Watford General Hospital, Vicarage Rd, Watford WD1 8HB, United Kingdom; email: vasanta.nanduri{at}whht.nhs.uk.
Purpose: Damage to the CNS, including the cerebellum, and to the hypothalamopituitary axis, is documented in Langerhans cell histiocytosis (LCH). Neuropsychologic deficits have been recognized, but this is the first study in which cognitive function has been systematically assessed in a cohort of patients. Patients and Methods: Twenty-eight long-term survivors of multisystem LCH (mean age, 15.1 years) were investigated for intelligence, memory and learning, language, and academic attainments. Results: The mean intelligence quotient (IQ) of the entire group was not significantly different from the mean of the population (ie, mean ± SD, 100 ± 1), but there were wide ranges (Full-Scale IQ [FSIQ]: mean, 93.6; range, 61.7 to 134; Performance IQ [PIQ]: mean, 92.2; range, 46 to 136; and Verbal IQ [VIQ]: mean, 93.7; range, 64.2 to 126). CNS involvement was a significant risk factor for lower scores, but sex, diabetes insipidus, and cranial radiotherapy were not. The CNS group had lower VIQ, PIQ, and FSIQ than patients with no CNS involvement (no CNS group: mean ± SD FSIQ, 102.3 ± 15.6; CNS group: mean ± SD FSIQ, 73.6 ± 7.7; P < .001). A similar pattern of results was obtained for all other cognitive measures. Even when effects of reduction in FSIQ were taken into account, specific deficits were found in patients in the CNS group. Conclusion: Long-term survivors of multisystem LCH, particularly patients with CNS involvement, may develop significant cognitive deficits. All patients should have formal, repeated neuropsychologic assessment as part of long-term follow-up, which will enable abnormalities to be detected early so that appropriate supportive measures can be offered.
LANGERHANS CELL histiocytosis (LCH) is a granulomatous disorder that has been known over the years by several names, including Letterer-Siwe disease, Hand-Schüller-Christian disease, eosinophilic granuloma, and histiocytosis X.1 It results from an abnormal proliferation of pathologic Langerhans cells, which accumulate along with other inflammatory cells such as eosinophils, neutrophils, macrophages, and lymphocytes in various tissues.2 The lesions are destructive, and healing results in scarring and fibrosis. Involvement of the brain by LCH has been recognized for many years. In a comprehensive review of CNS involvement by histiocytosis X, Kepes3 described the wide range of clinical presentations associated with abnormal histopathologic findings in different parts of the brain. Involvement of the hypothalamopituitary axis as a result of direct infiltration of granulomatous tissue is the most common manifestation described, with diabetes insipidus (DI) occurring in up to half of patients with multisystem LCH.4–7 Less is known about the incidence and pattern of involvement of the rest of the CNS, and at present, there is no universally accepted definition of CNS involvement. The LCH-CNS study of the Histiocyte Society collected information on 38 patients with abnormal neurologic symptoms and signs from 27 institutions worldwide.8 Although neurologic symptoms occasionally preceded the diagnosis of LCH, they usually appeared several years (median, 4 years 6 months) after the original diagnosis had been made. Gliosis and demyelination were the characteristic histologic findings. The cause of this late-presenting, demyelinating disease is not understood, although several explanations have been proposed. The infiltrate of active LCH may have been present at an earlier phase but has burnt out, leaving scarring as seen in other organs, such as the lung9 and the liver.10 Alternatively, it has been indicated that there could be similarities between the pathology of CNS involvement in LCH and the pathology of human immunodeficiency virus–related dementia,11 in which there is excessive activation of glutamate receptors by neurotoxins or cytokines associated with neuronal injury and cell death.12 At present, there is no substantive evidence to support any of these theories, and further studies of the clinical spectrum, natural history, and etiopathogenesis of CNS involvement in LCH are required. There are no systematic studies of cognitive outcome in patients with LCH. Previous studies have reported intellectual deficits in individual patients with CNS involvement13–15 or as just one aspect of long-term outcome investigations.6,16 There is only one published abstract17 and one case report18 available in which cognitive function is specifically addressed. In this study, we report the cognitive outcome in a large cohort (n = 28) of long-term survivors of multisystem LCH from a single institution.
Patients The assessments were performed as part of a long-term follow-up study of survivors of multisystem LCH, undertaken with ethical committee approval from our institution. To be included in the study, patients had to be registered at Great Ormond Street Hospital (London, United Kingdom), have biopsy-proven LCH, and be long-term survivors (> 5 years from end of treatment). Forty patients were eligible for this study, of whom 28 (16 males and 12 females) agreed to undergo neuropsychologic investigations. The 12 patients who did not complete cognitive testing (lived abroad, n = 1; were unable to attend because of work commitments or distance, n = 4; or failed to attend appointments, n = 7) were no different on diagnostic variables, treatment received, time elapsed, or sequelae in other systems. The age at diagnosis of LCH was 6 weeks to 15.6 years (mean age, 16 months; median, 18.2 months). Treatment had included one or more of the following: biopsy or curettage of lesions or both in 19 patients; oral corticosteroids in 21 patients, with cumulative doses of 38 mg/kg to 1,000 mg/kg (median dose, 150 mg/kg) over a duration of 2 months to several years; chemotherapy (etoposide, vinblastine, vincristine, cyclophosphamide, and methotrexate) in 13 patients; other immunosuppressive agents (azathioprine) in six patients; and radiotherapy to the region of the head in four patients. Two patients had received 12 Gy of radiotherapy to the pituitary region at ages 8 and 10 years, one patient had received 12 Gy to the orbit at age 6 years, and one patient had received 12 Gy to the pituitary and an additional 3 Gy boost to an occipital skull lesion at 6 years. At the time of these studies, patients ranged in age from 6 years to 27 years (mean ± SD age, 15.1 ± 4.8 years), and as required by the entry criteria, all were more than 5 years beyond the end of treatment for active disease (mean ± SD number of years after treatment, 13.3 ± 4.4 years).
Clinical Examination
Neuropsychologic Assessment Intelligence. The age-appropriate Wechsler Intelligence Scales were administered during each assessment to evaluate intellectual ability. Children aged 2 years to 5 years 11 months were assessed on the Wechsler Preschool and Primary Scale of Intelligence–Revised (WPPSI-R; n = 1).23 The Wechsler Intelligence Scale for Children–Third Edition (WISC-III)24 was used to assess all children aged 6 years to 16 years 11 months (n = 14), and children older than 17 years received the Wechsler Adult Intelligence–Revised Scale (WAIS-R; n = 13).25 Raw scores from the Information, Similarities, Arithmetic, Vocabulary, and Comprehension subtests on the WPPSI-R, WISC-III, and WAIS-R were standardized to provide estimates of Verbal Intelligence Quotient (VIQ). Raw scores from the Picture Completion, Block Design, Object Assembly, Coding (WISC-III/WAIS-R), and Picture Arrangement subtests (WISC-III/WAIS-R) were standardized to provide estimates of Performance IQ (PIQ). Both VIQ and PIQ have a mean of 100 and an SD of 15. Memory and learning. The Wechsler Memory Scale (WMS)26 was used to assess memory and learning. Age corrections similar to those used for individuals over 20 years old were used.27 In addition to the overall memory quotient (MQ), two subtests were selected to provide measures of immediate and delayed recall for verbal and visual information (Logical Memory and Visual Reproduction),27,28 and one subtest was selected to provide a measure of verbal learning and delayed recall (Paired Associate Learning). For the Logical Memory component, children recalled two stories immediately after presentation, and an average score was calculated. After a 90-minute interval filled with other tasks, delayed recall of each story was obtained, and an average score was calculated. Children less than 12 years of age (n = 6) received two stories developed by Kimura and McGlone,27 while children older than 12 years of age (n = 19) received the two standard Wechsler stories constituting the Logical Memory subtest (Form 1). Visual Reproduction involved the reproduction of the geometric designs as described in the WMS manual. After a 40-minute delay, subjects were asked to reproduce the designs from memory. Paired Associate Learning was assessed using the 10 word pairs described in the WMS manual. This set consisted of six related items (eg, up/down) and four unrelated pairs (eg, cabbage/pen). Word pairs were learned over three trials. After a 90-minute filled delay, subjects were again asked to recall the word pairs. The Children’s Auditory Verbal Learning Test–Second Edition29 was administered to provide standardized measures of immediate memory span, level of learning, interference with learning, and immediate and delayed recall. Raw scores for adults older than 18 years of age (n = 6) were standardized to age-appropriate scores for children aged 17 years to 17 years 11 months. Language. The British Picture Vocabulary Scale30 was administered to provide a measure of receptive vocabulary for words and concepts. The Wechsler Objective Language Dimensions31 measure includes the following three subtests: Listening Comprehension, Oral Expression, and Written Expression. Academic attainments. The Wechsler Objective Reading Dimensions32 measure includes the following three subtests: Basic Reading, Spelling, and Reading Comprehension. The Wechsler Objective Numerical Dimensions33 measure includes the following two subtests: Mathematical Reasoning and Numerical Operations.
Neuroimaging All patients had received either a computed tomography scan or MRI at diagnosis of DI. Because this long-term study spanned over 20 years, only computed tomography scanning was available for patients diagnosed in the earlier years.
Eight (28%) of the 28 patients had clinical abnormalities (ataxia and incoordination, psychologic problems, and learning difficulty) or MRI evidence of CNS involvement or both. These problems were not evident in any patient at diagnosis of LCH and were not present during the active phase of the illness. The time from diagnosis of LCH to the first symptom suggestive of CNS damage ranged from 8 to 168 months (mean, 78 months; median, 81 months). Patients with isolated DI, without neurologic or behavioral abnormalities, were not classified as having CNS involvement in our study.
Cerebellar Involvement
Behavioral and Psychologic Problems
MRI Changes
Cognitive Outcome
Examination of individual scores showed that all but three of the patients who had scores below the normal range had evidence of CNS involvement. Patients were then divided into groups with and without CNS involvement according to their clinical status or neuroimaging abnormalities or both. Although patients with hypothalamopituitary involvement have traditionally been classified as having CNS disease, we found that patients with isolated DI had a neuropsychologic profile similar to the patients without evidence of CNS involvement. As a consequence, we divided the patients into the three following groups: no CNS or hypothalamopituitary involvement (no CNS group, n = 14), isolated DI (DI group, n = 6), and other CNS involvement (cerebellar ataxia, behavioral abnormalities, and MRI changes of the brain, excluding the hypothalamopituitary region) with or without DI (CNS group, n = 8). Details of these eight patients are listed in Table 3  .
The assumptions of normality and homogeneity were checked for all data before parametric analyses were conducted on standardized scores and percentage correct results. One-way analysis of variance indicated no significant group differences on the variables of age at diagnosis, age at assessment, and time elapsed between diagnosis and assessment in the three groups.
There was a significant difference in neuropsychologic outcome between the groups. The patients in the group with CNS involvement had significantly lower FSIQ, PIQ, and VIQ (P < .001) compared with patients in the no CNS and DI groups. Details of IQ in the these three groups are listed in Table 4
There was a similar pattern of deficit in other domains of cognitive function, including immediate verbal memory, immediate and delayed visual memory, immediate memory span, interference in learning, receptive vocabulary, reading comprehension, oral expression, mathematical reasoning, and numerical operations. Details of these results are listed in Table 5 . The results of some of these tests are illustrated in Figs 1, 2, and 3. Post hoc investigation (Scheffe) revealed that, for each of these differences, the CNS group performed worse than both the no CNS and DI groups.
Because there was a significant linear relationship between FSIQ and test scores in other cognitive domains, it was appropriate to account for the effects of FSIQ and then examine group differences on these measures (analysis of covariance). The performance of patients in the CNS group remained significantly different from the performance of both the no CNS and DI groups on Immediate Visual Memory (F[2,21] = 5.66; P = .011), Immediate Memory Span (F[2,17] = 3.88; P = .041), Interference in Learning (F[2,17] = 4.63; P = .025), Oral Expression (F[2,11] = 8.54; P = .006), and Reading Comprehension (F[2,19] = 4.75; P = .021). This indicates that there is a specific impairment in performance in patients in the CNS group over and above the reduction in overall intelligence.
In this study, we have shown that there is significant impairment of cognitive function in a representative group of survivors of multisystem LCH who have clinical or MRI evidence of involvement of the CNS. Involvement of the CNS by LCH is not common but can cause serious handicap. The incidence, natural history, and functional consequences of CNS involvement in LCH are still not fully understood, and definitions of CNS involvement vary. Cerebellar damage, resulting in ataxia and incoordination, often several years after the initial disease has burnt out, is well recognized, but the pathogenesis is, at present, unclear.
There is increasing interest in neuropsychologic deficits in patients with LCH. However, the etiopathogenesis of cognitive deficit is poorly understood because it is rare for the cerebrum to be detectably affected by the acute disease. Cranial irradiation and intrathecal methotrexate, used for CNS-directed therapy in childhood leukemia, can cause cognitive deficits,35,36 but children with LCH do not often receive specific CNS-directed therapy. Moreover, the systemic agents used in these patients are not known to be neurotoxic in the doses administered. Cranial radiotherapy was administered in relatively low doses ( Reported neuropsychologic sequelae of LCH include intellectual loss, learning deficit, poor school performance, and emotional disturbance. Most publications report single cases or very small series or LCH.6,13–17 None of these studies gave details of the neuropsychometric testing performed. A recent report18 describes detailed neuropsychologic assessment of two children with LCH, both with neurologic involvement. Both patients were shown to have marked deficits in global intellectual functioning (measured by WISC-III). At final testing, both patients were below the twentieth percentile for FSIQ, with a drop from premorbid levels. They also had evidence of a short-term memory deficit and of behavioral abnormality. However, no study, other than ours, has assessed the neuropsychologic outcome in a cohort of survivors of multisystem LCH regardless of whether or not they have had neurologic symptoms or signs. In our study, a high proportion of patients (11 of 28 patients, 39%) had intellectual deficits with IQs below 85, and of these patients, eight had clinical or imaging evidence or both of CNS involvement. Only four of these patients had such grossly abnormal signs (severe ataxia) that they would have been easily recognized as having clinically significant CNS involvement, and the use of a specific ataxia assessment in this study has probably increased our awareness of subclinical cerebellar damage. The significance of this degree of abnormality is not understood at present, but it should be clarified by studies with larger numbers of patients. Involvement of the pituitary region has always traditionally been regarded as representing CNS disease in LCH patients, but cognitive function was within normal limits in our patients with isolated DI. Therefore, patients with DI should not necessarily be regarded as having CNS damage unless other abnormalities are also present. In this context, it would be worth recalling that the posterior pituitary is derived from Rathke’s pouch, which is derived from the fetal nasopharynx and not the developing brain. We have demonstrated that CNS involvement in LCH limits general cognitive development extending to both verbal and nonverbal abilities. Importantly, our results also indicate that there is a relatively specific vulnerability in regard to the development of both processing capacity and, possibly, attentional systems. This vulnerability is reflected in the low scores obtained by the CNS group on measures of immediate auditory verbal memory span and immediate recall of geometric designs. Similarly, the patients in this group showed increased vulnerability to interference during learning. Although there are published studies indicating that aspects of information processing and attention are subserved by the frontal and temporal lobes or their interactions,37,38 further investigation using quantified MRI techniques is necessary to confirm the exact locus of the underlying neuropathology in our patients. In this single-institution study, we have found that neurologic abnormality and cognitive deficit are more common in survivors of multisystem LCH than has previously been thought. Careful neuropsychologic testing is essential and should be included in the long-term follow-up assessment of all patients with multisystem disease, especially those with any clinical or MRI abnormality of the brain. Early recognition of any cognitive deficit will enable assessment for special needs assistance in school and, through remedial treatment, improvement of the patients’ long-term outcome. This study has been the pilot project on which further studies investigating long-term survivors of LCH could be based. Future studies are being planned as part of the international collaborative multicenter trials coordinated by the Histiocyte Society.
Supported by Ferring Pharmaceuticals and Jack’s Pack, London, United Kingdom (V.R.N.).
1. Lichtenstein L: Histiocytosis X: Integration of eosinophilic granuloma of bone, "Letterer-Siwe disease" and "Schuller-Christian disease" as related manifestations of a single nosologic entity. Arch Pathol 56:84–101, 1953 2. Pritchard J, Malone M: Histiocyte disorders, in Peckham M, Pinedo B, Veronesi U (eds): Oxford Textbook of Oncology (ed 2). Oxford, United Kingdom, Oxford University Press, 2000, pp 2457–2475 3. Kepes J: Histiocytosis X, in Vinken PJ, Bruyn GW (eds): Handbook of Neurology. New York, NY, Elsevier Publishing Company, 1979, pp 93–117 4. Grois N, Flucher Wolfram B, Heitger A, et al: Diabetes insipidus in Langerhans cell histiocytosis: Results from the DAL-HX 83 study. Med Pediatr Oncol 24:248–256, 1995[Medline]
5. Donadieu J, French LCH Study Group: A multi-centre retrospective survey of Langerhans’ cell histiocytosis: 348 cases observed between 1983 and 1993. Arch Dis Child 75:17–24, 1996
6. Sims DG: Histiocytosis X: Follow-up of 43 cases. Arch Dis Child 52:433–440, 1977 7. Dunger DB, Broadbent V, Yeoman E, et al: The frequency and natural history of diabetes insipidus in children with Langerhans-cell histiocytosis. N Engl J Med 321:1157–1162, 1989[Abstract] 8. Grois NG, Favara BE, Mostbeck GH, et al: Central nervous system disease in Langerhans cell histiocytosis. Hematol Oncol Clin North Am 12:287–305, 1998[CrossRef][Medline]
9. Ha SY, Helms P, Fletcher M, et al: Lung involvement in Langerhans’ cell histiocytosis: Prevalence, clinical features, and outcome. Pediatrics 89:466–469, 1992
10. Thompson HH, Pitt HA, Lewin KJ, et al: Sclerosing cholangitis and histiocytosis X. Gut 25:526–530, 1984 11. Grois N, Broadbent V, Favara BE, et al: Report of the Histiocyte Society Workshop on "Central Nervous System" (CNS) disease in Langerhans cell histiocytosis (LCH)." Med Pediatr Oncol 29:73–78, 1997[CrossRef][Medline]
12. Lipton SA, Gendelman HE: Seminars in medicine of the Beth Israel Hospital, Boston: Dementia associated with the acquired immunodeficiency syndrome. N Engl J Med 332:934–940, 1995
13. Braunstein GD, Whitaker JN, Kohler PO: Cerebellar dysfunction in Hand-Schuller-Christian disease. Arch Intern Med 132:387–390, 1973 14. Hayward J, Packer R, Finlay J: Central nervous system and Langerhans cell histiocytosis. Med Pediatr Oncol 18:325–328, 1990[Medline] 15. Cervera A, Madero L, Penas JJG, et al: CNS sequelae in Langerhans cell histiocytosis: Progressive spinocerebellar degeneration as a late manifestation of the disease. Pediatr Hematol Oncol 14:577–584, 1997[Medline] 16. Komp DM, El Mahdi A, Starling KA, et al: Quality of survival in histiocytosis X: A Southwest Oncology Group study. Med Pediatr Oncol 8:35–40, 1980[Medline] 17. Ransom JL, Powazek M, Goff JR, et al: Neuropsychological late sequelae of Histiocytosis X. Pediatr Res 12:472, 1978 (abstr 653) 18. Whitsett SF, Kneppers K, Coppes MJ, et al: Neuropsychological deficits in children with Langerhans cell histiocytosis. Med Pediatr Oncol 33:486–492, 1999[CrossRef][Medline] 19. Trouillas P, Takayanagi T, Hallett M, et al: International Cooperative Ataxia Rating Scale for pharmacological assessment of the cerebellar syndrome. The Ataxia Neuropharmacology Committee of the World Federation of Neurology. J Neurol Sci 145:205–211, 1997[CrossRef][Medline]
20. Vargha-Khadem F, Isaacs E, van-der-Werf S, et al: Development of intelligence and memory in children with hemiplegic cerebral palsy: The deleterious consequences of early seizures. Brain 115:315–329, 1992
21. Christie D, Battin M, Leiper AD, et al: Neuropsychological and neurological outcome after relapse of lymphoblastic leukaemia. Arch Dis Child 70:275–280, 1994
22. Christie D, Leiper AD, Chessells JM, et al: Intellectual performance after presymptomatic cranial radiotherapy for leukaemia: Effects of age and sex. Arch Dis Child 73:136–140, 1995 23. Wechsler D: Wechsler Preschool and Primary Scale of Intelligence-Revised. Kent, United Kingdom, The Psychological Corporation, 1990 24. Wechsler D: Wechsler Intelligence Scale for Children (ed 3). Kent, United Kingdom, The Psychological Corporation, 1992 25. Wechsler D: Wechsler Adult Intelligence Scale-Revised. Kent, United Kingdom, The Psychological Corporation, 1986 26. Wechsler D: Wechsler Memory Scale. San Antonio, TX, The Psychological Corporation, 1945 27. Kimura D, McGlone J: Children’s stories for testing LTM, in Kimura D, McGlone J (eds): Neuropsychology Test Manual. London, Ontario, Canada, DK Consultants, 1979, p 108 28. Milner B: Psychological aspects of focal epilepsy and its neurosurgical management. Adv Neurol 8:299–321, 1975[Medline] 29. Talley J: Children’s Auditory Verbal Learning Test-2. Odessa, FL, Psychological Assessment Resources Inc, 1993 30. Dunn LM: British Picture Vocabulary Scale. Berkshire, United Kingdom, NFER-NELSON, 1982 31. Wechsler D: Wechsler Objective Language Dimensions. Kent, United Kingdom, The Psychological Corporation, 1996 32. Wechsler D: Wechsler Objective Reading Dimensions. Kent, United Kingdom, The Psychological Corporation, 1993 33. Wechsler D: Wechsler Objective Numerical Dimensions. Kent, United Kingdom, The Psychological Corporation, 1996 34. Nanduri VR, Jarosz JM, Levitt G, et al: Basilar invagination as a sequela of multisystem Langerhans’ cell histiocytosis. J Pediatr 136:114–118, 2000[CrossRef][Medline] 35. Copeland DR: Neuropsychological and psychosocial effects of childhood leukemia and its treatment. CA Cancer J Clin 42:283–295, 1992[Medline] 36. Mulhern RK, Armstrong FD, Thompson SJ: Function-specific neuropsychological assessment. Med Pediatr Oncol S1:34–40, 1998 (suppl 1) 37. Markowitsch HJ: Memory and amnesia, in Mesulam M (ed): Principles of Behavioral and Cognitive Neurology (ed 2). Oxford, United Kingdom, Oxford University Press, 2000, pp 257–293 38. Gabrieli JDE, Brewer JB, Poldrack RA: Images of medial temporal lobe functions in human learning and memory. Neurobiol Learn Mem 70:275–283, 1998[CrossRef][Medline] Submitted May 6, 2002; accepted May 13, 2003.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
12 Gy) to four of our patients, but there was no statistical association between the use of radiotherapy and cognitive outcome (Tables 1
