Brain Tumors in Childhood and Adolescence - Pediatric Neurology

Brain Tumors in Childhood and Adolescence Daniel L. Keene, MD, MA*, Elizabeth Hsu, MD, CM†, and Enrique Ventureyra, MD‡ Brain tumors are the second most common neoplasm in childhood and adolescence. With the recent advances in technology, changes in tumor incidence have been reported. This study examines this statement. A 19-year retrospective case review of primary brain tumors in persons younger than 18 years of age at time of diagnosis, who had permanent residence in our catchment area, was performed. Data were examined for changes in presenting symptoms and signs and incidence rates for tumors on the basis of anatomic location and histologic tumor type. An incidence rate of 2.76 per 100,000 people younger than 18 years of age was found. During the period of this study a small, but significant, trend toward increasing incidence was evident. No changes in patterns of presentation or duration of symptoms before diagnosis was observed. The incidence rate based on histologic tumor diagnosis remained fairly constant during the study period. © 1999 by Elsevier Science Inc. All rights reserved. Keene DL, Hsu E, Ventureyra E. Brain tumors in childhood and adolescence. Pediatr Neurol 1999;20:198-203. Introduction

Results

Brain tumors are the second most common neoplasm in childhood. Recent advances in diagnostic technologies and treatment protocols have resulted in a significant improvement in survival rates. As a result of these advances the questions of whether changes in initial patterns of clinical presentation have changed or whether the incidence of diagnosis of primary brain tumors has changed have been asked. The authors report the results of a retrospective review of childhood brain tumors and the accompanying clinical patterns performed at the Children’s Hospital of Eastern Ontario during a 19-year period. Materials and Methods A retrospective chart review of all cases of primary brain tumors diagnosed at the Children’s Hospital of Eastern Ontario between 1975

From the *Division of Neurology; †Oncology Service; and ‡Division of Neurosurgery; Department of Pediatric Surgery; Children’s Hospital of Eastern Ontario; Ottawa, Ontario, Canada.

198

PEDIATRIC NEUROLOGY

and 1993 was performed. At the time of diagnosis, all patients had to be younger than 18 years of age, had to have a primary intracranial neoplasm, and had to have their recorded permanent address in the hospital catchment area to be included in this review. Except for tumors of the brainstem, all patients entered into this study had pathologically proven brain tumor. Because it was not the practice at the authors’ institution to biopsy brainstem tumors, only radiologic diagnosis was available for those tumors. For the basis of this study the catchment area was defined to include the following municipalities: Ottawa-Carleton, Rideau, Leeds-Grenville, Renfrew, Hull, Gatineau, Prescott-Russell, Pontiac, and Stormont-Dundas-Glengarry. Loss of patients from the study catchment area to other neurosurgical centers in the province and Quebec was checked for through consultation with provincial tumor board registries, cancer clinic records, and neurosurgery clinics. Data abstracted from the patients’ clinical records included sex, age at presentation, presenting clinical symptoms and signs, duration of symptoms before diagnosis, histologic diagnosis of the tumor, and tumor location. The incidence rate for all pediatric brain tumors was calculated. The different clinical patterns at the time of diagnosis for the various tumor groups and locations were evaluated. Population figures for the catchment area for persons younger than 18 years of age were obtained from Statistics Canada records at 5-year intervals for the period of the study. Incidence figures were calculated using these data. Chi-square tests with P ! 0.05 was used for proportional data. Continuous variables were analyzed with analysis of variance, using the SSPS-PC statistical package.

Vol. 20 No. 3

Between 1975 and 1993, 200 new cases of primary intracranial neoplasms in children were diagnosed at the Children’s Hospital of Eastern Ontario. After appropriate consultation, loss to other centers at time of diagnosis was believed to be nonexistent. One hundred four cases (52%) occurred in the supratentorial region; 96 cases (48%) occurred in the infratentorial region. Histologic tumor diagnoses were as follows: astrocytomas of various grades in 66 patients (33%), primitive neuroectodermal tumors in 32 patients (16%), pineal germ cell tumors in 17 patients (8.7%), gangliogliomas in 16 patients (8%), glioblastomas in 13 patients (6.5%), craniopharyngiomas in 11 patients (5.5%), pituitary adenoma in four patients (2%), chorioid plexus papillomas in three patients (1.5%), oligodendroglioma in three patients (1.5%), meningiomas in three Communications should be addressed to: Dr. Keene; Division of Neurology; Children’s Hospital of Eastern Ontario; 401 Smyth Road; Ottawa, Ontario K1H 8L1, Canada. Received May 27, 1998; accepted October 27, 1998.

© 1999 by Elsevier Science Inc. All rights reserved. PII S0887-8994(98)00139-8 ● 0887-8994/99/$20.00

patients (1.5%), melanoma of the central nervous system in one patient (0.5%), hemangioblastoma in one patient (0.5%), acoustic neuroma in one patient (0.5%), and teratoma in one patient (0.5%). Tumors in the brainstem region occurred in 19 patients (9.5%) and were considered to be brainstem gliomas. Mean age at time of diagnosis was 8 years, 11 months " 4 years, 11 months. The male-female ratio was 1.08:1. Hemispheric tumors occurred in 52 patients. They accounted for 50% of the tumors located in the supratentorial compartment. This group accounted for 26% of the total tumor series. The distribution of histologic tumor types among the hemispheric tumors was gangliogliomas 30%, low-grade astrocytomas 30%, high-grade astroglial tumors 20%, meningiomas 6%, ependymomas 4%, and oligodendroglioma 4%. Mean age at presentation for patients with this group of tumors was 8 years, 5 months " 5 years, 1 month. The presenting signs were as follows: seizures 60%, headache 37%, vomiting 23%, changes in behavior or personality 11%, facial asymmetry 9%, and visual difficulties 6%. The initial finding on examination included one or more of the following: no abnormalities 51%, hemiplegia 34%, signs of increased intracranial pressure 23%, cranial nerve dysfunction 3%, and macrocrania 3%. Supratentorial axial or midline tumors occurred in 50 patients. They accounted for 25% of the total tumor group and accounted for 50% of tumors located in the supratentorial compartment. The location of these tumors was 41% in the suprasellar region, 34% in the pineal region, 10% in the sellar region, 10% in the intraventricular region, and 5% arising from thalamic structures. Tumor types in the sellar and suprasellar regions included craniopharyngioma 50%, astrocytomas 25%, pituitary adenoma 20%, and germ cell tumors 5%. Tumors arising from the thalamic region were all astrocytic in nature. Pineal region tumors were of germ cell origin. Intraventricular tumors were 50% ependymomas, 25% teratomas, and 25% choroid plexus tumors. The presenting signs for tumors arising from axial structures included one or more of the following: nonspecific headache 60%, polyuria 35%, nonspecific malaise 10%, short stature 10%, and visual difficulties 5%. The findings on initial examination at the time of diagnosis for patients with this group of tumors included signs of increased intracranial pressure 30%, visual field disturbances 25%, optic atrophy 15%, and Parinaud’s sign 15%. Cerebellar tumors occurred in 74 patients. They accounted for 37% of the total tumor group and accounted for 80% of the infratentorial tumors. Histologic diagnosis of tumors occurring in this region included astrocytomas of various grades 44%, primitive ectodermal tumors 41%, and ependymomas 9%. The mean age at presentation of patients with cerebellar tumors was 7 years, 6 months " 9 years, 5 months with a mean length of time before diagnosis of 6.7 " 9.2 weeks (range # 0-30). The presenting symptoms included one or more of the following: vomiting 76%, headache 62%, and incoordination

55%. The frequency of clinical signs on initial examination included one or more of the following: ataxia 69%, signs of increased intracranial pressure 57%, nystagmus 31%, head tilt 14%, cranial nerve palsies 28%, and macrocrania 10%. Brainstem tumors occurred in 19 patients. They accounted for 9.5% of the total tumor group and 20% of infratentorial tumors. The mean age at presentation was 7 years, 4 months " 3 years, 4 months, with a mean duration of symptoms before diagnosis of 11 " 8.9 weeks. The frequency of symptoms at the time of initial diagnosis was gait difficulties 83%, “cross eyes” 50%, headaches 25%, vomiting 25%, and swallowing difficulties 8%. The findings on initial examination included one or more of the following: cranial nerve VI dysfunction 67%, ataxia 50%, cranial nerve VII dysfunction 42%, nystagmus 33%, hemiplegia 33%, and head tilt 33%. Over the time of this review, significant changes in the patterns of presenting symptoms and signs for each of the aforementioned groups was not observed. When the data were examined for age differences, 23 cases (11.5%) occurred in patients younger than 2 years of age; 41 cases (20.5%) 2-5 years of age; 46 cases (23%) 5-10 years of age, and 90 cases (45%) in patients older than 10 years of age. The mean age of the total group was 8 years; the mean age at diagnosis of supratentorial tumors was 9 years, 2 months compared with 6 years, 6 months for infratentorial tumors. Patients with malignant tumors had a mean age at time of diagnosis of 9 years, 6 months; the mean age at presentation of patients with malignant supratentorial tumors was 12.4 years compared with 7 years, 10 months for patients with infratentorial tumors. Fifty-six percent of tumors occurring in patients younger than 2 years of age were infratentorial tumors; 43% of these tumors were classified as malignant. The breakdown of histologic diagnosis was as follows: astrocytomas 30%, glioblastomas 13%, primitive neuroectodermal tumors 13%, gangliogliomas 13%, ependymomas 9%, choroid plexus tumors 9%, teratomas 4%, hemangioblastomas 4%, and meningiomas 4% (Fig 1). Sixty-eight percent of tumors occurring in patients between the ages of 2 and 5 years (inclusive) were infratentorial; 34% of these tumors were classified as malignant. The distribution of histologic tumor types was as follows: astrocytomas 41%, primitive neuroectodermal tumors 20%, brainstem gliomas 15%, gangliogliomas 12%, craniopharyngomas 8%, glioblastomas 7%, ependymomas 2%, and meningiomas 2% (Fig 2). Between the ages of 6 and 10 years (inclusive), 52% of the tumors were located in the infratentorial region; 36% of tumors in this age group were believed to be malignant. The frequency of histologic tumor types was as follows: astrocytomas 39%, brainstem tumors 17%, primitive neuroectodermal tumors 13%, gangliogliomas 10%, ependymomas 6%, pineal germ cell tumors 6%, glioblastomas 2%, oligodendrogliomas 2%, and craniopharygiomas 2% (Fig 3).

Keene et al: Brain Tumors 199

Figure 1. Comparison of age distribution with major histologic tumor diagnosis for patients younger than 2 years of age.

In children older than 10 years, 33% of the tumors were infratentorial tumors; 44% of tumors in this age group were classified as malignant. The frequency of various tumors based on histologic diagnosis was as follows: astrocytomas 22%, primitive neuroectodermal tumors 16%, pineal germ cell tumors 15%, craniopharyngiomas 8%, glioblastomas 7%, gangliogliomas 6%, brainstem tumors 5%, pituitary adenomas 4%, ependymomas 3%, and oligodendogliomas, acoustic neuromas, melanomas, meningiomas, and chorioid plexus tumors, 1% each (Fig 4). The mean yearly incidence rate for primary brain tumors in persons younger than 18 years was 2.76 per 100,000 people younger than 18 years of age. During the period of this review, there was an increase

in the yearly incidence rate of primary intracranial lesions in persons younger than 18 years of age, which occurred for both infratentorial and supratentorial tumors (Figs 5-7). Discussion The case material included in this review has many similarities to previously reported series of brain tumors in the age group under study. The mean yearly incidence rate for intracranial tumors in persons younger than 18 years observed in this study of 2.76 per 100,000 people younger than 18 years of age was similar to previously reported rates [1-6] but less than that reported for the same age group in Sweden by Lannering et al. [7] (3.49 per 100,000). The age-specific incidence rates reported in the

Figure 2. Comparison of age distribution with major histologic tumor diagnosis for patients 2-5 years of age.

200

PEDIATRIC NEUROLOGY

Vol. 20 No. 3

Figure 3. Comparison of age distribution with major histologic tumor diagnosis for patients 5-10 years of age.

authors’ study also mirror those previously reported [5,8, 9]. The mix of tumors based on histologic characteristics, the distribution of the tumors between the infratentorial and supratentorial compartments, and the initial clinical presentations observed in the authors’ study were similar to those previously reported [6,10-12]. This study demonstrated a small, but significant, trend toward an increasing yearly incidence rate of primary brain tumors in persons younger than 18 years of age for the years under study. This trend was noted for tumors located in both the infratentorial and supratentorial space. It was not as obvious when the tumors were divided into tumor-specific histologic groups. This trend toward an increasing incidence of tumor occurrence was similar to previously reported studies [1-3]; however, Miltenburg et al. [4] could not confirm this finding. They reported no increase in tumor incidence in their study population of persons younger than 18 years of age with the diagnosis of

a brain tumor. The shorter duration of their study period (7 years) and the smaller number of patients entered could account for this difference. Several reasons have been postulated to account for the trend the authors report here. These reasons have included changes in population demographics during the study period, bias of ascertainment, advances in diagnostic technologies, and changes in tumor biology [9]. Although the authors’ study was a retrospective hospital-based review rather than a prospective population study, it is believed by the authors that the results reflect the trend that has occurred in the pediatric population for this catchment area. Because the Children’s Hospital of Eastern Ontario was the only center for pediatric neurosurgical services for the area of Eastern Ontario and Western Quebec for the period of this study, it is unlikely that cases were missed. In the authors’ experience, if a referral out of the region occurred, it would usually occur

Figure 4. Comparison of age distribution with major histologic tumor diagnosis for patients older than 10 years of age.


Figure 5. Change in incidence of all brain tumors.

after the initial diagnosis was confirmed at the time when various treatment options were being discussed with patients and their guardians. The exclusion from the authors’ study of patients who did not have a permanent residence in the catchment area at the time of their initial diagnosis prevented bias of selection based on certain surgical and technologic advantages offered by the authors’ center to patients outside the catchment area. In theory, changes in population demographics could account for the changes in incidence rates with time. Immigration or emigration shifts have occurred in the catchment area under study. The rate of these changes has been small. There has been a steady, but small, growth of population of the catchment area, but the ethnic mix of the population remained fairly stable during the study period [13]. The advances in technology and its availability have

been suggested [4,14-16] as reasons for the changing incidence rate. There is no doubt that the advances in imaging studies, surgical technology, and pathologic analysis of brain tumor tissue specimens have led to a better understanding of the pathophysiology and treatment outcome of brain tumors. Whether these advances can account for the trend toward an increasing incidence rate is another matter. If this was the case, a decrease in the time between initial symptoms and confirmed clinical diagnosis should have occurred during the study period. Also, it would have been expected that a change in the patterns of signs at time of diagnosis would have occurred. Neither of these trends were found in this study. There was no significant change in the time from initial symptoms to diagnosis. In addition, no correlation between the time of introduction of new technology and major peak trends in

Figure 6. Change in incidence of supratentorial brain tumors.

202

PEDIATRIC NEUROLOGY

Vol. 20 No. 3

Figure 7. Change in incidence of infratentorial brain tumors.

incident rates was found in this study. Edgeworth et al. [17] reported a similar delay in diagnosis. It was their opinion that this delay was the result of the nonspecificity of symptoms and the high incidence of psychologic symptoms, which often confused the clinical picture at the time of presentation. The authors concur with their opinion. Imaging procedures are usually ordered only when there is a high clinical suspicion of an abnormality being found. The presenting symptoms of brain tumors are often quite subtle, particularly in the young age group and in the midline supratentorial tumor group. The authors’ inability, using the aforementioned reasons, to explain the trend toward an increasing primary intracranial tumor incidence rate found in this study suggests that other causes may be involved, such as changes in the environment [9,15] or genetic predisposition to central nervous system tumors [18]. The present study was not designed to examine these possibilities. This study confirmed the trend toward an increased incidence rate of primary intracranial neoplasms in persons younger 18 years of age. Although the reason for this trend is not clear, the implications for the provision of healthcare for this group is clear. The increased incidence, combined with the longer survival times for this group of patients, will result in the need for more diagnostic and supportive healthcare services. References [1] Birch J, Marsden H, Swindell ●●. Incidence of malignant disease in childhood: A 24-year review or the Manchester children’s tumour registry data. Br J Cancer 1980;42:215-23. [2] Mosso M, Colomb R, Giordano, et al. Childhood cancer registry of the province of Torino, Italy. Cancer 1992;69:1300-6.

[3] Walker A, Robins M, Weinfeld F. Epidemiology of brain tumours. Neurology 1985;35:219-26. [4] Miltenburg D, Louw D, Sutherland G. Epidemiology of childhood brain tumours. Can J Neurol Sci 1996;23:118-22. [5] Mueller B, Gurney J. Epidemiology of pediatric brain tumors. Neurosurg Clin N Am 1992;3:715-21. [6] Gjerris F. Clinical aspects and long-term prognosis of intracranial tumours in infancy and childhood. Dev Med Child Neurol 1976;18: 145-59. [7] Lannering B, Marky I, Nordborg C. Brain tumours in childhood and adolescence in West Sweden 1970-1984. Cancer 1990;66:604-9. [8] Preston-Martin S, Staples M, Farrugia H, et al. Primary tumours of the brain, cranial nerves and cranial meninges in Victoria, Australia, 1983-1990: Patterns of incidence and survival. Neuroepidemiology 1993;12:270-9. [9] Leviton A. Principles of epidemiology. In: Cohen N, Duffner P, eds. Brain tumours in children: Principles of diagnosis and treatment, 2nd ed. New York: Raven Press, 1994:27-49. [10] Farwell J, Dohrman G, Flannery J. Central nervous system tumours in children. Cancer 1977;40:3123-32. [11] Siqueira E, Rahm B, Kanaan I, Jallu A. Brain tumours in pediatric patients at King Faisal Specialist Hospital and Research Centre. Surg Neurol 1993;39:443-50. [12] Geyer J, Berger M. Central nervous system malignancies in children. NY State J Med 1990;90:601-8. [13] Statistics Canada, Ministry of Industry, Ottawa, Ontario, Canada. [14] Schiffer D, Giordana M, Vigliani M. Brain tumours of childhood: Nosological and diagnostic problems. Childs Nerv Syst 1989;5: 220-9. [15] Brain tumour incidence rising: Researchers ask why. J Natl Cancer Inst 1993;85:1024-5. [16] Desmeules M, Mikkelsen T, Mao Y. Increasing incidence of primary malignant brain tumours: Influence of diagnostic methods. J Natl Cancer Inst 1992;84:442-5. [17] Edgeworth J, Bullock P, Bilay A, et al. Why are brain tumours still being missed? Arch Dis Child 1996;74:148-51. [18] Giuffre R. Biological aspects of brain tumours in infancy and childhood. Childs Nerv Syst 1989;5:55-9.
