Keywords
Astrocytoma - cerebellum - craniopharyngioma - ependymoma - medulloblastoma - supratentorial
Introduction
After hematological malignancies as a group, primary brain tumors (PBTs) are the second
most common malignancy among children. However, among all solid tumors in children,
they are the most common form and also the leading cause of cancer-related deaths.
They constitute approximately 15%–25% of all pediatric malignancies.[1]
[2]
[3] Childhood brain tumors are biologically distinct entity as compared to those that
present in later life about their sites of origin, clinical presentation, histological
features, biological behavior, and prognosis. In adults, the most common tumor site
is the meninges and the most common primary histology being the meningioma. However,
in children, the most common tumor site is the cerebellum with astrocytoma being the
most common primary histology isolated.[4] A lot of development has happened in the field of pediatric brain tumors with regard
to characterization of various molecular and genetic pathways, with change in classification
dependent only on histogenesis (the World Health Organization [WHO] 2007 classification)
to incorporation of both, genotypic and phenotypic parameters, as in the latest WHO
2016 classification.[5]
[6] There is no dearth of epidemiologic information on pediatric brain tumors in the
Western literature; however, such knowledge is scarce in developing nation like India.
Hence, this hospital-based study was conducted to characterize the demographic profile
and histopathological subtypes of PBTs among the pediatric patients who presented
to us.
Materials and Methods
Data regarding age, gender, anatomical site, and histopathology (according to the
WHO classification prevalent at the time of the diagnosis) of 242 patients with brain
tumors (0–18 years) operated over 10 years (January 1, 2006 to December 31, 2015)
were collected retrospectively, and analyzed using GraphPad. Patients with metastatic
brain tumor, benign cystic lesion (arachnoid cysts, epidermoid cysts, and colloid
cysts), space occupying lesion of infectious etiology, and vascular malformations
were excluded from the study. Only patients residing in Gujarat and attending Gujarat
Cancer and Research Institute, Ahmedabad, were enrolled in the study.
Results
Age- wise and gender-wise distribution
Of total 242 patients presenting with PBTs, 189 (78.1%) were from 5 to 14 years age
group, with an almost equal number of patients in 5–9 (n = 94/189) and 10–14 (n = 95/189) years age group. Children (0–14 years) accounted for 95% of the cohort
(n = 230/242) while adolescents (15–18 years) accounted for only 5% of the cohort (n = 12/242) [Table 1]. The mean age of presentation was 9.38 ± 3.82 years (95% confidence interval). PBTs
were more common in males (64.1%) (n = 155/242) as compared to females (35.9%) (n = 87/242) with a male-to-female (M: F) ratio of 1.78:1. However, M: F ratio varied
according to the histopathological diagnosis with M:F ratio being 3.33:1, 3:1, and
2.81:1 in patients with ependymal tumors, meningiomas, and medulloblastomas, respectively.
Equal sex distribution was seen in patients with craniopharyngiomas and pineal gland
tumors. Only female representation was seen in patients with oligodendroglial tumors
(n = 2), choroid plexus tumors and ganglioglioma (n = 1 each). Among children, males accounted for 64.8% (n = 149/230) while females accounted for 35.2% (n = 81/230) of the cohort. Among adolescents, there was equal representation of the
sexes.
Table 1
Age-wise and gender-wise distribution of the patients in this study
|
Age group (years)
|
Gender
|
|
Male
|
Female
|
|
<1
|
7
|
1
|
|
1-4
|
22
|
11
|
|
5-9
|
62
|
32
|
|
10-14
|
58
|
37
|
|
15-18
|
6
|
6
|
|
Total
|
155
|
87
|
Location wise distribution
The most common anatomical site, as shown in [Figure 1], was cerebellum (46.3%) (n = 112/242), followed by cerebral hemispheres (14.1%) (n = 34/242), brainstem (13.6%) (n = 33/242) ventricles (9.9%) (n = 24/242), and sellar region (7.9%) (n = 19/242). Involvement of optic apparatus, hypothalamus, and thalamus (grouped as
a diencephalic region) was seen in 7.4% of the patients. Infratentorial tumors (60.9%)
were predominant than supratentorial tumors (39.1%) in 0–14 years age group while
supratentorial tumors (58.3%) were predominant than infratentorial tumors (41.7%)
in 15–18 years age group.
Figure 1: Distribution of primary brain tumors by the anatomical site in our study
Histopathological distribution
The most common histological subtype isolated were medulloblastomas accounting for
33.1% (n = 80/242) of all PBTs, followed by ependymomas (16.1%) (n = 39/242), Grade I astrocytomas (14.9%) (n = 36/242), Grade II astrocytomas (12.8%) (n = 31/242), and Grade IV astrocytomas (9.5%) (n = 23/242). The distribution of other histological subtypes in this study and other
national and international studies is shown in [Table 2] and [3]. Among astrocytic tumors, lower grade histology (Grade I and II) was seen in 71.3%
of the patients. However, the histopathological distribution also varied according
to age groups. [Table 4] shows the most common and second most common brain tumor histologies by age at the
occurrence.
Table 2
Frequency of various histological subtypes of pediatric primary brain tumors in Indian
studies (%)
|
Tumour
|
AIIMS[3]
|
NIMHANS[3]
|
GB pant[3]
|
TMH[3]
|
CSMMU[3]
|
CMC[3]
|
PGIMER[3]
|
BJMC[2]
|
LTMMC[7]
|
Our study
|
|
AIIMS – All India Institute of Medical Sciences; NIMHANS – National Institute of Mental
Health and Neurosciences; TMH: Tata Memorial Hospital; CSMMU – Chhatrapati Shahuji
Maharaj Medical University; CMC – Christian Medical College; PGIMER – Post Graduate
Institute of Medical Education and Research; BJMC – B. J. Medical College and Civil
Hospital; LTMMC – Lokmanya Tilak Memorial Medical College; PNET – Primitive neuroectodermal
tumor; AT/RT – Atypical teratoid/rhabdoid teratomas; NA – Not available
|
|
Medulloblastoma
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
33.1
|
|
Ependymoma
|
8.5
|
8.5
|
12.2
|
19.1
|
9.4
|
4.8
|
6.3
|
6.6
|
12.5
|
16.1
|
|
Grade I astrocytoma
|
23
|
29
|
19.4
|
17.4
|
14.6
|
34.9
|
NA
|
NA
|
NA
|
14.9
|
|
Grade II astrocytoma
|
2.7
|
1.1
|
1.1
|
7.7
|
12.4
|
5.3
|
NA
|
NA
|
NA
|
12.8
|
|
Grade III astrocytoma
|
2.4
|
4.8
|
0
|
1.4
|
0
|
2.9
|
NA
|
NA
|
NA
|
1.7
|
|
Grade IV astrocytoma
|
5.6
|
9.2
|
1.8
|
2.1
|
3.6
|
3.6
|
NA
|
NA
|
NA
|
9.5
|
|
Total astrocytic tumours
|
33.7
|
44.1
|
22.3
|
28.6
|
30.6
|
46.7
|
37
|
29
|
46.8
|
38.9
|
|
Craniopharyngioma
|
12.7
|
7.7
|
13.5
|
4.5
|
13.1
|
8.5
|
11.5
|
11.8
|
9.2
|
5
|
|
Meningioma
|
5.6
|
4.3
|
0.3
|
3.4
|
2.2
|
3.5
|
NA
|
1.3
|
2.9
|
1.7
|
|
Schwannoma
|
7
|
4.3
|
1.3
|
2.4
|
2.2
|
4.6
|
NA
|
2.6
|
2.9
|
1.2
|
|
PNET
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
1.2
|
|
Oligodendroglioma
|
0.7
|
0.9
|
2.9
|
1.4
|
1.5
|
0
|
0
|
2.6
|
10
|
0.8
|
|
Pineal gland tumours
|
0.7
|
1.4
|
1.3
|
1
|
3
|
NA
|
NA
|
1.3
|
0.8
|
0.8
|
|
Choroid plexus tumours
|
1.5
|
2.6
|
1.6
|
1.7
|
1.5
|
NA
|
3.5
|
2.6
|
0.8
|
0.4
|
|
Neuronal and mixed neuronal glial
|
4.1
|
2.8
|
5.2
|
2.1
|
0
|
NA
|
NA
|
1.3
|
1.7
|
0.4
|
|
AT/RT
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
NA
|
0.4
|
|
Total embryonal tumours
|
16.8
|
19.7
|
32
|
29
|
27.7
|
10.3
|
21.6
|
29
|
18.4
|
34.7
|
|
Germ cell tumors
|
2.2
|
2.2
|
3.3
|
1.7
|
2.2
|
NA
|
NA
|
0
|
1.3
|
NA
|
|
Lymphoma
|
1
|
0.5
|
0.3
|
0
|
0
|
1.1
|
NA
|
NA
|
0.8
|
NA
|
|
Total number of paediatric tumors
|
819
|
648
|
378
|
288
|
137
|
1297
|
369
|
76
|
239
|
242
|
Table 3
Frequency of various histological subtypes of pediatric primary brain tumors in international
studies (%)
|
Tumour
|
Brazil[8]
|
Korea[9]
|
Germany[10]
|
Canada[11]
|
Beijing[12]
|
Morocco[13]
|
Japan[14]
|
Hong Kong[15]
|
Our study
|
|
PNETs – Primitive neuroectodermal tumors; NA – Not available
|
|
Astrocytoma
|
32.5
|
27.8
|
41.7
|
39.4
|
30.5
|
37.1
|
35.7
|
57
|
38.9
|
|
Medulloblastomas and PNETs
|
13.9
|
19
|
25.7
|
15.4
|
14.6
|
28.9
|
10
|
23
|
34.3
|
|
Ependymoma
|
7.4
|
8.1
|
10.4
|
7
|
5.6
|
12
|
4.8
|
8
|
16.1
|
|
Oligodendroglioma
|
0.9
|
2.6
|
1.1
|
1.7
|
6.2
|
1.7
|
0
|
NA
|
0.8
|
|
Craniopharyngioma
|
11
|
9.2
|
4.4
|
6.8
|
18.4
|
6.6
|
10.5
|
6
|
5
|
|
Choroid plexus tumour
|
3
|
2.2
|
NA
|
2.3
|
1.8
|
NA
|
0
|
NA
|
0.4
|
|
Neuronal and mixed neuronal glial
|
7.6
|
6.2
|
3.2
|
<2
|
3.1
|
1.3
|
0
|
NA
|
0.4
|
|
Meningioma
|
3
|
2.6
|
1.2
|
<2
|
3.1
|
2.2
|
1.9
|
NA
|
1.7
|
|
Schwanomma
|
NA
|
0.4
|
NA
|
3.1
|
2.8
|
NA
|
0
|
NA
|
1.2
|
|
Germ cell tumor
|
3.6
|
8.1
|
NA
|
3.1
|
7.9
|
0.9
|
14.3
|
2
|
NA
|
|
Pineal tumors
|
NA
|
NA
|
1.3
|
0.5
|
0.6
|
0.7
|
0
|
NA
|
0.8
|
Table 4
First and second most common histopathological distribution by age group at the occurrence
|
Age group (years)
|
Most common histology
|
Second most common histology
|
|
<1
|
Ependymal tumors
|
Medulloblastomas
|
|
1-4
|
Medulloblastomas
|
Ependymal tumors
|
|
5-9
|
Astrocytic tumors
|
Medulloblastomas
|
|
10-14
|
Astrocytic tumors
|
Medulloblastomas
|
|
15-18
|
Astrocytic tumors
|
Craniopharyngiomas and schwannoma (equal frequencies)
|
Discussion
The incidence and mortality of PBTs among pediatric population is high; however, its
tumor burden is underestimated in developing countries like ours due to the lack of
complete registration of newly diagnosed cases with local cancer registries. In such
scenario, hospital- based prevalence data becomes the primary source for estimating
the disease load. This data are essential for assessing geographical differences in
phenotypic and genotypic profiles of PBTs and also ascertaining the required neuro-oncological
health-care infrastructure for their management. Demographic and histopathological
profiles of Indian pediatric PBTs are available from nine tertiary healthcare centers-GB
Pant Hospital, New Delhi; Christian Medical College, Vellore; Post Graduate Institute
of Medical Education and Research, Chandigarh; National Institute of Mental Health
and Neurosciences, Bangalore; Tata Memorial Hospital, Mumbai; Chhatrapati Shahuji
Maharaj Medical University, Lucknow; All India Institute of Medical Sciences, New
Delhi; and B. J. Medical College and Civil Hospital (BJMC), Ahmedabad and Lokmanya
Tilak Memorial Medical College, Mumbai, via studies conducted by Jain et al.,[3] Shah et al.,[2] and Sangita et al.[7] This study has analyzed similar data of patients residing only in Gujarat. The results
were compared to other national and international studies.
The frequency of pediatric patients with PBTs was highest in 5–14 years age group
(78.1%) which is in line with other studies.[2]
[7]
[16]
[17] The frequencies of PBTs were higher among children (0–14 years) as compared to the
adolescents (15–18 years). The mean age at diagnosis in this study was 9.38 years
while it was 10.69 years in BJMC study.[2] However, the difference was statistically insignificant (P = 0.0625, Unpaired t-test with Welch correction). Mean ages reported from Morocco,[13] Pakistan,[18] and China [19] were 9.3, 8.8, and 12.68 years, respectively. A higher proportion of brain tumors
were found in males as compared to in females in this study, with a male-to-female
ratio of 1.78:1, consistent with findings of other studies,[12]
[13]
[14]
[19]
[20] with Pakistan reporting the highest ratio of 2.52:1.[18] The ratio also varied according to the histological subtype with male predominance
seen in subtypes ependymal tumors, meningiomas, and medulloblastomas. As observed
in other studies,[2]
[7]
[11]
[16] males and females with PBT were evenly distributed among different age groups in
our study and observed differences were not statistically significant (P = 0.3586, Fischer’s exact test).
The most common anatomical site in this study was cerebellum followed by cerebral
hemispheres, brainstem, and ventricles. The topographical predominance of cerebellum
and cerebral hemispheres was due to the maximum isolation of medulloblastomas and
astrocytomas at those sites. These findings were in line with the results of other
published data.[2]
[7]
[8]
[10]
[16] In this study, infratentorial tumors were bit predominant in childhood age group
whereas, in adolescent age group, supratentorial tumors were more common, which was
consistent with the findings of other studies.[2]
[7]
[16]
[17] However, there was no statistically significant difference in the distribution of
supratentorial and infratentorial tumors among children and adolescents (P = 0.2309, Fischer’s exact test).
In the present study, medulloblastoma was the most common histological subtype (33.1%),
followed by ependymomas (16.1%) and Grade I astrocytomas (14.9%). However, overall,
astrocytomas (38.9%) were the most common tumors followed by embryonal tumors including
medulloblastomas, supratentorial primitive neuroectodermal tumors, and atypical teratoid/rhabdoid
teratomas (34.7%), together constituting almost three-quarters of all PBTs in the
study population. Ependymoma appeared to be the third most common tumor in this study
which contradicted the results of two Indian [2]
[3] and two international studies [8]
[9] (which showed craniopharyngioma as the third most common tumor) but was in line
with major international studies.[10]
[11]
[13]
[16]
[17] Craniopharyngioma was the fourth most common tumor in this study which is in line
with other studies [10]
[11]
[13] which contradicted with two Indian studies.[2]
[3] The frequency of various histological subtypes of pediatric PBTs in this study as
compared to other national and international studies is shown in [Table 2] and [3]. Histopathological distribution also varied according to age groups as shown in
[Table 4]; however, it was different to that of Central Brain Tumor Registry of the United
States report.[4] The primary limitation of our study was that it was restricted to only patients
residing in Gujarat, and treated at a single institution. Hence, caution is required
when extrapolating the results of this data to different geographic regions of our
country.
Conclusions
Histopathological profiles of cohort in this study do not differ substantially from
other hospital-based and population-based studies. Astrocytomas and medulloblastomas,
which form the major histologic subtypes in children residing in Gujarat, needs special
attention with respect to the distribution of infrastructure and resources. Hospital-
based studies like ours play a major role in planning the distribution of neurosurgical
infrastructure and radiation and medical oncological resources toward the disease
management and preventive programs.
