| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Institute of Neuropathology,*
the Department of
Pediatric Oncology,
the Department of
Neurosurgery,§
and the Gerhard-Domagk-Institute
of Pathology,¶
Westfälische
Wilhelms-Universität Münster, Münster; and the German
Childhood Cancer Registry,
Institute for
Medical Statistics and Documentation, Johannes
Gutenberg-Universität, Mainz, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Comparative genomic hybridization (CGH) is able to identify imbalances of the entire genome in terms of DNA copy number changes. Its main advantage is that it bypasses the need for laborious cell culture to harvest metaphase spreads, and that it can be applied to archival material. Although it has previously been used in a few studies on adult high-grade astrocytomas, no CGH study has hitherto been undertaken on pediatric high-grade astrocytomas. To screen for DNA copy number changes that may be involved in tumorigenesis of pediatric malignant astrocytomas, we applied CGH on anaplastic astrocytomas and glioblastomas from 23 patients under 18 years of age.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Formalin-fixed and paraffin-embedded biopsy specimens of 23
primary pediatric high-grade astrocytomas were investigated (Table 1)
. These consisted of 10 anaplastic
astrocytomas World Health Organization grade III (three males, seven
females; mean age, 7.1 years; range, 8 months to 17 years) and 13
glioblastomas World Health Organization grade IV [six males, seven
females (including two sisters; cases 12 and 13); mean age, 8.9 years;
range, 1 to 16 years]. Of these, 20 tumors were located
supratentorially (9 anaplastic astrocytomas, 11 glioblastomas), two in
the cerebellum (one anaplastic astrocytoma, one glioblastoma), and one
in the brain stem (one glioblastoma). The pediatric age group was
defined as patients under 18 years of age at the time of operation. All
patients underwent surgery or stereotactic biopsy. A combination of
postoperative radiation therapy and chemotherapy was applied to eight
patients, whereas 10 patients underwent radiation therapy and three
patients underwent chemotherapy alone; two children did not receive any
adjuvant therapy (Table 1)
. Clinical follow-up data were available for
all patients.
|
CGH Analysis
DNA was isolated by phenol-chloroform extraction according to standard protocols. With minor modifications, CGH analysis was performed as described previously.3 Briefly, tumor DNA was labeled with biotin-16-dUTP (Boehringer Mannheim, Mannheim, Germany) and reference DNA from a healthy male donor with digoxigenin-11-dUTP (Boehringer Mannheim) in a standard nick translation reaction. The DNase concentration in the labeling reaction was adjusted to reveal an average fragment size of 200 to 500 bp. The labeled DNA fragments were purified from remaining nucleotides by column chromatography.
For CGH, 500 ng of tumor DNA, 300 ng of reference DNA, and 30 µg of human Cot1 DNA (Gibco, Karlsruhe, Germany) were co-precipitated and redissolved in 10 µl of hybridization buffer. Denaturation of DNA (75°C for 5 minutes) was followed by a pre-annealing time of 45 minutes at 37°C. Target metaphase spreads (46,XY), which had been prepared following standard procedures, were denatured separately in 70% formamide/2x standard saline citrate for 2 minutes at 72°C. Hybridization was allowed to proceed for 3 to 4 days. Posthybridization washes were performed to a stringency of 50% formamide/2x standard saline citrate at 45°C and 0.1x standard saline citrate at 60°C. Biotinylated and digoxygenated sequences were detected simultaneously, using avidin-fluorescein isothiocyanate (1:200; Boehringer Mannheim) and anti-digoxigenin-rhodamine (1:40; Boehringer Mannheim). The slides were counterstained with 4,6-diamidino-2-phenylindole (DAPI) and mounted in an antifade solution (Vectashield; Vector Laboratories, Burlingame, CA).
Microscopy and Digital Image Analysis
Separate digitized gray level images of DAPI, fluorescein isothiocyanate, and rhodamine fluorescence were taken with a charge-couple device camera connected to a Leica DMRBE microscope (Leica, Wetzler, Germany). The image processing was performed by use of Applied Imaging Software (Applied Imaging, Sunderland, UK). Average green-red ratios were calculated for each chromosome in at least 10 metaphases.
Chromosomal regions with CGH ratio profiles surpassing the 50% CGH thresholds (upper threshold, 1.25; lower threshold, 0.75) were defined as loci with copy number gains or losses. Based on experiments with normal control DNA, these thresholds have been shown to eliminate false-positive results. These values have been used in several studies comparing CGH data with results obtained by other cytogenetic methods and have proven to provide robust criteria for the diagnosis of chromosomal gains and losses. Overrepresentations were diagnosed as high-level gains or amplifications when the fluorescence intensity levels exceeded 1.5 or when the fluorescein isothiocyanate fluorescence showed strong focal signals. For the assignment of these high-level amplifications to chromosomal bands, the signal intensities were compared to the DAPI banding on individual chromosomes. As tumor specimens and normal DNA were not sex matched, X and Y chromosomes were excluded. Also excluded were centromeric and satellite regions of the acrocentric chromosomes and chromosome 19 because of the abundance of highly repetitive DNA sequences as well as the frequent occurrence of false-positive CGH results as shown by interphase fluorescence in situ hybridization using suitable DNA probes.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
.
Every chromosome was shown to carry imbalances (Figure 1)
; most imbalances were found on
chromosomes 1 (14 cases), 6 (11 cases), and 17 (10 cases), the fewest
on chromosomes 14 and 15 (2 each). The most DNA copy number changes
were detected in an anaplastic astrocytoma (19 changes), the fewest in
two glioblastomas (none). Anaplastic astrocytomas showed an average of
6.6 chromosomal changes per tumor (range, 1 to 19; 2.9 gains
versus 3.7 losses), which were more common than in
glioblastomas with an average of 6.1 changes per tumor (not
significant; range, 0 to 18; 2.8 gains versus 3.3 losses)
(Table 1
, Figure 1
). Losses were more common than gains in both
entities (not significant).
|
|
, Figure 1
).
Minimal common regions found in at least five tumors consisted of gains
of 1q21-41 (10 cases), 3q27-qter (6 cases), 2q31-32, and 5q14-22 (5
cases) as well as losses of 22q12-qter (8 cases), 10q23-25, 6q25-qter
(7 cases), 9q34.2, 10q, 11q14-22, 16q22-qter, and 17p (5 cases).
High-level gains affecting whole chromosomes, as indicative of trisomy,
were found on chromosomes 7 (3 cases), 1, 2, and 9 (2 cases), as well
as on 18 (1 case). High-level gains affecting parts of a chromosome or
chromosome arms were found on chromosomes 1q (7 cases), 2q, 7q (4
cases), 3q (3 cases), 17q (2 cases) as well as 4q, 8q, and 20q (1 case)
(Figure 1)
. Between one and five high-level gains were found in four of
the anaplastic astrocytomas (cases 3, 4, 7, and 10), and between one
and six in five of the glioblastomas (cases 15, 16, 19, 20, and 21)
(Table 1)
.
The mean MIB-1 proliferation indices were highly significantly
different between anaplastic astrocytomas and glioblastomas and were
17.0% (range, 3.0 to 38.5%; SD, 12.2%) for anaplastic astrocytomas
and 31.8% (range, 11.7 to 51.1%; SD, 14.6%) for glioblastomas
(P < 0.01). All children could be followed-up,
and the median overall survival was 15.0 months (SD, 3.8 months). At
the time of reporting, 6 of 10 children with anaplastic astrocytomas
were still alive (no median overall survival calculable because >50%
of patients were still alive at time of reporting; mean overall
survival, 64.2 months; SD, 14.7 months) whereas only 2 of 13 children
with glioblastomas were still alive (median overall survival, 11.0
months; SD, 1.7 months; mean overall survival, 13.5 months; SD, 2.7
months); the difference in survival was statistically significant
(P < 0.05). The respective 1-, 2-, and 3-year
survival rates were 55, 33, and 21% for all malignant astrocytomas;
70, 60, and 50% for anaplastic astrocytomas; and 46, 9, and 0% for
glioblastomas. A highly significantly shorter survival was found for
all high-grade astrocytomas showing a MIB-1 proliferation index of
>25% (median overall survival, 8.0 months; SD, 2.4 months; mean
survival, 8.4 months; SD, 1.6 months) compared with tumors with a MIB-1
proliferation index of <25% (no median overall survival was
calculable because >50% of patients were still alive at time of
reporting; mean survival, 63.0 months; SD, 13.4 months;
P < 0.001). Although the number of aberrations overall
as well as of gains and losses on their own bore no significance on
survival, a significantly shorter survival was found among pediatric
anaplastic astrocytomas showing gains of 1q (median overall survival,
9.0 months; SD, 2.4 months; mean survival, 12.3 months; SD, 2.7 months)
compared with cases without this imbalance (no median overall survival
calculable because >50% of patients were still alive at time of
reporting; mean survival, 87.9 months; SD, 13.3 months;
P < 0.05) (Figure 2a)
although no such correlation could be found for pediatric glioblastomas
in which cases with +1q as well as those without +1q showed a median
overall survival of 11.0 months (P = 1.00;
Figure 2b
). No correlation could be found between prognosis and age,
gender, or year of diagnosis.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Pediatric High-Grade Astrocytomas
Although a normal karyotype was found in 14 out of 22 pediatric
anaplastic astrocytomas,14
anaplastic astrocytomas all
showed DNA copy number changes in our series. These were mainly
characterized by gains of 5q and losses of 22q and 9q, whereas
glioblastomas predominantly showed gains of 1q and 3q as well as losses
of 6q, 10q, 13q, and 17p (Figures 1 and 3)
. Some of these imbalances
have previously been reported in pediatric malignant astrocytomas,
namely losses of chromosomes 13,9
17p,5,6,9,15
and 22q.5,6,14
Furthermore,
losses of heterozygosity at 17p13.1, 9p21, and 10q23-25 were found in
50, 83, and 78% of high-grade pediatric astrocytomas,
respectively.13
TP53 tumor suppressor gene mutations
(mapping to 17p13.1) have been found in 38% of pediatric high-grade
astrocytomas,13
25% of glioblastomas,12,16
and 11% of anaplastic astrocytomas16
as well as 71% of
brain stem glioblastomas,17
underscoring the proposed
preferential inactivation of the p53 tumor suppressor pathway in >95%
of pediatric astrocytomas.18
On the other
hand, mutations of the PTEN gene (mapping to 10q23-25) are rare in
pediatric cases in which they are only found in 
8% of
tumors;11,13
in our series chromosomal material of 10q was
lost in 31% of tumors. Furthermore, no MDM2 (12q13-21)15
or epidermal growth factor receptor (EGFR, mapping to 7q21-32) gene
amplifications were detected in several
investigations,12,13,17-19
which is corroborated by the
few CGH changes found by us at the aforementioned sites.
|
200 pediatric glioblastomas
demonstrated gains of chromosome 7 to be the most common imbalance
encountered in 45% of tumors, followed by losses of 10, 9p, and 22q in
35, 20, and 17% of cases, respectively; very similar cytogenetic
changes were reported for 43 pediatric anaplastic
astrocytomas.14
However, these aberrations as well as
losses of other implicated chromosome regions like 1q17
and 10p5,6
played little or no role among our cases,
except for losses of 22q in anaplastic astrocytomas and 10q in
glioblastomas (Figure 1)
. In contrast, our cases showed frequent and
prognostically relevant gains of 1q, the most commonly amplified
chromosomal region overall and associated with a poorer outcome, and
one that has previously been implicated in pediatric low-grade but not
hitherto in high-grade astrocytomas,20
and losses of 9q
rather than 9p (Figure 1)
. CGH in Pediatric and Adult High-Grade Astrocytomas
CGH results on primary adult high-grade astrocytomas
have been presented in nine studies consisting of 22 anaplastic
astrocytomas21-24
and 184
glioblastomas.21-29
Loss of genetic material of 10q was
the most common chromosomal change overall and similarly
frequent in the pediatric primary anaplastic astrocytomas
and glioblastomas of this series (30 and 31%, respectively) and adults
(32 and 59%) alike (Figure 3)
. Multiple tumor suppressor genes on the
long arm of chromosome 10 have been implicated in the development of
astrocytic gliomas:30
among glioblastoma mutations,
deletions, and loss of heterozygosity on 10q have been found more
frequently in primary compared with secondary glioblastomas and have
predominantly been associated with the PTEN gene.31,32
Gains of chromosome 7, containing the EGFR gene that is amplified in
>30% of adult malignant astrocytomas,33
and losses of
9p, although, were much more frequently found in adult malignant
astrocytomas (32 to 58%) compared with pediatric cases (8 to 20%),
whereas losses of 10p exclusively occurred among adult anaplastic
astrocytomas and glioblastomas (27 and 51%) but not in any of our
pediatric cases. In contrast, there was a marked prevalence for
pediatric anaplastic astrocytomas and glioblastomas to show gains of
chromosomes 1p (20 and 23% versus 5 and 9%), 2q (20 and
23% versus 9 and 3%), and 21q (20 and 16%
versus 0 and 3%) as well as losses of 6q (40 and 31%
versus 9 and 18%), 11q (20 and 23% versus 5 and
6%), and 16q (20 and 23% versus 5 and 5%). However, a
recent microsatellite study on astrocytic tumors found losses affecting
chromosome 6q in 38% of anaplastic astrocytomas and 37% of
glioblastomas.34
Chromosomal imbalances that seemed to be
more characteristic for a specific tumor entity rather than a
particular age group were gains of 9p for anaplastic astrocytomas (20%
of pediatric and 14% of adult anaplastic astrocytomas
versus 0 and 5% for glioblastomas) and losses of 13q,
containing the retinoblastoma tumor suppressor gene, for glioblastomas
(31% of pediatric and 38% of adult glioblastomas versus 10
and 23% for anaplastic astrocytomas). Furthermore, characteristic
aberrations distinguishing the two tumor grades within the pediatric
age group consisted of gains of 5q and losses of 9q, 12q, and 22q for
anaplastic astrocytomas as well as gains of 1q, 3q, and 16p and losses
of 8q and 17p for glioblastomas. Loss of 22q is a frequent event in
several tumor entities, eg, ependymomas, pheochromocytomas, rhabdoid
tumors, and hepatocellular, colorectal, ovarian, breast, and oral
squamous cell carcinomas, and was observed in 17, 31, and 38% of adult
low-grade, anaplastic astrocytomas and glioblastomas, respectively,
consistent with a role in astrocytoma tumorigenesis and
progression.35
However, among our pediatric cases loss of
22q was most frequently encountered in anaplastic astrocytomas rather
than glioblastomas (Figure 1)
. Moreover, amplifications of CDK4 (12q),
CCND1 (11q13), and CCND3 (6p21) that have been found in adult malignant
gliomas36
do not seem to play a role among pediatric
high-grade astrocytomas (Figure 1)
. Of the minimal common regions found
in our series most are novel and have not previously been put forward.
They consisted of gains of 1q21-41 (containing the proto-oncogenes SK1,
PBX1, TRK/TRKC, ABL, and ELK4), 2q31-32, 3q27-qter (LAZ3/MLF1), and
5q14-22 (GAP) as well as losses of 6q25-qter (containing the tumor
suppressor genes MAS and IGF2R), 9q34.2 (TSC1), 10q23-25 (PTEN),
11q14-22, 16q22-qter (CMAR/CAR and E-cadherin), 17p (TP53 and OVCA1/2),
and 22q12-qter (BAM22 and NF2). Furthermore, the whole q-arm of
chromosome 10 was lost in five cases, indicating that not only PTEN but
additional putative tumor suppressor genes may be involved in the
development of glioblastomas, eg, DMBT1, FGFR2, and/or MXI1 that are
located at the distal end of 10q.37-39
Apart from PTEN,
DMBT1, FGFR2, and MXI1 at 10q, TP53 at 17p13.1 and NF2 at 22q12, none
of the known tumor oncogenes or suppressor genes mentioned above have
yet been associated with astrocytoma tumorigenesis.
Tumor Proliferation, Genetics, and Prognosis
MIB-1 proliferation indices in our series were significantly lower in anaplastic astrocytomas than in glioblastomas corroborating previous investigations.40,41 A significantly shorter survival was found for all high-grade astrocytomas with a MIB-1 proliferation index of >25% compared with tumors with a MIB-1 proliferation index of <25% reflecting a similar difference in outcome presented for MIB-1 proliferation cut-off levels of 11%41 and 12%.40 Although age was not prognostically significant, the three youngest children in this series which were <2 years of age were still alive 8, 27, and 107 months after diagnosis. The 1-, 2-, and 3-year survival rates overall in our series were 55, 33, and 21%, respectively, which are similar to the 2- and 3-year survival rates of 43% and 35 to 36% encountered elsewhere.42-44 Outcome was significantly better for anaplastic astrocytomas with 1-, 2-, and 3-year survival rates of 70, 60, and 50%, respectively, compared with 46, 9, and 0% for pediatric glioblastomas.
When correlating prognosis with specific chromosomal aberrations, a
significantly shorter survival among children with anaplastic
astrocytomas showing gains of 1q compared to cases without this change
could be found; this correlation could not be established for pediatric
glioblastomas (Figure 2)
. Although an association between TP53 mutation
and a shorter progression-free survival has been proposed in one
investigation on pediatric malignant gliomas,45
no
association between presence of TP53 mutation and patient survival was
found in two series of adult glioblastomas;8,46
in our
study, the five children with -17p all died between 4 and 15 months
after diagnosis. Correlations between prognosis, tumor dynamics and
chromosomal imbalances have previously been presented for adult
high-grade gliomas and consisted of more frequent gains of 12q14-21 and
19 among slower growing glioblastomas compared with gains of 6q16-qter,
13, and 20 among faster growing subgroup.26
Another study
found higher chemosensitivity, longer recurrence-free survival after
chemotherapy, and longer overall survival among anaplastic
oligodendrogliomas to be associated with loss of chromosomes 1p and 19q
whereas CDKN2A gene deletions had a worse prognosis.47
On
the other hand, loss of the INK4a-ARF locus on chromosome 9p21 and loss
of 14pARF expression was speculated to contribute to the highly
malignant behavior and treatment resistance of adult high-grade
astrocytomas;48
loss of material on 9p, however, only
affected two children with anaplastic astrocytomas, who were still
alive at time of reporting, and one with glioblastoma who died after 12
months.
In conclusion, our results show that although there may be some common genetic steps in the malignant progression of both pediatric and adult astrocytomas there are also a number of nonrandom aberrations that have not been observed in adult high-grade astrocytomas, supporting a different genetic pathway and contradicting the notion that high-grade astrocytic tumors in children differ from those in adults by lacking consistent numerical and structural deviations.10 Furthermore, gains of chromosomal material on 1q seem to be correlated with a worse prognosis among pediatric anaplastic astrocytomas but not glioblastomas. However, as these histological groups consisted of only 10 and 13 cases, respectively, this finding will have to be corroborated among a larger cohort.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Accepted for publication January 5, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. E. Warren Pediatric High-grade Gliomas ASCO Educational Book, January 1, 2008; 2008(1): 472 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Faury, A. Nantel, S. E. Dunn, M.-C. Guiot, T. Haque, P. Hauser, M. Garami, L. Bognar, Z. Hanzely, P. P. Liberski, et al. Molecular Profiling Identifies Prognostic Subgroups of Pediatric Glioblastoma and Shows Increased YB-1 Expression in Tumors J. Clin. Oncol., April 1, 2007; 25(10): 1196 - 1208. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Broniscer and A. Gajjar Supratentorial High-Grade Astrocytoma and Diffuse Brainstem Glioma: Two Challenges for the Pediatric Oncologist Oncologist, April 1, 2004; 9(2): 197 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. Rickert, O. D. Wiestler, and W. Paulus Chromosomal Imbalances in Choroid Plexus Tumors Am. J. Pathol., March 1, 2002; 160(3): 1105 - 1113. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
