Introduction
Ependymal tumours are relatively uncommon primary neoplasms of the central nervous system (CNS) (3-9% of gliomas) arising from the lining of the ventricular system and from the remnants of the central canal of the spinal cord [37]. They are the third most common CNS malignancy in childhood, after astrocytomas and medulloblastomas. The current WHO classification recognizes ependymoma (WHO grade II), anaplastic ependymoma (WHO grade III), myxopapillary ependymoma (WHO grade I), and subependymoma (WHO grade I) [37]. Ependymomas may develop at any site along cerebral ventricles and the spinal canal. However, there is an interesting correlation between the site of their development and the age of patients: about 90% of paediatric ependymomas arise within the cranium, whereas most adult ependymomas develop in the spinal cord [9,20,23]. Myxopapillary ependymomas and subependymomas are rare variants of ependymal tumours and they usually develop in the cauda equina and at the ventricular wall, respectively [23,44].
Ependymomas affect mostly children and young adults, and are characterized by tremendous variability of their clinical behaviour. The overall 3-year survival rate in paediatric tumours is approximately 75% [70], and the overall 5-year survival rate for the adults is 60-70% [50]. Patients’ age, anatomical location of the tumour, and the extent of surgical excision are all parameters of prognostic significance. In contrast, the prognostic significance of the specific microscopic features of those tumours, including their grading system, remains a controversial issue. In this setting, elucidation of the complex molecular changes may result in more precise understanding of their biology and, as a consequence, it may be of predictive and prognostic value. Therefore, we have decided to review the current state of knowledge on molecular alterations in ependymomas, with special regard to the pathological and clinical consequences of these aberrations.
Chromosomal abnormalities
Cytogenetic studies have shown that chromosomal abnormalities are relatively common in ependymomas [5,23,41,66,75]. Because of the rarity of ependymomas, and their clinical heterogeneity, the role of specific molecular alterations in the biological behaviour of these tumours remains unclear. Furthermore, when alterations are detected, they do not consistently help in distinguishing between low and high grade tumours [23].
Numerical aberrations of chromosomes in ependymomas show important differences between tumours developing in children and adults. About 40% of childhood ependymomas show a balanced chromosomal profile, in contrast to approximately 9% of adult tumours [11,15,48,78]. Likewise, there are differences between intracranial and spinal tumours: balanced chromosomal profiles are evident in 21-32% of intracranial ependymomas, and only in up to 3% of spinal ependymomas [11,24,68]. A common pattern of abnormalities in spinal (64%) and adult (56%) ependymomas is gain of multiple whole chromosomes.
A widespread imbalance was shown by comparative genomic hybridization (CGH) only in the myxopapillary ependymomas [11,48,61].
In total, abnormalities of the copy number of chromosomes in ependymomas as detected by classical cytogenetics and by CGH include chromosomes 1, 6, 7, 9, 10, 13, 17, 19 and 22. Deletions are more common, and losses of chromosome 22 are one of the most frequent (20-26%) [11,16,78]. The other chromosomal losses occurred at 1p, 4q, 6q, 9p, 10, 11q, 13q, 16, 17, 19q and 20q [4,6,16,24,35,36,41,48, 60,61,66,73,77,78].
Chromosome 22 and mutations of NF2 gene
This is one of the most common chromosomal alterations in ependymomas. The role of chromosome 22 alterations in the pathogenesis of spinal ependymomas is emphasized by frequent development of that specific tumour in the setting of neurofibromatosis type 2 (NF2) syndrome. The molecular background of that family tumour syndrome depends on the germline mutations of the NF2 gene, whose locus resides at chromosome 22q12. Indeed, cytogenetic studies of ependymomas have implicated chromosome 22 as an important site of nonrandom losses. By classical karyotyping, deletions and translocations involving chromosome 22q were identified in 56% of the adult and 31% of paediatric tumours [5,22,28,38,38,47,53,75,80,81].
By means of other molecular techniques (loss of heterozygosity, CGH), frequency of allelic losses of chromosome 22 varied according to histological variant of ependymomas, their anatomical site, and age of the patient. In a large series of ependymal tumours, allelic losses on 22q were found in 0-100% of cases [6,11,16,25,28,30,47]. Loss of chromosome 22 was significantly associated with a spinal rather than an intracranial location [1,11,16,24]. It is not surprising that, due to the fact of a close relationship between the age of the patients and the location of the tumour (see above), analyses of that alteration in paediatric ependymomas revealed much lower frequency (9-28%) [16,34,77] than in the adult patients (54-56%) [16,38,39]. However, this observation and correlation has not been confirmed by Zheng et al., who identified only a slightly higher frequency of chromosome 22 loss in the intracranial than in the spinal ependymomas (78% vs. 60%) by means of microsatellite and CGH analysis [82]. More recent analysis disclosed preferential 22q loss in the adult infratentorial ependymomas in contrast to supratentorial and spinal ones, which are characterized by –9 and +2/+7/+12/–14q alterations, respectively [30].
NF2 mutations
Despite initial controversies, the NF2 gene is clearly involved in ependymoma tumorigenesis, especially those tumours developing in the spinal cord [7,13,16,36,55,65,77]. The studies on paediatric and intracranial tumours failed to disclose NF2 involvement [13,55,65,77]. However, in a large study of 62 ependymomas, Ebert et al. [16] showed NF2 mutations in 43% of intramedullary tumours in contrast to none of the intracranial ependymomas. Interestingly, all the tumours bearing NF2 mutations disclosed LOH 22 [16,36]; this indicates that NF2 plays an important role in the oncogenesis of spinal ependymomas and shows genetically distinct subsets among WHO Grade II ependymomas [16].
Mutations of NF2 in ependymomas affected splice sites in two tumours, frame shift mutations (two deletions and one insertion) with the introduction of premature stop codons in three tumours, and a nonsense mutation creating an immediate stop codon in one tumour [16]. These NF2 changes affected exons 1, 5, 7 (two instances), 10, and 13. A mutation reported by Alonso et al. [1] represented the first sequence duplication of NF2 gene.
Other putative antioncogenes at chromosome 22
Analyses of non-NF2 families with ependymomas suggested a putative involvement of other tumour suppressor genes in the pathogenesis of these tumours independently of the NF2 gene [27,42,59]. In support of this view, a case of anaplastic ependymoma was reported in a 5-year-old boy with a balanced reciprocal translocation of his constitutional karyotype t(1;22) (p22;q11.2) [45]. As the chromosomal breakpoint was located proximally to the NF2 locus, it seemingly did not alter the gene itself. The putative role of the hSNF5/INI1 gene in the evolution of ependymomas was excluded [35]. Molecular analysis of 53 ependymal tumours from 48 patients failed to identify mutations or homozygous deletions of the hSNF5/INI1 gene [35]. These findings corroborate the results of a study by Sevenet et al. [63], who did not detect alterations of the hSNF5/INI1 gene in 25 ependymomas.
Chromosome 1
Gain of chromosome 1q was a frequent finding in intracranial ependymomas and this alteration was significantly associated with posterior fossa location and anaplastic histological features (WHO grade III) [11,15,24,39,48,61]. Recently, fluorescence in situ hybridization (FISH) analysis determined gain of 1q25 as an independent prognostic marker for either recurrence-free survival or overall survival in ependymomas [39].
Chromosome 6
Rearrangements and loss of chromosome 6q are common findings in a number of cases of adult and paediatric ependymomas [22,26,34,38,41,48,53,75]. No tumour suppressor gene has yet been identified at that locus. Structural abnormalities of 6q in ependymomas occurred in association with other chromosome abnormalities [34,41,53,65]. A strong association of loss of chromosome 6q with infratentorial location was reported by Hirose et al. [24] and later confirmed by Carter et al. [11]; all of the tumours with that alteration were located in the posterior fossa.
Allelotyping studies of ependymomas defined a hot spot deletion region at chromosome 6 (6q25.2-27) [73]. Frequent aberrations were also detected at other chromosomal regions: 6q15-16, 6q24 and 6q21-22.1 [26].
Chromosome 9
Several conventional cytogenetic studies [4,5,12, 14,41,47,53,66,69,81] described gains of chromosome 9 or translocations involving chromosome 9 in approximately 15% of ependymomas. Gain of 9p24.3-qter was identified as one of the most common alterations in ependymomas (58%) [39]. Comparing cases with alterations of chromosome 6q and/or chromosome 9, there appears to be a mutually exclusive correlation between these chromosomal aberrations [24,26]. On the other hand, loss of the whole chromosome 9 was associated with gain on 1q [24]. Furthermore, loss of whole 9 and 6q were identified in ependymomas with different anatomical location; the former alteration was seen in supratentorial lesions, while the latter was found in infratentorial tumours [24]. Loss of chromosome 9 is regarded as a hallmark of clear cell ependymomas, which preferably show this alteration in 40% of WHO grade II and 100% of WHO grade III lesions [51].
The INK4A/ARF/INK4B (CDKN2A/ARF/CDKN2B) locus is mapped to chromosome 9p21 and it is mutated in many cancers. It encodes three polypeptides that regulate cell proliferation via the RB and P53 tumour suppressor pathways. CDKN2A and CDKN2B encode the P16INK4a and P15INK4b polypeptides, respectively. These genes have a highly conserved amino acid sequence and seemingly they result from the duplication of the same gene [64]. They are inhibitors of CDK4 and CDK6 and, thereby, they block phosphorylation of RB [62]. The third product derived from that locus is ARF (for Alternative Reading Frame). As its name implies, it has no isoforms and structural homology with P16 and P15 [19,64]. It regulates P53 activity in response to unscheduled growth response signals generated by oncogenes. The alternative exons designated as 1α and 1β are spliced into common exons 2 and 3. P16INK4a is composed of the transcript exon 1α-exon 2-exon3, while ARF is encoded by exon 1β-exon 2-exon3 transcript and it has its own promoter [64].
Deletions and mutations in the CDKN2A gene are uncommon in ependymomas [10,58]. Loss of P16 expression was uncommonly found by immunohistochemistry (6.25%) [8]. The results presented by Bouvier et al. [10] suggest that in ependymomas, lack of P16INK4a is not associated with anaplasia and is inversely correlated with the Ki-67 labelling index (LI). It was also shown that P16INK4a was expressed only when cellular proliferation reached a threshold level [10]. Despite these studies that suggested an insignificant role of P16 alterations in ependymomas, Taylor et al. identified a preferential CDKN2A deletion in supratentorial tumours by array comparative genomic hybridization and FISH [68].
Although the significance of P16 inactivation for the pathogenesis of ependymomas is not clear, inactivation of P14ARF appears to play a role in ependymoma progression as it was shown in about 30% of these tumours [2,32].
Chromosome 5
CGH analysis revealed high incidence of gains on chromosome 5 (46%) with an overlapping region of DNA gain mapped to 5q21-22 [82]. Recurrent gains at 5p15.33 were determined as an adverse prognostic factor with resultant overexpression of hTERT leading to an increase of telomerase activity [39].
Chromosome 7
In contrast to glioblastomas, gain of chromosome 7 has been less commonly reported in ependymal tumours. Karyotyping revealed gain of chromosome 7 in a number of cases of ependymomas; most of them were anaplastic (WHO grade III) [4,21,47,53,60,75,81]. The frequency of whole chromosome 7 gains differed significantly between spinal and intracranial ependymomas; furthermore, intraspinal location was preferentially seen in adult patients [24]. A recent study confirmed these results, as high frequency of gains on 7q11.23-22.1 (58%) was identified in spinal tumours; gains of chromosome 7 were also one of the most common chromosomal imbalances independently of the anatomical location [39].
Gains of chromosome 7 were shown as a common genetic characteristic not only of spinal WHO grade II/III ependymomas but myxopapillary ependymomas as well [24]. These tumours differed in the profile of other chromosomal changes, as loss on 22q and gains of 15q and 12 had not occurred in myxopapillary tumours, in contrast to losses of chromosomes 1, 2, and 10, which occurred solely in the myxopapillary group [24].
Chromosome 10
Losses of chromosome 10 were reported in about 9-19% of ependymomas [3,21,53,60,61,66,81]. Similarly to oligodendrogliomas, it has been suggested that chromosome 10 loss may represent a final step in the malignant evolution of ependymomas [23].
In a study of spinal ependymomas, losses on chromosome 10 were seen only in myxopapillary tumours [11].
Chromosome 11
Monosomy of chromosome 11 has been described uncommonly in ependymomas [3,29,30,47,60,80,81]. Rearrangements involving 11q13 were described in a few paediatric cases [12,41,57]. The locus 11q13 is known to contain the oncogenes BCL1, HST and INT2, which are amplified in some human cancers [12], and it is likely that one of these genes plays a role in pathogenesis of ependymomas. MEN1 mutation at 11q13 was identified in the recurrences of ependymoma WHO Grade II, that presented with LOH11q only [36]. This finding suggests a possible role of that alteration in ependymoma progression to higher grades [36,74].
Chromosome 13
With conventional cytogenetics, losses of chromosome 13 were described in approximately 5% of ependymomas [47,60,66].
Chromosome 16
Loss of chromosome 16 has not been reported as a consistent marker in ependymomas and the data on that subject are controversial. Monosomy of chromosome 16 was reported in one out of four ependymomas [60]. Much higher frequency (50-57%) of chromosome 16p loss was reported in more recent publications with the overlapped deletion regions mapped at loci 16p13.1-13.3 and 16q22-q24 [30,82].
Chromosome 17
Deletion of chromosome 17 is of particular interest because of the presence of a well-known tumour suppressor gene, TP53 (17p13), and NF1 (17q11.2). Monosomy 17 is one of the most common chromosomal abnormalities in ependymomas, especially in paediatric patients [21,47,53,66,81]. In a microsatellite analysis, von Haken et al. demonstrated that 50% of ependymomas harboured 17p arm loss, preferentially at the terminal end of 17p [77]. CGH data reported by Zheng et al. [82] indicated DNA losses in both arms of chromosome 17. However, other studies did not find convincing evidence of chromosome 17 abnormalities in ependymomas [6,24,28,30,34,38,47]. A recent study by Mendrzyk et al. suggested a candidate gene PRKCA responsible for ependymoma development, that is lost from the chromosome locus 17q24.2 [39].
Although a TP53 germline mutation has been described in one patient with anaplastic ependymoma [40], somatic mutations of TP53, mapped to 17p13.1, are rarely affected by LOH or point mutation [17,34,43,71,77], indicating that tumour suppressor genes other than TP53 are most likely involved in the aetiology of ependymomas.
Other genetic abnormalities
Ependymal tumours do not show amplification at classical amplified loci (MYCC, MYCN and EGFR) [22]. One study detected low accumulation of MYCN transcript in ependymoma without elevated MYCN gene copy number [18]. In another small series of ependymomas no amplification of MYCN was identified by Southern blotting [79].
EGFR overexpression, as determined by RT-PCR, was observed in 3 of 3 spinal and 6 of 7 intracranial ependymomas at similar levels and independently of DNA copy number [39]. Immunohistochemical EGFR overexpression was frequently detected and was correlated with adverse outcome in intracranial tumours. No correlation between EGFR overexpression and overall survival was observed in the spinal ependymomas [39].
Data on MDM2 amplification in ependymomas are contradictory. The gene was mapped to chromosome 12q13-q14. It encodes the protein that specifically binds and inactivates P53. Suzuki et al. [67] detected MDM2 gene amplification in 35% of ependymomas by differential PCR. Using the same technique, Tong et al. found MDM2 gene amplification in only one case in their series of 26 ependymomas [72]. Another study performed by Southern blotting in 8 ependymomas did not reveal MDM2 gene amplification in any of the tumours [49].
Gene silencing by CpG island hypermethylation seems to be rather non-operative in ependymomas as it was uncommonly identified in the ten genes analyzed by Alonso et al. [2]: 28% for MGMT; 28% for GSTP1; 57% for DAPK; 28% for TP14ARF; 0% for THBS1; 28% for TIMP3; 14% for TP73; 0% for CDKN2A/ /P16INK4A; 14% for RB1; and 0% for TP53. In another study, promoter methylation for CDKN2A, CDKN2B and P14ARF was identified in 21%, 32% and 21% of ependymomas, respectively [54]. In posterior fossa ependymomas all three genes were less frequently methylated in paediatric patients than in the adults. For CDKN2B, extracranial tumours were more frequently methylated (50%) than intracranial ones (23%). For CDKN2B and P14ARF, methylation was more frequent in low-grade tumours; the reverse was observed for CDKN2A [54].
Histogenesis of ependymoma subsets as defined by gene expression profiles
A new technique of gene expression profiling enables simultaneous estimation of thousands of genes at the mRNA level. In a recent study, Taylor et al. have shown that ependymomas from various anatomical locations (the supratentorial region, the posterior fossa and the spinal cord) exhibit distinct patterns of gene expression and chromosomal losses and gains [68]. Interestingly, neither clinical nor histological features correlated with these molecular profiles. An important molecular hallmark of the supratentorial ependymomas was identified as an increase of expression of the members of EPHB-EPHRIN and NOTCH signalling pathways. In contrast, spinal ependymomas showed preferential expression of homeobox (HOX) family members. Interestingly, the genetic signature of these subgroups precisely correlates with the gene expression profiles of the normal ependymal cells developing from the embryonic radial glial cells (RGCs) in the subventricular zone of the lateral ventricles and spinal canal, respectively. This confirms earlier observations that supratentorial and spinal ependymomas may arise from different populations of neural progenitor cells [68]. Taylor et al. have defined these progenitor cells more closely. The subset of RGCs was identified in the population of ependymoma cells (phenotype CD133+/RC2+/BLPB+) and the orthotopic transplants composed of these cells were capable of tumour formation, in contrast to CD133-negative and unsorted ependymoma cells. Taylor et al. [69] concluded that in ependymomas these stem cells have properties of self-renewal and multipotency and may represent the cellular targets of primary mutations that promote disease.
Prognostic significance of molecular markers
Several prognostic studies indicate that ependymomas developing in children fare worse than in the adults [37]. This difference may derive to some extent from the preferential location of paediatric ependymomas in the infratentorial compartment in contrast to the spinal ependymomas that prevail in adults.
The recent histological classification of ependymomas has thus far proven to be an unreliable predictor of clinical outcome. Likewise, the relationship between ependymoma grade and specific chromosomal aberrations is also controversial [11,15,24,30,31]. Although some age-related immunohistochemical patterns [33] and genetic alterations [11,15,24] have been found to be associated with clinical outcome in ependymoma patients, the underlying biological mechanisms remain unclear.
The relationships between ependymoma grade, specific chromosomal aberrations [11,24,30,61] and clinical outcome [33,46,52] are also controversial. Some reports indicate that gain of 1q may be a potential marker of poor prognosis in paediatric ependymomas [11,15]. The effect of gain of lq on survival of patients with intracranial ependymomas was examined. Survival curves of intracranial tumours split into classic and anaplastic groups, and those of intracranial tumours with and without gain of lq also showed clear differences. The difference between patients with anaplastic ependymomas showing gain of 1q and other tumours was even more significant [11].
Dyer et al. distinguished ependymal tumours according to the number of the chromosomal imbalances into “numerical” tumours (a total of 13 or more chromosome imbalances), “structural” tumours (a to- tal of six or fewer imbalances) and a “balanced” group (no genetic imbalances) [15]. In multivariate analysis the structural tumours had a significantly worse outcome when compared with the other two genetic groups.
TP53 gene mutations were rarely detected in ependymal tumours [6,17,23], whereas aberrant P53 overexpression was closely correlated with both high-grade ependymomas and poor prognosis [56,76].
A high level of P14ARF expression is independently associated with prolonged progression-free survival in high-grade ependymomas [33].
Conclusions
Overall, the data indicate that spinal ependymomas, which present almost exclusively in adult patients, and intracranial childhood tumours differ significantly in their genetic profiles. Categorization of these tumours by cytogenetic aberrations may help establish a classification system that predicts patient outcome. Intracranial ependymomas may also be discriminated by molecular analyses into supratentorial and infratentorial lesions [24]. The former show preferentially loss of the whole chromosome 9, while loss on 6q is a hallmark of the latter tumours. Among spinal ependymal tumours, molecular studies disclosed basic differences between WHO grade II/III lesions and myxopapillary ependymomas (WHO grade I), despite commonly shared chromosomal 7 gains. The latter tumours did not show loss of chromosome 22 or gains of 15q and 12, but had losses of chromosomes 1, 2 and 10 [24].
Acknowledgements
This paper was supported by the Medical University of Gdansk (ST-542), Poland.
References
1. Alonso ME, Bello MJ, Arjona D, Gonzalez-Gomez P, Lomas J, de Campos JM, Kusak ME, Isla A, Rey JA. Analysis of the NF2 gene in oligodendrogliomas and ependymomas. Cancer Genet Cytogenet 2002; 134: 1-5.
2. Alonso ME, Bello MJ, Gonzalez-Gomez P, Arjona D, Lomas J, de Campos JM, Isla A, Sarasa JL, Rey JA. Aberrant promoter methylation of multiple genes in oligodendrogliomas and ependymomas. Cancer Genet Cytogenet 2003; 144: 134-142.
3. Arnoldus EP, Wolters LB, Voormolen JH, van Duinen SG, Raap AK, van der Ploeg M, Peters AC. Interphase cytogenetics: a new tool for the study of genetic changes in brain tumors. J Neurosurg 1992; 76: 997-1003.
4. Bhattacharjee MB, Armstrong DD, Vogel H, Cooley LD. Cytogenetic analysis of 120 primary pediatric brain tumors and literature review. Cancer Genet Cytogenet 1997; 97: 39-53.
5. Bigner SH, McLendon RE, Fuchs H, McKeever PE, Friedman HS. Chromosomal characteristics of childhood brain tumors. Cancer Genet Cytogenet 1997; 97: 125-134.
6. Bijlsma EK, Voesten AM, Bijleveld EH, Troost D, Westerveld A, Mérel P, Thomas G, Hulsebos TJ. Molecular analysis of genetic changes in ependymomas. Genes Chromosomes Cancer 1995; 13: 272-277.
7. Birch BD, Johnson JP, Parsa A, Desai RD, Yoon JT, Lycette CA, Li YM, Bruce JN. Frequent type 2 neurofibromatosis gene transcript mutations in sporadic intramedullary spinal cord ependymomas. Neurosurgery 1996; 39: 135-140.
8. Bortolotto S, Chiadò-Piat L, Cavalla P, Bosone I, Mauro A, Schiffer D. CDKN2A/p16 in ependymomas. J Neurooncol 2001; 54: 9-13.
9. Bouffet E, Perilongo G, Canete A, Massimino M. Intracranial ependymomas in children: a critical review of prognostic factors and a plea for cooperation. Med Pediatr Oncol 1998; 30: 319-329.
10. Bouvier-Labit C, Civatte M, Bartoli C, Renaud W, Pellissier JF, Figarella-Branger D. p16INK4a and p19INK4d mRNA expression in neuroglial tumours: correlation with Ki67 proliferation index. Neuropathol Appl Neurobiol 1999; 25: 408-416.
11. Carter M, Nicholson J, Ross F, Crolla J, Allibone R, Balaji V, Perry R, Walker D, Gilbertson R, Ellison DW. Genetic abnormalities detected in ependymomas by comparative genomic hybridisation. Br J Cancer 2002; 86: 929-939.
12. Dal Cin P, Sandberg AA. Cytogenetic findings in a supratentorial ependymoma. Cancer Genet Cytogenet 1988; 30: 289-293.
13. De Vitis LR, Tedde A, Vitelli F, Ammannati F, Mennonna P, Bono P, Grammatico B, Grammatico P, Radice P, Bigozzi U, Montali E, Papi L. Analysis of the neurofibromatosis type 2 gene in different human tumors of neuroectodermal origin. Hum Genet 1996; 97: 638-641.
14. Debiec-Rychter M, Lasota J, Alwasiak J, Liberski PP. Recurrent anaplastic ependymoma with an abnormal karyotype and c-myc proto-oncogene overexpression. Acta Neuropathol (Berl) 1995; 89: 270-274.
15. Dyer S, Prebble E, Davison V, Davies P, Ramani P, Ellison D, Grundy R. Genomic imbalances in pediatric intracranial ependymomas define clinically relevant groups. Am J Pathol 2002; 161: 2133-2141.
16. Ebert C, von Haken M, Meyer-Puttlitz B, Wiestler OD, Reifenberger G, Pietsch T, von Deimling A. Molecular genetic analysis of ependymal tumors. NF2 mutations and chromosome 22q loss occur preferentially in intramedullary spinal ependymomas. Am J Pathol 1999; 155: 627-632.
17. Fink KL, Rushing EJ, Schold SC Jr, Nisen PD. Infrequency of p53 gene mutations in ependymomas. J Neurooncol 1996; 27: 111-115.
18. Fujimoto M, Sheridan PJ, Sharp ZD, Weaker FJ, Kagan-Hallet S, Story JL. Proto-oncogene analyses in brain tumors. J Neurosurg 1989; 70: 910-915.
19. Gil J, Peters G. Regulation of the INK4b-ARF-INK4a tumour suppressor locus: all for one or one for all. Nat Rev Mol Cell Biol 2006; 7: 667-677.
20. Gilles FH, Sobel EL, Tavaré CJ, Leviton A, Hedley-Whyte ET. Age-related changes in diagnoses, histological features, and survival in children with brain tumors: 1930-1979. The Childhood Brain Tumor Consortium. Neurosurgery 1995; 37: 1056-1068.
21. Griffin CA, Long PP, Carson BS, Brem H. Chromosome abnormalities in low-grade central nervous system tumors. Cancer Genet Cytogenet 1992; 60: 67-73.
22. Grill J, Avet-Loiseau H, Lellouch-Tubiana A, Sévenet N, Terrier-Lacombe MJ, Vénuat AM, Doz F, Sainte-Rose C, Kalifa C, Vassal G. Comparative genomic hybridization detects specific cytogenetic abnormalities in pediatric ependymomas and choroid plexus papillomas. Cancer Genet Cytogenet 2002; 136: 121-125.
23. Hamilton RL, Pollack IF. The molecular biology of ependymomas. Brain Pathol 1997; 7: 807-822.
24. Hirose Y, Aldape K, Bollen A, James CD, Brat D, Lamborn K, Berger M, Feuerstein BG. Chromosomal abnormalities subdivide ependymal tumors into clinically relevant groups. Am J Pathol 2001; 158: 1137-1143.
25. Huang B, Starostik P, Kühl J, Tonn JC, Roggendorf W. Loss of heterozygosity on chromosome 22 in human ependymomas. Acta Neuropathol (Berl) 2002; 103: 415-420.
26. Huang B, Starostik P, Schraut H, Krauss J, Sörensen N, Roggendorf W. Human ependymomas reveal frequent deletions on chromosomes 6 and 9. Acta Neuropathol (Berl) 2003; 106: 357-362.
27. Hulsebos TJ, Oskam NT, Bijleveld EH, Westerveld A, Hermsen MA, van den Ouweland AM, Hamel BC, Tijssen CC. Evidence for an ependymoma tumour suppressor gene in chromosome region 22pter-22q11.2. Br J Cancer 1999; 81: 1150-1154.
28. James CD, He J, Carlbom E, Mikkelsen T, Ridderheim PA, Cavenee WK, Collins VP. Loss of genetic information in central nervous system tumors common to children and young adults. Genes Chromosomes Cancer 1990; 2: 94-102.
29. Jenkins RB, Kimmel DW, Moertel CA, Schultz CG, Scheithauer BW, Kelly PJ, Dewald GW. A cytogenetic study of 53 human gliomas. Cancer Genet Cytogenet 1989; 39: 253-279.
30. Jeuken JW, Sprenger SH, Gilhuis J, Teepen HL, Grotenhuis AJ, Wesseling P. Correlation between localization, age, and chromosomal imbalances in ependymal tumours as detected by CGH. J Pathol 2002; 197: 238-244.
31. Kleihues P, Cavenee WK. World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Nervous System. IARC Press, Lyon, 2000.
32. Korshunov A, Golanov A, Timirgaz V. p14ARF protein (FL-132) immunoreactivity in intracranial ependymomas and its prognostic significance: an analysis of 103 cases. Acta Neuropathol (Berl) 2001; 102: 271-277.
33. Korshunov A, Golanov A, Timirgaz V. Immunohistochemical markers for prognosis of ependymal neoplasms. J Neurooncol 2002; 58: 255-270.
34. Kramer DL, Parmiter AH, Rorke LB, Sutton LN, Biegel JA. Molecular cytogenetic studies of pediatric ependymomas. J Neurooncol 1998; 37: 25-33.
35. Kraus JA, de Millas W, Sörensen N, Herbold C, Schichor C, Tonn JC, Wiestler OD, von Deimling A, Pietsch T. Indications for a tumor suppressor gene at 22q11 involved in the pathogenesis of ependymal tumors and distinct from hSNF5/INI1. Acta Neuropathol (Berl) 2001; 102: 69-74.
36. Lamszus K, Lachenmayer L, Heinemann U, Kluwe L, Finckh U, Höppner W, Stavrou D, Fillbrandt R, Westphal M. Molecular genetic alterations on chromosomes 11 and 22 in ependymomas. Int J Cancer 2001; 91: 803-808.
37. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK WHO Classification of Tumours of the Central Nervous System. IARC, Lyon; 2007.
38. Mazewski C, Soukup S, Ballard E, Gotwals B, Lampkin B.Karyotype studies in 18 ependymomas with literature review of 107 cases. Cancer Genet Cytogenet 1999; 113: 1-8.
39. Mendrzyk F, Korshunov A, Benner A, Toedt G, Pfister S, Radlwimmer B, Lichter P. Identification of gains on 1q and epidermal growth factor receptor overexpression as independent prognostic markers in intracranial ependymoma. Clin Cancer Res 2006; 12: 2070-2079.
40. Metzger AK, Sheffield VC, Duyk G, Daneshvar L, Edwards MS, Cogen PH. Identification of a germ-line mutation in the p53 gene in a patient with an intracranial ependymoma. Proc Natl Acad Sci U S A 1991; 88: 7825-7829.
41. Neumann E, Kalousek DK, Norman MG, Steinbok P, Cochrane DD, Goddard K. Cytogenetic analysis of 109 pediatric central nervous system tumors. Cancer Genet Cytogenet 1993; 71: 40-49.
42. Nijssen PC, Deprez RH, Tijssen CC, Hagemeijer A, Arnoldus EP, Teepen JL, Holl R, Niermeyer MF. Familial anaplastic ependymoma: evidence of loss of chromosome 22 in tumour cells. J Neurol Neurosurg Psychiatry 1994; 57: 1245-1248.
43. Nozaki M, Tada M, Matsumoto R, Sawamura Y, Abe H, Iggo RD. Rare occurrence of inactivating p53 gene mutations in primary non-astrocytic tumors of the central nervous system: reappraisal by yeast functional assay. Acta Neuropathol (Berl) 1998; 95: 291-296.
44. Packer RJ. Ependymomas in children. J Neurosurg 2000; 93: 721-723.
45. Park JP, Chaffee S, Noll WW, Rhodes CH. Constitutional de novo t(1;22)(p22;q11.2) and ependymoma. Cancer Genet Cytogenet 1996; 86: 150-152.
46. Pollack IF, Gerszten PC, Martinez AJ, Lo KH, Shultz B, Albright AL, Janosky J, Deutsch M. Intracranial ependymomas of childhood: long-term outcome and prognostic factors. Neurosurgery 1995; 37: 655-666.
47. Ransom DT, Ritland SR, Kimmel DW, Moertel CA, Dahl RJ, Scheithauer BW, Kelly PJ, Jenkins RB. Cytogenetic and loss of heterozygosity studies in ependymomas, pilocytic astrocytomas, and oligodendrogliomas. Genes Chromosomes Cancer 1992; 5: 348-356.
48. Reardon DA, Entrekin RE, Sublett J, Ragsdale S, Li H, Boyett J, Kepner JL, Look AT. Chromosome arm 6q loss is the most common recurrent autosomal alteration detected in primary pediatric ependymoma. Genes Chromosomes Cancer 1999; 24: 230-237.
49. Reifenberger G, Liu L, Ichimura K, Schmidt EE, Collins VP. Amplification and overexpression of the MDM2 gene in a subset of human malignant gliomas without p53 mutations. Cancer Res 1993; 53: 2736-2739.
50. Reni M, Brandes AA, Vavassori V, Cavallo G, Casagrande F, Vastola F, Magli A, Franzin A, Basso U, Villa E. A multicenter study of the prognosis and treatment of adult brain ependymal tumors. Cancer 2004; 100: 1221-1229.
51. Rickert CH, Korshunov A, Paulus W. Chromosomal imbalances in clear cell ependymomas. Mod Pathol 2006; 19: 958-962.
52. Robertson PL, Zeltzer PM, Boyett JM, Rorke LB, Allen JC, Geyer JR, Stanley P, Li H, Albright AL, McGuire-Cullen P, Finlay JL, Stevens KR Jr, Milstein JM, Packer RJ, Wisoff J. Survival and prognostic factors following radiation therapy and chemotherapy for ependymomas in children: a report of the Children’s Cancer Group. J Neurosurg 1998; 88: 695-703.
53. Rogatto SR, Casartelli C, Rainho CA, Barbieri-Neto J. Chromosomes in the genesis and progression of ependymomas. Cancer Genet Cytogenet 1993; 69: 146-152.
54. Rousseau E, Ruchoux MM, Scaravilli F, Chapon F, Vinchon M, De Smet C, Godfraind C, Vikkula M. CDKN2A, CDKN2B and p14ARF are frequently and differentially methylated in ependymal tumours. Neuropathol Appl Neurobiol 2003; 29: 574-583.
55. Rubio MP, Correa KM, Ramesh V, MacCollin MM, Jacoby LB, von Deimling A, Gusella JF, Louis DN. Analysis of the neurofibromatosis 2 gene in human ependymomas and astrocytomas. Cancer Res 1994; 54: 45-47.
56. Rushing EJ, Brown DF, Hladik CL, Risser RC, Mickey BE, White CL 3rd. Correlation of bcl-2, p53, and MIB-1 expression with ependymoma grade and subtype. Mod Pathol 1998; 11: 464-470.
57. Sainati L, Montaldi A, Putti MC, Giangaspero F, Rigobello L, Stella M, Zanesco L, Basso G. Cytogenetic t(11;17)(q13;q21) in a pediatric ependymoma. Is 11q13 a recurring breakpoint in ependymomas? Cancer Genet Cytogenet 1992; 59: 213-216.
58. Sato K, Schäuble B, Kleihues P, Ohgaki H. Infrequent alterations of the p15, p16, CDK4 and cyclin D1 genes in non-astrocytic human brain tumors. Int J Cancer 1996; 66: 305-308.
59. Sato T, Shimoda A, Takahashi T, Kurokawa H, Ando M, Goto S, Takamura H. Congenital anaplastic ependymoma: a case report of familial glioma. Childs Brain 1984; 11: 342-348.
60. Sawyer JR, Sammartino G, Husain M, Boop FA, Chadduck WM. Chromosome aberrations in four ependymomas. Cancer Genet Cytogenet 1994; 74: 132-138.
61. Scheil S, Brüderlein S, Eicker M, Herms J, Herold-Mende C, Steiner HH, Barth TF, Möller P. Low frequency of chromosomal imbalances in anaplastic ependymomas as detected by comparative genomic hybridization. Brain Pathol 2001; 11: 133-143.
62. Serrano M. The INK4a/ARF locus in murine tumorigenesis. Carcinogenesis 2000; 21: 865-869.
63. Sévenet N, Sheridan E, Amram D, Schneider P, Handgretinger R, Delattre O. Constitutional mutations of the hSNF5/INI1 gene predispose to a variety of cancers. Am J Hum Genet 1999; 65: 1342-1348.
64. Sharpless NE. INK4a/ARF: a multifunctional tumor suppressor locus. Mutat Res 2005; 576: 22-38.
65. Slavc I, MacCollin MM, Dunn M, Jones S, Sutton L, Gusella JF, Biegel JA. Exon scanning for mutations of the NF2 gene in pediatric ependymomas, rhabdoid tumors and meningiomas. Int J Cancer 1995; 64: 243-247.
66. Stratton MR, Darling J, Lantos PL, Cooper CS, Reeves BR. Cytogenetic abnormalities in human ependymomas. Int J Cancer 1989; 44: 579-581.
67. Suzuki SO, Iwaki T. Amplification and overexpression of mdm2 gene in ependymomas. Mod Pathol 2000; 13: 548-553.
68. Taylor MD, Poppleton H, Fuller C, Su X, Liu Y, Jensen P, Magdaleno S, Dalton J, Calabrese C, Board J, Macdonald T, Rutka J, Guha A, Gajjar A, Curran T, Gilbertson RJ. Radial glia cells are candidate stem cells of ependymoma. Cancer Cell 2005; 8: 323-335.
69. Thiel G, Losanowa T, Kintzel D, Nisch G, Martin H, Vorpahl K, Witkowski R. Karyotypes in 90 human gliomas. Cancer Genet Cytogenet 1992; 58: 109-120.
70. Timmermann B, Kortmann RD, Kühl J, Meisner C, Slavc I, Pietsch T, Bamberg M. Combined postoperative irradiation and chemotherapy for anaplastic ependymomas in childhood: results of the German prospective trials HIT 88/89 and HIT 91. Int J Radiat Oncol Biol Phys 2000; 46: 287-295.
71. Tominaga T, Kayama T, Kumabe T, Sonoda Y, Yoshimoto T. Anaplastic ependymomas: clinical features and tumour suppressor gene p53 analysis. Acta Neurochir (Wien ) 1995; 135: 163-170.
72. Tong CY, Ng HK, Pang JC, Hui AB, Ko HC, Lee JC. Molecular genetic analysis of non-astrocytic gliomas. Histopathology 1999; 34: 331-341.
73. Tong CY, Zheng PP, Pang JC, Poon WS, Chang AR, Ng HK. I Identification of novel regions of allelic loss in ependymomas by high-resolution allelotyping with 384 microsatellite markers. J Neurosurg 2001; 95: 9-14.
74. Urioste M, Martínez-Ramírez A, Cigudosa JC, Colmenero I, Madero L, Robledo M, Martínez-Delgado B, Benítez J. Complex cytogenetic abnormalities including telomeric associations and MEN1 mutation in a pediatric ependymoma. Cancer Genet Cytogenet 2002; 138: 107-110.
75. Vagner-Capodano AM, Zattara-Cannoni H, Gambarelli D, Figarella-Branger D, Lena G, Dufour H, Grisoli F, Choux M. Cytogenetic study of 33 ependymomas. Cancer Genet Cytogenet 1999; 115: 96-99.
76. Verstegen MJ, Leenstra DT, Ijlst-Keizers H, Bosch DA. Proliferation- and apoptosis-related proteins in intracranial ependymomas: an immunohistochemical analysis. J Neurooncol 2002; 56: 21-28.
77. von Haken MS, White EC, Daneshvar-Shyesther L, Sih S, Choi E, Kalra R, Cogen PH. Molecular genetic analysis of chromosome arm 17p and chromosome arm 22q DNA sequences in sporadic pediatric ependymomas. Genes Chromosomes Cancer 1996; 17: 37-44.
78. Ward S, Harding B, Wilkins P, Harkness W, Hayward R, Darling JL, Thomas DG, Warr T. Gain of 1q and loss of 22 are the most common changes detected by comparative genomic hybridisation in paediatric ependymoma. Genes Chromosomes Cancer 2001; 32: 59-66.
79. Wasson JC, Saylors RL 3rd, Zeltzer P, Friedman HS, Bigner SH, Burger PC, Bigner DD, Look AT, Douglass EC, Brodeur GM. Oncogene amplification in pediatric brain tumors. Cancer Res 1990; 50: 2987-2990.
80. Weremowicz S, Kupsky WJ, Morton CC, Fletcher JA. Cytogenetic evidence for a chromosome 22 tumor suppressor gene in ependymoma. Cancer Genet Cytogenet 1992; 61: 193-196.
81. Wernicke C, Thiel G, Lozanova T, Vogel S, Kintzel D, Jänisch W, Lehmann K, Witkowski R. Involvement of chromosome 22 in ependymomas. Cancer Genet Cytogenet 1995; 79: 173-176.
82. Zheng PP, Pang JC, Hui AB, Ng HK. Comparative genomic hybridization detects losses of chromosomes 22 and 16 as the most common recurrent genetic alterations in primary ependymomas. Cancer Genet Cytogenet 2000; 122: 18-25.
