Genomic and Protein Expression Profiling Identifies CDK6 As Novel Independent Prognostic Marker in Medulloblastoma
http://www.100md.com
《临床肿瘤学》
the Division of Molecular Genetics, German Cancer Research Center
Central Unit Biostatistics, German Cancer Research Center, Heidelberg
Department of Neuropathology, Heinrich Heine University, Duesseldorf, Germany
Wellcome Trust Genome Campus, The Wellcome Trust Sanger Institute, Hinxton, United Kingdom
Department of Neuropathology, Burdenko Neurosurgical Institute, Moscow, Russia.
ABSTRACT
PURPOSE: Medulloblastoma is the most common malignant brain tumor in children. Despite multimodal aggressive treatment, nearly half of the patients die as a result of this tumor. Identification of molecular markers for prognosis and development of novel pathogenesis-based therapies depends crucially on a better understanding of medulloblastoma pathomechanisms.
PATIENTS AND METHODS: We performed genome-wide analysis of DNA copy number imbalances in 47 medulloblastomas using comparative genomic hybridization to large insert DNA microarrays (matrix-CGH). The expression of selected candidate genes identified by matrix-CGH was analyzed immunohistochemically on tissue microarrays representing medulloblastomas from 189 clinically well-documented patients. To identify novel prognostic markers, genomic findings and protein expression data were correlated to patient survival.
RESULTS: Matrix-CGH analysis revealed frequent DNA copy number alterations of several novel candidate regions. Among these, gains at 17q23.2-qter (P < .01) and losses at 17p13.1 to 17p13.3 (P = .04) were significantly correlated to poor prognosis. Within 17q23.2-qter and 7q21.2, two of the most frequently gained chromosomal regions, confined amplicons were identified that contained the PPM1D and CDK6 genes, respectively. Immunohistochemistry revealed strong expression of PPM1D in 148 (88%) of 168 and CDK6 in 50 (30%) of 169 medulloblastomas. Overexpression of CDK6 correlated significantly with poor prognosis (P < .01) and represented an independent prognostic marker of overall survival on multivariate analysis (P = .02).
CONCLUSION: We identified CDK6 as a novel molecular marker that can be determined by immunohistochemistry on routinely processed tissue specimens and may facilitate the prognostic assessment of medulloblastoma patients. Furthermore, increased protein-levels of PPM1D and CDK6 may link the TP53 and RB1 tumor suppressor pathways to medulloblastoma pathomechanisms.
INTRODUCTION
Medulloblastoma is a highly malignant embryonal tumor of the cerebellum that accounts for 20% to 25% of all primary brain tumors in children; the incidence peaks at 7 years of age.1 Several clinical, histologic, and molecular parameters have been identified as predictive factors for patient outcome.2-6 Despite aggressive multimodal therapy, including surgery, irradiation, and chemotherapy, disease dissemination is common and the 5-year survival rate is only 50% to 60%.7
To develop novel therapeutic strategies for the improvement of clinical outcome, it is of paramount importance to increase our understanding of the cellular and molecular basis of this disease. Previous molecular studies showed that the sonic hedgehog signaling pathway is aberrantly activated by genetic alterations in the PTCH, SMOH, or SUFU genes in up to 20% of medulloblastomas, affecting mostly tumors of the desmoplastic subtype.8 A smaller fraction of medulloblastomas demonstrate alterations in genes involved in wingless/WNT (wingless-type MMTV integration site family) signaling, such as CNNTB1, APC, and AXIN.8,9 Studies using conventional karyotyping or comparative genomic hybridization (CGH) identified trisomy 7 and isochromosome 17 [i(17q)] as frequent chromosomal aberrations in medulloblastoma.10-16 Furthermore, amplification of the proto-oncogenes MYC or MYCN is common in large-cell anaplastic medulloblastoma—a rare but highly malignant subtype of medulloblastoma.15-19
To overcome the restricted resolution of chromosomal CGH,20 matrix-based CGH (matrix-CGH) was developed using hybridization targets immobilized on microarrays.21-23 For genome-wide profiling of copy number imbalances, we constructed a DNA microarray consisting of approximately 6,000 large-insert genomic clones covering the whole genome with an average resolution of 0.5 megabases (Mb).24 In this study, we investigated 47 sporadic medulloblastomas using this microarray to identify recurrent DNA copy number imbalances and novel medulloblastoma-associated candidate genes. The matrix-CGH findings were compared with real-time quantitative reverse transcriptase polymerase chain reaction (RQ-PCR), interphase fluorescent in situ hybridization (FISH), and immunohistochemistry (IHC) using a tissue microarray (TMA) composed of medulloblastomas from 189 patients with available clinical follow-up. Genomic imbalances were clustered, combined with protein expression data, and correlated to clinical outcome to search for novel prognostic markers.
PATIENTS AND METHODS
Tumor Material and Patient Characteristics
All samples used in this study were collected at the Department of Neuropathology, Burdenko Neurosurgical Institute (Moscow, Russia), between 1993 and 2003. All diagnoses were confirmed by histologic assessment of specimens obtained at surgery by at least two neuropathologists according to the criteria of the WHO classification.25 The tumors were classified by histology as classic, desmoplastic, or anaplastic medulloblastomas; the latter diagnosis was made only when signs of moderate or severe anaplasia were present.3
All samples were provided from an existing tumor bank for research purposes and approval to link laboratory data to clinical data was obtained by the institutional review board. This procedure is according to the official statement of the National Ethics Council of the federal government of Germany (March 2004). Tumor and patient characteristics are summarized in \?\Appendix Table 1. In the matrix-CGH and TMA study, 53 and 189 tumor samples were analyzed, respectively. All patients received craniospinal irradiation after surgery consisting of 36 Gy to the craniospinal axis and a boost of 53 to 56 Gy to the posterior fossa. Adjuvant chemotherapy using lomustine, cisplatin, and vincristine7,26 was administered routinely to all patients treated after 1995, including the patients of the matrix-CGH study and 110 patients (58%) in the TMA study. Before 1995, no chemotherapy had been applied. On May 1, 2004, the 189 medulloblastoma patients included in the TMA study had a median follow-up time of 43 months (range, 3 to 130 months), estimated according to Korn.27 There were 118 survivors and 71 patients who died during follow-up. Sixty-two deceased patients exhibited disease dissemination along the neuroaxis, whereas the remaining nine patients displayed isolated local tumor progression. The 53 medulloblastoma patients included in the matrix-CGH study had a median follow-up time of 30 months (range, 3 to 50 months).
Labeling and Hybridization to Microarrays
Selection of genomic clones, isolation of BAC DNA, performance of degenerate oligonucleotide primer-PCR, and preparation of microarrays were described previously.24,28,29 Genomic DNA from tumor tissue and blood of healthy donors was isolated using the Blood and Cell Culture Kit (Qiagen, Hilden, Germany) following the instructions of the suppliers. Labeling, hybridization, and washing procedure were performed as reported previously.24
Image and Data Analysis
Arrays were scanned with an Axon 4000B scanner (Axon Instruments, Burlingame, CA) and images were analyzed using GenePix Pro 4.0 software (Axon Instruments). Fluorescence intensities of all spots were filtered consistently (intensity/local background > 3; mean/median intensity < 1.3; standard deviation [SD] of clone log ratios < 0.2) and normalized globally. For each tumor two hybridizations were performed with reversed dyes (dye swap) and averaged. In every experiment, individual thresholds for gains/losses were defined as ± 3x SD estimated from the mean ratios of all clones in balanced regions. A region was scored as imbalanced if at least two adjacent clones reached the threshold ratios. Imbalances with linear ratios of less than 0.5 were scored as homozygous deletions because this threshold corresponds to an average copy number of less than 1, indicating the presence of at least a subpopulation of cells with a homozygous deletion. Copy number gains with linear ratios higher than 2 were scored as amplifications. Six tumors that showed a high degree of imbalances could not be normalized unambiguously and were excluded from the data set for the calculation of frequency of copy number changes and the statistical analysis. The chromosomal mapping information is based on the Ensembl (version 17) or the University of California at Santa Cruz genome database (Freeze, July 2003) and was updated for the identified DNA copy number hot spots. All data sets will be made publicly available at the National Center for Biotechnology Information Gene Expression Omnibus (GSE2139, GPL1432).
Preparation of TMAs
Sections from each paraffin block were stained with hematoxylin and eosin and prepared to define representative tumor regions. Microarray preparation was performed as described previously.30 After preparation, the tissue sections were again stained with hematoxylin and eosin and rechecked by pathologists.
FISH
To validate MYCN amplifications, dual-colored cocktail probes were used (Q-Biogene, Irvine, CA). To confirm chromosomal amplification of CDK6, FISH to sections of the TMA was performed using a fluorescein isothiocyanate–labeled locus-specific probe (RP5-850G1) in combination with a rhodamine-labeled centromere-specific probe (P7T1).31 Pretreatment of slides, hybridization, posthybridization processing, and signal detection were performed as described previously.32 Metaphase FISH for verifying clone mapping position was performed using peripheral-blood cell cultures of healthy donors as outlined previously.33
RNA Isolation
Total RNA was isolated according to a protocol that applies both Trizol reagent (Invitrogen, Karlsruhe, Germany) and RNeasy Midi spin columns (Qiagen). Total RNA quality and concentration were determined by the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA) and absorption spectroscopy between 220 and 320 nm (1x OD260 = 40 μg/mL single-stranded RNA).
RQ-PCR
All primers were tested to exclude amplification from genomic DNA. Estimation of correct PCR product sizes from cDNA was done by polyacrylamide gel electrophoresis combined with melting curve analysis. For tissue-specific normalization, a reference pool of total RNA (catalog No. 636535) obtained from human cerebellum of 24 individuals (age range, 16 to 70 years) was used (BD Bioscience, Heidelberg, Germany). Each cDNA sample was analyzed in triplicate (aliquots of 2 μL each; 25 ng/μL) using the ABI PRISM 7700 Sequence Detector (Applied Biosystems, Weiterstadt, Germany) with Absolute SYBR Green ROX Mix (ABgene, Epsom, United Kingdom) according to the manufacturers' instructions. To standardize the amount of sample cDNA, two endogenous control amplicons were used as housekeeping genes (PGK1, LMNB1). Sequences of oligonucleotides used for RQ-PCR are available on request. The relative quantification of each target gene (CDK6, PPM1D) in comparison with the housekeeping genes was calculated according to a previously published algorithm.34
IHC
Staining for detection of bound antibody was performed according to standard protocols using the VECTASTAIN Elite ABC Kit (Linaris, Wertheim-Bettingen, Germany) and DAB Substrate Kit (Linaris). For CDK6 protein, a 1:10 dilution of the monoclonal mouse antibody (clone DCS-83) was used (Progen, Heidelberg, Germany). Detection of PPM1D protein was performed using a 1:250 diluted (2380-MC-100) monoclonal mouse antibody (Biozol, Eching, Germany). Evaluation of immunostaining was carried out in a double-blinded fashion. Of 189 tumors, 169 for CDK6 and 168 for PPM1D could be analyzed for protein expression. Four semiquantitative categories of scores were defined according to the proportion of cells displaying positive staining for CDK6 and PPM1D: (–) no staining, (+) 5% to 30% of cells (low), (++) 30% to 60% of cells (moderate), and (+++) 60% to 100% (high), respectively.
Statistical Analysis
Estimation of the survival time distribution of patients with primary tumors was done according to the method of Kaplan and Meier. For pairwise comparisons of survival time distributions, the log-rank test was used.35 Pairwise comparisons of patient characteristics were performed by the Mann-Whitney test for continuous variables and by Fisher's exact test for categoric variables. Unsupervised clustering was performed by the use of Cluster 3.0 (Michiel de Hoon, Toyko, Japan) and the following parameters: array-wise, hierarchical clustering, and complete linkage of the data sets including all genomic imbalances. The similarities between log ratios were measured by uncentered correlation.36 Analysis of interactions between CDK6 expression or MYC and MYCN amplification as well as chromosome 17 imbalances was done by computing Pearson's correlation coefficient. Multivariate analysis of the dependence of survival times on CDK6 protein levels and relevant clinical parameters (age, sex, histology, stage of metastasis, and chemotherapy, as listed in Appendix Table 1) was performed by Cox proportional hazards regression. To provide quantitative information about the relevance of results of the statistical analysis, hazard ratios and their corresponding 95% CIs were computed. All statistical analyses were performed using the statistical software environment R, version 1.9.1 (http://www.r-project.org).
RESULTS
Genomic Imbalances in Medulloblastomas
In this study, 47 tumors were analyzed by high-resolution matrix-CGH (Fig 1). The most recurrent aberrations affecting entire chromosome bands were gains of 17q23.2-qter (51%), 7q11.21-qter (32%), chromosome 7 (30%), and 1q21.1 to q44 (13%), as well as losses of 17p13.1 to 13.3 (43%), 11p12pter (21%), 10q24.33-qter (17%), and chromosome 8 (17%). Each tumor displayed DNA copy number imbalances in at least one region. On median, 2.5 imbalances were scored per patient in desmoplastic tumors (n = 8), seven in classic tumors (n = 27), and six in anaplastic tumors (n = 12).
Hot Spots of DNA Copy Number Aberrations
By superimposing the genomic profiles of the 47 medulloblastomas, minimally overlapping regions of DNA copy number imbalances were identified. Regions smaller than 3 Mb, which were present in at least five tumors, and all loci exhibiting amplification or homozygous deletion are listed in Table 1. Homozygous deletions were detected once each at 1p36.21 and 2q37.1 to 37.3. Recurrent amplifications were found at 8q24.21 (MYC) in two tumors, 2p24.3 (MYCN) in three tumors, and 7q21.2 in three tumors. MYC and MYCN amplifications (n = 5) were associated with metastasis (n = 5; 0 of 31 M0 v 5/16 M1-3) and poor clinical outcome (estimated survival probability 12 months after surgery 0.20 v 0.86). The novel amplicon at 7q21.2 spanned a minimally overlapping region of 250 kb and contained CDK6 as the only known gene. Furthermore, previously reported amplicons17,37 at 3p21 to 23, 5p15, and 12p13 were narrowed in size to 1.18, 2.30, and 9.65 Mb, respectively, and new amplicons were detected once each at 1q32.1, 7q36.2 to 36.3, 8q21.3 to 22.2, and 17q23.2.
Results of FISH Analyses
To independently confirm the matrix-CGH data, we performed FISH analyses for the MYCN and CDK6 loci on the TMA. Figure 2A shows high-level amplification of MYCN in the tumor sample 1m10. Of the 15 tumors that showed single-copy gain of the CDK6 locus in matrix-CGH analysis, 11 were represented on the TMA. In each of these tumors, FISH analysis confirmed single-copy gains. The three tumors showing CDK6 amplification were not represented on the TMA; however, in tumor 1m26 the high-level amplification could be confirmed using a cytospin preparation (Fig 2B). CDK6 copy number was estimated as more than 20 copies per cell (range, 24 to 66 copies per cell). Two more tumors carrying CDK6 amplifications were identified on the TMA by FISH screening of 60 tumors, which were not in the series of samples analyzed by matrix-CGH. The amplifications in these tumors were present over large tumor areas and focal amplifications were not detectable (Fig 2C).
Correlation of Chromosome 17 Aberrations With Poor Survival
When log-rank tests were used, genomic imbalances on chromosome 17 were found to be related significantly to poor overall survival in the 47 patients included in the matrix-CGH study (Fig 2D). The loss of about 9 Mb on 17p and gain of about 23.8 Mb on 17q cover three of the previously identified hot spots of DNA copy number aberration (Table 1).
Unsupervised clustering divided our tumor set into two about equally large groups (Fig 3A). The most striking difference between these groups was the almost complete absence (group I) or presence (group II) of imbalances on chromosome 17 including an apparent i(17q), gain of chromosome 17 and isolated gain on 17q. The most common of these imbalances was gain of 17q23.3-qter, occurring in 24 of 25 group II tumors. Other imbalances more frequently observed in group II than group I included gain of 7q11.21-qter, loss of 10q, and loss of 11p12-qter. Groups I and II again could be divided into two subgroups each. Tumors in groups Ia and Ib were characterized by the presence or absence of chromosome 6. Group IIa showed a gain of 8q13.2-qter and contained the two MYC amplifications, whereas tumors of group IIb frequently displayed loss of chromosome 8 and included all CDK6 (n = 3) and MYCN (n = 2) amplifications. Correlation analyses of clinical and DNA copy number data identified significant differences between the subgroups (Fig 3B). Patients with group II tumors showed significantly worse overall survival compared with patients with group I tumors. Furthermore, the tumors of group II were more often of anaplastic and less commonly of desmoplastic histology, and had more multicopy number aberrations (amplifications/homozygous deletions).
High Protein Expression Levels of PPM1D and CDK6 in Medulloblastomas
The strong correlation between the presence of multicopy number imbalances and adverse prognosis lead us to investigate more closely the novel DNA amplicons at 17q23.2 (n = 1) and 7q21.2 (n = 3). These amplicons were located within the two most frequently gained chromosomal regions in our study (Table 1), and potentially indicate the map position of proto-oncogenes that might be relevant for medulloblastoma pathogenesis. These particular amplicons are defined by three (17q23.2; Fig 4A) and two (7q21.2; Fig 4B) clones, which contain the APPBP2 and PPM1D genes or the CDK6 gene. To evaluate the potential pathogenic importance of these amplicons, we analyzed mRNA and protein expression of PPM1D and CDK6. When compared with normal cerebellum reference, PPM1D mRNA was moderately overexpressed in seven of the nine tumors (Fig 4C), whereas CDK6 mRNA was strongly overexpressed in nine of the nine tumors (Fig 4D). The mean of two individual experiments was plotted and error bars represent 1 x SD. Tumors were selected to include different histologic subtypes and varying complexity of genetic aberrations. Protein expression was determined by IHC analysis of a large series of tumors using TMAs (Appendix Table 1). Strong and widespread immunoreactivity, defined as nuclear staining in more than 30% of the tumor cells, was observed for PPM1D in 148 of 168 (88%) tumors (Figs 4E and 4F) and for CDK6 in 50 of 169 (30%) tumors (Figs 4G and 4H).
CDK6 Expression Status Is an Independent Predictor of Prognosis
Comparison of tumors with no or low CDK6 staining versus tumors with moderate or high staining (Fig 5A) identified CDK6 overexpression as a prognostic marker that was associated significantly with poor clinical outcome using univariate analysis (Fig 5B; log-rank test, P < .01). The proportional hazards regression model identified four significant prognostic factors: chemotherapy (P < .01), stage of metastasis (P < .01), histology (P = .02), and CDK6 overexpression (P = .02). Age (P = .57), sex (P = .09), and level of resection (P = .49) were not statistically significant. Results are illustrated in Figure 5C.
The relationships of CDK6 overexpression and the presence of the potential molecular prognostic markers loss 17p13.3 to 13.1 and amplification of MYC or MYCN also were evaluated using informative tumor samples of the matrix-CGH study (n = 29). CDK6 overexpression is not associated with loss of 17p (P = .41) and tumors with MYC or MYCN amplification did not necessarily show CDK6 overexpression (one of two and one of three, respectively).
DISCUSSION
Matrix-CGH analysis identified DNA copy number alterations of several novel candidate genes that may play a role in medulloblastoma pathogenesis. The PPM1D gene at 17q23.2 was amplified in one medulloblastoma and showed a lower copy number gain in 24 other tumors, making it the most frequent genomic aberration detected in our tumor series (25 of 47; 53%). Notably, the region 7q21.2, containing CDK6, was gained or amplified in 32% (15 of 47) of the tumors. Regarding the frequent losses at 17p13.1, our findings suggest DLG4 as an interesting candidate gene, in addition to other known candidates, such as TP53, LIS1, HIC1, and REN.38-43 Candidate tumor suppressor genes located in the deleted regions on chromosome arm 10q include UTF1, the gene product of which seems to function as transcriptional regulator of DNA-polymerase II promoters, as well as PTEN, which encodes a known tumor suppressor that functions as an inhibitor of the phosphoinositol-3-kinase/AKT signaling pathway.44 Detailed mutational analyses and functional studies will be required to elucidate the role that each of these genes might play in medulloblastoma pathogenesis.
In a previous study,45 expression analysis was performed on a subset (n = 26) of the tumor samples used here. When correlating candidate genes from genomic amplicons identified by matrix-CGH with their respective expression ratios, gene dosage effects are observed for three of four loci (CDK6, MYCN, and CCND2). Only MYC amplification did not result in increased gene expression.
The high frequency of aberrations on chromosome 17 supports its central importance in medulloblastoma pathogenesis. We confirmed frequent loss on 17p and identified a region of recurrent copy number gain on 17q, both of which were correlated with poor overall survival in univariate analyses. These analyses have to be viewed with caution, however, given that loss of 17p and gain of 17q are not independent events in tumors with i(17q). Chromosome 17 aberrations also had prominent impact in the unsupervised cluster analysis (ie, characterized the tumors of a group of patients with poor prognosis [group II]). Molecular classification correlated significantly with histology mainly because of seven of eight desmoplastic tumors clustering in the group with good prognosis (group I) and eight of 12 anaplastic tumors clustering in the group with poor prognosis. Tumors with classic histology were evenly distributed between groups I and II. Amplifications of MYC and MYCN, which were found to be mutually exclusive in a recent study of medulloblastoma,19 clustered in subgroups IIa and IIb, respectively. In addition to all MYCN amplifications, group IIb tumors contained all of the amplifications involving the CDK6 locus. Recently, MYCN and the CDK6 partner cyclin D1 were identified as targets of the sonic hedgehog pathway involved in the proliferation of neuronal precursors in medulloblastoma.46 Finally, group IIa, which only contained pediatric tumors, represented a clinically defined group of patients with poor prognosis and anaplastic tumors carrying MYC amplifications.6
The amplicon detected at 17q23.2 contained the genes APPBP2 and PPM1D, which were also reported to be coamplified in neuroblastoma47 as well as breast48 and ovarian carcinoma.49 Overexpression of PPM1D but not APPBP2 was reported to cooperate with rat sarcoma viral oncogene homolog in transforming primary mouse fibroblasts.48 Recently, it was shown that overexpression of PPM1D in mouse embryo fibroblasts abolished TP53 tumor-suppressor activity,50 whereas PPM1D deficiency led to resistance against oncogenic transformation mediated by the p16INK4A and p19ARF (p14ARF in human) pathways.51 Although increased PPM1D protein is not associated with poor prognosis, it might contribute to escape from the TP53-dependent control of cell cycle regulation, apoptosis, and DNA repair. PPM1D, which itself is induced by TP53, takes part in a negative-feedback loop downregulating TP53 directly via p38 MAPK.52 Our results, together with published data,53 indicate that different levels of deregulation of the TP53 pathway might exist in medulloblastoma, including amplification or gain and strong expression of PPM1D, copy number gain/amplification of MDM4, frequent monoallelic loss of TP53 and, less commonly, TP53 mutation.
The novel amplicon detected at 7q21.2 contained CDK6 as the only gene, and genomic amplification was associated with increased CDK6 protein levels. However, CDK6 overexpression was not restricted to cases with gene amplification, thus suggesting that additional mechanisms exist, resulting in high protein levels of CDK6 in approximately 30% of medulloblastomas. At least in part, a direct connection between PPM1D and CDK6 protein expression in medulloblastoma could be mediated by p38 MAPK52,54 (Fig 6). Overexpression of CDK6 was reported previously in T-cell lymphoma55 and a small subset of malignant gliomas.56 Similar to CDK4, the CDK6 protein is activated by binding to D-type cyclins. The cyclin D/CDK6 complexes phosphorylate the retinoblastoma protein RB1 and the related proteins RBL1 and RBL2, leading to the release of E2F transcription factors.57 In line with our findings, data from mouse models support the importance of retinoblastoma pathway alterations in medulloblastoma development.58-61 High levels of CDK6 protein may result in constitutive phosphorylation of RB1, which may override inhibition by p16INK4A, as shown in murine embryonal stem cells.62 Recent studies also showed that CDK6 overexpression blocks differentiation in leukemic cells63 and osteoblasts,54 suggesting that CDK6 is a key regulator of the RB1 pathway affecting proliferation and differentiation in medulloblastoma and other tumors.
For better risk stratification and the planning of individual therapeutic regimens, it is important to have reliable markers for the assessment of prognosis and response to therapy. At present, prognostic markers for medulloblastoma patients include certain clinical factors; the histologic subtype; and a few molecular parameters, such as loss of 17p, amplification and overexpression of MYC or MYCN, and expression levels of TRKC, ERBB2, and PDGFRB.64-68 In this study, we identified a novel prognostic marker for medulloblastoma patients, namely overexpression of CDK6. In multivariate regression analysis, we found protein overexpression to be an independent prognostic factor for poor overall survival of medulloblastoma patients. Given its higher frequency as compared with the occurrence of DNA amplifications, we consider CDK6 overexpression to be a more informative marker than either MYC or MYCN amplification. No statistically significant overlap was observed between CDK6 expression and 17p13.3 to 13.1 loss in the matrix-CGH data set, suggesting that these are independent prognostic factors that can complement each other in diagnostics. Given that CDK6 expression can be determined by means of IHC on formalin-fixed paraffin sections, this marker is well suited for the routine diagnostic setting as well as an evaluation in controlled clinical studies. Moreover, therapeutic approaches specifically targeting cyclin-dependent kinases are under development and may provide a novel option for the treatment of this tumor.69
Appendix
The Appendix is included in the full-text version of this article, available online at www.jco.org. It is not included in the PDF (via Adobe? Acrobat Reader?) version.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Stefanie Hofmann, Andrea Wittmann, Frauke Devens, Bj?rn Tews, and Gunnar Wrobel for excellent technical support. We also thank the Mapping Groups of the Wellcome Trust Sanger Institute for initial clone supply and verification.
NOTES
Supported by Bundesministerium fr Bildungund Forschung Grants No. 01GS0460 and 01GRO417 (P.L.), and by a Kekulé-fellowship of the Fonds der Chemischen Industrie (F.M.).
Both F.M. and B.R. contributed equally to this work.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Gurney JG, Kadan-Lottick N: Brain and other central nervous system tumors: Rates, trends, and epidemiology. Curr Opin Oncol 13:160-166, 2001
Fisher PG, Burger PC, Eberhart CG: Biologic risk stratification of medulloblastoma: The real time is now. J Clin Oncol 22:971-974, 2004
Eberhart CG, Kepner JL, Goldthwaite PT, et al: Histopathologic grading of medulloblastomas: A Pediatric Oncology Group study. Cancer 94:552-560, 2002
Giangaspero F, Perilongo G, Fondelli MP, et al: Medulloblastoma with extensive nodularity: A variant with favorable prognosis. J Neurosurg 91:971-977, 1999
Giangaspero F, Rigobello L, Badiali M, et al: Large-cell medulloblastomas: A distinct variant with highly aggressive behavior. Am J Surg Pathol 16:687-693, 1992
Brown HG, Kepner JL, Perlman EJ, et al: "Large cell/anaplastic" medulloblastomas: A Pediatric Oncology Group Study. J Neuropathol Exp Neurol 59:857-865, 2000
Zeltzer PM, Boyett JM, Finlay JL, et al: Metastasis stage, adjuvant treatment, and residual tumor are prognostic factors for medulloblastoma in children: Conclusions from the Children's Cancer Group 921 randomized phase III study. J Clin Oncol 17:832-845, 1999
Gilbertson R: Paediatric embryonic brain tumours: Biological and clinical relevance of molecular genetic abnormalities. Eur J Cancer 38:675-685, 2002
Wechsler-Reya RJ: Analysis of gene expression in the normal and malignant cerebellum. Recent Prog Horm Res 58:227-248, 2003
Bayani J, Zielenska M, Marrano P, et al: Molecular cytogenetic analysis of medulloblastomas and supratentorial primitive neuroectodermal tumors by using conventional banding, comparative genomic hybridization, and spectral karyotyping. J Neurosurg 93:437-448, 2000
Pan E, Pellarin M, Holmes E, et al: Isochromosome 17q is a negative prognostic factor in poor-risk childhood medulloblastoma patients. Clin Cancer Res 11:4733-4740, 2005
Nicholson JC, Ross FM, Kohler JA, et al: Comparative genomic hybridization and histological variation in primitive neuroectodermal tumours. Br J Cancer 80:1322-1331, 1999
Reardon DA, Michalkiewicz E, Boyett JM, et al: Extensive genomic abnormalities in childhood medulloblastoma by comparative genomic hybridization. Cancer Res 57:4042-4047, 1997
Avet-Loiseau H, Venuat AM, Terrier-Lacombe MJ, et al: Comparative genomic hybridization detects many recurrent imbalances in central nervous system primitive neuroectodermal tumours in children. Br J Cancer 79:1843-1847, 1999
Schutz BR, Scheurlen W, Krauss J, et al: Mapping of chromosomal gains and losses in primitive neuroectodermal tumors by comparative genomic hybridization. Genes Chromosomes Cancer 16:196-203, 1996
Gilhuis HJ, Anderl KL, Boerman RH, et al: Comparative genomic hybridization of medulloblastomas and clinical relevance: Eleven new cases and a review of the literature. Clin Neurol Neurosurg 102:203-209, 2000
Eberhart CG, Kratz JE, Schuster A, et al: Comparative genomic hybridization detects an increased number of chromosomal alterations in large cell/anaplastic medulloblastomas. Brain Pathol 12:36-44, 2002
Hui AB, Takano H, Lo KW, et al: Identification of a novel homozygous deletion region at 6q23.1 in medulloblastomas using high-resolution array comparative genomic hybridization analysis. Clin Cancer Res 11:4707-4716, 2005
Lamont JM, McManamy CS, Pearson AD, et al: Combined histopathological and molecular cytogenetic stratification of medulloblastoma patients. Clin Cancer Res 10:5482-5493, 2004
Lichter P, Joos S, Bentz M, et al: Comparative genomic hybridization: Uses and limitations. Semin Hematol 37:348-357, 2000
Solinas-Toldo S, Lampel S, Stilgenbauer S, et al: Matrix-based comparative genomic hybridization: Biochips to screen for genomic imbalances. Genes Chromosomes Cancer 20:399-407, 1997
Pinkel D, Segraves R, Sudar D, et al: High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 20:207-211, 1998
Schwaenen C, Nessling M, Wessendorf S, et al: Automated array-based genomic profiling in chronic lymphocytic leukemia: Development of a clinical tool and discovery of recurrent genomic alterations. Proc Natl Acad Sci U S A 101:1039-1044, 2004
Zielinski B, Gratias S, Toedt G, et al: Detection of chromosomal imbalances in retinoblastoma by matrix-based comparative genomic hybridization. Genes Chromosomes Cancer 43:294-301, 2005
Giangaspero F, Bigner SH, Kleihues P, et al: Medulloblastoma, in Cavanee WK, Kleihues P (eds): Pathology and Genetics of Tumours of the Nervous System. Lyons, France, IARC Press, 2000, pp 129-137
Kortmann RD, Kuhl J, Timmermann B, et al: Postoperative neoadjuvant chemotherapy before radiotherapy as compared to immediate radiotherapy followed by maintenance chemotherapy in the treatment of medulloblastoma in childhood: Results of the German prospective randomized trial HIT '91. Int J Radiat Oncol Biol Phys 46:269-279, 2000
Korn EL: Censoring distributions as a measure of follow-up in survival analysis. Stat Med 5:255-260, 1986
Fiegler H, Carr P, Douglas EJ, et al: DNA microarrays for comparative genomic hybridization based on DOP-PCR amplification of BAC and PAC clones. Genes Chromosomes Cancer 36:361-374, 2003
Knight SJ, Lese CM, Precht KS, et al: An optimized set of human telomere clones for studying telomere integrity and architecture. Am J Hum Genet 67:320-332, 2000
Freier K, Joos S, Flechtenmacher C, et al: Tissue microarray analysis reveals site-specific prevalence of oncogene amplifications in head and neck squamous cell carcinoma. Cancer Res 63:1179-1182, 2003
Waye JS, England SB, Willard HF: Genomic organization of alpha satellite DNA on human chromosome 7: Evidence for two distinct alphoid domains on a single chromosome. Mol Cell Biol 7:349-356, 1987
Bubendorf L, Kononen J, Koivisto P, et al: Survey of gene amplifications during prostate cancer progression by high-throughout fluorescence in situ hybridization on tissue microarrays. Cancer Res 59:803-806, 1999
Lichter P, Cremer T, Borden J, et al: Delineation of individual human chromosomes in metaphase and interphase cells by in situ suppression hybridization using recombinant DNA libraries. Hum Genet 80:224-234, 1988
Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29:e45, 2001
Harrell FE Jr, Shih YC: Using full probability models to compute probabilities of actual interest to decision makers. Int J Technol Assess Health Care 17:17-26, 2001
de Hoon MJ, Imoto S, Nolan J, et al: Open source clustering software. Bioinformatics 20:1453-1454, 2004
Fan X, Wang Y, Kratz J, et al: HTERT gene amplification and increased mRNA expression in central nervous system embryonal tumors. Am J Pathol 162:1763-1769, 2003
Adesina AM, Nalbantoglu J, Cavenee WK: p53 gene mutation and mdm2 gene amplification are uncommon in medulloblastoma. Cancer Res 54:5649-5651, 1994
Biegel JA, Burk CD, Barr FG, et al: Evidence for a 17p tumor related locus distinct from p53 in pediatric primitive neuroectodermal tumors. Cancer Res 52:3391-3395, 1992
Di Marcotullio L, Ferretti E, De Smaele E, et al: REN(KCTD11) is a suppressor of Hedgehog signaling and is deleted in human medulloblastoma. Proc Natl Acad Sci U S A 101:10833-10838, 2004
Koch A, Tonn J, Kraus JA, et al: Molecular analysis of the lissencephaly gene 1 (LIS-1) in medulloblastomas. Neuropathol Appl Neurobiol 22:233-242, 1996
McDonald JD, Daneshvar L, Willert JR, et al: Physical mapping of chromosome 17p13.3 in the region of a putative tumor suppressor gene important in medulloblastoma. Genomics 23:229-232, 1994
Ohgaki H, Eibl RH, Schwab M, et al: Mutations of the p53 tumor suppressor gene in neoplasms of the human nervous system. Mol Carcinog 8:74-80, 1993
Wu X, Senechal K, Neshat MS, et al: The PTEN/MMAC1 tumor suppressor phosphatase functions as a negative regulator of the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci U S A 95:15587-15591, 1998
Neben K, Korshunov A, Benner A, et al: Microarray-based screening for molecular markers in medulloblastoma revealed STK15 as independent predictor for survival. Cancer Res 64:3103-3111, 2004
Oliver TG, Grasfeder LL, Carroll AL, et al: Transcriptional profiling of the Sonic hedgehog response: A critical role for N-myc in proliferation of neuronal precursors. Proc Natl Acad Sci U S A 100:7331-7336, 2003
Saito-Ohara F, Imoto I, Inoue J, et al: PPM1D is a potential target for 17q gain in neuroblastoma. Cancer Res 63:1876-1883, 2003
Li J, Yang Y, Peng Y, et al: Oncogenic properties of PPM1D located within a breast cancer amplification epicenter at 17q23. Nat Genet 31:133-134, 2002
Hirasawa A, Saito-Ohara F, Inoue J, et al: Association of 17q21-q24 gain in ovarian clear cell adenocarcinomas with poor prognosis and identification of PPM1D and APPBP2 as likely amplification targets. Clin Cancer Res 9:1995-2004, 2003
Bulavin DV, Demidov ON, Saito S, et al: Amplification of PPM1D in human tumors abrogates p53 tumor-suppressor activity. Nat Genet 31:210-215, 2002
Bulavin DV, Phillips C, Nannenga B, et al: Inactivation of the Wip1 phosphatase inhibits mammary tumorigenesis through p38 MAPK-mediated activation of the p16(Ink4a)-p19(Arf) pathway. Nat Genet 36:343-350, 2004
Takekawa M, Adachi M, Nakahata A, et al: p53-inducible wip1 phosphatase mediates a negative feedback regulation of p38 MAPK-p53 signaling in response to UV radiation. Embo J 19:6517-6526, 2000
Portwine C, Chilton-MacNeill S, Brown C, et al: Absence of germline and somatic p53 alterations in children with sporadic brain tumors. J Neurooncol 52:227-235, 2001
Moro T, Ogasawara T, Chikuda H, et al: Inhibition of Cdk6 expression through p38 MAP kinase is involved in differentiation of mouse prechondrocyte ATDC5. J Cell Physiol 204:927-933, 2005
Chilosi M, Doglioni C, Yan Z, et al: Differential expression of cyclin-dependent kinase 6 in cortical thymocytes and T-cell lymphoblastic lymphoma/leukemia. Am J Pathol 152:209-217, 1998
Costello JF, Plass C, Arap W, et al: Cyclin-dependent kinase 6 (CDK6) amplification in human gliomas identified using two-dimensional separation of genomic DNA. Cancer Res 57:1250-1254, 1997
Ortega S, Malumbres M, Barbacid M: Cyclin D-dependent kinases, INK4 inhibitors and cancer. Biochim Biophys Acta 1602:73-87, 2002
Zindy F, Nilsson LM, Nguyen L, et al: Hemangiosarcomas, medulloblastomas, and other tumors in Ink4c/p53-null mice. Cancer Res 63:5420-5427, 2003
Wetmore C, Eberhart DE, Curran T: Loss of p53 but not ARF accelerates medulloblastoma in mice heterozygous for patched. Cancer Res 61:513-516, 2001
Marino S, Vooijs M, van Der Gulden H, et al: Induction of medulloblastomas in p53-null mutant mice by somatic inactivation of Rb in the external granular layer cells of the cerebellum. Genes Dev 14:994-1004, 2000
Dyer MA: Mouse models of childhood cancer of the nervous system. J Clin Pathol 57:561-576, 2004
Faast R, White J, Cartwright P, et al: Cdk6-cyclin D3 activity in murine ES cells is resistant to inhibition by p16(INK4a). Oncogene 23:491-502, 2004
Matushansky I, Radparvar F, Skoultchi AI: CDK6 blocks differentiation: Coupling cell proliferation to the block to differentiation in leukemic cells. Oncogene 22:4143-4149, 2003
Eberhart CG, Kratz J, Wang Y, et al: Histopathological and molecular prognostic markers in medulloblastoma: C-myc, N-myc, TrkC, and anaplasia. J Neuropathol Exp Neurol 63:441-449, 2004
Gajjar A, Hernan R, Kocak M, et al: Clinical, histopathologic, and molecular markers of prognosis: Toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 22:984-993, 2004
Grotzer MA, Hogarty MD, Janss AJ, et al: MYC messenger RNA expression predicts survival outcome in childhood primitive neuroectodermal tumor/medulloblastoma. Clin Cancer Res 7:2425-2433, 2001
Gilbertson RJ, Clifford SC: PDGFRB is overexpressed in metastatic medulloblastoma. Nat Genet 35:197-198, 2003
MacDonald TJ, Brown KM, LaFleur B, et al: Expression profiling of medulloblastoma: PDGFRA and the RAS/MAPK pathway as therapeutic targets for metastatic disease. Nat Genet 29:143-152, 2001
Fry DW, Harvey PJ, Keller PR, et al: Specific inhibition of cyclin-dependent kinase 4/6 by PD 0332991 and associated antitumor activity in human tumor xenografts. Mol Cancer Ther 3:1427-1438, 2004(Frank Mendrzyk, Bernhard )
Central Unit Biostatistics, German Cancer Research Center, Heidelberg
Department of Neuropathology, Heinrich Heine University, Duesseldorf, Germany
Wellcome Trust Genome Campus, The Wellcome Trust Sanger Institute, Hinxton, United Kingdom
Department of Neuropathology, Burdenko Neurosurgical Institute, Moscow, Russia.
ABSTRACT
PURPOSE: Medulloblastoma is the most common malignant brain tumor in children. Despite multimodal aggressive treatment, nearly half of the patients die as a result of this tumor. Identification of molecular markers for prognosis and development of novel pathogenesis-based therapies depends crucially on a better understanding of medulloblastoma pathomechanisms.
PATIENTS AND METHODS: We performed genome-wide analysis of DNA copy number imbalances in 47 medulloblastomas using comparative genomic hybridization to large insert DNA microarrays (matrix-CGH). The expression of selected candidate genes identified by matrix-CGH was analyzed immunohistochemically on tissue microarrays representing medulloblastomas from 189 clinically well-documented patients. To identify novel prognostic markers, genomic findings and protein expression data were correlated to patient survival.
RESULTS: Matrix-CGH analysis revealed frequent DNA copy number alterations of several novel candidate regions. Among these, gains at 17q23.2-qter (P < .01) and losses at 17p13.1 to 17p13.3 (P = .04) were significantly correlated to poor prognosis. Within 17q23.2-qter and 7q21.2, two of the most frequently gained chromosomal regions, confined amplicons were identified that contained the PPM1D and CDK6 genes, respectively. Immunohistochemistry revealed strong expression of PPM1D in 148 (88%) of 168 and CDK6 in 50 (30%) of 169 medulloblastomas. Overexpression of CDK6 correlated significantly with poor prognosis (P < .01) and represented an independent prognostic marker of overall survival on multivariate analysis (P = .02).
CONCLUSION: We identified CDK6 as a novel molecular marker that can be determined by immunohistochemistry on routinely processed tissue specimens and may facilitate the prognostic assessment of medulloblastoma patients. Furthermore, increased protein-levels of PPM1D and CDK6 may link the TP53 and RB1 tumor suppressor pathways to medulloblastoma pathomechanisms.
INTRODUCTION
Medulloblastoma is a highly malignant embryonal tumor of the cerebellum that accounts for 20% to 25% of all primary brain tumors in children; the incidence peaks at 7 years of age.1 Several clinical, histologic, and molecular parameters have been identified as predictive factors for patient outcome.2-6 Despite aggressive multimodal therapy, including surgery, irradiation, and chemotherapy, disease dissemination is common and the 5-year survival rate is only 50% to 60%.7
To develop novel therapeutic strategies for the improvement of clinical outcome, it is of paramount importance to increase our understanding of the cellular and molecular basis of this disease. Previous molecular studies showed that the sonic hedgehog signaling pathway is aberrantly activated by genetic alterations in the PTCH, SMOH, or SUFU genes in up to 20% of medulloblastomas, affecting mostly tumors of the desmoplastic subtype.8 A smaller fraction of medulloblastomas demonstrate alterations in genes involved in wingless/WNT (wingless-type MMTV integration site family) signaling, such as CNNTB1, APC, and AXIN.8,9 Studies using conventional karyotyping or comparative genomic hybridization (CGH) identified trisomy 7 and isochromosome 17 [i(17q)] as frequent chromosomal aberrations in medulloblastoma.10-16 Furthermore, amplification of the proto-oncogenes MYC or MYCN is common in large-cell anaplastic medulloblastoma—a rare but highly malignant subtype of medulloblastoma.15-19
To overcome the restricted resolution of chromosomal CGH,20 matrix-based CGH (matrix-CGH) was developed using hybridization targets immobilized on microarrays.21-23 For genome-wide profiling of copy number imbalances, we constructed a DNA microarray consisting of approximately 6,000 large-insert genomic clones covering the whole genome with an average resolution of 0.5 megabases (Mb).24 In this study, we investigated 47 sporadic medulloblastomas using this microarray to identify recurrent DNA copy number imbalances and novel medulloblastoma-associated candidate genes. The matrix-CGH findings were compared with real-time quantitative reverse transcriptase polymerase chain reaction (RQ-PCR), interphase fluorescent in situ hybridization (FISH), and immunohistochemistry (IHC) using a tissue microarray (TMA) composed of medulloblastomas from 189 patients with available clinical follow-up. Genomic imbalances were clustered, combined with protein expression data, and correlated to clinical outcome to search for novel prognostic markers.
PATIENTS AND METHODS
Tumor Material and Patient Characteristics
All samples used in this study were collected at the Department of Neuropathology, Burdenko Neurosurgical Institute (Moscow, Russia), between 1993 and 2003. All diagnoses were confirmed by histologic assessment of specimens obtained at surgery by at least two neuropathologists according to the criteria of the WHO classification.25 The tumors were classified by histology as classic, desmoplastic, or anaplastic medulloblastomas; the latter diagnosis was made only when signs of moderate or severe anaplasia were present.3
All samples were provided from an existing tumor bank for research purposes and approval to link laboratory data to clinical data was obtained by the institutional review board. This procedure is according to the official statement of the National Ethics Council of the federal government of Germany (March 2004). Tumor and patient characteristics are summarized in \?\Appendix Table 1. In the matrix-CGH and TMA study, 53 and 189 tumor samples were analyzed, respectively. All patients received craniospinal irradiation after surgery consisting of 36 Gy to the craniospinal axis and a boost of 53 to 56 Gy to the posterior fossa. Adjuvant chemotherapy using lomustine, cisplatin, and vincristine7,26 was administered routinely to all patients treated after 1995, including the patients of the matrix-CGH study and 110 patients (58%) in the TMA study. Before 1995, no chemotherapy had been applied. On May 1, 2004, the 189 medulloblastoma patients included in the TMA study had a median follow-up time of 43 months (range, 3 to 130 months), estimated according to Korn.27 There were 118 survivors and 71 patients who died during follow-up. Sixty-two deceased patients exhibited disease dissemination along the neuroaxis, whereas the remaining nine patients displayed isolated local tumor progression. The 53 medulloblastoma patients included in the matrix-CGH study had a median follow-up time of 30 months (range, 3 to 50 months).
Labeling and Hybridization to Microarrays
Selection of genomic clones, isolation of BAC DNA, performance of degenerate oligonucleotide primer-PCR, and preparation of microarrays were described previously.24,28,29 Genomic DNA from tumor tissue and blood of healthy donors was isolated using the Blood and Cell Culture Kit (Qiagen, Hilden, Germany) following the instructions of the suppliers. Labeling, hybridization, and washing procedure were performed as reported previously.24
Image and Data Analysis
Arrays were scanned with an Axon 4000B scanner (Axon Instruments, Burlingame, CA) and images were analyzed using GenePix Pro 4.0 software (Axon Instruments). Fluorescence intensities of all spots were filtered consistently (intensity/local background > 3; mean/median intensity < 1.3; standard deviation [SD] of clone log ratios < 0.2) and normalized globally. For each tumor two hybridizations were performed with reversed dyes (dye swap) and averaged. In every experiment, individual thresholds for gains/losses were defined as ± 3x SD estimated from the mean ratios of all clones in balanced regions. A region was scored as imbalanced if at least two adjacent clones reached the threshold ratios. Imbalances with linear ratios of less than 0.5 were scored as homozygous deletions because this threshold corresponds to an average copy number of less than 1, indicating the presence of at least a subpopulation of cells with a homozygous deletion. Copy number gains with linear ratios higher than 2 were scored as amplifications. Six tumors that showed a high degree of imbalances could not be normalized unambiguously and were excluded from the data set for the calculation of frequency of copy number changes and the statistical analysis. The chromosomal mapping information is based on the Ensembl (version 17) or the University of California at Santa Cruz genome database (Freeze, July 2003) and was updated for the identified DNA copy number hot spots. All data sets will be made publicly available at the National Center for Biotechnology Information Gene Expression Omnibus (GSE2139, GPL1432).
Preparation of TMAs
Sections from each paraffin block were stained with hematoxylin and eosin and prepared to define representative tumor regions. Microarray preparation was performed as described previously.30 After preparation, the tissue sections were again stained with hematoxylin and eosin and rechecked by pathologists.
FISH
To validate MYCN amplifications, dual-colored cocktail probes were used (Q-Biogene, Irvine, CA). To confirm chromosomal amplification of CDK6, FISH to sections of the TMA was performed using a fluorescein isothiocyanate–labeled locus-specific probe (RP5-850G1) in combination with a rhodamine-labeled centromere-specific probe (P7T1).31 Pretreatment of slides, hybridization, posthybridization processing, and signal detection were performed as described previously.32 Metaphase FISH for verifying clone mapping position was performed using peripheral-blood cell cultures of healthy donors as outlined previously.33
RNA Isolation
Total RNA was isolated according to a protocol that applies both Trizol reagent (Invitrogen, Karlsruhe, Germany) and RNeasy Midi spin columns (Qiagen). Total RNA quality and concentration were determined by the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA) and absorption spectroscopy between 220 and 320 nm (1x OD260 = 40 μg/mL single-stranded RNA).
RQ-PCR
All primers were tested to exclude amplification from genomic DNA. Estimation of correct PCR product sizes from cDNA was done by polyacrylamide gel electrophoresis combined with melting curve analysis. For tissue-specific normalization, a reference pool of total RNA (catalog No. 636535) obtained from human cerebellum of 24 individuals (age range, 16 to 70 years) was used (BD Bioscience, Heidelberg, Germany). Each cDNA sample was analyzed in triplicate (aliquots of 2 μL each; 25 ng/μL) using the ABI PRISM 7700 Sequence Detector (Applied Biosystems, Weiterstadt, Germany) with Absolute SYBR Green ROX Mix (ABgene, Epsom, United Kingdom) according to the manufacturers' instructions. To standardize the amount of sample cDNA, two endogenous control amplicons were used as housekeeping genes (PGK1, LMNB1). Sequences of oligonucleotides used for RQ-PCR are available on request. The relative quantification of each target gene (CDK6, PPM1D) in comparison with the housekeeping genes was calculated according to a previously published algorithm.34
IHC
Staining for detection of bound antibody was performed according to standard protocols using the VECTASTAIN Elite ABC Kit (Linaris, Wertheim-Bettingen, Germany) and DAB Substrate Kit (Linaris). For CDK6 protein, a 1:10 dilution of the monoclonal mouse antibody (clone DCS-83) was used (Progen, Heidelberg, Germany). Detection of PPM1D protein was performed using a 1:250 diluted (2380-MC-100) monoclonal mouse antibody (Biozol, Eching, Germany). Evaluation of immunostaining was carried out in a double-blinded fashion. Of 189 tumors, 169 for CDK6 and 168 for PPM1D could be analyzed for protein expression. Four semiquantitative categories of scores were defined according to the proportion of cells displaying positive staining for CDK6 and PPM1D: (–) no staining, (+) 5% to 30% of cells (low), (++) 30% to 60% of cells (moderate), and (+++) 60% to 100% (high), respectively.
Statistical Analysis
Estimation of the survival time distribution of patients with primary tumors was done according to the method of Kaplan and Meier. For pairwise comparisons of survival time distributions, the log-rank test was used.35 Pairwise comparisons of patient characteristics were performed by the Mann-Whitney test for continuous variables and by Fisher's exact test for categoric variables. Unsupervised clustering was performed by the use of Cluster 3.0 (Michiel de Hoon, Toyko, Japan) and the following parameters: array-wise, hierarchical clustering, and complete linkage of the data sets including all genomic imbalances. The similarities between log ratios were measured by uncentered correlation.36 Analysis of interactions between CDK6 expression or MYC and MYCN amplification as well as chromosome 17 imbalances was done by computing Pearson's correlation coefficient. Multivariate analysis of the dependence of survival times on CDK6 protein levels and relevant clinical parameters (age, sex, histology, stage of metastasis, and chemotherapy, as listed in Appendix Table 1) was performed by Cox proportional hazards regression. To provide quantitative information about the relevance of results of the statistical analysis, hazard ratios and their corresponding 95% CIs were computed. All statistical analyses were performed using the statistical software environment R, version 1.9.1 (http://www.r-project.org).
RESULTS
Genomic Imbalances in Medulloblastomas
In this study, 47 tumors were analyzed by high-resolution matrix-CGH (Fig 1). The most recurrent aberrations affecting entire chromosome bands were gains of 17q23.2-qter (51%), 7q11.21-qter (32%), chromosome 7 (30%), and 1q21.1 to q44 (13%), as well as losses of 17p13.1 to 13.3 (43%), 11p12pter (21%), 10q24.33-qter (17%), and chromosome 8 (17%). Each tumor displayed DNA copy number imbalances in at least one region. On median, 2.5 imbalances were scored per patient in desmoplastic tumors (n = 8), seven in classic tumors (n = 27), and six in anaplastic tumors (n = 12).
Hot Spots of DNA Copy Number Aberrations
By superimposing the genomic profiles of the 47 medulloblastomas, minimally overlapping regions of DNA copy number imbalances were identified. Regions smaller than 3 Mb, which were present in at least five tumors, and all loci exhibiting amplification or homozygous deletion are listed in Table 1. Homozygous deletions were detected once each at 1p36.21 and 2q37.1 to 37.3. Recurrent amplifications were found at 8q24.21 (MYC) in two tumors, 2p24.3 (MYCN) in three tumors, and 7q21.2 in three tumors. MYC and MYCN amplifications (n = 5) were associated with metastasis (n = 5; 0 of 31 M0 v 5/16 M1-3) and poor clinical outcome (estimated survival probability 12 months after surgery 0.20 v 0.86). The novel amplicon at 7q21.2 spanned a minimally overlapping region of 250 kb and contained CDK6 as the only known gene. Furthermore, previously reported amplicons17,37 at 3p21 to 23, 5p15, and 12p13 were narrowed in size to 1.18, 2.30, and 9.65 Mb, respectively, and new amplicons were detected once each at 1q32.1, 7q36.2 to 36.3, 8q21.3 to 22.2, and 17q23.2.
Results of FISH Analyses
To independently confirm the matrix-CGH data, we performed FISH analyses for the MYCN and CDK6 loci on the TMA. Figure 2A shows high-level amplification of MYCN in the tumor sample 1m10. Of the 15 tumors that showed single-copy gain of the CDK6 locus in matrix-CGH analysis, 11 were represented on the TMA. In each of these tumors, FISH analysis confirmed single-copy gains. The three tumors showing CDK6 amplification were not represented on the TMA; however, in tumor 1m26 the high-level amplification could be confirmed using a cytospin preparation (Fig 2B). CDK6 copy number was estimated as more than 20 copies per cell (range, 24 to 66 copies per cell). Two more tumors carrying CDK6 amplifications were identified on the TMA by FISH screening of 60 tumors, which were not in the series of samples analyzed by matrix-CGH. The amplifications in these tumors were present over large tumor areas and focal amplifications were not detectable (Fig 2C).
Correlation of Chromosome 17 Aberrations With Poor Survival
When log-rank tests were used, genomic imbalances on chromosome 17 were found to be related significantly to poor overall survival in the 47 patients included in the matrix-CGH study (Fig 2D). The loss of about 9 Mb on 17p and gain of about 23.8 Mb on 17q cover three of the previously identified hot spots of DNA copy number aberration (Table 1).
Unsupervised clustering divided our tumor set into two about equally large groups (Fig 3A). The most striking difference between these groups was the almost complete absence (group I) or presence (group II) of imbalances on chromosome 17 including an apparent i(17q), gain of chromosome 17 and isolated gain on 17q. The most common of these imbalances was gain of 17q23.3-qter, occurring in 24 of 25 group II tumors. Other imbalances more frequently observed in group II than group I included gain of 7q11.21-qter, loss of 10q, and loss of 11p12-qter. Groups I and II again could be divided into two subgroups each. Tumors in groups Ia and Ib were characterized by the presence or absence of chromosome 6. Group IIa showed a gain of 8q13.2-qter and contained the two MYC amplifications, whereas tumors of group IIb frequently displayed loss of chromosome 8 and included all CDK6 (n = 3) and MYCN (n = 2) amplifications. Correlation analyses of clinical and DNA copy number data identified significant differences between the subgroups (Fig 3B). Patients with group II tumors showed significantly worse overall survival compared with patients with group I tumors. Furthermore, the tumors of group II were more often of anaplastic and less commonly of desmoplastic histology, and had more multicopy number aberrations (amplifications/homozygous deletions).
High Protein Expression Levels of PPM1D and CDK6 in Medulloblastomas
The strong correlation between the presence of multicopy number imbalances and adverse prognosis lead us to investigate more closely the novel DNA amplicons at 17q23.2 (n = 1) and 7q21.2 (n = 3). These amplicons were located within the two most frequently gained chromosomal regions in our study (Table 1), and potentially indicate the map position of proto-oncogenes that might be relevant for medulloblastoma pathogenesis. These particular amplicons are defined by three (17q23.2; Fig 4A) and two (7q21.2; Fig 4B) clones, which contain the APPBP2 and PPM1D genes or the CDK6 gene. To evaluate the potential pathogenic importance of these amplicons, we analyzed mRNA and protein expression of PPM1D and CDK6. When compared with normal cerebellum reference, PPM1D mRNA was moderately overexpressed in seven of the nine tumors (Fig 4C), whereas CDK6 mRNA was strongly overexpressed in nine of the nine tumors (Fig 4D). The mean of two individual experiments was plotted and error bars represent 1 x SD. Tumors were selected to include different histologic subtypes and varying complexity of genetic aberrations. Protein expression was determined by IHC analysis of a large series of tumors using TMAs (Appendix Table 1). Strong and widespread immunoreactivity, defined as nuclear staining in more than 30% of the tumor cells, was observed for PPM1D in 148 of 168 (88%) tumors (Figs 4E and 4F) and for CDK6 in 50 of 169 (30%) tumors (Figs 4G and 4H).
CDK6 Expression Status Is an Independent Predictor of Prognosis
Comparison of tumors with no or low CDK6 staining versus tumors with moderate or high staining (Fig 5A) identified CDK6 overexpression as a prognostic marker that was associated significantly with poor clinical outcome using univariate analysis (Fig 5B; log-rank test, P < .01). The proportional hazards regression model identified four significant prognostic factors: chemotherapy (P < .01), stage of metastasis (P < .01), histology (P = .02), and CDK6 overexpression (P = .02). Age (P = .57), sex (P = .09), and level of resection (P = .49) were not statistically significant. Results are illustrated in Figure 5C.
The relationships of CDK6 overexpression and the presence of the potential molecular prognostic markers loss 17p13.3 to 13.1 and amplification of MYC or MYCN also were evaluated using informative tumor samples of the matrix-CGH study (n = 29). CDK6 overexpression is not associated with loss of 17p (P = .41) and tumors with MYC or MYCN amplification did not necessarily show CDK6 overexpression (one of two and one of three, respectively).
DISCUSSION
Matrix-CGH analysis identified DNA copy number alterations of several novel candidate genes that may play a role in medulloblastoma pathogenesis. The PPM1D gene at 17q23.2 was amplified in one medulloblastoma and showed a lower copy number gain in 24 other tumors, making it the most frequent genomic aberration detected in our tumor series (25 of 47; 53%). Notably, the region 7q21.2, containing CDK6, was gained or amplified in 32% (15 of 47) of the tumors. Regarding the frequent losses at 17p13.1, our findings suggest DLG4 as an interesting candidate gene, in addition to other known candidates, such as TP53, LIS1, HIC1, and REN.38-43 Candidate tumor suppressor genes located in the deleted regions on chromosome arm 10q include UTF1, the gene product of which seems to function as transcriptional regulator of DNA-polymerase II promoters, as well as PTEN, which encodes a known tumor suppressor that functions as an inhibitor of the phosphoinositol-3-kinase/AKT signaling pathway.44 Detailed mutational analyses and functional studies will be required to elucidate the role that each of these genes might play in medulloblastoma pathogenesis.
In a previous study,45 expression analysis was performed on a subset (n = 26) of the tumor samples used here. When correlating candidate genes from genomic amplicons identified by matrix-CGH with their respective expression ratios, gene dosage effects are observed for three of four loci (CDK6, MYCN, and CCND2). Only MYC amplification did not result in increased gene expression.
The high frequency of aberrations on chromosome 17 supports its central importance in medulloblastoma pathogenesis. We confirmed frequent loss on 17p and identified a region of recurrent copy number gain on 17q, both of which were correlated with poor overall survival in univariate analyses. These analyses have to be viewed with caution, however, given that loss of 17p and gain of 17q are not independent events in tumors with i(17q). Chromosome 17 aberrations also had prominent impact in the unsupervised cluster analysis (ie, characterized the tumors of a group of patients with poor prognosis [group II]). Molecular classification correlated significantly with histology mainly because of seven of eight desmoplastic tumors clustering in the group with good prognosis (group I) and eight of 12 anaplastic tumors clustering in the group with poor prognosis. Tumors with classic histology were evenly distributed between groups I and II. Amplifications of MYC and MYCN, which were found to be mutually exclusive in a recent study of medulloblastoma,19 clustered in subgroups IIa and IIb, respectively. In addition to all MYCN amplifications, group IIb tumors contained all of the amplifications involving the CDK6 locus. Recently, MYCN and the CDK6 partner cyclin D1 were identified as targets of the sonic hedgehog pathway involved in the proliferation of neuronal precursors in medulloblastoma.46 Finally, group IIa, which only contained pediatric tumors, represented a clinically defined group of patients with poor prognosis and anaplastic tumors carrying MYC amplifications.6
The amplicon detected at 17q23.2 contained the genes APPBP2 and PPM1D, which were also reported to be coamplified in neuroblastoma47 as well as breast48 and ovarian carcinoma.49 Overexpression of PPM1D but not APPBP2 was reported to cooperate with rat sarcoma viral oncogene homolog in transforming primary mouse fibroblasts.48 Recently, it was shown that overexpression of PPM1D in mouse embryo fibroblasts abolished TP53 tumor-suppressor activity,50 whereas PPM1D deficiency led to resistance against oncogenic transformation mediated by the p16INK4A and p19ARF (p14ARF in human) pathways.51 Although increased PPM1D protein is not associated with poor prognosis, it might contribute to escape from the TP53-dependent control of cell cycle regulation, apoptosis, and DNA repair. PPM1D, which itself is induced by TP53, takes part in a negative-feedback loop downregulating TP53 directly via p38 MAPK.52 Our results, together with published data,53 indicate that different levels of deregulation of the TP53 pathway might exist in medulloblastoma, including amplification or gain and strong expression of PPM1D, copy number gain/amplification of MDM4, frequent monoallelic loss of TP53 and, less commonly, TP53 mutation.
The novel amplicon detected at 7q21.2 contained CDK6 as the only gene, and genomic amplification was associated with increased CDK6 protein levels. However, CDK6 overexpression was not restricted to cases with gene amplification, thus suggesting that additional mechanisms exist, resulting in high protein levels of CDK6 in approximately 30% of medulloblastomas. At least in part, a direct connection between PPM1D and CDK6 protein expression in medulloblastoma could be mediated by p38 MAPK52,54 (Fig 6). Overexpression of CDK6 was reported previously in T-cell lymphoma55 and a small subset of malignant gliomas.56 Similar to CDK4, the CDK6 protein is activated by binding to D-type cyclins. The cyclin D/CDK6 complexes phosphorylate the retinoblastoma protein RB1 and the related proteins RBL1 and RBL2, leading to the release of E2F transcription factors.57 In line with our findings, data from mouse models support the importance of retinoblastoma pathway alterations in medulloblastoma development.58-61 High levels of CDK6 protein may result in constitutive phosphorylation of RB1, which may override inhibition by p16INK4A, as shown in murine embryonal stem cells.62 Recent studies also showed that CDK6 overexpression blocks differentiation in leukemic cells63 and osteoblasts,54 suggesting that CDK6 is a key regulator of the RB1 pathway affecting proliferation and differentiation in medulloblastoma and other tumors.
For better risk stratification and the planning of individual therapeutic regimens, it is important to have reliable markers for the assessment of prognosis and response to therapy. At present, prognostic markers for medulloblastoma patients include certain clinical factors; the histologic subtype; and a few molecular parameters, such as loss of 17p, amplification and overexpression of MYC or MYCN, and expression levels of TRKC, ERBB2, and PDGFRB.64-68 In this study, we identified a novel prognostic marker for medulloblastoma patients, namely overexpression of CDK6. In multivariate regression analysis, we found protein overexpression to be an independent prognostic factor for poor overall survival of medulloblastoma patients. Given its higher frequency as compared with the occurrence of DNA amplifications, we consider CDK6 overexpression to be a more informative marker than either MYC or MYCN amplification. No statistically significant overlap was observed between CDK6 expression and 17p13.3 to 13.1 loss in the matrix-CGH data set, suggesting that these are independent prognostic factors that can complement each other in diagnostics. Given that CDK6 expression can be determined by means of IHC on formalin-fixed paraffin sections, this marker is well suited for the routine diagnostic setting as well as an evaluation in controlled clinical studies. Moreover, therapeutic approaches specifically targeting cyclin-dependent kinases are under development and may provide a novel option for the treatment of this tumor.69
Appendix
The Appendix is included in the full-text version of this article, available online at www.jco.org. It is not included in the PDF (via Adobe? Acrobat Reader?) version.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Stefanie Hofmann, Andrea Wittmann, Frauke Devens, Bj?rn Tews, and Gunnar Wrobel for excellent technical support. We also thank the Mapping Groups of the Wellcome Trust Sanger Institute for initial clone supply and verification.
NOTES
Supported by Bundesministerium fr Bildungund Forschung Grants No. 01GS0460 and 01GRO417 (P.L.), and by a Kekulé-fellowship of the Fonds der Chemischen Industrie (F.M.).
Both F.M. and B.R. contributed equally to this work.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Gurney JG, Kadan-Lottick N: Brain and other central nervous system tumors: Rates, trends, and epidemiology. Curr Opin Oncol 13:160-166, 2001
Fisher PG, Burger PC, Eberhart CG: Biologic risk stratification of medulloblastoma: The real time is now. J Clin Oncol 22:971-974, 2004
Eberhart CG, Kepner JL, Goldthwaite PT, et al: Histopathologic grading of medulloblastomas: A Pediatric Oncology Group study. Cancer 94:552-560, 2002
Giangaspero F, Perilongo G, Fondelli MP, et al: Medulloblastoma with extensive nodularity: A variant with favorable prognosis. J Neurosurg 91:971-977, 1999
Giangaspero F, Rigobello L, Badiali M, et al: Large-cell medulloblastomas: A distinct variant with highly aggressive behavior. Am J Surg Pathol 16:687-693, 1992
Brown HG, Kepner JL, Perlman EJ, et al: "Large cell/anaplastic" medulloblastomas: A Pediatric Oncology Group Study. J Neuropathol Exp Neurol 59:857-865, 2000
Zeltzer PM, Boyett JM, Finlay JL, et al: Metastasis stage, adjuvant treatment, and residual tumor are prognostic factors for medulloblastoma in children: Conclusions from the Children's Cancer Group 921 randomized phase III study. J Clin Oncol 17:832-845, 1999
Gilbertson R: Paediatric embryonic brain tumours: Biological and clinical relevance of molecular genetic abnormalities. Eur J Cancer 38:675-685, 2002
Wechsler-Reya RJ: Analysis of gene expression in the normal and malignant cerebellum. Recent Prog Horm Res 58:227-248, 2003
Bayani J, Zielenska M, Marrano P, et al: Molecular cytogenetic analysis of medulloblastomas and supratentorial primitive neuroectodermal tumors by using conventional banding, comparative genomic hybridization, and spectral karyotyping. J Neurosurg 93:437-448, 2000
Pan E, Pellarin M, Holmes E, et al: Isochromosome 17q is a negative prognostic factor in poor-risk childhood medulloblastoma patients. Clin Cancer Res 11:4733-4740, 2005
Nicholson JC, Ross FM, Kohler JA, et al: Comparative genomic hybridization and histological variation in primitive neuroectodermal tumours. Br J Cancer 80:1322-1331, 1999
Reardon DA, Michalkiewicz E, Boyett JM, et al: Extensive genomic abnormalities in childhood medulloblastoma by comparative genomic hybridization. Cancer Res 57:4042-4047, 1997
Avet-Loiseau H, Venuat AM, Terrier-Lacombe MJ, et al: Comparative genomic hybridization detects many recurrent imbalances in central nervous system primitive neuroectodermal tumours in children. Br J Cancer 79:1843-1847, 1999
Schutz BR, Scheurlen W, Krauss J, et al: Mapping of chromosomal gains and losses in primitive neuroectodermal tumors by comparative genomic hybridization. Genes Chromosomes Cancer 16:196-203, 1996
Gilhuis HJ, Anderl KL, Boerman RH, et al: Comparative genomic hybridization of medulloblastomas and clinical relevance: Eleven new cases and a review of the literature. Clin Neurol Neurosurg 102:203-209, 2000
Eberhart CG, Kratz JE, Schuster A, et al: Comparative genomic hybridization detects an increased number of chromosomal alterations in large cell/anaplastic medulloblastomas. Brain Pathol 12:36-44, 2002
Hui AB, Takano H, Lo KW, et al: Identification of a novel homozygous deletion region at 6q23.1 in medulloblastomas using high-resolution array comparative genomic hybridization analysis. Clin Cancer Res 11:4707-4716, 2005
Lamont JM, McManamy CS, Pearson AD, et al: Combined histopathological and molecular cytogenetic stratification of medulloblastoma patients. Clin Cancer Res 10:5482-5493, 2004
Lichter P, Joos S, Bentz M, et al: Comparative genomic hybridization: Uses and limitations. Semin Hematol 37:348-357, 2000
Solinas-Toldo S, Lampel S, Stilgenbauer S, et al: Matrix-based comparative genomic hybridization: Biochips to screen for genomic imbalances. Genes Chromosomes Cancer 20:399-407, 1997
Pinkel D, Segraves R, Sudar D, et al: High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 20:207-211, 1998
Schwaenen C, Nessling M, Wessendorf S, et al: Automated array-based genomic profiling in chronic lymphocytic leukemia: Development of a clinical tool and discovery of recurrent genomic alterations. Proc Natl Acad Sci U S A 101:1039-1044, 2004
Zielinski B, Gratias S, Toedt G, et al: Detection of chromosomal imbalances in retinoblastoma by matrix-based comparative genomic hybridization. Genes Chromosomes Cancer 43:294-301, 2005
Giangaspero F, Bigner SH, Kleihues P, et al: Medulloblastoma, in Cavanee WK, Kleihues P (eds): Pathology and Genetics of Tumours of the Nervous System. Lyons, France, IARC Press, 2000, pp 129-137
Kortmann RD, Kuhl J, Timmermann B, et al: Postoperative neoadjuvant chemotherapy before radiotherapy as compared to immediate radiotherapy followed by maintenance chemotherapy in the treatment of medulloblastoma in childhood: Results of the German prospective randomized trial HIT '91. Int J Radiat Oncol Biol Phys 46:269-279, 2000
Korn EL: Censoring distributions as a measure of follow-up in survival analysis. Stat Med 5:255-260, 1986
Fiegler H, Carr P, Douglas EJ, et al: DNA microarrays for comparative genomic hybridization based on DOP-PCR amplification of BAC and PAC clones. Genes Chromosomes Cancer 36:361-374, 2003
Knight SJ, Lese CM, Precht KS, et al: An optimized set of human telomere clones for studying telomere integrity and architecture. Am J Hum Genet 67:320-332, 2000
Freier K, Joos S, Flechtenmacher C, et al: Tissue microarray analysis reveals site-specific prevalence of oncogene amplifications in head and neck squamous cell carcinoma. Cancer Res 63:1179-1182, 2003
Waye JS, England SB, Willard HF: Genomic organization of alpha satellite DNA on human chromosome 7: Evidence for two distinct alphoid domains on a single chromosome. Mol Cell Biol 7:349-356, 1987
Bubendorf L, Kononen J, Koivisto P, et al: Survey of gene amplifications during prostate cancer progression by high-throughout fluorescence in situ hybridization on tissue microarrays. Cancer Res 59:803-806, 1999
Lichter P, Cremer T, Borden J, et al: Delineation of individual human chromosomes in metaphase and interphase cells by in situ suppression hybridization using recombinant DNA libraries. Hum Genet 80:224-234, 1988
Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29:e45, 2001
Harrell FE Jr, Shih YC: Using full probability models to compute probabilities of actual interest to decision makers. Int J Technol Assess Health Care 17:17-26, 2001
de Hoon MJ, Imoto S, Nolan J, et al: Open source clustering software. Bioinformatics 20:1453-1454, 2004
Fan X, Wang Y, Kratz J, et al: HTERT gene amplification and increased mRNA expression in central nervous system embryonal tumors. Am J Pathol 162:1763-1769, 2003
Adesina AM, Nalbantoglu J, Cavenee WK: p53 gene mutation and mdm2 gene amplification are uncommon in medulloblastoma. Cancer Res 54:5649-5651, 1994
Biegel JA, Burk CD, Barr FG, et al: Evidence for a 17p tumor related locus distinct from p53 in pediatric primitive neuroectodermal tumors. Cancer Res 52:3391-3395, 1992
Di Marcotullio L, Ferretti E, De Smaele E, et al: REN(KCTD11) is a suppressor of Hedgehog signaling and is deleted in human medulloblastoma. Proc Natl Acad Sci U S A 101:10833-10838, 2004
Koch A, Tonn J, Kraus JA, et al: Molecular analysis of the lissencephaly gene 1 (LIS-1) in medulloblastomas. Neuropathol Appl Neurobiol 22:233-242, 1996
McDonald JD, Daneshvar L, Willert JR, et al: Physical mapping of chromosome 17p13.3 in the region of a putative tumor suppressor gene important in medulloblastoma. Genomics 23:229-232, 1994
Ohgaki H, Eibl RH, Schwab M, et al: Mutations of the p53 tumor suppressor gene in neoplasms of the human nervous system. Mol Carcinog 8:74-80, 1993
Wu X, Senechal K, Neshat MS, et al: The PTEN/MMAC1 tumor suppressor phosphatase functions as a negative regulator of the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci U S A 95:15587-15591, 1998
Neben K, Korshunov A, Benner A, et al: Microarray-based screening for molecular markers in medulloblastoma revealed STK15 as independent predictor for survival. Cancer Res 64:3103-3111, 2004
Oliver TG, Grasfeder LL, Carroll AL, et al: Transcriptional profiling of the Sonic hedgehog response: A critical role for N-myc in proliferation of neuronal precursors. Proc Natl Acad Sci U S A 100:7331-7336, 2003
Saito-Ohara F, Imoto I, Inoue J, et al: PPM1D is a potential target for 17q gain in neuroblastoma. Cancer Res 63:1876-1883, 2003
Li J, Yang Y, Peng Y, et al: Oncogenic properties of PPM1D located within a breast cancer amplification epicenter at 17q23. Nat Genet 31:133-134, 2002
Hirasawa A, Saito-Ohara F, Inoue J, et al: Association of 17q21-q24 gain in ovarian clear cell adenocarcinomas with poor prognosis and identification of PPM1D and APPBP2 as likely amplification targets. Clin Cancer Res 9:1995-2004, 2003
Bulavin DV, Demidov ON, Saito S, et al: Amplification of PPM1D in human tumors abrogates p53 tumor-suppressor activity. Nat Genet 31:210-215, 2002
Bulavin DV, Phillips C, Nannenga B, et al: Inactivation of the Wip1 phosphatase inhibits mammary tumorigenesis through p38 MAPK-mediated activation of the p16(Ink4a)-p19(Arf) pathway. Nat Genet 36:343-350, 2004
Takekawa M, Adachi M, Nakahata A, et al: p53-inducible wip1 phosphatase mediates a negative feedback regulation of p38 MAPK-p53 signaling in response to UV radiation. Embo J 19:6517-6526, 2000
Portwine C, Chilton-MacNeill S, Brown C, et al: Absence of germline and somatic p53 alterations in children with sporadic brain tumors. J Neurooncol 52:227-235, 2001
Moro T, Ogasawara T, Chikuda H, et al: Inhibition of Cdk6 expression through p38 MAP kinase is involved in differentiation of mouse prechondrocyte ATDC5. J Cell Physiol 204:927-933, 2005
Chilosi M, Doglioni C, Yan Z, et al: Differential expression of cyclin-dependent kinase 6 in cortical thymocytes and T-cell lymphoblastic lymphoma/leukemia. Am J Pathol 152:209-217, 1998
Costello JF, Plass C, Arap W, et al: Cyclin-dependent kinase 6 (CDK6) amplification in human gliomas identified using two-dimensional separation of genomic DNA. Cancer Res 57:1250-1254, 1997
Ortega S, Malumbres M, Barbacid M: Cyclin D-dependent kinases, INK4 inhibitors and cancer. Biochim Biophys Acta 1602:73-87, 2002
Zindy F, Nilsson LM, Nguyen L, et al: Hemangiosarcomas, medulloblastomas, and other tumors in Ink4c/p53-null mice. Cancer Res 63:5420-5427, 2003
Wetmore C, Eberhart DE, Curran T: Loss of p53 but not ARF accelerates medulloblastoma in mice heterozygous for patched. Cancer Res 61:513-516, 2001
Marino S, Vooijs M, van Der Gulden H, et al: Induction of medulloblastomas in p53-null mutant mice by somatic inactivation of Rb in the external granular layer cells of the cerebellum. Genes Dev 14:994-1004, 2000
Dyer MA: Mouse models of childhood cancer of the nervous system. J Clin Pathol 57:561-576, 2004
Faast R, White J, Cartwright P, et al: Cdk6-cyclin D3 activity in murine ES cells is resistant to inhibition by p16(INK4a). Oncogene 23:491-502, 2004
Matushansky I, Radparvar F, Skoultchi AI: CDK6 blocks differentiation: Coupling cell proliferation to the block to differentiation in leukemic cells. Oncogene 22:4143-4149, 2003
Eberhart CG, Kratz J, Wang Y, et al: Histopathological and molecular prognostic markers in medulloblastoma: C-myc, N-myc, TrkC, and anaplasia. J Neuropathol Exp Neurol 63:441-449, 2004
Gajjar A, Hernan R, Kocak M, et al: Clinical, histopathologic, and molecular markers of prognosis: Toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 22:984-993, 2004
Grotzer MA, Hogarty MD, Janss AJ, et al: MYC messenger RNA expression predicts survival outcome in childhood primitive neuroectodermal tumor/medulloblastoma. Clin Cancer Res 7:2425-2433, 2001
Gilbertson RJ, Clifford SC: PDGFRB is overexpressed in metastatic medulloblastoma. Nat Genet 35:197-198, 2003
MacDonald TJ, Brown KM, LaFleur B, et al: Expression profiling of medulloblastoma: PDGFRA and the RAS/MAPK pathway as therapeutic targets for metastatic disease. Nat Genet 29:143-152, 2001
Fry DW, Harvey PJ, Keller PR, et al: Specific inhibition of cyclin-dependent kinase 4/6 by PD 0332991 and associated antitumor activity in human tumor xenografts. Mol Cancer Ther 3:1427-1438, 2004(Frank Mendrzyk, Bernhard )