Microsoft Word - 2.KetterRalf_PrognosticGenetic.doc 388 Ralf Ketter et al Prognostic genetic markers in malignant gliomas Prognostic genetic markers in malignant gliomas Ralf Ketter1, Silke Wemmert2, Stefan Linsler1, Wolf-Ingo Steudel1, Steffi Urbschat1 1Department of Neurosurgery, Saarland University, 66421 Homburg/Saar, Germany 2Department of Otolaryngology, Saarland University, Homburg/ Saar, Germany Abstract Glioblastomas are the most frequent and malignant brain tumors in adults. Surgical cure is virtually impossible and despite of radiation and chemotherapy the clinical course is very poor. Epigenetic silencing of MGMT has been associated with a better response to temozolomide-chemotherapy. We previously showed that temozolomide increases the median survival time of patients with tumors harbouring deletions on 9p within the region for p15(INK4b), p16(INK4a), and 10q (MGMT). The aim of this study was to investigate the methylation status of p15, p16, 14ARF and MGMT in glioblastomas and to correlate the results with the clinical data. Only patients with KPS > 70, radical tumor resection, radiation and temozolomide-chemotherapy after recurrence were included. We observed promoter methylation of MGMT in 56% (15/27) and of p15 in 37% (10/27) of the tumors, whereas methylation of p16 and p14ARF were rare. Interestingly, methylation of p15 emerged as a significant predictor of shorter overall survival (16.9 vs. 23.8 months, p=0.025), whereas MGMT promoter methylation had no significant effect on median overall survival under this treatment regimen (22.5 vs. 22.1 months, p=0.49). In the presence of other clinically relevant factors, p15 methylation remains the only significant predictor (p=0.021; Cox regression). Although these results need to be confirmed in larger series and under different treatment conditions, our retrospective study shows clear evidence that p15 methylation can act as an additional prognostic factor for survival and underlines that this tumor suppressor, involved in cell cycle control, can act as an attractive candidate for therapeutic approaches in glioblastomas. Keywords: glioblastoma, O(6)- methylguanine-DNA methyltransferase (MGMT), methylated MGMT, temozolomide Introduction Glioblastomas [World Health Organization (WHO) grade IV] are the most frequent and the most malignant brain tumors in adults. They arise either de novo without recognizable precursor lesions or develop from lower grade astrocytomas (secondary GBM). Despite of multimodal therapy approaches, the prognosis is generally poor but varies markedly between the malignancy grades and even between individuals with the same malignancy grade. Less than half of the Romanian Neurosurgery (2010) XVII 4: 388 – 395 389 patients survive more than a year. (1) Besides radical surgery, a higher preoperative Karnofsky Performance Score (KPS) and younger age are predictors of a more favourable clinical course. (2-5) Over the past decades, genetic abnormalities involved in pathogenesis and progression of these tumors were identified. Several alterations were also shown to be correlated with prognosis. In novel studies progression-associated genetic markers in differing tumor entities as well as in gliomas are increasingly observed. Because of the mostly focal appearance of the changes to start with, the molecular-cytogenetic methods with their high dissolving ability are on this occasion essential. In a study which has just been finished, we compared the genetic aberration profile of the diffuse gliomas with the response rate of temozolomide chemotherapy. Because of the observation, we assume that there is minority of patients who survive for longer than 5 years after the recognition of a glioblastoma, surgery and a combined radio- and chemotherapy. We could show an until now unknown correlation of the patterns of genetic changes with the varying response rate of patients on this chemotherapy. Using CGH and LOH we could prove that gliomas patients with a tumor which shows a loss of 9p or 10q significantly benefit from temozolomide treatment. The gene for the O (6) - methylguanine-methyltransferase (MGMT) is localised on chromosome 10 (10q26) and encoded as the DNA repair- enzyme O(6) -alkylguanine-DNA- alkyltransferase (OGAT). The function of OGAT is the protection of the cells from alkylating substances by the removal of damaging DNA-methylisations i.e. for the demethylisation of methylguanine. The effect of the alkylating cytostaticum temozolomide is based exactly on these damaging DNA methylisations. The effect of the chemotherapeuticum is therefore impeded by MGMT and a loss of this would because of this be of advantage for the treated patients. Interesting for secondary studies is the question if the appropriate chromosomal regions on the apparently intact second chromosome 10 are mutated, methylated or still active. Therefore the expression status and the methylisation pattern of MGMT in the post-characterised gliomas be analysed on a DNA-level. For that reason it should be clarified if the MGMT gene on the apparently intact chromosome 10 is still active of if it has been made inactive by the promotor-hypermethylisation and if a reduced dose of the gene will already be sufficient for therapeutic success. Current studies on anaplastic astrocytomas and glioblastomas have shown that patients with a promotor hypermethylisation of the MGMT gene benefit significantly from a nitrosoharn substance and temozolomide chemotherapy in comparison to patients whose tumors do not show any methylisation in the MGMT gene (27, 28). Deletion or mutation of the p16(INK4a)/ARF/p15(INK4b) locus on chromosome 9p21 is among the most common alterations seen in human cancer and in human gliomas. (6-9) The INK4a locus encodes two gene products that are involved in cell cycle regulation through inhibition of CDK4-mediated RB phosphorylation (p16) and binding to MDM2 leading to p53 stabilization (p14ARF). The tumor suppressor gene products p16 and p15 are both capable of binding to CDK4 and CDK6, these kinases associate with D-type cyclines and these 390 Ralf Ketter et al Prognostic genetic markers in malignant gliomas binary complexes are responsible for phosphorylation of RB-protein at mid G1 of the cell cycle. The phosphorylation of RB is assumed to be critical for progression through G1 and entry into S-phase of the cell cycle. Binding of INK4 to CDK4/6 inhibits its kinase activity and thereby arrests progression through the cell cycle in mid-late G1. (10) Over half of the high grade gliomas lack a functional INK4a/ARF locus. Gliomas with intact INK4a/ARF carry mutations in other components of the RB and p53 pathways implicating these two pathways as being absolutely critical in cell growth and death control. (11-14) Previous studies showed that deletion of 9p including the INK4a/b locus is a significant unfavorable prognostic factor for survival of glioblastoma patients. (15, 16) Further on, this alteration has been reported to be inversely correlated with the chemosensitivity of malignant gliomas. (17) Over the past years aberrant DNA methylation was shown to be a common molecular lesion in human tumors as well, which had also impact on patient prognosis and treatment response. Epigenetic silencing of p16 and p15 was shown in a variety of human neoplasms, in glioma patients hypermethylation is reported in about 30% of the cases. (18-21) Whereas in other tumors inactivation of both tumor suppressor genes was associated with prognosis and response to chemotherapy, a prognostic and predictive role in gliomas is not shown. (22-25) On the other hand, epigenetic silencing of the DNA repair gene O6- methylguanine-DNA methyltransferase (MGMT) on chromosome 10q26 by hypermethylation has been linked to a better prognosis for glioblastoma patients treated with alkylating agents like temozolomide. (26-29) A benefit was not observed for patients with MGMT promoter methylation and BCNU- chemotherapy. (30) Another study comparing different treatment regimens showed that the prognostic effect was only significant when patients were treated simultaneously with radio - and chemotherapy. (31) These results suggest that the reported impact of MGMT methylation is strongly dependent on therapeutic modalities and schedules. In our previous study we identified the negative prognostic impact for deletions on 9p and 10q, which can be compensated by temozolomide treatment. (16) In the current setting, our aim was to investigate the methylation status of p15, p16, p14ARF and MGMT, and correlate the results with the clinical data of the glioblastoma patients. Materials and Methods Patients and tumor samples The retrospective study included 27 glioblastoma patients (23 primary and 4 secondary GBM) who underwent surgery at the Department of Neurosurgery of the Saarland University, Homburg, Germany. After radical tumor surgery, all patients received standard radiation therapy (RT) (1.8-2 Gy, total dose of 60 Gy) and adjuvant temozolomide chemotherapy in case of recurrence. The doses were 150 mg/m² for 5 days in 4 week cycles. Specimens of resected tumor were immediately shock frozen in liquid nitrogen and stored at – 80°C or fixed in formalin and embedded in paraffin. All patients gave written informed consent for the use of the tumor samples for genetic analysis. Romanian Neurosurgery (2010) XVII 4: 388 – 395 391 Methylation-specific polymerase chain reaction (MS-PCR) DNA of the tumor samples was isolated following standard protocols with chloroforme followed by sodium bisulfite modification. Promoter hypermethylation of the MGMT, p15, p16 and p14ARF genes were determined by MS-PCR as described previously. (18, 26, 32) The amplified products were electrophoresed on 3.5% agarose gels and visualized with ethidium bromide. Methylated blood DNA was included in each PCR set as methylated and unmethylated controls, respectively. Statistical analyses Comparison of survival times between groups defined by methylation status was performed by Kaplan-Meier curves and with two-sided log rank tests. Multivariate Cox regression analysis was performed to identify significant predictors for survival. Effects in these models were quantified by hazard ratio estimates with 95% confidence intervals. Median survival rates were calculated using the Kaplan-Meier method. Results Clinical Data Median age at surgery was 49 years (range 26-70), the sex ratio was 2.375 (19 men / 8 women) (Table 1) and median post operative KPS was 100 (range 80-100, 18 patients with KPS 100). Case Histologic age/ ST MGMT- p15- p16- p14ARF- type sex [months] status status status status T4789 GBM 38/m 55,3 M U U U 596/98 GBM 32/m 21,1 U U U U 1099/98 GBM 49/m 23,8 U U U U 1782/98 sGBM 26/m 24,2 M U U U 896/99 GBM 51/m 13 U M U U 1326/99 GBM 56/m 19,2 U U U U 1349/99 GBM 54/m 72,6 M U U U 1460/99 GBM 31/f 11,9 M M U U 1795/99 sGBM 31/m 6,9 M U U U 265/00 GBM 46/f 23,1 U M n.a. U 497/00 GBM 58/f 16,9 M M U U 1106/00 GBM 70/m 19,8 M M U U 1691/00 sGBM 39/f 80,2 M U U U 1707/00 GBM 47/m 29,4 U U U U 643/01 GBM 54/f 26,6 U U U U 662/01 GBM 70/m 13,4 U M U U 1405/01 GBM 53/m 7,4 U M M U 6/02 GBM 41/m 26,2 M U U U 369/02 sGBM 37/m 20,2 M U U U 947/02 GBM 46/m 17,7 M M U U 1536/02 GBM 53/m 15,1 U U U U 1940/02 GBM 54/f 14,3 M M U U XXL/02 GBM 45/m 8,6 M U U U 392 Ralf Ketter et al Prognostic genetic markers in malignant gliomas 316/03 GBM 63/f 23,7 U U n.a. U 784/03 GBM 52/m 22,5 M U U U 831/03 GBM 58/m 53 U U n.a. U 1457/03 GBM 37/m 49,1 M M n.a. U Methylation analyses The MGMT promoter was methylated in 15/27 cases (55.6%), the p15 promoter was methylated in 10/27 (37%) glioblastomas. All secondary GBM (4/4) showed a methylated MGMT promoter and an unmethylated p15 promoter. Methylation of p14ARF was absent in all (27/27) investigated GBM. Methylation status of p16 was available for 23/27 glioblastomas. Hypermethylation of p16 was detected in only 1/23 cases (4.3%) (Table 1). Clinical outcome Overall median survival was 22.5 months with a two-year survival rate of 35.0%. In univariate analyses, MGMT methylation had no impact on overall survival (22.5 vs. 22.1 months, p=0.49, log- rank test, Fig. 1A), whereas p15 methylation was associated significantly with a shorter overall survival (16.9 vs. 23.8 months; p=0.0252, log-rank test, Figure 1B). Table 2 contains estimated hazard ratios and p- values for univariate analyses for all examined variables. We also performed a multivariate analysis including parameters previously identified as significant. As shown in Table 3, only p15 methylation emerged as a significant prognostic factor after adjusting for KPS, sex, age and MGMT. In the first analysis, the predictors KPS and age enter as numerical variables in the model, in the second analysis KPS and age are dichotomized with cutoffs 90 and 50 in order to reduce model complexity. The gender variable sex is set to 1 for females and 0 for males. Both analyses yield very similar results, identifying p15 methylation as only significant predictor. Discussion A better understanding of the genetic alterations predicting disease outcome and therapy response in patients with high grade gliomas will help to optimize both treatment and overall outcome. In our study setting, MGMT promoter methylation had no significant impact on survival, in contrast with other studies that observed a strong effect on patient response and on survival in larger cohorts treated with radio- and temozolomide- chemotherapy. (26, 28) This discrepancy might result, besides our smaller patient cohort, from the different treatment schedules. Actually, comparing various treatment modalities in glioblastomas, the prognostic effect of MGMT methylation was observed only when simultaneous chemoirradiation was administered. (26, 31) Further on, our PCR-results showed besides the methylated band also unmethylated bands in a large number of tumors. Taking into account this heterogeneity and that we are dealing with diffusely growing gliomas, (33, 34) this observation arises most likely from a different tumor cell population and from normal cells contaminating the tumor sample. Nevertheless, this heterogeneous methylation pattern may also result in different amounts of MGMT and therefore affect chemotherapy response. Romanian Neurosurgery (2010) XVII 4: 388 – 395 393 All evidence collected to date implicates that the INK4a/ARF gene products are critically important in control of growth arrest and senescence. Loss of p16 and ARF expression is associated with many human cancers, particularly gliomas. (20) A number of studies have shown that reconstitution of INK4a/ARF expression in glioma cells altered growth characteristics, reduced tumorigenicity and decreased invasive potential. These studies demonstrate the importance of the INK4a/ARF pathway in suppression of the neoplastic phenotype and suggest that restoration of a functional INK4a/ARF locus will be an important means of controlling the growth of gliomas. (36, 37) A striking observation of our study is the significant correlation of p15 methylation with a poorer clinical course. Loss of p15 had not been widely investigated previously as a potential determinant of chemo- and radiosensitivity or as a prognostic factor. To our knowledge, this is the first study showing inactivation of p15 by promoter hypermethylation to be a predictor for an unfavorable clinical course. Cyclin dependent kinase inhibitors (p16, p21, p27) were shown to exhibit an antitumor effect in malignant gliomas inducing growth arrest and apoptosis in cell culture. (36, 37) Further on, retrovirus mediated transfer of INK4a halts glioma formation in a rat model. (38) This corroborates the idea that retrovirus mediated gene transfer of INK4a/b may also be an effective means to arrest human gliomas. Therefore, restoring the normal function of p15 by gene therapy is an attractive goal in the treatment of human gliomas. Furthermore, our finding that glioblastomas have not simultaneously hypermethylated the investigated tumor suppressor genes on 9p implicates that these tumors carry no general defect in their pattern of CpG island methylation. Interestingly, all secondary GBM (4/4) showed a methylated MGMT promoter and an unmethylated p15 promoter which indicates that distinct molecular pathways constitute for primary and for secondary glioblastomas and let them differ both in biological behavior and in clinical outcome. This corresponds to the finding that methylation of p14ARF is associated with a shorter patient survival and is mutually exclusive of MGMT promoter methylation except of one case in low grade astrocytoma who underwent progression or recurrence. (35) In our patient cohort that mainly consists of primary glioblastoma, p14ARF was not observed, supporting that this alteration is mainly restricted to secondary glioblastoma and therefore in the pathway of astrocytoma progression. Although these results need to be confirmed in larger series, our retrospective study suggests that p15 hypermethylation can act as an additional important prognostic factor for survival in glioblastomas. Further investigations have to clarify if p15 methylation is a predictive factor for temozolomide treatment response and can act as a prognostic parameter for survival, independent of therapy. References 1. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK. WHO classification of tumours of the central nervous system. 4thed. Lyon: International Agency for Research on Cancer, 2007 2. Burger PC, Green SB. Patient age, histologic features, and length of survival in patients with glioblastoma multiforme. Cancer 1987;59:1617-25 3. Kreth FW, Warnke PC, Scheremet R, Ostertag CB. Surgical resection and radiation therapy versus biopsy and radiation therapy in the treatment of glioblastoma multiforme. J Neurosurg 1993;78:762-6 394 Ralf Ketter et al Prognostic genetic markers in malignant gliomas 4. Lacroix M, Abi-Said D, Fourney DR, Gokaslan ZL, Shi W, DeMonte F, Lang FF, McCutcheon IE, Hassenbusch SJ, Holland E, Hess K, Michael C, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190-8 5. Stummer W, Reulen HJ, Meinel T, Pichlmeier U, Schumacker W, Tonn JC, Rohde V, Oppel F, Turowski B, Woiciechowsky C, Franz K, Pietsch T. Extent of resection and survival in glioblastoma multiforme: Identification of and adjustment for bias. Neurosurgery 2008;62:564-76 6. Jen J, Harper JW, Bigner SH, Bigner DD, Papadopoulos N, Markowitz S, Willson JK, Kinzler KW, Vogelstein B. Deletion of p16 and p15 Genes in Brain Tumors. Cancer Res 1994;54:6353-8 7. Schmidt EE, Ichimura K, Reifenberger G, Collins VP. CDKN2 (p16/MTS1) gene deletion or CDK4 amplification occurs in the majority of glioblastomas. Cancer Res 1994;54:6321-4 8. Simon M, Köster G, Menon AG, Schramm J. Functional evidence for a role of combined CDKN2A(p16/p14(arf))/CDKN2B (p15) gene inactivation in malignant gliomas. Acta Neuropathol 1999;98:444-52 9. Rasheed A, Herndon JE, Stenzel TT, Raetz JG, Kendelhardt J, Friedman HS, Friedman AH, Bigner DD, Bigner SH, McLendon RE. Molecular markers of prognosis in astrocytic tumors. Cancer 2002;94:2688-97 10. 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;7:667-77 11. He J; Olson JJ, James CD. Lack of p16INK4 or retinoblastoma protein (pRb), or amplification- associated overexpression of cdk4 is observed in distinct subsets of malignant glial tumors and cell lines. Cancer Res 1995;55:4833-6 12. Ichimura K, Schmidt EE, Goike HM, Collins VP. Human glioblastomas with no alterations of the CDKN2A (p16INK4A, MTS1) and CDK4 genes have frequent mutations of the retinoblastoma gene. Oncogene 1996;13:1065-72 13. Ichimura K, Bolin MB, Goike HM, Schmidt EE, Moshref A, Collins VP. Deregulation of the p14ARF/MDM2/p53 pathway is a prerequisite for human astrocytic gliomas with G1-S transition control gene abnormalities. Cancer Res 2000;60:417-24 14. Fulci G, Labuhn M, Maier D, Lachat Y, Hausmann O, Hegi ME, Janzer RC, Merlo A, Van Meir EG. p53 gene mutation and ink4a-arf deletion appear to be two mutually exclusive events in human glioblastoma. Oncogene 2000;19:3816-22 15. Kamiryo T, Tada K, Shiraishi S, Shinojima N, Nakamura H, Kochi M, Kuratsu J, Saya H, Ushio Y. Analysis of homozygous deletion of the p16 gene and correlation with survival in patients with glioblastoma multiforme. J Neurosurg 2002;96:815-22 16. Wemmert S, Ketter R, Rahnenführer J, Beerenwinkel N, Strowitzki M, Feiden W, Hartmann C, Lengauer T, Stockhammer F, Zang KD, Meese E, Steudel WI, et al. Patients with high-grade gliomas harboring deletions of chromosomes 9p and 10q benefit from temozolomide treatment. Neoplasia 2005;7:883- 93 17. Simon M, Voss D, Park-Simon TW, Mahlberg R, Köster G. Role of p16 and p14ARF in radio-and chemosensitivity of malignant gliomas. Oncol Rep 2006;16:127-32 18. Herman JG, Graff JR, Myöhänen S, Nelkin BD, Baylin SB. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci U S A 1996;93:9821-6 19. Herman JG, Jen J, Merlo A, Baylin SB. Hypermethylation-associated inactivation indicates a tumor suppressor role for p15INK4B1. Cancer Res 1996;56:722-7 20. Esteller M, Corn PG, Baylin SB, Herman JG. A gene hypermethylation profile of human cancer. Cancer Res 2001;61:3225-9 21. Uhlmann K, Rohde K, Zeller C, Szymas J, Vogel S, Marczinek K, Thiel G, Nürnberg P, Laird PW. Distinct methylation profiles of glioma subtypes. Int J Cancer2003;106(1):52-9 22. Uchida T, Kinoshita T, Nagai H, Nakahara Y, Saito H, Hotta T, Murate T. Hypermethylation of the p15INK4B gene in myelodysplastic syndromes. Blood 1997;90:1403-9 23. Kudoh K, Ichikawa Y, Yoshida S, Hirai M, Kikuchi Y, Nagata I, Miwa M, Uchida K. Inactivation of p16/CDKN2 and p15/MTS2 is associated with prognosis and response to chemotherapy in ovarian cancer. Int J Cancer 2002;99:579-82 24. Lee M, Sup Han W, Kyoung Kim O, Hee Sung S, Sun Cho M, Lee SN, Koo H. Prognostic value of p16INK4a and p14ARF gene hypermethylation in human colon cancer. Pathol Res Pract 2006;202:415-24 25. Rasheed A, Herndon JE, Stenzel TT, Raetz JG, Kendelhardt J, Friedman HS, Friedman AH, Bigner DD, Bigner SH, McLendon RE. Molecular markers of prognosis in astrocytic tumors. Cancer 2002;94:2688-97 26. Paz MF, Yaya-Tur R, Rojas-Marcos I, Reynes G, Pollan M, Aguirre-Cruz L, García-Lopez JL, Piquer J, Safont MJ, Balaña C, Sanchez-Cespedes M, García- Villanueva M, et al. CpG island hypermethylation of the DNA repair enzyme methyltransferase predicts response to temozolomide in primary gliomas. Clin Cancer Res 2004;10:4933-8 27. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, Belanger K, Brandes AA, Marosi C, Bogdahn U, Curschmann J, Janzer RC, et al. Romanian Neurosurgery (2010) XVII 4: 388 – 395 395 European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups; National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-96 28. Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, Kros JM, Hainfellner JA, Mason W, Mariani L, Bromberg JE, Hau P, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005;352:997-1003 29. Watanabe T, Katayama Y, Komine C, Yoshino A, Ogino A, Ohta T, Fukushima T. O6-methylguanine- DNA methyltransferase methylation and TP53 mutation in malignant astrocytomas and their relationships with clinical course. Int J Cancer. 2005 Feb 10;113:581-7 30. Blanc JL, Wager M, Guilhot J, Kusy S, Bataille B, Chantereau T, Lapierre F, Larsen CJ, Karayan-Tapon L. Correlation of clinical features and methylation status of MGMT gene promoter in glioblastomas. J Neurooncol 2004;68:275-83 31. Crinière E, Kaloshi G, Laigle-Donadey F, Lejeune J, Auger N, Benouaich-Amiel A, Everhard S, Mokhtari K, Polivka M, Delattre JY, Hoang-Xuan K, Thillet J, et al. MGMT prognostic impact on glioblastoma is dependent on therapeutic modalities. J Neurooncol 2007;83:173-9 32. Sato F, Harpaz N, Shibata D, Xu Y, Yin J, Mori Y, Zou TT, Wang S, Desai K, Leytin A, Selaru FM, Abraham JM, et al. Hypermethylation of the p14(ARF) gene in ulcerative colitis-associated colorectal carcinogenesis. Cancer Res 2002;62:1148-51 33. Loeper S, Romeike BF, Heckmann N, Jung V, Henn W, Feiden W, Zang KD, Urbschat S. Frequent mitotic errors in tumor cells of genetically micro- heterogeneous glioblastomas. Cytogenet Cell Genet 2001;94:1-8 34. Jung V, Romeike BF, Henn W, Feiden W, Moringlane JR, Zang KD, Urbschat S. Evidence of focal genetic microheterogeneity in glioblastoma multiforme by area-specific CGH on microdissected tumor cells. J Neuropathol Exp Neurol 1999;58:993-9 35. Watanabe T, Katayama Y, Yoshino A, Yachi K, Ohta T, Ogino A, Komine C, Fukushima T. Aberrant hypermethylation of p14ARF and O6-methylguanine- DNA methyltransferase genes in astrocytoma progression. Brain Pathol 2007;17:5-10 36. Komata T, Kanzawa T, Takeuchi H, Germano IM, Schreiber M, Kondo Y, Kondo S. Antitumor effect of cyclin-dependent kinase inhibitors (p16(INK4A),p18(INK4C), p19(INK4D). p21 (WAF1(CIP1) and p27 (KIP1)) on malignant glioma cells. Br J Cancer 2003;88:1277-80 37. Inoue R, Moghaddam KA, Ranasinghe M, Saeki Y, Chiocca EA, Wade-Martins R. Infectious delivery of the 132 kb CDKN2A/CDKN2B genomic DNA region results in correctly spiced gene expression and growth suppression in glioma cells. Gene Ther 2004;11:1195- 204 38. Strauss BE, Fontes RB, Lotfi CF, Skorupa A, Bartol I, Cipolla-Neto J, Costanzi-Strauss E. Retroviral transfer of the p16INK4a cDNA inhibits C6 glioma formation in Wistar rats. Cancer Cell Int 2002;2:1-9