MOLECULAR BIOLOGY OF GLIOMA

CHAPTER 1MOLECULAR BIOLOGY OF GLIOMA Tomotoshi Marumoto1 and Hideyuki Saya*,21Department of Molecular Genetics, Division of Molecular and Clinical Genetics, Medical Instituteof Bioregulation, Fukuoka, Japan; 2Division of Gene Regulation, Institute for Advanced Medical Research,Keio University School of Medicine, Tokyo, Japan*Corresponding Author: Hideyuki Saya—Email: hsaya@a5.keio.jpAbstract:Glioblastoma (GBM) is the most aggressive form of glioma. Despite ceaseless efforts by researchers and physicians to find new therapeutic strategies, there havebeen no significant advances in the treatment of GBMs for several decades and mostpatients with GBM die within one and half years of diagnosis. Undoubtedly, onereason for this is the insufficient understanding of the initiation and progression ofGBMs at the molecular level. However, recent information regarding the geneticand epigenetic alterations and the microRNAs that are aberrantly activated orinactivated in GBMs has helped elucidate the formation of GBM in more detail.Here, we describe recent advances in the understanding of the biology of GBMs. INTRODUCTIONGliomas are the most common central nervous system tumors in adults. They are histologically classified as astrocytomas, oligodendrogliomas and oligoastrocytomas. Gliomas are graded on the World Health Organization (WHO) grading system according to their degree of malignancy. The most malignant Grade IV gliomas are termed glioblastomas (GBMs) and exhibit advanced features of malignancy, which include necrosis, vascular proliferation and pleomorphism. Patients with GBMs generally die within one and half years from the time of diagnosis because of their strong resistance to conventional therapies, which include surgery, chemotherapy and irradiation. Although until recently chemotherapy for GBMs did not substantially improve disease outcome when combined with other treatment methods, a randomized clinical trial showed that combination therapy using temozolomide and radiation is superior to radiation therapy Glioma: Immunotherapeutic Approaches, edited by Ryuya Yamanaka.©2012 Landes Bioscience and Springer Science+Business Media.23 MOLECULAR BIOLOGY OF GLIOMAalone.1 The median survival of the patients who received this combination treatment was 14.6 months.1 This is 2.5 months longer than the median survival time of patients who received radiation therapy alone. Therefore, combination therapy using temozolomide and radiation is now widely used worldwide; however, physicians keep seeking for a more effective treatment for GBMs. An understanding of the genetic and epigenetic alterations and resulting aberrant molecular pathways leading to the initiation and progression of GBMs is essential and indispensable for the identification of new therapeutic strategies for GBM.PRIMARY AND SECONDARY GBMsGBMs have been subdivided into primary or secondary GBM subtypes on the basis of their clinical presentation. Primary GBMs are generally found in older patients, while secondary GBMs are found in younger patients (below the age of 45). Primary GBMs are not associated with prior symptoms or with evidence of detectable antecedent lower grade tumors. In contrast, secondary GBMs arise from lower grade gliomas within 5-10 years of diagnosis.2 Although they have distinctive clinical histories, primary and secondary GBMs exhibit features that are histologically similar and are generally associated with an equally poor prognosis.Several genes, such as Tumor protein 53 (TP53), p16Ink4a, Phosphatase and Tensin Homolog Deleted from Chromosome 10 (PTEN) and Epidermal Growth Factor Receptor (EGFR), are altered in both primary and secondary GBMs.3-7 These alterations are thought to occur in a sequential order during the malignant progression of gliomas (Fig. 1). Loss or mutation of the PTEN tumor suppressor gene and amplification or overexpression of the EGFR oncogene are thought to be a characteristic of primary GBMs, whereas mutation of the TP53 tumor suppressor gene appears to be an early event during the development of secondary GBMs8-10 (Fig. 1). Most importantly, abnormal activation of growth factor signaling and dysregulation of cell cycle checkpoints are found in both types of GBMs (Figs. 1 and 2).ACTIVATED GROWTH FACTOR SIGNALING PATHWAYSVarious growth factors, such as platelet-derived growth factor (PDGF)-A, PDGF-B, epidermal growth factor (EGF), transforming growth factor (TGF)-_, insulin growth factor (IGF)-1 and fibroblast growth factor (FGF) are often produced and secreted in GBMs. In addition, the various growth factor receptors corresponding to each specific growth factor, such as EGFR, PDGFR-A, PDGFR-B and FGFR2, are often overexpressed in GBMs. Consequently, the autocrine/paracrine loops between ligand and receptor enhance their impact in gliomagenesis.3,9,11-13Several important signaling pathways, such as Ras-mitogen-activated protein kinase (MAPK), phosphoinositide 3-kinase (PI3K)—AKT and Phospholipase-C gamma (PLC a)—Protein Kinase C (PKC) are activated downstream of growth factor receptors. Recent reports using genetically engineered mice suggest that these growth factor signaling pathways cause the formation of gliomas directly. Weissenberger et al reported that the astrocyte-specific expression of v-src, which is one of the oncogenes expressed in GBMs, causes the formation of gliomas.14 In their experiments, abnormal astrogliosis was found in4GLIOMA: IMMUNOTHERAPEUTIC APPROACHES all the astrocyte-specific v-src expressing mice; however, only 14.4% of mice developed astrocytomas within 65 weeks. Therefore, the authors speculated that factors other than v-src are needed for the malignant transformation of astrocytes.DYSREGULATION OF CELL CYCLE CHECKPOINTSThe eukaryotic cell cycle is composed of four distinct phases: G1 phase, S phase (DNA synthesis), G2 phase and M phase (mitosis). Through the cell cycle, the genetic materials are correctly replicated and transmitted to the two daughter cells. Accurate regulation of the cell cycle is crucial for the survival of cells. Each phase has to progress in a sequential fashion and, to activate each phase, proper progression and completion of the previous phase is required 15-17 (Fig. 2).The engine of the cell cycle is the cyclin—cyclin-dependent kinase (CDK) complex. Cyclins constitute the regulatory subunit and CDK is the catalytic subunit of an activated heterodimeric kinase complex. CDKs are inactive in the absence of cyclin. When activated, CDKs bound to cyclin catalyze protein phosphorylation reactions that mediate the coordination of the progression of the cell cycle.15-17 When cells receive promitotic extracellular signals, G1 cyclin—CDK complexes, such as cyclinD/cdk4,6, are activated and promote the expression of transcription factors, which leads to the increase of S-phase cyclin expression.15-17The disruption of cell cycle checkpoints caused by the dysregulation of major tumor suppressors appears to be crucial for gliomagenesis. Firstly, the TP53 gene is either mutated or deleted in certain gliomas, particularly in secondary GBMs.3-5Secondly, the Figure 1. Distinct molecular alterations between primary and secondary GBMs. Molecular alterations during the malignant progression of gliomas between primary GBMs and secondary GBMs are different. It should be noted that amplification or overexpression of EGFR is more frequently found in primary GBMs whereas mutation of TP53 is an early event in secondary GBMs. MDM2: murine double minute 2, LOH : loss of heterozygosity, RB: retinoblastoma, DCC: deleted in colorectal carcinoma.5MOLECULAR BIOLOGY OF GLIOMA retinoblastoma (RB ) gene is mutated in 10-25% of high-grade astrocytomas. In addition, RB is functionally silenced in another 15% of astrocytomas via the amplification of its antagonist CDK4.3,5,18 Lastly, the CDKN2A locus encoding p16INK4A and p14Arf (which positively regulate the RB and TP53 pathways, respectively) is deleted in approximately half of high-grade astrocytomas.3,5,6 Similar effects on the RB and TP53 pathways are also frequently found in anaplastic oligodendrogliomas.19ISOCITRATE DEHYDROGENASE 1 (IDH1) AND IDH2 MUTATIONS IN GBMsRecently, a somatic mutation at amino acid 132 of the isocitrate dehydrogenase 1 (IDH1) protein was identified in more than 70% of WHO Grade II and III astrocytomas, oligodendrogliomas and secondary GBMs.5,20 Surprisingly, tumors expressing an intact IDH1 gene often have mutations at the analogous amino acid 172 of the IDH2 protein. Interestingly, the R132 residue of IDH1 and the R172 residue of IDH2 are located in the active site of these enzymes and form hydrogen bonds with the isocitrate substrate.21A recent report showed that forced expression of a mutated IDH1 gene in cultured cells results in the reduction of the enzyme product _-ketoglutarate (_-KG), which leads to the increase of the levels of the hypoxia-induced factor-1_ (HIF-1_), the stability of which is regulated by _-KG.22 Because HIF-1_ is thought to promote tumor growth at low oxygen levels, mutation of the IDH1 gene may facilitate tumor progression in GBMs.It should be noted that mutations in the IDH1 or IDH2 genes are identified in the vast majority of WHO Grade II or III gliomas and secondary GBMs.20,23 It has been pointed out that approximately 80% of anaplastic astrocytomas and GBMs with mutated IDH1Figure 2. RB and TP53 pathways controlling G1-S transition. Phosphorylation of RB by CDK4,6/CyclinD activated by mitogenic signals leads to the activation of transcription factor E2F, which results in the activation of S phase genes. Due to the inactivation of tumor suppressors such as TP53 and RB, G1-S transition is dysregulated in GBMs.6GLIOMA: IMMUNOTHERAPEUTIC APPROACHES or IDH2 genes also express mutated TP53, whereas only 3% express mutated PTEN, EGFR, or p16Ink4a.20 On the other hand, only 18% of anaplastic astrocytomas and GBMs with intact IDH1 or IDH2 genes express mutated TP53; however, more than 70% of these tumors exhibit mutations in PTEN, EGFR, or p16Ink4a.20 These findings suggest that gliomas with mutated IDH genes are genetically different from gliomas with intact IDH genes.ABERRANT FUNCTIONS OF MicroRNAS IN GBMIt has been speculated that at least one-third of all protein-coding genes in the human genome are partially regulated by microRNAs.24 MicroRNAs, which are single-stranded RNA molecules that comprise approximately 20-22 nucleotides, have been shown to play a central role in biological processes and distinct patterns of microRNA expression are seen in different normal cells and tissues.25-27MicroRNAs are noncoding RNAs from various chromosomal origins, which include intergenic, intronic, or exonic regions. The primary transcripts of microRNAs (Pri-miR) are transcribed from the genome by RNA polymerase II and are processed through a conserved mechanism to yield mature microRNAs (Fig. 3). Mature microRNAs recognize partially complementary sequences in the 3v—UTR of their target genes and decrease the mRNA levels of the target by cleavage or by inhibiting translation initiation directly.28 Other reported mechanisms of microRNA-mediated translational repression are the destabilization of mRNA via deadenylation of the poly-A tail and recruitment of theFigure 3. Maturation of microRNAs in normal cells. The primary transcripts of microRNAs (Pri-miR) are transcribed from the genome by RNA polymerase II and are processed through a conserved mechanism to yield mature microRNAs. Initially, the hairpin—loop sequences in Pri-miRs are cleaved by a nuclear protein complex that includes Drosha, which is a Type III RNase and Pre-miRs 60-80 base pairs in length are formed. After being exported to the cytoplasm by the Exportin 5 mediated mechanisms, Pre-miRs are further processed by a protein complex containing Dicer, a Type III RNase, which leads to the generation of mature microRNAs. The mature microRNAs are then loaded onto the RNA-induced silencing complex (RISC) and target mRNAs, thus reducing protein translation.7 MOLECULAR BIOLOGY OF GLIOMAribosome inhibitory protein eLF6.29,30 Surprisingly, it has also been reported that certain types of microRNAs upregulate the translation of their target genes.31 These reports on the functions of microRNAs suggest that microRNAs have hundreds of potential targets and regulate multiple signaling pathways.24Various levels of microRNA expression have been observed in cancers, including GBMs.32-34 It has been reported that microRNA-21 (miR-21) is highly expressed in GBMs.35 Inhibition of miR-21 in glioma cell lines results in the increase of caspase-dependent apoptosis, which suggests that miR-21 acts as an oncogene by inhibiting apoptosis.35 MiR-21 has also been reported to promote the cellular invasion by directly downregulating metalloprotease inhibitors, such as tissue inhibitor of metalloproteinases-3 (TIMP3) and Reversion-inducing cysteine-rich protein with Kazal motifs (RECK).36 A computer-based analysis suggested that miR-21 targets a network of TP53, TGF-` and mitochondrial apoptosis factors in GBMs.37 MiR-7 is downregulated in GBMs and targets EGFR.38 MiR-7 has also been reported to target insulin receptor substrate-2 (IRS-2), which is an adaptor protein that mediates PI3K signaling downstream of receptor tyrosine kinases and suppresses AKT activation.38 These facts suggest that the oncogenes responsible for GBM formation may be regulated by microRNAs. MiR-124 and miR-137, which target CDK6, are also downregulated in GBMs when compared with normal brain tissues.39 MiR-128 is downregulated in GBMs and targets B-cell-specific Moloney murine leukemia virus integration site 1 (Bim-1), which is thought to be one of the oncogenes responsible for GBM formation.36DRUG RESISTANCE IN GBMDrug resistance in GBM is one of the major factors that render this type of tumor incurable. EGFR is frequently amplified, overexpressed, or mutated in GBMs.9,11-13 An activated form of EGFR mutant caused by the deletion of exons 2-7 (known as EGFRvIII) occurs in 20-30% of GBMs; EGFRvIII reduces apoptosis and increases the proliferation of GBM cells. In addition, EGFRvIII has the capacity to malignantly transform murine Ink4a/Arf null neural stem cells (NSCs) or astrocytes.40,41 Therefore, EGFR is a strongly validated molecular target for GBM treatment and the use of EGFR inhibitors is a reasonable therapeutic strategy for this disease. However, only 10-20% of GBM patients respond to EGFR inhibitors.42 Although the mechanism of the resistance of GBMs to EGFR inhibitors is not known, the loss of PTEN is suggested to be associated with this mechanism. PTEN is commonly lost in GBMs and is an inhibitor of the PI3K signaling pathway, which is activated downstream of EGFR. It is suggested that loss of PTEN may dissociate EGFR inhibition from the PI3K pathway inhibition, which may result in the resistance of GBMs to EGFR inhibitors.42Alkylating agents, which include the chloroethylnitrosoureas (nimustine (ACNU), carmustine (BCNU) and lomustine (CCNU)), procarbazine and temozolomide, are commonly used for the treatment of GBMs.43-45 These drugs cause DNA damage by adding alkyl groups to DNA (i.e., alkylation) and induce apoptosis in tumor cells. O6-methylguanine-methyltransferase (MGMT) is a DNA repair protein that can reverse alkylation at the O6 position of guanine. The neutralization of the cytotoxic effects of alkylating agents by the MGMT has been demonstrated in cell lines, in xenograft models and in an MGMT transgenic mouse.46-518GLIOMA: IMMUNOTHERAPEUTIC APPROACHES High levels of MGMT activity in tumor tissues are associated with resistance to alkylating agents and the extent of methylation of the promoter region of the MGMT gene, which results in the decreased expression of MGMT in GBMs, correlates with the outcome of treatment with alkylating agents.52-54Despite the information described above, there has been no significant improvement in overcoming the drug resistance in GBM for several decades. Therefore, the drug resistance mechanism in GBMs still warrants investigation.CONCLUSIONAs described above, there is accumulating information regarding the molecular mechanisms responsible for the initiation and progression of GBMs (Fig. 1); however, the prognosis of GBMs remains dismal. Some crucial questions remain unsolved. For example, it is well known that the loss of chromosomes 1p and 19q is frequently associated with the prognosis of the patients with oligodendrogliomas.55 The deletion of chromosome 1p is most commonly combined with loss of chromosome 19q (“whole arm” deletion). Oligodendrogliomas carrying the codeletion of 1p/19q are sensitive to both chemotherapy and radiotherapy, which leads to a better prognosis in patients with this type of tumor.56,57 However, this genetic abnormality has not been modeled in mice, which is partly because of the lack of knowledge regarding the precise genes involved in these regions.55 The elucidation of the molecular mechanisms leading to the high sensitivity of tumors lacking chromosomes 1p/19q to chemotherapy and radiotherapy may provide new and improved therapies for oligodendrogliomas.Another crucial question that remains unsolved is the cellular origin of GBMs. It is now believed that brain tumors, including GBMs, arise from brain tumor stem/ initiating cells (BTSCs).58,59 BTSCs have been identified as a rare population of cells in tumor tissues and have a robust tumor initiating capability when injected into immunocompromised mice.58,59 BTSCs are positive for CD133, which is a marker of stem or progenitor cells in various organs.60,61 CD133 cells from GBMs have the ability to form neurospheres in vitro and as few as 100 CD133 cells are able to form tumors in immunocompromised mice, while most human GBM cell lines require 105 to 106 cells for tumor engraftment and formation.59,62,63 Several reports from different laboratories support this contention; however, more recent evidence questions the reliability of CD133 as a marker for BTSCs. Wang et al showed that transplantation of CD133< cells from human GBMs results in the formation of tumors in nude rats and that those cells give rise to CD133 cells.58,64 Another report indicates that the expression of CD133 in GBMs may just reflect a response of the tumor cells to environmental stress.65 Therefore, it may be necessary to identify additional markers that more consistently represent BTSCs.As described above, knowledge on the molecular signaling pathways responsible for GBM formation and the BTSC hypothesis have significantly advanced our understanding of CNS malignancies. H owever, as several key questions remain unanswered, the continuous efforts of researchers and physicians are necessary to improve the management of GBMs.MOLECULAR BIOLOGY OF GLIOMA9 REFERENCES1. Stupp R, Mason WP, van den Bent MJ et al. Radiotherapy plus concomitant and adjuvant temozolomidefor glioblastoma. N Engl J Med 2005; 352:987-996.2. Kleihues P, Ohgaki H. Primary and secondary glioblastomas: from concept to clinical diagnosis. NeuroOncol 1999; 1:44-51.3. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature2008; 455:1061-1068.4. Louis DN. The p53 gene and protein in human brain tumors. J Neuropathol Exp Neurol 1994; 53:11-21.5. Parsons DW, Jones S, Zhang X et al. An integrated genomic analysis of human glioblastoma multiforme.Science 2008; 321:1807-1812.6. Ueki K, Ono Y, Henson JW et al. CDKN2/p16 or RB alterations occur in the majority of glioblastomas andare inversely correlated. Cancer Res 1996; 56:150-153.7. Wong AJ, Bigner SH, Bigner DD et al. Increased expression of the epidermal growth factor receptor genein malignant gliomas is invariably associated with gene amplification. Proc Natl Acad Sci USA 1987;84:6899-6903.8. Ohgaki H, Kleihues P. Genetic pathways to primary and secondary glioblastoma. Am J Pathol 2007;170:1445-1453.9. Furnari FB, Fenton T, Bachoo RM et al. Malignant astrocytic glioma: genetics, biology and paths to treatment.Genes Dev 2007; 21:2683-2710.10. Weber RG, Sabel M, Reifenberger J et al. Characterization of genomic alterations associated with gliomaprogression by comparative genomic hybridization. Oncogene 1996; 13:983-994.11. Huse JT, Holland EC. Genetically engineered mouse models of brain cancer and the promise of preclinicaltesting. Brain Pathol 2009; 19:132-143.12. Di Rocco F, Carroll RS, Zhang J et al. Platelet-derived growth factor and its receptor expression in humanoligodendrogliomas. Neurosurgery 1998; 42:341-346.13. Hesselager G, Holland EC. Using mice to decipher the molecular genetics of brain tumors. Neurosurgery2003; 53:685-694; discussion 95.14. Weissenberger J, Steinbach JP, Malin G et al. Development and malignant progression of astrocytomas inGFAP-v-src transgenic mice. Oncogene 1997; 14:2005-2013.15. Hartwell L. Defects in a cell cycle checkpoint may be responsible for the genomic instability of cancercells. Cell 1992; 71:543-546.16. Nurse P. Checkpoint pathways come of age. Cell 1997; 91:865-867.17. Kastan MB, Bartek J. Cell-cycle checkpoints and cancer. Nature 2004; 432:316-323.18. Henson JW, Schnitker BL, Correa KM et al. The retinoblastoma gene is involved in malignant progressionof astrocytomas. Ann Neurol 1994; 36:714-721.19. Watanabe T, Yokoo H, Yokoo M et al. Concurrent inactivation of RB1 and TP53 pathways in anaplasticoligodendrogliomas. J Neuropathol Exp Neurol 2001; 60:1181-1189.20. Yan H, Parsons DW, Jin G et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med 2009; 360:765-773.21. Xu X, Zhao J, Xu Z et al. Structures of human cytosolic NADP-dependent isocitrate dehydrogenase reveala novel self-regulatory mechanism of activity. J Biol Chem 2004; 279:33946-33957.22. Zhao S, Lin Y, Xu W et al. Glioma-derived mutations in IDH1 dominantly inhibit IDH1 catalytic activityand induce HIF-1alpha. Science 2009; 324:261-265.23. Watanabe T, Nobusawa S, Kleihues P et al. IDH1 mutations are early events in the development ofastrocytomas and oligodendrogliomas. Am J Pathol 2009; 174:1149-1153.24. Lim LP, Lau NC, Garrett-Engele P et al. Microarray analysis shows that some microRNAs downregulatelarge numbers of target mRNAs. Nature 2005; 433:769-773.25. Lee RC, Feinbaum RL, Ambros V. The C. elegans heterochronic gene lin-4 encodes small RNAs withantisense complementarity to lin-14. Cell 1993; 75:843-854.26. Wightman B, Ha I, Ruvkun G. Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4mediates temporal pattern formation in C. elegans. Cell 1993; 75:855-862.27. Baulcombe D. DNA events. An RNA microcosm. Science 2002; 297:2002-2003.28. Doench JG, Sharp PA. Specificity of microRNA target selection in translational repression. Genes Dev2004; 18:504-511.29. Wu L, Fan J, Belasco JG. MicroRNAs direct rapid deadenylation of mRNA. Proc Natl Acad Sci USA2006; 103:4034-9.30. Chendrimada TP, Finn KJ, Ji X et al. MicroRNA silencing through RISC recruitment of eIF6. Nature2007; 447:823-838.10GLIOMA: IMMUNOTHERAPEUTIC APPROACHES 31. Vasudevan S, Tong Y, Steitz JA. Switching from repression to activation: microRNAs can up-regulatetranslation. Science 2007; 318:1931-1934.32. Volinia S, Calin GA, Liu CG et al. A microRNA expression signature of human solid tumors defines cancergene targets. Proc Natl Acad Sci USA 2006; 103:2257-2261.33. Calin GA, Ferracin M, Cimmino A et al. A MicroRNA signature associated with prognosis and progressionin chronic lymphocytic leukemia. N Engl J Med 2005; 353:1793-1801.34. Lawler S, Chiocca EA. Emerging functions of microRNAs in glioblastoma. J Neurooncol 2009; 92:297-306.35. Corsten MF, Miranda R, Kasmieh R et al. MicroRNA-21 knockdown disrupts glioma growth in vivo anddisplays synergistic cytotoxicity with neural precursor cell delivered S-TRAIL in human gliomas. Cancer Res 2007; 67:8994-9000.36. Godlewski J, Nowicki MO, Bronisz A et al. Targeting of the Bmi-1 oncogene/stem cell renewal factor bymicroRNA-128 inhibits glioma proliferation and self-renewal. Cancer Res 2008; 68:9125-9130.37. Papagiannakopoulos T, Shapiro A, Kosik KS. MicroRNA-21 targets a network of key tumor-suppressivepathways in glioblastoma cells. Cancer Res 2008; 68:8164-8172.38. Kefas B, Godlewski J, Comeau L et al. microRNA-7 inhibits the epidermal growth factor receptor and theAkt pathway and is down-regulated in glioblastoma. Cancer Res 2008; 68:3566-3572.39. Silber J, Lim DA, Petritsch C et al. miR-124 and miR-137 inhibit proliferation of glioblastoma multiformecells and induce differentiation of brain tumor stem cells. BMC Med 2008; 6:14.40. H olland EC, H ively WP, DePinho RA et al. A constitutively active epidermal growth factor receptorcooperates with disruption of G1 cell-cycle arrest pathways to induce glioma-like lesions in mice. Genes Dev 1998; 12:3675-3685.41. Bachoo RM, Maher EA, Ligon KL et al. Epidermal growth factor receptor and Ink4a/Arf: convergentmechanisms governing terminal differentiation and transformation along the neural stem cell to astrocyte axis. Cancer Cell 2002; 1:269-277.42. Mellinghoff IK, Wang MY, Vivanco I et al. Molecular determinants of the response of glioblastomas toEGFR kinase inhibitors. N Engl J Med 2005; 353:2012-2024.43. Kreisl TN. Chemotherapy for malignant gliomas. Semin Radiat Oncol 2009; 19:150-154.44. Sarkaria JN, Kitange GJ, James CD et al. Mechanisms of chemoresistance to alkylating agents inmalignant glioma. Clin Cancer Res 2008; 14:2900-2908.45. Engelhard HH. The role of interstitial BCNU chemotherapy in the treatment of malignant glioma. SurgNeurol 2000; 53:458-464.46. Dumenco LL, Allay E, Norton K et al. The prevention of thymic lymphomas in transgenic mice by humanO6-alkylguanine-DNA alkyltransferase. Science 1993; 259:219-222.47. Gerson SL. Clinical relevance of MGMT in the treatment of cancer. J Clin Oncol 2002; 20:2388-2399.48. Kaina B, Fritz G, Mitra S et al. Transfection and expression of human O6-methylguanine-DNAmethyltransferase (MGMT) cDNA in Chinese hamster cells: the role of MGMT in protection against the genotoxic effects of alkylating agents. Carcinogenesis 1991; 12:1857-1867.49. Baer JC, Freeman AA, Newlands ES et al. Depletion of O6-alkylguanine-DNA alkyltransferase correlates withpotentiation of temozolomide and CCNU toxicity in human tumour cells. Br J Cancer 1993; 67:1299-1302.50. Redmond SM, Joncourt F, Buser K et al. Assessment of P-glycoprotein, glutathione-based detoxifyingenzymes and O6-alkylguanine-DNA alkyltransferase as potential indicators of constitutive drug resistance in human colorectal tumors. Cancer Res 1991; 51:2092-2097.51. Hermisson M, Klumpp A, Wick W et al. O6-methylguanine DNA methyltransferase and p53 status predicttemozolomide sensitivity in human malignant glioma cells. J Neurochem 2006; 96:766-776.52. Watts GS, Pieper RO, Costello JF et al. Methylation of discrete regions of the O6-methylguanine DNAmethyltransferase (MGMT) CpG island is associated with heterochromatinization of the MGMT transcription start site and silencing of the gene. Mol Cell Biol 1997; 17:5612-5619.53. H egi ME, Diserens AC, Godard S et al. Clinical trial substantiates the predictive value ofO-6-methylguanine-DNA methyltransferase promoter methylation in glioblastoma patients treated with temozolomide. Clin Cancer Res 2004; 10:1871-1874.54. H egi ME, Diserens AC, Gorlia T et al. MGMT gene silencing and benefit from temozolomide in glioblastoma.N Engl J Med 2005; 352:997-1003.55. Reifenberger J, Reifenberger G, Liu L et al. Molecular genetic analysis of oligodendroglial tumors showspreferential allelic deletions on 19q and 1p. Am J Pathol 1994; 145:1175-1190.56. Cairncross JG, Ueki K, Zlatescu MC et al. Specific genetic predictors of chemotherapeutic response andsurvival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998; 90:1473-1479.57. Cairncross G, Berkey B, Shaw E et al. Phase III trial of chemotherapy plus radiotherapy compared withradiotherapy alone for pure and mixed anaplastic oligodendroglioma: Intergroup Radiation Therapy Oncology Group Trial 9402. J Clin Oncol 2006; 24:2707-2714.。