Chromosome arm 1q gain associated with good response to chemotherapy in a malignant glioma

Transcription

1 J Neurosurg 106: , 2007 Chromosome arm 1q gain associated with good response to chemotherapy in a malignant glioma Case report SATOSHI TAKAHASHI, M.D., YUICHI HIROSE, M.D., PH.D., EIJI IKEDA, M.D., PH.D., RAITA FUKAYA, M.D., AND TAKESHI KAWASE, M.D., PH.D. Departments of Neurosurgery and Pathology, Keio University School of Medicine, Tokyo, Japan The authors describe the case of a patient with a glioblastoma multiforme who showed remarkably good response to chemotherapy. A genetic analysis using comparative genomic hybridization (CGH) revealed that the tumor had a gain on the q arm of chromosome 1 (1q). Using CGH for a series of genetic analyses of more than 180 patients with gliomas, six were found to have a demonstrated 1q gain. Although the tumors in all six of these cases were histopathologically diagnosed as high-grade gliomas, compared with other malignant gliomas they demonstrated a good prognosis because of their favorable chemotherapeutic sensitivity. In immunohistochemical tests, most of the tumor cells in these cases were negative for O 6 -methylguanine DNA methyltransferase, which antagonizes the effect of DNA-alkylating chemotherapeutic agents. The authors believed that a gain of 1q could be produced through the genetic events that cause loss of 1p, because these chromosomal aberrations have an imbalance of DNA copy number in common (1p 1q). A gain of 1q is an infrequent chromosomal aberration and its clinical importance should be investigated in a larger study; however, patients with malignant gliomas demonstrating a 1q gain possibly show longer survival and good response to chemotherapy similar to patients with tumors demonstrating 1p loss. The importance of using genetic analysis for gliomas is emphasized in this report because it may help in selecting cases responsive to chemotherapy and because appropriate treatment for these patients will lead to progress in the treatment of malignant gliomas. KEY WORDS glioblastoma multiforme comparative genomic hybridization chromosome arm 1q chemotherapy D Abbreviations used in this paper: CGH = comparative genomic hybridization; DIG dutp = digooxygenin deoxyuridine 5 triphosphate; DOP-PCR = degenerate oligonucleotide-primed polymerase chain reaction; FISH = fluorescence in situ hybridization; GBM = glioblastoma multiforme; LOH = loss of heterozygosity; MGMT = O 6 -methylguanine DNA methyltransferase; MR = magnetic resonance; PAV = procarbazine, nimustine hydrochloride, and vincristine. ESPITE current therapeutic modalities including surgery, chemotherapy, and radiotherapy, only limited progress has been achieved in the treatment of GBM. In most patients, GBM leads to death within approximately 12 months, 26 but authors of some reports have described the clinical features of patients with gliomas who survived for a longer duration. 19,27,28 Genetic analyses of tumors have been developed, and correlations between genetic subgrouping and clinical information for gliomas have been investigated. These investigations show that allelic loss on chromosomal arms 1p and 19q serves as a powerful predictor of both sensitivity to chemotherapy and prolonged survival in patients with oligodendrogliomas. 5,29 According to recent reports, this predictor also occurs in patients with astrocytomas, although less commonly. 9,12 Identification of genetic aberrations closely associated with good prognosis and response to chemotherapy appears to be important, as this information can assist in the selection of appropriate treatment for patients. Comparative genomic hybridization is a cytogenetic analysis method that detects and maps relative DNA-sequence copy number aberrations on each chromosome. 13,25 Authors of previous studies have demonstrated that CGH can be used to divide tumors of the same histopathological entity into clinically relevant subgroups, and several reports on genetic prognostic markers of gliomas using CGH have been published. 2,4,15,31 More recently, genetic analysis with array-based CGH has been conducted in search of specific genetic prognostic markers of gliomas. 6,11,17,21 In this report, we validate the possibility of prolonged survival and good response to chemotherapy in patients with malignant gliomas demonstrating a q-arm copy gain of chromosome J. Neurosurg. / Volume 106 / March, 2007

2 Chromosome arm 1q gain and chemotherapy response in a GBM Case Report Examination and Operation. This 79-year-old man with a history of aortic valve regurgitation complained of numbness in his right leg, and 3 weeks later motor weakness of the right arm and mild motor aphasia developed. He was admitted to our hospital, where computed tomography scans revealed a heterogeneous mass in his left frontal lobe (Fig. 1A). His vital signs were stable, and his consciousness level was graded as 13 (E4V3M6) on the Glasgow Coma Scale. Radiography of his chest did not show a lung lesion. Magnetic resonance imaging with intravenous infusion of gadolinium showed that the mass in his left frontal lobe enhanced heterogeneously, suggesting a diagnosis of malignant glioma (Fig. 1B and C). After admission, the patient s level of consciousness progressively deteriorated, and he underwent a left frontal craniotomy to remove the tumor. The tumor was found to be soft and moderately hemorrhagic, and was not well demarcated from the surrounding brain tissue in the deep portion. The tumor mass was gross totally resected with the aid of a surgical microscope. Histopathological Findings. Electron microscopy revealed the surgical specimen to be a hypercellular glial tumor. A prominent population of tumor cells demonstrated mitotic figures. Vascular proliferation and necrosis were also identified (Fig. 2). Subpopulations of tumor cells were positive for glial fibrillary acidic protein and S100 protein. The DNA repair protein MGMT antagonizes the effect of DNAalkylating chemotherapeutic agents; 10,22,23 the proportion of tumor cells positive for MGMT in relation to the total number of cells (the MGMT score) was only 1.7% in this sample. More than 40% of the neoplastic cells expressed proliferation-associated nuclear antigen Ki 67 (using MIB-1 staining). The histopathological diagnosis of the tumor was a GBM. The CGH Analysis. The CGH analysis was performed according to a protocol described elsewhere. 8 Briefly, tumor DNA was extracted from a microdissected piece of formalin-fixed and paraffin-embedded tissue section (5 m thick) and amplified using DOP-PCR. To select the region of truly neoplastic tissue, tissue microdissection was performed after referring to the pathological findings of consecutive tissue sections that were stained with H & E or MIB-1; DNA was labeled with another DOP-PCR reaction using DIG dutp (Roche), and the reference DNA was amplified from 50 ng normal male DNA and labeled as described earlier except that rhodamine-dutp (Roche) was used instead of DIG dutp. The probe mixture was denatured and hybridized to normal metaphase spreads (Vysis, Inc.). After unhybridized probes were washed, the metaphase spread was incubated using fluorescein isothiocyanate-conjugated anti- DIG-antibody (Roche). The preparations were washed and counterstained using 4,6-diamino-2-phenylinodole in antifade solution. Red, green, and blue images were acquired and the ratios of fluorescence intensity along the chromosomes were quantitated using the CytoVision analysis system (Applied Imaging Co.). The CGH profile is shown in Fig. 3. Chromosomal copy number aberrations (gains [ ] and losses [ ]) that were detected using CGH analysis were as follows: 1q, 2q21-32, 3, 7, 8q21.2-ter, 9pter-23, 9p-22, 10q22-ter, 12, 13, 14, 15q24-ter, 16q , and 18q22- ter. Postoperative Course. Although MR imaging performed 19 days after surgery showed a small residual tumor (Fig. 4B), another MR image obtained 3 weeks later showed remarkable progression of the tumor (Fig. 4C). The CGH analysis of the tumor found a 1p LOH pattern similar to that seen in oligodendrogliomas (although the actual chromosomal abnormality was distinct), and the immunohistochemical test for MGMT revealed that the tumor did not express a high level of the enzyme. For these reasons, chemotherapy using PAV was started 3 days after the second MR image was obtained. The consciousness of the patient improved gradually after administration of chemotherapeutic agents, and an MR image obtained on Day 25 of PAV showed remarkable reduction of the tumor (Fig. 4D). Furthermore, the tumor disappeared almost completely 7 days after the end of the first PAV course (Fig. 4E). Two months later, however, tumor development was recognized (Fig. 4F), and the second course of PAV was started in combination with local radiotherapy (60 Gy). The tumor then disappeared again (Fig. 4G and H). The patient remained free from further tumor development and was discharged from our hospital. Eight months after the initial surgery, he died of heart failure at another hospital. His last MR image obtained 10 days before his death revealed no sign of tumor recurrence. Summary of Additional Cases We have analyzed the genetic profiles of more than 180 FIG. 1. Computed tomography scan (A) and MR images (B and C) obtained in a 79-year-old man at admission. A: A heterogeneous mass in the left frontal lobe is evident. B and C: Axial (B) and sagittal (C) Gd-enhanced images demonstrating a heterogeneously enhanced brain tumor with corpus callosum involvement, suggesting a diagnosis of malignant glioma. J. Neurosurg. / Volume 106 / March,

3 S. Takahashi et al. FIG. 2. Photomicrographs of tumor specimens obtained during the operation demonstrating the histopathogical features of a GBM. A and B: The tumor cells showed pseudopalisading necrosis and vascular proliferation (A) and highly mitotic figures (B). C: More than 40% of the neoplastic cells expressed the proliferation-associated nuclear antigen Ki 67 after MIB-1 staining. D: Tumor cells negative for MGMT. H & E (A and B); MIB-1 staining (C); and MGMT staining (D). Original magnifications 100 (A) and 400 (B D). glioma tissues using CGH (unpublished data), and six of these cases were found to have a copy gain of the q arm of chromosome 1 ( 1q). Clinical and genetic features of these six cases are summarized in Table 1. In general, these patients showed a tendency to have a favorable clinical course. In Case 2, the patient s family did not allow her to undergo any adjuvant therapy because of her age. In the other five cases, chemotherapy and radiotherapy were performed. Because our CGH analysis was performed retrospectively, genetic information was obtained after the initial treatment for these patients (except for Case 6 described earlier), and did not help in determining the chemotherapeutic regimen. Immunostaining of tumor cells to detect MGMT was also performed retrospectively. All of the tumors with 1q gain contained a small number of MGMT-positive cells (MGMT scores were less than 15% in all cases), suggesting a good response to chemotherapy (Table 1). Case 1. This 53-year-old woman underwent a left frontal partial lobectomy, and a GBM tumor was diagnosed. Adjuvant chemotherapy using ranimustine and vincristine was administered as well as 60-Gy whole-brain radiotherapy. She did not undergo additional chemotherapy, and the tumor recurred within 8 months after surgery. She died of the disease 10 months after surgery. Case 2. This 77-year-old woman had a history of surgery for removal of a right frontal tumor, which was diagnosed as an oligodendroglioma. She developed a tumor again after remaining disease free for 7 years and underwent a second craniotomy. This time the tumor was diagnosed as an anaplastic oligoastrocytoma, and we performed a CGH analysis on the tumor sample. We could not obtain a CGH profile from the first tumor because of the age of the sample. Instead, FISH was performed using 1p36 and 1q25 probes according to the method described previously. 29 The FISH analysis confirmed 1p loss in 61.5% of the cells of the primary tumor, and this tumor was genetically diagnosed as a tumor with 1p loss. Magnetic resonance images obtained 3 months after the second surgery showed no sign of tumor recurrence, but another MR image obtained 1 year after surgery showed remarkable tumor regrowth. Her family decided that, due to her age, she should not undergo any adjuvant therapy. She died of the disease 17 months postoperatively. Case 3. This 14-year-old boy underwent a right frontal partial lobectomy for a relatively well-circumscribed tumor that was demonstrated using MR imaging with gadolinium enhancement. The tumor was histologically diagnosed as a GBM with nuclear atypism, mitosis, and vascular proliferation. Adjuvant therapy included chemotherapy with ranimustine and 60-Gy whole-brain radiotherapy. After 6 years, he remains free from recurrence. He did not undergo any additional chemotherapy after the initial cycle. In this case, the tumor was relatively well circumscribed, and total resection of the tumor might have led to a good outcome. Case 4. This 47-year-old woman underwent radiotherapy alone as adjuvant therapy, based on the patient s own preference. The tumor was diagnosed as anaplastic astrocytoma, and reduction of the residual tumor after radiotherapy was confirmed using MR imaging. Tumor regrowth was not recognized on neuroimages for 27 months until MR imaging revealed recurrence. After recognition of tumor recurrence, the patient refused to undergo any adjuvant chemotherapy. Case 5. The tumor in this 66-year-old man was diagnosed as a GBM, and was treated with radiotherapy and chemotherapy (ranimustine). Magnetic resonance imaging 490 J. Neurosurg. / Volume 106 / March, 2007

4 Chromosome arm 1q gain and chemotherapy response in a GBM FIG. 3. Graphic profiles of the CGH results for each chromosome of the present case. The chromosome number is shown under the appropriate ideogram. The mean (dark lines) and standard deviation (light lines) of the fluorescence intensity ratios for the indicated chromosomes are shown; n represents the number of metaphase spread from which data were collected. Lines to the left of each chromosome ideogram represent regions of relative loss, and lines to the right represent regions of relative gain. The following chromosomal copy aberrations were detected: 1q, 2q21-32, 3, 7, 8q21.2- ter, 9pter-23, 9p22, 10q22-ter, 12, 13, 14, 15q24-ter, 16q , and 18q22-ter. revealed necrosis within the residual tumor; however, the patient died of interstitial pneumonia 8 months after the surgery without signs of residual tumor. In all cases except Case 6 (the present case), patients received only one cycle of chemotherapy (if any). Discussion J. Neurosurg. / Volume 106 / March, 2007 Cairncross and colleagues 5 reported longer survival and better response to chemotherapy in patients with oligodendroglioma demonstrating chromosomal 1p/19q loss, which has drawn attention to identifying genetic subgroups of malignant brain tumors that show sensitivity to chemotherapy and longer survival. Ino and coworkers 12 reported longer survival times and better therapeutic responses in patients with malignant astrocytic gliomas demonstrating 1p loss. The importance of categorizing malignant gliomas into genetic subgroups for the purpose of predicting their prognosis and chemotherapeutic sensitivity has received increasing attention. Subgrouping malignant gliomas according to genetic aberration could be utilized to classify GBMs into clinically relevant subgroups, and furthermore could help in determining the adjuvant chemotherapeutic regimen. Investigators in previous studies reported on genetic subgrouping of malignant astrocytic gliomas in an effort to establish clinically relevant classifications. 2,4,15,31 In those studies, malignant glioma with a 1q gain is relatively rare; Kunwar et al. 15 found 1q gain in only approximately 8.5% of patients with an anaplastic astrocytoma. The incidence of 1p loss is approximately 10 to 12% in malignant gliomas, 2,12 and longer survival times and better chemotherapeutic sensitivity in patients with gliomas demonstrating 1p loss have been thoroughly investigated. To our knowledge, however, no study has focused on 1q gain in terms of longer survival and better chemotherapeutic sensitivity. The patient in the present case demonstrated a 1q copy gain. We believed that this 1q gain was a similar type of genetic aberration as a 1p loss, because the DNA copy number on 1p was decreased in comparison to that on 1q. We therefore believed that the tumor might show a good response to a chemotherapeutic regimen of PAV (similar to cases with 1p loss). An additional reason we used PAV treatment was because the tumor did not express a high level of MGMT during immunohistochemical tests, which was true with all of the other tumors with a 1q gain. Only one of our patients with a 1q gain (Case 1) did not show a good response to adjuvant therapy even with low MGMT expression; however, this patient underwent chemotherapy with ranimustine alone, and additional chemotherapy might have 491

5 S. Takahashi et al. FIG. 4. Axial Gd-enhanced MR images obtained in the patient in the present case during chemotherapy. A: Image obtained before craniotomy (same as Fig. 1B). B: Image obtained on postoperative Day 19 showing a small residual tumor. C: Image obtained on postoperative Day 40 before PAV treatment showing remarkable progression of the tumor. D and E: Images obtained on Days 25 (D) and 53 (E) of PAV treatment demonstrating remarkable reduction of the tumor. F: Image obtained on postoperative Day 125 before the second course of PAV treatment showing tumor development. G and H: Images obtained 1 (G) and 2 (H) months after the last PAV treatment showing the tumor has disappeared again. been effective, as was seen in the present case (Case 6). Multiple cycles of adjuvant chemotherapy might be needed to control the development of tumors with a 1q copy gain. Investigators in several studies reported a low expression of MGMT in oligodendrogliomas with a 1p loss, 14,16,18 and our analysis suggests that tumors with a 1q gain share this common epigenetic feature of blunting MGMT expression. Our data suggest that malignant gliomas with a 1q gain contain only a small proportion of cells with high MGMT expression and show good chemotherapeutic sensitivity. In our study, no case of a supratentorial malignant glioma with a 1q gain showed a high level of MGMT expression in histopathological tests. Further investigation of this finding is needed, however. Because tumors with a 1q gain and those with a 1p loss share common clinical and genetic features, we hypothesized that these aberrations developed, at least in part, through a common genetic pathway (Fig. 5). In the development of a 1q gain and 1p loss, there must be a chromosomal breakpoint at (or close to) the centromere of chromosome 1, and a gain of the entire chromosome 1 is associated with promotion of a 1q gain while maintaining the imbalance between 1p and 1q. It is noteworthy that in our data, the histopathological diagnosis for all tumors with a 1q gain was malignant glioma, which is expected to have genomic instability compared with low-grade tumors. In such tumors with genetic instability, chromosomal aneuploidy develops frequently, and gain of the entire chromosome 1 can occur relatively easily. Interestingly, in one of our patients with a 1q gain (Case 2) anaplastic oligoastrocytoma was diagnosed, which typically shows a 1p loss; the primary tumor in this patient was a low-grade oligodendroglioma with a 1p loss confirmed using FISH. In this case, it is possible that the 1p loss occurred first, and then a gain of the entire chromosome 1 developed. However, our FISH analysis also showed that approximately 17% of the primary TABLE 1 Summary of the clinical and genetic features of six patients with a 1q copy gain on chromosome 1* Case Age (yrs), FU Radiation Histopathological MGMT No. Sex (mos) Outcome Op Therapy Chemo Finding Score (%) Chromosomal Copy Number Aberrations 1 53, F 10 died PR yes ranimustine GBM 9.4 1q, 5q32 ter, 7, 9p, 10, 12q15-21, 13, 14q21-23, 16q, 19q, 20, 22q 2 77, F 17 died biop no none AOA 3.3 1q, 3q, 4, 7, 9, 11, 12, 14q, 15q, 16, 18, 19q, , M 72 remission TR yes ranimustine GBM q, 2p21-11, 2q , 5p, 7p, 9q13 33, 11p, 19q 4 47, F 27 remission STR yes none AA 3.4 1q, 2, 4, 5, 7, 8, 9q13-33, 10, 14, 21, X 5 66, M 12 died PR yes ranimustine GBM 3.4 1q, 7, 10, 16p, 17q, 19q, , M 8 died STR yes PAV GBM 1.7 (see Case Report) * AA = anaplastic astrocytoma; AOA = anaplastic oligoastrocytoma; biop = biopsy procedure; chemo = chemotherapy; FU = follow up; PR = partial removal; STR = subtotal removal; TR = total removal. Present case. 492 J. Neurosurg. / Volume 106 / March, 2007

6 Chromosome arm 1q gain and chemotherapy response in a GBM tumor cells had a 1p/19q imbalance with more than three copies of 1q25 (data not shown), and it is simultaneously possible that these cells developed due to the secondary (recurrent) tumor. Regardless of the genetic development pathway, our data from this case suggest that tumors with a 1q gain share common characteristics with those with a 1p loss. After his initial surgery, the patient in the present case experienced remarkable tumor progression, with high Ki 67 expression, and we could not prevent fatal progression of the tumor without adjuvant therapy. We do not have data on the radiosensitivity of our cases; however, we believe that the prognosis of the patients is directly related to tumor chemotherapeutic sensitivity. The patient in whom anaplastic oligoastrocytoma was diagnosed who had a 1q gain did not undergo any adjuvant therapy because of her family s decision (Case 2), and she died of tumor progression. None of our cases with a 1q gain expressed a high level of MGMT. Because our CGH analysis was performed retrospectively, genetic information was obtained after the initial treatment of the patients, except for the present case, and did not help determine the chemotherapeutic regimen. If we could have obtained genetic information prior to adjuvant chemotherapy, DNA-alkylating chemotherapeutic agents such as temozolomide 1,7,30 may have been the most suitable for this adjuvant chemotherapy in patients with malignant gliomas and a 1q gain. In this study, we used CGH analysis to examine genetic aberrations in gliomas. This powerful analytic method can reveal information regarding all 23 pairs of chromosomes in a single analysis. It can also utilize paraffin-embedded specimens, and allow us to perform a retrospective analysis. In FISH analysis and LOH analysis, only information on specified chromosomes is available, and these analyses cannot reveal information on all 23 pairs of chromosomes simultaneously. Thus, to identify novel genetic aberrations, CGH analysis is a more useful method than FISH and LOH analyses. Loss of heterozygosity analysis has also been reported to have limitations in the genetic classification of tumors. 20 Because CGH analysis can detect highly amplified sequences greater than 2 to 5 megabases in length, 3 a more 6,11,17, 21,24 sensitive assay system, such as microarray CGH, may help to identify the region of interest. The conventional (metaphase) CGH we used, however, has the advantage of using archival paraffin-embedded tissue, because the more sensitive the assay system is, the higher the possibility of developing an artifact produced from a long-preserved sample in which DNA degradation is inevitable to some degree. We believe that CGH analysis of gliomas is useful because it can help subgroup tumors that conventional histopathological examination cannot characterize in a clinically relevant fashion. In this report we present a case that demonstrates the importance of genetic analysis of gliomas. FIG. 5. Diagram representing the hypothetical pathway of a 1q copy gain. In the development of a 1q gain and 1p loss, a chromosomal breakpoint arises at (or close to) the centromere of chromosome 1, and a gain of the entire chromosome 1 (ch1) is associated with promotion of a 1q gain while maintaining an imbalance between 1p and 1q (1p 1q). J. Neurosurg. / Volume 106 / March, 2007 Conclusions We report on a patient with a GBM demonstrating chromosome 1q-arm copy gain in CGH analysis. This patient showed a good response to a chemotherapy regimen of PAV and a good prognosis. We also report five other cases that demonstrated the same genetic aberration. All of the tumors in these cases expressed low-level MGMT. Although further investigation is needed, there is a possibility that patients with gliomas showing a chromosome 1q gain also demonstrate good response to chemotherapy, similar to patients with oligodendrogliomas demonstrating chromosome 1p/19q loss, and thus achieve longer survival durations. We believe genetic analysis of gliomas using CGH is useful because it helps select patients with good chemotherapeutic sensitivity and avoids omitting appropriate treatment for those patients. These efforts will lead to progress in the treatment of patients with malignant gliomas. References 1. Athanassiou H, Synodinou M, Maragoudakis E, Paraskevaidis M, Verigos C, Misailidou D, et al: Randomized phase II study of temozolomide and radiotherapy compared with radiotherapy alone in newly diagnosed glioblastoma multiforme. J Clin Oncol 23: , Batchelor TT, Betensky RA, Esposito JM, Pham LD, Dorfman MV, Piscatelli N, et al: Age-dependent prognostic effects of genetic alterations in glioblastoma. Clin Cancer Res 10: , Bentz M, Plesch A, Stilgenbauer S, Döhner H, Lichter P: Minimal sizes of deletions detected by comparative genomic hybridization. Genes Chromosomes Cancer 21: , Burton EC, Lamborn KR, Feuerstein BG, Prados M, Scott J, Forsyth P, et al: Genetic aberrations defined by comparative genomic hybridization distinguish long-term from typical survivors of glioblastoma. Cancer Res 62: , Cairncross JG, Ueki K, Zlatescu MC, Lisle DK, Finkelstein DM, Hammond RR, et al: Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 90: , de Stahl TD, Hartmann C, de Bustos C, Piotrowski A, Benetkiewicz M, Mantripragada KK, et al: Chromosome 22 tiling-path array-cgh analysis identifies germ-line- and tumor-specific aberrations in patients with glioblastoma multiforme. Genes Chromosomes Cancer 44: , Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, et al: MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 352: ,

7 S. Takahashi et al. 8. Hirose Y, Aldape K, Takahashi M, Berger MS, Feuerstein BG: Tissue microdissection and degenerate oligonucleotide-primed polymerase chain reaction (DOP-PCR) is an effective method to analyze genetic aberrations in invasive tumors. J Mol Diagn 3: 62 67, Hirose Y, Aldape KD, Chang S, Lamborn K, Berger MS, Feuerstein BG: Grade II astrocytomas are subgrouped by chromosome aberrations. Cancer Genet Cytogenet 142:1 7, Hirose Y, Kreklau EL, Erickson LC, Berger MS, Pieper RO: Delayed repletion of O6-methylguanine-DNA methyltransferase resulting in failure to protect the human glioblastoma cell line SF767 from temozolomide-induced cytotoxicity. J Neurosurg 98: , Hui AB, Lo KW, Yin XL, Poon WS, Ng HK: Detection of multiple gene amplifications in glioblastoma multiforme using array-based comparative genomic hybridization. Lab Invest 81: , Ino Y, Zlatescu MC, Sasaki H, Macdonald DR, Stemmer-Rachamimov AO, Jhung S, et al: Long survival and therapeutic responses in patients with histologically disparate high-grade gliomas demonstrating chromosome 1p loss. J Neurosurg 92: , Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, et al: Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258: , Kim SH, Kim H, Kim TS: Clinical, histological, and immunohistochemical features predicting 1p/19q loss of heterozygosity in oligodendroglial tumors. Acta Neuropathol (Berl) 110:27 38, Kunwar S, Mohapatra G, Bollen A, Lamborn KR, Prados M, Feuerstein BG: Genetic subgroups of anaplastic astrocytomas correlate with patient age and survival. Cancer Res 61: , McLendon RE, Herndon JE II, West B, Reardon D, Wiltshire R, Rasheed BK, et al: Survival analysis of presumptive prognostic markers among oligodendrogliomas. Cancer 104: , Misra A, Pellarin M, Nigro J, Smirnov I, Moore D, Lamborn KR, et al: Array comparative genomic hybridization identifies genetic subgroups in grade 4 human astrocytoma. Clin Cancer Res 11: , Mollemann M, Wolter M, Felsberg J, Collins VP, Reifenberger G: Frequent promoter hypermethylation and low expression of the MGMT gene in oligodendroglial tumors. Int J Cancer 113: , Morita M, Rosenblum MK, Bilsky MH, Fraser RA, Rosenfeld MR: Long-term survivors of glioblastoma multiforme: clinical and molecular characteristics. J Neurooncol 27: , Mrak RE, Yaşargil MG, Mohapatra G, Earel J Jr, Louis DN: Atypical extraventricular neurocytoma with oligodendrogliomalike spread and an unusual pattern of chromosome 1p and 19q loss. Hum Pathol 35: , Nigro JM, Misra A, Zhang L, Smirnov I, Colman H, Griffin C, et al: Integrated array-comparative genomic hybridization and expression array profiles identify clinically relevant molecular subtypes of glioblastoma. Cancer Res 65: , Pegg AE, Wiest L, Foote RS, Mitra S, Perry W: Purification and properties of O-6-methylguanine-DNA transmethylase from rat liver. J Biol Chem 258: , Pieper RO: Understanding and manipulating O 6 -methylguanine- DNA methyltransferase expression. Pharmacol Ther 74: , Pinkel D, Segraves R, Sudar D, Clark S, Poole I, Kowbel D, et al: High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 20: , Piper J, Rutovitz D, Sudar D, Kallioniemi A, Kallioniemi OP, Waldman FM, et al: Computer image analysis of comparative genomic hybridization. Cytometry 19:10 26, Prados MD, Levin V: Biology and treatment of malignant glioma. Semin Oncol 27 (Suppl 6):1 10, Sabel M, Reifenberger J, Weber RG, Reifenberger G, Schmitt HP: Long-term survival of a patient with giant cell glioblastoma. Case report. J Neurosurg 94: , Salvati M, Cervoni L, Artico M, Caruso R, Gagliardi FM: Longterm survival in patients with supratentorial glioblastoma. J Neurooncol 36:61 64, Smith JS, Perry A, Borell TJ, Lee HK, O Fallon J, Hosek SM, et al: Alterations of chromosome arms 1p and 19q as predictors of survival in oligodendrogliomas, astrocytomas, and mixed oligoastrocytomas. J Clin Oncol 18: , Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, et al: Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352: , Wiltshire RN, Herndon JE II, Lloyd A, Friedman HS, Bigner DD, Bigner SH, et al: Comparative genomic hybridization analysis of astrocytomas: prognostic and diagnostic implications. J Mol Diagn 6: , 2004 Manuscript submitted January 16, Accepted June 27, Address reprint requests to: Satoshi Takahashi, M.D., Department of Neurosurgery, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo , Japan. stakahashi@2002.jukuin.keio.ac.jp. 494 J. Neurosurg. / Volume 106 / March, 2007