This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Silber, J. R. Articles by Berger, M. S. Search for Related Content PubMed PubMed Citation Articles by Silber, J. R. Articles by Berger, M. S.

O⁶-Methylguanine-DNA Methyltransferase-deficient Phenotype in Human Gliomas: Frequency and Time to Tumor Progression after Alkylating Agent-based Chemotherapy¹

Department of Neurological Surgery [J. R. S., M. S. Bo., S. G., D. D. K., M. S. Be.], Box 356470, and Department of Pathology [A. B.], Box 357705, University of Washington, Seattle, Washington 98195-6470; Division of Neurosurgery, Children's Hospital and Medical Center, Seattle, Washington 98105 [M. S. Bo.]

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The DNA repair protein O⁶-methylguanine-DNA methyltransferase (MGMT) contributes to the resistance of human brain tumor cell lines and xenografts to methylating and chloroethylating agents. We assayed MGMT in 174 newly diagnosed or recurrent gliomas to (a) quantitate changes in MGMT activity associated with alkylating agent-based chemotherapy; and (b) assess the contribution of MGMT to clinical outcome. Glioma MGMT activity ranged 300-fold, averaging 3,800 ± 7,200 molecules/cell. Twenty-four percent of tumors lacked detectable activity [Methyl repair-deficient (Mer^-) phenotype, defined here as <151 molecules/cell or <0.25 fmol/10⁶ cells]. Tumors treated with surgery alone and tumors recurring after surgery and radiotherapy did not differ significantly in frequency of the Mer^- phenotype (29% versus 24%). However, the frequency of the Mer^- phenotype among tumors recurring after surgery, radiation, and alkylating agent-based chemotherapy was 7-fold lower than in tumors treated with surgery alone (4.3% versus 29%; P 0.02) and 6-fold lower than in tumors recurring after surgery and radiation (4.3% versus 24%; P 0.05). In contrast to gliomas, there was no relationship of alkylating agent-based therapy with the frequency of the Mer^- phenotype in paired histologically normal brain. These data suggest that alkylating agents, either alone or synergistically with radiotherapy, selectively kill Mer^- glioma cells in situ. Importantly, Mer^- and Mer⁺ tumors did not differ in time to tumor progression following treatment with alkylating agents, indicating that although Mer^- glioma cells may be differentially killed by alkylators, factors other than Mer phenotype were the principal determinants of time to clinical progression. Nonetheless, our results support the possibility that complete ablation of glioma MGMT with substrate analogue inhibitors could improve the efficacy of alkylating agent-based chemotherapy.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Despite advances in surgical technique and radiation therapy, the prognosis for malignant gliomas remains poor, with 2-year survival rates <20% (1 , 2) . The chloroethylating agents carmustine (BCNU)⁴ and lomustine (CCNU) and the methylating agents procarbazine and temozolomide, in single agent or combination chemotherapy, have been shown to modestly increase response rates and survival times when used as an adjuvant to surgery and radiation (1 , 2) . The clinical effectiveness of these agents is attributed, in part, to their relatively high yields of cytotoxic O⁶-alkylguanine adducts in DNA (3 , 4) .

Intrinsic and acquired resistance to alkylating agents limits the efficacy of these drugs in the therapy of gliomas. A large body of evidence demonstrates that the DNA repair protein MGMT contributes to alkylating agent resistance in human brain tumor cell lines and xenografts (5, 6, 7, 8, 9, 10) . MGMT mediates resistance by removing cytotoxic alkyl adducts from the O⁶ atom of guanine in DNA to an internal cysteine, yielding guanine and S-alkylcysteine. Because the alkyl receptor site in MGMT is not regenerated, the number of adducts that can be removed from DNA in vivo is limited by the number of MGMT molecules and the rate of synthesis of the protein (5)

Most human neoplasms (5) , including those from brain (11, 12, 13, 14, 15, 16, 17, 18) , express MGMT activity, suggesting that MGMT contributes to the alkylating agent resistance of tumors in vivo. We have observed a 300-fold range of detectable activity among 152 adult brain tumors (18) . Importantly, 25% of specimens had no detectable activity (Mer^-), implying that an appreciable fraction of brain tumors may have heightened sensitivity to alkylating agents as a consequence of lacking MGMT. Evidence of preferential killing of Mer^- glioma cells in situ would support this idea and would strengthen the rationale for antiresistance strategies using inhibitors of MGMT, such as O⁶-benzylguanine (19)

In the present study, we analyzed the relationship of prior alkylating agent-based therapy to MGMT activity in recurrent gliomas, and the relationship of MGMT activity to response to alkylator therapy. We observed a 6- to 7-fold reduction in the frequency of the Mer^- phenotype among tumors that recurred after treatment with alkylators, whereas the frequency of the Mer^- phenotype in adjacent, microscopically normal brain was unchanged. We interpret these findings to suggest that Mer^- glioma cells are selectively killed in situ by alkylating agent-based chemotherapy. However, despite the putative selective killing of Mer^- cells, we found that the Mer^- phenotype was not associated with a longer time to tumor progression after alkylating agent-based therapy. This result indicates that variables other than Mer phenotype can be overriding determinants of clinical response to current treatment regimens that include alkylators. Nonetheless, our observations support the idea that complete ablation of glioma MGMT activity with substrate analogue inhibitors might improve clinical response.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Tissue.
Tumors were resected at the University of Washington Medical Center from 1991 to 1996. Subcortical normal brain adjacent to tumor was obtained from 104 patients. The specimens included 152 tumors and 70 normal brain samples for which MGMT activity had been reported in earlier studies of the relationship of activity to patient and tumor characteristics (15 , 18 , 20) . Diagnosis was obtained from the final neuropathology report. Normal tissue was microscopically free of hypercellularity, infiltrating tumor, endothelial proliferation, edema, and gliosis. Demographic information together with course and response to therapy was obtained from medical records

Immediately upon resection, tissue was placed in ice-cold DMEM supplemented with 15% fetal bovine serum and was transported to the laboratory within minutes. We have successfully established cell lines from tumors held overnight on ice in supplemented DMEM. We have also found identical MGMT activities in aliquots of brain tumor and normal brain processed either immediately upon arrival in the laboratory or after being held overnight on ice in supplemented medium, demonstrating that our protocol for transporting specimens preserves cell viability and MGMT activity. To determine cell number, a small piece of tissue (0.05–0.1 g) was finely minced with scalpel blades, after blood vessels were removed, and suspended in PBS. The suspension was serially passed through 18, 20, and 22 gauge needles to completely disrupt the tissue. After filtration of the suspension through a 60 µm mesh, cells were pelleted by centrifugation. Contaminating erythrocytes were eliminated by hypotonic lysis, and debris was removed by repeated washes with PBS. After the washed pellet was resuspended in PBS, cells in a 10-µl aliquot were stained with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, and the total cell number was determined by counting with a hemacytometer. The remaining tissue (0.5–5 g) was divided into two or more aliquots, frozen in liquid nitrogen, and stored at -80°C.

MGMT Assay.
The MGMT content of extracts (i.e., high speed supernatants of whole tissue sonicates) was assayed by quantitating the transfer of radioactivity from a DNA substrate containing [methyl³H]O⁶-methylguanine to protein, as described previously in detail (18 , 20) . MGMT content is the mean of at least five determinations that generally differed by no more than 20%. We have validated this assay for extracts prepared from tumor and normal brain by demonstrating that (a) the increase in radioactivity transferred to protein is linear over a 3- to 5-fold range of added extract; (b) the transfer of radioactivity is sensitive to protease digestion; (c) transfer of radioactivity is prevented by O⁶-benzylguanine, a potent inhibitor of MGMT (19) ; and (d) transferred radioactivity migrates on a SDS-polyacrylamide gel at M_r 22,000, identical to that of human MGMT (5)

Additional controls indicated that the wide range of MGMT activity observed is unlikely to be due to degradation of MGMT and/or its [³H]DNA substrate during extraction and assay, or to a diffusible inhibitor in extracts. (a) The protease inhibitors aprotinin, leupeptin, and pepstatin did not increase activity when present together during extraction and assay. (b) All Mer⁺ samples displayed linearity of activity with added extract. (c) Additive amounts of activity were found for mixed extracts of Mer⁺ specimens with high and low MGMT activities. (d) Assay of an 8:1 (v/v) mixture of every Mer^- extract with a Mer⁺ extract yielded activity expected for the Mer⁺ sample. (e) Extracts prepared from mixtures of Mer^- brain tissue and Mer⁺ cultured medulloblastoma cells yielded activity expected of the cultured cells

We define the Mer^- phenotype here as MGMT activity <0.25 fmol/10⁶ cells or 151 molecules/cell. Mer^- phenotype is a functional term that refers to the limit of detection in the particular assay used. In our biochemical assay, the definition of Mer^- phenotype was established by (a) the specific activity of the [methyl 3H]-O⁶-methylguanine moiety in the DNA substrate (20 Ci/mmol); (b) counting efficiency equal to 14%; (c) the requirement that extract from at least 7.7 x 10⁶ cells be included in the assay; and (d) the requirement that extract from 7.7 x 10⁶ cells yield less than 12 cpm (86 dpm) above an unincubated control that displays 25–40 cpm when counted for 10 min. On average, the Mer^- tumor and normal brain samples reported here yielded 2.4 ± 1.6 cpm (mean ± SD) above the unincubated control; these counts were sporadic and were not proportional to added extract. Notably, the limit of detection in our assay is lower than that of histocytochemical procedures [4000–30,000 molecules/cell (17 , 21) ] and comparable to that of oligonucleotide-based assays [200 molecules/cell (22) ] and Western assay [80 molecules/cell if 7.7 x 10⁶ cell-equivalents could be analyzed (23) ].

Alkylating Agent-based Chemotherapy.
Sixty-two patients received alkylating agent-based chemotherapy. Eight of these individuals were treated with BCNU. Of the eight, five received 200 mg/m² BCNU i.v. every 6–8 weeks; three received 80 mg/m² BCNU and 33 mg/m² cisplatin i.v. daily for 3 days every 6 weeks for three cycles and then 80 mg/m² BCNU i.v. daily for 3 days every 6 weeks for two cycles. Fifty-four patients were treated with procarbazine and CCNU in combination with other drugs. Of the 54, 41 (PCV group) received 110 mg/m² CCNU p.o. on day 1, 60 mg/m² procarbazine p.o. on days 8 through 21 and 1.4 mg/m² vincristine i.v. on days 8 and 29. This cycle was repeated every 6 to 8 weeks for up to 8 cycles. Nine patients in the PCV group also received 300 mg/m² hydroxyurea p.o. every 6 h during radiotherapy. Thirteen of the 54 patients treated with procarbazine and CCNU (6-drug group) received 30 mg/m² 6-thioguanine every 6 h for 12 doses, 50 mg/m² procarbazine every 6 h for 4 doses, 1 dose of 400 mg/m² dibromodulcitol, and 1 dose of 100 mg/m² CCNU (all p.o.) in 3 days, followed by a continuous i.v. infusion of 1000 mg/m² of 5-fluorouracil on days 14 and 15 and 1000 mg/m² hydroxyurea every 4 h for 4 doses on day 14. This cycle was repeated every 6–8 weeks. Chemotherapy was terminated upon unacceptable levels of toxicity or evidence of tumor progression

All 62 patients treated with alkylators also received prior radiotherapy (52–62 Gy) in 4 to 6 weeks). Of the 62, 28 individuals had newly operated tumors that were treated with alkylators within 2 to 4 weeks of completing radiotherapy. Another six individuals had tumors recurring after initial surgery only and were treated with radiation and alkylators after a second surgery. The remaining 28 individuals had tumors recurring after initial surgery and radiation and were given alkylating agents after a second surgery.

Statistical Analysis.
Standard statistical procedures (24) were applied as specified by using Microsoft Excel (Microsoft, Redmond, WA). For purposes of calculation, Mer samples were assigned a value of 0.125 fmol/10⁶ cells, i.e., one-half the lower limit of detection.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patient and Tumor Characteristics.
One hundred seventy-four glial tumors were obtained from 166 informed patients (Table 1) . Patient age ranged from 18 to 79 years (mean ± SD, 45 ± 13 years), and the ratio of males to females was 1.9, in accord with the known gender bias for gliomas (25) . One hundred twenty-two tumors (70%) were astrocytic gliomas, the most frequent glial tumor diagnosis; they included 16 astrocytomas, 23 anaplastic astrocytomas, and 83 glioblastomas. Other tumors included 30 oligodendrogliomas and 22 mixed oligodendroglioma-astrocytomas

MGMT in Tumors.
The MGMT activity of tumor extracts was quantitated in our standard biochemical assay, which yields the average activity of many millions of cells. Tumors with activity <0.25 fmol/10⁶ cells or 151 molecules/cell, the limit of detection of the assay, were assigned the Mer^- phenotype ("Materials and Methods" and Refs. 18 and 20 ). The MGMT activity of all 174 tumors covered a 300-fold range, from 0.30 to 89 fmol/10⁶ cells (180 to 54,000 molecules/cell), the mean being 6.3 ± 12 fmol/10⁶ cells (3,800 ± 7,200 molecules/cell). Twenty-four percent of tumors had no detectable activity (Mer^- phenotype). Males and females did not differ significantly in mean activity (5.8 ± 11 versus 7.2 ± 15 fmol/10⁶ cells) or fraction of Mer^- tumors (24% versus 25%). The wide variation in activity is similar to previous observations in brain tumors and other neoplastic human tissues (11, 12, 13, 14, 15, 16, 17, 18) . A detailed analysis of the relationships of MGMT activity to patient and tumor characteristics (e.g., age, diagnosis, ploidy, and proliferative rate) for 152 of the specimens has been presented elsewhere (18)

The MGMT activity in 174 tumors grouped by treatment regimen and diagnosis is summarized in Table 2 . There was no statistically significant difference in mean MGMT (6.1 ± 10 versus 5.9 ± 13 fmol/10⁶ cells) or fraction of Mer^- specimens (29% versus 24%) between tumors treated with surgery alone and tumors recurrent after surgery and adjuvant radiotherapy. In contrast, the frequency of the Mer^- phenotype among tumors recurring after surgery, radiation, and alkylating agent-based chemotherapy was 7-fold lower than for tumors treated with surgery alone (4.3% versus 29%; P 0.02; ² = 6.14) and 6-fold lower than for tumors recurring after surgery and radiotherapy (4.3% versus 24%; P 0.05; ² = 3.94). Comparable findings applied to the large astrocytic glioma subgroup where the frequency of the Mer^- phenotype among tumors recurring after surgery, radiation, and alkylating agent-based chemotherapy was 6-fold lower than for tumors treated with surgery alone (4.5% versus 27%; P 0.05; ² = 4.43) and 5-fold lower than for tumors recurring after surgery and radiotherapy (4.5% versus 24%; P 0.05; ² = 3.86). Thus, we infer that, whereas adjuvant radiotherapy was not associated with an altered proportion of Mer^- tumors, treatment with radiation and alkylating agents was associated with reduced frequency of the Mer^- phenotype in recurrent tumors. We believe the likeliest explanation for the observed frequencies is that Mer^- glioma cells are preferentially killed by alkylating agents, acting alone or synergistically with radiotherapy (see "Discussion")

View larger version (40K): [in this window] [in a new window]   Fig. 1. Distribution of MGMT activity in Mer+ tumors grouped by treatment regimen. , tumors treated with surgery alone; , tumors recurring after surgery and radiotherapy; , tumors recurring after surgery, radiotherapy, and alkylating agent therapy. ANOVA revealed no significant difference in the distribution of activity in the three groups.

As summarized in Table 3 , treatment with alkylating agents had no statistically significant relationship with mean MGMT activity or frequency of the Mer^- phenotype in normal tissue. In particular, the frequency of the Mer^- phenotype in normal brain from individuals treated with surgery alone, with surgery and radiation, and with surgery, radiation, and alkylating agents was 60, 63, and 63%, respectively. In contrast, the frequency of the Mer^- phenotype among the paired tumors recurring after surgery, radiation and alkylating agents (6.3%) was 5-fold lower than the frequency in tumors treated with surgery alone (34%; P 0.02; ² = 5.43) and 5-fold lower than in tumors recurring after surgery and radiotherapy (30%; P 0.05; ² = 3.92). These latter data confirm that the subgroup of 104 tumors paired with normal brain (Table 3) showed a correlation of alkylating agent-based therapy with Mer^- frequency comparable to that observed in the entire tumor population (Table 2) .

Tumor MGMT and Clinical Response to Alkylating Agents.
The foregoing results suggest that the Mer phenotype could be a determinant of the alkylating agent survival of glioma cells in situ. To assess the role of the Mer^- phenotype in clinical response to alkylators, we examined the relationship of tumor MGMT activity to TTP, i.e., the interval from initiation of alkylating agent-based chemotherapy to the appearance of radiological and/or symptomatic evidence of tumor progression. Sixty-two patients were treated with alkylating agents. The majority of cases (42 of 62) were anaplastic astrocytoma and glioblastoma, the remainder were malignant oligodendroglioma and malignant mixed oligodendroglioma-astrocytoma. Forty-five patients had shown progression, as indicated in Table 4 . Of the 45, 7 received a chloroethylating agent alone (BCNU) and 38 received both a methylating and a chloroethylating agent (procarbazine in combination with CCNU). Twenty-six patients treated with procarbazine/CCNU also received vincristine (PCV protocol). The remaining 12 patients treated with procarbazine/CCNU were also given 6-thioguanine, dibromodulcitol, hydroxyurea, and 5-fluorouracil (6-drug protocol)

Lastly, we analyzed the relationship of the Mer phenotype to the proportion of progression-free patients by using the Kaplan-Meier method, which allowed inclusion of the 17 individuals who were progression free at the end of the study. Log-rank analysis revealed no significant difference between 20 Mer^- and 42 Mer⁺ tumors in the proportion of progression-free patients (Fig. 2A) . Grouping of the 62 tumors on the basis of other MGMT activities (i.e., the mean, one-half the mean, and 1.5 times the mean) likewise revealed no significant difference. Comparable results were obtained if analysis was limited to anaplastic astrocytomas and glioblastomas (Fig. 2B ) or to malignant oligodendrogliomas and mixed oligodendroglioma-astrocytomas (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 2. Proportion of progression-free patients as a function of time after initiation of alkylating agent therapy. A, results for 62 gliomas, 20 Mer-, and 42 Mer+. B, results for 19 anaplastic astrocytomas and 24 glioblastomas, 15 Mer-, and 28 Mer+.

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Ionizing radiation is a mainstay of therapy for malignant brain tumors and, as an adjuvant to surgery, substantially improves survival of adult and pediatric patients (1 , 2) . We found no difference in mean MGMT activity or in frequency of the Mer^- phenotype between gliomas treated with surgery alone and gliomas recurring after surgery and radiation therapy (Table 2) , nor did we find any difference in the normal tissue adjacent to tumors (Table 3) . Notably, it is possible that radiation reduces the cytotoxicity of subsequent alkylator therapy, as well as the differential susceptibility of Mer⁺ and Mer^- cells, by selecting for variants that tolerate increased levels of DNA damage without lethality (28 , 29)

In contrast to radiotherapy, we found a strong association between alkylating agent-based chemotherapy, acting either alone or synergistically with radiation, and MGMT activity in tumors (Tables 2 and 3 ). Whereas 29% of tumors from patients treated with surgery alone and 24% of tumors recurring after surgery and radiotherapy were Mer^-, only 4.3% of tumors recurring after surgery, radiotherapy, and alkylating agents had the Mer^- phenotype. Importantly, there was no association of alkylator therapy with the frequency of the Mer^- phenotype in adjacent, histologically normal brains (Table 3)

The reduction in frequency of the Mer^- phenotype among tumors exposed to alkylators was statistically significant for all gliomas and for the astrocytic glioma subgroup, and is based on the assumption of equal Mer^- frequencies in the three treatment groups at the start of therapy. This assumption is supported by consideration of the indications for the three therapeutic regimens and by MGMT measurements of relevant tumor populations. With respect to the surgery versus surgery plus radiotherapy groups, tumor grade is the primary indication for radiation, with high-grade, but not all low-grade, gliomas receiving radiotherapy. This criterion would not be expected to bias Mer^- frequencies, because we have observed that the proportion of Mer^- tumors among astrocytomas, anaplastic astrocytomas, and glioblastomas did not differ statistically (18) . The primary indications for alkylating agent therapy are recurrence and high grade. Recurrence per se would not be expected to bias Mer^- frequencies because, as documented above, tumors recurrent after resection without adjuvant therapy did not differ from newly operated tumors in the Mer^- fraction. And as mentioned, tumor grade would not be expected to be a source of bias because the three astroglial diagnoses were found to be associated with similar Mer^- frequencies (18) . Although advanced age is a contraindication for alkylator therapy, the three treatment groups had very similar mean ages and standard deviations (Table 1) . Finally, with respect to possible bias in assignment of Mer^- tumors to alkylator regimens, the Mer phenotype was not known by the attending physicians

We believe the likeliest explanation for the reduced frequency of the Mer^- phenotype in recurrent gliomas from patients treated with alkylators is selective killing of Mer^- cells and differential regrowth by surviving Mer⁺ cells. This explanation is consonant with the large body of data documenting the greater susceptibility of MGMT-deficient prokaryotic and eukaryotic cells to alkylating agent-induced killing (5 , 26 , 27) . Importantly, intercellular heterogeneity of MGMT content has been observed in gliomas, including the occurrence of a minority of Mer⁺ cells within Mer^- tumors (30) . Such heterogeneity is consistent with the epigenetic regulation of MGMT activity observed in cultured human cells (31) and the frequent alterations of epigenetic control in tumors (32) . The foregoing explanation is also in accord with the results of Phillips et al. (33) who observed selective expansion of Mer⁺ cells following BCNU treatment of heterogeneous xenografts composed of Mer^- and Mer⁺ cells. It also accounts for the unchanged frequency of the Mer^- phenotype we observed in normal brain adjacent to tumors because selection would be absent or relatively weak in tissue where proliferation is rare or nonexistent. A significant point is that acquisition of the Mer⁺ phenotype by a formerly Mer^- tumor could reflect increased MGMT activity in only a fraction of the cell population. This is possible because our assay measures a cellular average, and increased activity in even a small minority of cells could elevate the mean into Mer⁺ range

Our data are also consistent with an epigenetically controlled, alkylation-induced increase in MGMT activity that converts tumors from the Mer^- to the Mer⁺ phenotype. However, we do not favor this interpretation for three reasons. First, there was no alkylation-associated difference in the frequency of the Mer^- phenotype in normal brain tissue (Table 3) . Second, the 1.3-fold greater mean MGMT activity among alkylating agent-treated tumors (Table 2) was fully accounted for by the decreased frequency of Mer^- tumors and, as shown in Fig. 1 , was not accompanied by a shift toward higher MGMT activity among Mer⁺ tumors. Third, in contrast to the well-documented, selective killing of MGMT-deficient prokaryotic and eukaryotic cells, we are unaware of any clear precedent for a long-term, epigenetically regulated increase in MGMT activity in response to alkylators. Although these reasons do not exclude an epigenetic mechanism, they reduce its likelihood in our view

Importantly, Mer⁺ tumors that were recurrent after alkylator therapy did not exhibit elevated levels of MGMT (Fig. 1) . We interpret this observation to suggest that widely different MGMT levels in Mer⁺ tumors prior to alkylation (e.g., 180–54,000 molecules/cell; Table 2 ) were associated with comparable cell survival. In other words, an MGMT level of at least 151 molecules/cell (our definition here of the Mer⁺ phenotype) enhanced survival, but beyond that, increased activity did not further protect cells from killing. Failure of higher MGMT levels to promote greater survival would occur when factors other than O⁶-alkylguanine intervene to become the limiting determinants of cell killing. Examples of such factors include other potentially lethal DNA lesions caused by alkylators, e.g., 3-methyladenine (34) . Our results for gliomas in situ are consistent with the finding that resistance to alkylating agents in brain tumor cell lines is determined by factors in addition to MGMT (6, 7, 8) and provide strong rationale for the identification of such factors

Despite our inference that alkylating agents selectively kill Mer^- cells, we observed that the Mer^- phenotype was not associated with longer TTP (Table 4) or higher probability of remaining progression free after alkylating agent therapy (Fig. 2) . One would expect a longer TTP for Mer^- tumors if (1) the Mer phenotype was the primary determinant of alkylating agent survival, and hence of residual tumor burden, and (2) if residual tumor burden was the primary determinant of TTP. Our findings are consistent with at least two possibilities. The first is that the putative difference in the killing of Mer^-versus Mer⁺ cells was not large enough to confer a clinically manifest difference between Mer^- and Mer⁺ tumors in the number of cells surviving alkylator therapy. This possibility is consistent with the conclusion that resistance to chloroethylating and methylating agents in brain tumor cell lines is determined by factors in addition to MGMT (6, 7, 8) . The second possibility is that other variables limit TTP, e.g., vascularization in areas of surviving cells

Belanich et al. (17) , in a retrospective study of malignant astrocytic gliomas treated with radiation and BCNU, found that MGMT levels 60,000 molecules/nucleus, quantitated in individual cells by immunocytochemistry, were associated with a longer TTP. There are major differences in clinical and analytic protocols between our work and that of Belanich et al. One difference is the alkylating agent therapy, i.e., the majority of patients in our study received the methylating agent procarbazine in addition to the chloroethylating agent CCNU. A second difference is the method of MGMT quantitation, i.e., biochemical assay of tissue extracts that measures the average of millions of cells versus quantitative immunofluorescence microscopy of individual cells in fixed tumor sections. Direct comparison of the two studies is precluded by these differences

Finally, our evidence suggesting that Mer^- glioma cells in situ are more susceptible to killing by alkylators supports the ongoing clinical trials of MGMT inhibitors such as O⁶-benzylguanine (19) . If chemical inhibitors can be used to convert all tumors cells to the Mer^- phenotype, the number of surviving cells may be reduced to well below the levels achievable at present, and this reduction may lead to improved response in at least some cases.

	ACKNOWLEDGMENTS

 
We thank Dr. Lawrence Loeb for the use of his facilities and his ongoing support. We thank Alex K. Lee, Peter L. Kennedy, and the Undergraduate Brain DNA Repair Group (Chris Dorr, Aaron Hanly, Darcie Porter, Bobby Stevens, Matthew Studer, Brian Whitmire, and David Wong) for expert technical assistance.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ This work was supported by grants from NIH (Grants CA 70790 and OIG R35-CA39903), the American Cancer Society (Grants EDT-53 and CN84214), and the American Federation for Aging Research. Additional support was from the John Gallagher Fund, the Doris Schiffman Fund, and the Brain Tumor Research Fund of the Department of Neurological Surgery, University of Washington, and the Neurooncology Gift Fund and Jessie’s Perfect Peach Fund of Children’s Hospital and Regional Medical Center, Seattle, Washington.

² To whom requests for reprints should be addressed, at Department of Neurological Surgery, Box 356470, University of Washington, 1959 NE Pacific Street, Seattle, WA 98195-6470. Phone: (206) 685-8642; Fax: (206) 543-8315; E-mail: jrsilber{at}u.washington.edu

³ Present address: Department of Neurological Surgery, University of California San Francisco, San Francisco, CA 94143-0112.

⁴ The abbreviations used are: BCNU, 1,3-bis(2-chloroethyl)-1-nitrosourea; CCNU, 1-(2-chloroethyl)-3-cyclohexyl-1-nitrososurea; MGMT, O⁶-methylguanine-DNA methyltransferase; TTP, time to tumor progression.

Received 7/17/98; revised 1/26/99; accepted 1/27/99.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Berger M. S., Spence A. M., Stelzer K. J. Brain tumors Brain M. C. Carbone P. P. eds. . Current Therapy in Hematology/Oncology, : 525-542, Mosby Publishers Philadelphia, PA 1995.
2. Prados M. D., Russo C. Chemotherapy for brain tumors. Semin. Surg. Oncol., 14: 88-95, 1998.[Medline]
3. Beranek D. T. Distribution of methyl and ethyl adducts following alkylation with monofunctional alkylating agents. Mutat. Res., 231: 11-30, 1990.[Medline]
4. Ludlum D. B. The chloroethylnitrosoureas: sensitivity and resistance to cancer chemotherapy at the molecular level. Cancer Invest., 15: 588-598, 1997.[Medline]
5. Pegg A. E., Dolan M. E., Moschel R. C. Structure, function and inhibition of O⁶-alkylguanine-DNA alkyltransferase. Prog. Nucleic Acid Res. Mol. Biol., 51: 167-223, 1995.[Medline]
6. Bobola M. S., Blank A., Berger M. S., Silber J. R. Contribution of O⁶-methylguanine-DNA methyltransferase to monofunctional alkylating agent resistance in human brain tumor-derived cell lines. Mol. Carcinog., 13: 70-80, 1995.[Medline]
7. Bobola M. S., Berger M. S., Silber J. R. Contribution of O⁶-methylguanine-DNA methyltransferase to resistance to 1,3-bis-(2-chloroethyl)-1-nitrosourea in human brain tumor-derived cell lines. Mol. Carcinog., 13: 81-88, 1995.[Medline]
8. Bobola M. S., Tseng S. H., Blank A., Berger M. S., Silber J. R. Role of O⁶-methylguanine-DNA methyltransferase in resistance of human brain tumor cell lines to the clinically relevant methylating agents temozolomide and streptozotocin. Clin. Cancer Res., 2: 735-741, 1996.[Abstract]
9. Friedman H. S., Dolan M. E., Pegg A. E., Marcelli S., Keir S., Catino J. J., Bigner D. D., Schold S. C. Activity of temozolomide in the treatment of central nervous system tumor xenografts. Cancer Res., 55: 2853-2857, 1995.[Abstract/Free Full Text]
10. Schold S. C., Kokkinakis D. M., Rudy J. L., Moschel R. C., Pegg A. E. Treatment of human brain tumor xenografts with O⁶-benzyl-2'-deoxyguanosine and BCNU. Cancer Res., 56: 2076-2081, 1996.[Abstract/Free Full Text]
11. Wiestler O., Kleihues P., Pegg A. E. O⁶-alkyltransferase activity in human brain and brain tumors. Carcinogenesis (Lond. ), 5: 121-124, 1984.[Abstract/Free Full Text]
12. Citron M., Decker R., Chen S., Schneider S., Graver M., Kleynerman L., Kahn L. B., White A., Schoenhaus M., Yarosh D. O⁶-methylguanine-DNA methyltransferase in human normal and tumor tissue from brain, lung and ovary. Cancer Res., 51: 4131-4134, 1991.[Abstract/Free Full Text]
13. Gerson S. L., Trey J. E, Miller K., Berger N. A. Comparison of O⁶-alkylguanine-DNA alkyltransferase activity based on cellular DNA content in human, rat and mouse tissues. Carcinogenesis (Lond.), 7: 745-749, 1986.[Abstract/Free Full Text]
14. Frosina G., Rossi O., Arena G., Gentile S. L., Bruzzone E., Abbondandolo A. O⁶-alkylguanine-DNA alkyltransferase activity in human brain tumors. Cancer Lett., 55: 153-158, 1990.[Medline]
15. Silber J. R., Mueller B. A., Ewers T. G., Berger M. S. Comparison of O⁶-methylguanine-DNA methyltransferase activity in brain tumors and adjacent normal brain. Cancer Res., 53: 3416-3420, 1993.[Abstract/Free Full Text]
16. Citron M., White A., Decker R., Wasserman P., Li B., Randall T., Guerra D., Belanich M., Yarosh D. O⁶-methylguanine-DNA methyltransferase in human brain tumors detected by activity assay and monoclonal antibodies. Oncol. Res., 7: 49-55, 1995.[Medline]
17. Belanich M., Pastor M., Randall T., Guerra D., Kibitel J., Alas L., Li B., Citron M., Wasserman P., White A., Eyre H., Jaeckle K., Schulman S., Rector D., Prados M., Coons S., Shapiro W., Yarosh D. Retrospective study of the correlation between the DNA repair protein alkyltransferase and survival of brain tumor patients treated with carmustine. Cancer Res., 56: 783-788, 1996.[Abstract/Free Full Text]
18. Silber J. R., Bobola M. S., Ghatan S., Kolstoe D. D., Blank A., Berger M. S. O⁶- methylguanine-DNA methyltransferase activity in adult gliomas: relation to patient and tumor characteristics. Cancer Res., 58: 1068-1073, 1998.[Abstract/Free Full Text]
19. Dolan M. E., Pegg A. E. O⁶-benzylguanine and its role in chemotherapy. Clin. Cancer Res., 6: 837-847, 1997.
20. Silber J. R., Blank A., Bobola M. S., Mueller B. A., Kolstoe D. D., Ojemann G. A., Berger M. S. Lack of the DNA repair protein O⁶-methylguanine-DNA methyltransferase in histologically normal brain adjacent to primary human brain tumors. Proc. Natl. Acad. Sci. USA, 93: 6941-6946, 1996.[Abstract/Free Full Text]
21. Brent T. P., von Wronski M. A., Edwards C. C., Bromley M., Margison G. P., Rafferty J. A., Pegram C. N., Bigner D. D. Identification of nitrosourea-resistant human rhabdomyosarcomas by in situ immunostaining of O⁶-methylguanine-DNA methyltransferase. Oncol. Res., 5: 83-86, 1993.[Medline]
22. Wu R. S., Hurst-Calderone S., Kohn K. Measurement of O⁶-alkylguanine-DNA alkyltransferase activity in human cells and tumor tissues by restriction endonuclease inhibition. Cancer Res., 47: 6229-6235, 1987.[Abstract/Free Full Text]
23. Ostrowski L. E., Pegram C. N., von Wronski M. A., Humphrey P. A., He X., Shiota S., Mitra S., Brent T. P., Bigner D. D. Production and characterization of antipeptide antibodies against human O⁶-methylguanine-DNA methyltransferase. Cancer Res., 51: 3339-3344, 1991.[Abstract/Free Full Text]
24. Bland M An Introduction to Medical Statistics Oxford University Press New York, NY 1995.
25. Preston-Martin S. Epidemiology of primary CNS neoplasms. Neuroepidemiology, 14: 273-290, 1996.
26. Xiao W., Derfler B., Chen J., Samson L. Primary sequence and biological functions of a Saccharomyces cerevisiaeO⁶-methylguanine/O⁴-methylthymine DNA repair methyltransferase gene. EMBO J., 10: 2179-2186, 1991.[Medline]
27. Rebeck G. W., Samson L. Increased spontaneous mutation and alkylation sensitivity of Escherichia coli strains lacking ogtO⁶-methylguanine DNA repair methyltransferase. J. Bacteriol., 173: 2068-2076, 1991.[Abstract/Free Full Text]
28. Walker K. A., Jennings C. D., Pulliam J., Ogburn C., Martin G. M., Urano M., Turker M. S. A cell line selected for resistance to ionizing radiation exhibits cross resistance to other genotoxic agents and a mutator phenotype for loss of heterozygosity events. Somat. Cell Mol. Genet., 23: 111-121, 1997.[Medline]
29. Yount G. L., Haas-Kogan D. A., Levine K. S., Aldape K. D., Israel M. A. Ionizing radiation inhibits chemotherapy-induced apoptosis in cultured glioma cells: implications for combined modality therapy. Cancer Res., 58: 3819-3825, 1998.[Abstract/Free Full Text]
30. Lee S. M., Reid H., Elder R. H., Thatcher N., Margison G. P. Inter- and intracellular heterogeneity of O⁶-alkylguanine-DNA alkyltransferase expression in human brain tumors: possible significance in nitrosourea therapy. Carcinogenesis (Lond.), 17: 637-641, 1996.[Abstract/Free Full Text]
31. Watts G. S., Pieper R. O., Costello J. F., Peng Y.-M., Dalton W. S., Futscher B. W. Methylation of discrete regions of the O⁶-methylguanine-DNA methyltransferase (MGMT) CpG island is associated with heterochromatinization of the MGMT transcription start site and silencing of the gene. Mol. Cell Biol., 17: 5612-5619, 1997.[Abstract]
32. Issa J-P. J., Baylin S. B. Epigenetics and human disease. Nat Med., 2: 281-282, 1996.[Medline]
33. Phillips W. P., Willson J. K. V., Markowitz S. D., Zborowska E., Zaidi N. H., Liu L., Gordon N. H., Gerson S. L. O⁶-methyl-guanine-DNA methyltransferase (MGMT) transfectants of a 1,3-bis-(2-chloroethyl)-1-nitrosourea (BCNU)-sensitive colon cancer cell line selectively repopulate heterogeneous MGMT⁺/MGMT^- xenografts after BCNU and O⁶-benzylguanine plus BCNU. Cancer Res., 57: 4817-4823, 1997.[Abstract/Free Full Text]
34. Engleward B. P., Dreslin A., Christensen J., Huzar D., Kurachara C., Samson L. Repair-deficient 3-methyladenine glycosylase homozygous mutant mouse cells have increased sensitivity to alkylation-induced chromosome damage and cell killing. EMBO J., 15: 945-952, 1996.[Medline]

This article has been cited by other articles:

				D. Fontijn, A. D. Adema, K. K. Bhakat, H. M. Pinedo, G. J. Peters, and E. Boven O6-Methylguanine-DNA-methyltransferase promoter demethylation is involved in basic fibroblast growth factor induced resistance against temozolomide in human melanoma cells Mol. Cancer Ther., October 1, 2007; 6(10): 2807 - 2815. [Abstract] [Full Text] [PDF]

				D. P. Cahill, K. K. Levine, R. A. Betensky, P. J. Codd, C. A. Romany, L. B. Reavie, T. T. Batchelor, P. A. Futreal, M. R. Stratton, W. T. Curry, et al. Loss of the Mismatch Repair Protein MSH6 in Human Glioblastomas Is Associated with Tumor Progression during Temozolomide Treatment Clin. Cancer Res., April 1, 2007; 13(7): 2038 - 2045. [Abstract] [Full Text] [PDF]

				M. S. Bobola, S. Varadarajan, N. W. Smith, R. D. Goff, D. D. Kolstoe, A. Blank, B. Gold, and J. R. Silber Human Glioma Cell Sensitivity to the Sequence-Specific Alkylating Agent Methyl-Lexitropsin Clin. Cancer Res., January 15, 2007; 13(2): 612 - 620. [Abstract] [Full Text] [PDF]

				O. K. Mirzoeva, T. Kawaguchi, and R. O. Pieper The Mre11/Rad50/Nbs1 complex interacts with the mismatch repair system and contributes to temozolomide-induced G2 arrest and cytotoxicity. Mol. Cancer Ther., November 1, 2006; 5(11): 2757 - 2766. [Abstract] [Full Text] [PDF]

				J. A. Maxwell, S. P. Johnson, J. A. Quinn, R. E. McLendon, F. Ali-Osman, A. H. Friedman, J. E. Herndon II, K. Bierau, J. Bigley, D. D. Bigner, et al. Quantitative analysis of O6-alkylguanine-DNA alkyltransferase in malignant glioma. Mol. Cancer Ther., October 1, 2006; 5(10): 2531 - 2539. [Abstract] [Full Text] [PDF]

				J. P. Fruehauf, H. Brem, S. Brem, A. Sloan, G. Barger, W. Huang, and R. Parker In vitro Drug Response and Molecular Markers Associated with Drug Resistance in Malignant Gliomas Clin. Cancer Res., August 1, 2006; 12(15): 4523 - 4532. [Abstract] [Full Text] [PDF]

				I. F. Pollack, R. L. Hamilton, R. W. Sobol, J. Burnham, A. J. Yates, E. J. Holmes, T. Zhou, and J. L. Finlay O6-Methylguanine-DNA Methyltransferase Expression Strongly Correlates With Outcome in Childhood Malignant Gliomas: Results From the CCG-945 Cohort J. Clin. Oncol., July 20, 2006; 24(21): 3431 - 3437. [Abstract] [Full Text] [PDF]

				R. Stupp, M. E. Hegi, M. J. van den Bent, W. P. Mason, M. Weller, R. O. Mirimanoff, J. G. Cairncross, and on behalf of the European Organisation for Researc Changing paradigms--an update on the multidisciplinary management of malignant glioma. Oncologist, February 1, 2006; 11(2): 165 - 180. [Abstract] [Full Text] [PDF]

				M. S. Bobola, J. R. Silber, R. G. Ellenbogen, J. R. Geyer, A. Blank, and R. D. Goff O6-Methylguanine-DNA Methyltransferase, O6-Benzylguanine, and Resistance to Clinical Alkylators in Pediatric Primary Brain Tumor Cell Lines Clin. Cancer Res., April 1, 2005; 11(7): 2747 - 2755. [Abstract] [Full Text] [PDF]

				M. E. Hegi, A.-C. Diserens, T. Gorlia, M.-F. Hamou, N. de Tribolet, M. Weller, J. M. Kros, J. A. Hainfellner, W. Mason, L. Mariani, et al. MGMT Gene Silencing and Benefit from Temozolomide in Glioblastoma N. Engl. J. Med., March 10, 2005; 352(10): 997 - 1003. [Abstract] [Full Text] [PDF]

				M. F. Paz, R. Yaya-Tur, I. Rojas-Marcos, G. Reynes, M. Pollan, L. Aguirre-Cruz, J. L. Garcia-Lopez, J. Piquer, M.-J. Safont, C. Balana, et al. CpG Island Hypermethylation of the DNA Repair Enzyme Methyltransferase Predicts Response to Temozolomide in Primary Gliomas Clin. Cancer Res., August 1, 2004; 10(15): 4933 - 4938. [Abstract] [Full Text] [PDF]

				C. Balana, J. L. Ramirez, M. Taron, Y. Roussos, A. Ariza, R. Ballester, C. Sarries, P. Mendez, J. J. Sanchez, and R. Rosell O6-methyl-guanine-DNA methyltransferase Methylation in Serum and Tumor DNA Predicts Response to 1,3-Bis(2-Chloroethyl)-1-Nitrosourea but not to Temozolamide Plus Cisplatin in Glioblastoma Multiforme Clin. Cancer Res., April 1, 2003; 9(4): 1461 - 1468. [Abstract] [Full Text] [PDF]

				J. R. Silber, M. S. Bobola, A. Blank, K. D. Schoeler, P. D. Haroldson, M. B. Huynh, and D. D. Kolstoe The Apurinic/Apyrimidinic Endonuclease Activity of Ape1/Ref-1 Contributes to Human Glioma Cell Resistance to Alkylating Agents and Is Elevated by Oxidative Stress Clin. Cancer Res., September 1, 2002; 8(9): 3008 - 3018. [Abstract] [Full Text] [PDF]

				S. L. Gerson Clinical Relevance of MGMT in the Treatment of Cancer J. Clin. Oncol., May 1, 2002; 20(9): 2388 - 2399. [Abstract] [Full Text] [PDF]

				M. Esteller, G. Gaidano, S. N. Goodman, V. Zagonel, D. Capello, B. Botto, D. Rossi, A. Gloghini, U. Vitolo, A. Carbone, et al. Hypermethylation of the DNA Repair Gene O6-Methylguanine DNA Methyltransferase and Survival of Patients With Diffuse Large B-Cell Lymphoma J Natl Cancer Inst, January 2, 2002; 94(1): 26 - 32. [Abstract] [Full Text] [PDF]

				M. Nakamura, T. Watanabe, Y. Yonekawa, P. Kleihues, and H. Ohgaki Promoter methylation of the DNA repair gene MGMT in astrocytomas is frequently associated with G:C {->} A:T mutations of the TP53 tumor suppressor gene Carcinogenesis, October 1, 2001; 22(10): 1715 - 1719. [Abstract] [Full Text] [PDF]

J. C. Buckner, T. J. Moynihan, D. I. Quinn, U. Schlegel, F. Ali-Osman, K. Srivenugopal, R. Sawaya, M. Esteller, J. G. Herman, and J. N. Weinstein The DNA-Repair Gene MGMT and the Clinical Response of Gliomas to Alkylating Agents N. Engl. J. Med., March 1, 2001; 344(9): 686 - 688. [Full Text] [PDF]