[Frontiers in Bioscience E3, 1182-1191, June 1, 2011]

Prognostic value of H-MLH1 after adjusting for RPA class in GBM patients

Ali Choucair1, Jennifer Moughan2, Chris Schultz3, Alan Schulsinger4, Minesh Mehta5, Walter Curran6

1Intermountain Medical Center, 5171 South Cottonwood Street, Neuroscience Center, Suite 810, Murray, Utah 84107, 2Radiation Therapy Oncology Group, American College of Radiology, 1818 Market Street, Suite 1600, Philadelphia, Pa. 19103-3604, 3Department of Radiation Oncology, Medical College of Wisconsin, 9200 W. Wisconsin Avenue, Milwaukee, WI 53226, 4SUNY, 339 Hicks Str, Brooklyn, N.Y. 11201, 5University of Wisconsin, Human Oncology/Radiation Oncology, 600 Highland Avenue K4/310-3684, Madison, WI 53792, 6Emory University School of Medicine, Radiation Oncology, 1365 Clifton Road NE Rm. A1316, Atlanta, GA 30322


1. Abstract
2. Introduction
3. Materials and methods
3.1. Immnohistochemistry (IHC)
3.2. Statistical methods
4. Results
5. Discussion
5.1. MMR
5.2. MGMT
5.3. p53
5.4. NER
6. Acknowledgements
7. References


Repair of DNA adducts appears to be an important mechanism in chemotherapy responsiveness in glioblastoma multiforme (GBM). Meta-analyses have suggested that the addition of chemotherapy increases the percentage of long-term survivors. Because GBM is characterized by multiplicity of pathways that characterize growth and treatment resistance, we hypothesized probing a multiplicity of repair factors may be able to identify more than one prognostic factor that may be utilized in molecularly targeted therapy that might improve survival and QOL. Seven DNA repair factors showed statistical significance when added to the initial logistic model of RPA class on length of survival status. After adjusting for RPA class the only statistically significant result of the multivariable logistic regressions for these 7 DNA repair factors was that as hMLH1-MF1 increased, the odds of being a short-term survivor versus a long-term survivor decreased (OR: 0.913, 95% CI: 0.838-0.995, p=0.0385), multivariable analysis showed no associations between survival status and MGMT and p53 status, and the only statistically significant prognostic DNA repair factor was human Mut L Homologue 1 (hMLH1).


Annually in the United States there are 22,500 cases of newly diagnosed malignant primary brain tumors in the adults, of which 70% are malignant gliomas. Approximately two thirds of malignant gliomas are GBMs that are characterized by devastating morbidity and a limited median survival with combined treatment of surgery, radiation and chemotherapy of 12-15 months. (1-2)

Meta-analyses have suggested that the addition of chemotherapy to surgery and radiation therapy may increase the one and two year survivals by 5-10% (3-4). A recent randomized trial by the European Organization for Research and Treatment of Cancer (EORTC) and the National Cancer Institute of Canada (NCIC) has established radiotherapy with concomitant and adjuvant temozolomide as the current standard of care for newly diagnosed GBM (5). In this trial the median and 2-year survival rate for radiation alone versus radiation plus temozolomide was 12.1 months versus 14.6 month (p less than 0.001) and 10.4% versus 26.5%, respectively. It has long been clear that considerable variation in tumor response and survival are observed among patients with glioblastoma multiforme (GBM) receiving similar therapy. The Radiation Therapy Oncology Group (RTOG) has previously combined several prognostic factors (age, Karnofsky performance status (KPS), and extent of surgical resection) in formulating prognostic categories utilizing recursive partitioning analysis (RPA) classes (6). While such clinical observations are valuable, there is clearly a need for further work aimed at elucidating the molecular mechanisms (molecular heterogeneity, invasive behavior, and multiplicity of growth pathways) underlying the differences among patients by identifying molecular factors with prognostic and predictive value. It is anticipated that such research may lead to increased refinement in prognostication and tailoring of therapy with resulting improvement in the outcomes of survival, toxicity, and quality of life. Methylating agents produce cytotoxic product O6 methylguanine DNA adduct, which initiates mismatch repair (MMR) pathway cycling, resulting in apoptotic cell death. Intact MMR is required for tumor cell kill by the two most commonly utilized classes of chemotherapy in the treatment of GBM; namely methylating (e.g. temozolomide and procarbazine) and alkylating (e.g. nitrosurea) agents.

Previous work has suggested that the ability of tumor cells to repair nitrosourea induced DNA adducts (see Figure 1) may be an important mechanism in predicting resistance to these agents. As shown in Table 2, there are six potentially important repair factors for nitrosurea-induced DNA damage (7). Clinical studies have correlated tumor cell levels of MGMT, a repair factor that can remove the chloroethyl adduct on the O-6 position of guanine, with response to nitrosurea and temozolmide and survival in patients with malignant gliomas (8-11). In vitro studies utilizing a variety of cell lines have suggested that deficiencies in any of several repair pathways render cells more sensitive to nitrosoureas. These pathways include MGMT, base excision repair (BER), nucleotide excision repair (NER), mismatch repair (MMR), and p53 (12-16). Unfortunately, the repair pathways involved in the nitrosourea induced guanine-cytosine DNA interstrand cross-link have not been elucidated. However, studies of other types of interstrand cross-links have suggested that homologous recombination repair (HHR) may be important (17).

In this study, we have examined the predictive value for survival of four factors that have been associated with nitrosourea sensitivity (MGMT; the DNA mismatch repair factors, human MutS homolog (hMSH2) and human MutL homolog (hMLH1); and p53) and four factors that have been associated with sensitivity to DNA interstrand cross-linking agents (excision repair cross-complimenting 1 and 4 (ERCC1, ERCC4 (XPF)), x-ray repair cross-complimenting 2 and 3 (XRCC2, XRCC3)) by comparing their expression in the tumor cells of 28 long term survivors ( greater than 18 months) with those of 35 short term survivors (less than 6 months) (16).


We retrieved from the RTOG tumor repository 63 pretreatment tumor samples from patients with newly diagnosed GBM who were enrolled on five Radiation Therapy Oncology Group (RTOG) protocols and had been treated with combined radiation and nitrosurea chemotherapy. Survival time was calculated from the date of study registration to the time of death for short-term survivors, and to the time of death or last follow-up for long-term survivors. The tissue microarray was conducted on the basis of statistical review of protocols from which they were derived to identify patients who had survived either longer than 18 months (n=28) or less than 6 month (n=35). Data on patient prognostic factors and treatment were available from the RTOG clinical database but were not provided to the investigators during the study. Tumor microarrays were prepared from random areas of the tumor block selected by a pathologist for inclusion. Pathologist selected and marked all areas of block with viable tumor of greater than 75% without significant necrosis or inflammation. Three random areas from each block were selected to construct three replicate arrays, based on our previously published work on appropriate sampling techniques (18).

3.1. Immnohistochemistry (IHC)

Sections of the paraffin-embedded tumor tissue array were immunostained using an immunoperoxidase method with 3,3 diaminobenzidine as the substrate. Staining was performed on a DAKO Autostainer. Mouse monoclonal antibodies against human MGMT (Chemicon); MSH2, MLH1 (PharMingen); p53 (Dako); ERCC1, ERCC4, XRCC2, and XRCC3 (NeoMarkers) were used as the primary antibody (Table 1). The secondary antibody was a biotinylated goat antimouse antibody (DAKO; prediluted). Antibody binding was detected using horseradish peroxidase (LSAB2 System; DAKO). Appropriate controls were stained simultaneously. The slides were counterstained with hematoxylin (DAKO), dehydrated, and mounted. The percentage of positively stained cells and the mean intensity of staining were determined using an automated cell imaging system (ACIS; Chroma Vision). We have previously published several studies using this system (19-20). The tissue array stained with each antibody is digitally captured and presented to the pathologist. Area of tumor is selected for counting based on morphology. The threshold to optimize scoring was previously defined using digitally captured positive control materials. Negative controls for each assay were also evaluated. Table 1 shows the antibody clones, pretreatments, dilutions and controls used. The tissue microarray was enriched to include patients who had survived either longer than 18 months (long-term survival (LTS), n=28) or less than 6 months (short-term survival (STS), n = 35). Clinical data were available from the RTOG database. Samples were evaluated for expression of MGMT, hMLH1, hMSH2, p53, ERCC1, ERCC4, XRCC3 as determined by quantitative IHC.

Immunostaining was carried out as follows: Heat slides for 10 minutes in an 85-90 degrees C oven, Deparaffinize (rehydrate) through two 5 minute changes xylene, 2 rinses absolute ethanol, several dips in 95% ethanol, then rinse well in water. Place slides into citrate buffer for antigen retrieval into the pressure cooker with for 50 minutes. Rinse slides with tap water. Tap off excess water and carefully wipe around specimen and apply enough 3% hydrogen peroxide to cover specimen for 5 minutes at room temperature. Rinse slides in Tris Buffer. Wipe excess buffer from slides and set slides onto humidity chamber. Overlay PRIMARY antibody (See Table 1). Rinse in Tris Buffer. Overlay LINK (Biotin) for 10 minutes. (Dako LSAB II Kit) Rinse slides in Tris Buffer. Overlay LABEL (Streptavidin) and incubate for 10 minutes. (Dako LSAB II Kit) Rinse with Tris. Apply DAB (chromogen) for 5 minutes. (Research Genetics) Cover slides with commercially prepared hematoxylin (Dako) for approximately 30 seconds. Dehydrate and cover slip with Permount. All stains were performed on the Dako Autostainer using the same method as stated above.

3.2. Statistical methods

Frequency tables with counts and percentages and summary statistics (median and range) were used to describe pretreatment characteristics and the DNA repair factors for each group. Logistic regression models were used to identify associations of DNA repair factors with length of survival status (short-term vs. long-term). Since RPA class is a combination of several independent prognostic factors (age, KPS, neurological function, extent of resection, RT dose, and mental status), all multivariate models were adjusted for RPA class. The initial model is the logistic regression of RPA class on length of survival status. To select potentially important variables, the DNA repair factors were tested one at a time in the logistic model with RPA class. The difference between the initial model and the expanded model with the DNA repair factor has an approximate Chi-square distribution with one degree of freedom. Using a criterion of 0.10 for variable selection, those DNA repair factors that were found to be statistically significant were investigated. The odds ratios (OR) for each variable in the final multivariable logistic regression model along with their 95% confidence intervals (C.I.) and p-values are reported.


Patient pretreatment characteristics for the short-term (STS) and long-term survivors (LTS) are shown in Table 3. All patients received a nitrosourea, mostly BCNU (1,3-bis (2-chloroethyl)-1-nitrosourea) as part of a multidisciplinary therapeutic approach for their GBM. The descriptive statistics of tumor cells expressing DNA repair factors and the intensity of expression for the positive cells between the two groups are shown in Table 4. There was a wide range for each factor among the individual tumor specimens. We hypothesized that because nitosurea DNA alkylation is a sequential process, which begins at guanine O-6 chlorethyl alkylation and concludes with DNA interstrand cross-linking (see Figure 1), the effects of downstream repair factors may be more difficult to detect in the presence of high levels of upstream repair factors.

Table 5 shows the results of the model fitting when each DNA repair factor was added one at a time to the initial logistic model of RPA class on length of survival status. Seven DNA repair factors showed statistical significance at the 0.10 level when added to the initial model. Table 6 contains the results of the multivariable logistic regression models for these 7 DNA repair factors. The only statistically significant result was that as hMLH1-MF1 increased, the odds of being a short-term survivor versus a long-term survivor decreased after adjusting for RPA class (OR: 0.913, 95% CI: 0.838-0.995, p=0.0385). Specifically, an increase in the hMLH1-MF1 by 1 unit meant a decrease in the odds of being in the STS group by 9% after adjusting for RPA class (Table 6). Multivariable analysis showed no associations between survival status and MGMT and p53 status. The small number of patients in each survivor group made it neither possible nor meaningful to look at the DNA repair factors jointly.


In this study we have examined the possible roles of several DNA repair proteins as predictive factors for survival among patients with GBM receiving post-operative irradiation and nitrosoureas. Nitrosoureas are felt to produce their tumor cell cytotoxic effects by DNA damage, which leads to cell death through necrosis or induction of apoptosis. There are several normal cellular responses to DNA alkylation which include cell cycle arrest, attempts at repair, and an upregulation or heightened sensitization of the cellular apoptotic apparatus. Important mechanisms of tumor cell drug resistance include alterations in these normal pathways such as an increased efficiency of lesion repair or a decreased ability to signal apoptosis. DNA repair factors that have higher expression in short term GBM survivors are most likely acting through enhanced repair while those higher in the long survivors may be primarily inducing tumor cell death. Thus, if inclusion of chemotherapy increases the chances of long term survival for patients with GBM, it is reasonable to investigate chemotherapy resistance mechanisms as possible predictive factors for survival.

One strategy for selective and efficient tumor therapy is for DNA repair modulation to be targeted against tumor cells with suboptimal DNA repair. The multiplicity of GBM's molecular pathways necessitates the evaluation of more than one predictive or molecular marker. A feature that is common to human cell DNA repair is the redundancy it has for the removal of many lesions. Loss of one or more repair pathway does not fully disable the repair process but makes the tumor more dependent upon the remaining pathways for its growth. This could lead to exploiting the "Achilles' heel" of certain tumors: targeting the remaining pathways, upon which the tumor's growth is dependent, should improve the chances for better response (21).

5.1. MMR

One of the mechanisms of resistance to alkylating agents involves DNA mismatch repair pathways. Deficiency in MMR pathways can endow tumor cells with resistance to the cytotoxic effects of alkylating agents: This cytotoxic resistance can be explained as follows: The O-6-methylgunanine lesion can initiate apoptosis through sensing by the DNA MMR apparatus and signaling through the mitochondrial apoptosis pathway in a p53-dependent and independent manner (22). The MMR pathway is critical in mediating the cytotoxic effect of O-6-methylguanine. The MMR pathway, which is composed of several proteins (hMLH1, hMSH2, hMSH3, hMSH6, and hPMS2 (see Table 2) is programmed to correct errors in DNA base pairing that arise during DNA replication. During DNA replication, DNA polymerase mispairs O-6-methylguanine with thymine. This mispairing triggers MMR-dependent removal of the mispaired thymine leaving the O-6-methylguanine, which causes subsequent mispairing with yet another thymine leading to futile cycling of the MMR system, which causes accumulation of DNS double-strand breaks, which trigger p53-dependent cell cycle arrest and apoptosis (23). Thus tumors that are deficient in MMR are relatively resistant to alkylating agents such as nitrosureas, temozolomide, and procarbazine. MMR deficiency can be caused by mutations in the hMLH1 or hMSH2 genes (hereditary nonpolyposis colon cancer) or due to methylation of the hMLH1 gene promoter, which is not known to occur in malignant gliomas and is known to occur in only five of the 60 cell lines in the National Cancer Institute tumor panel (24).

A paradoxical relationship between MMR and drug sensitivity has been observed in studies in MMR-deficient cell lines. Some tumor cell lines defective in MMR have been reported to show greater sensitivity to nitrosoureas. These findings have led to the conclusion that MMR may be active in nitrosourea lesion repair, although other explanations such as effects of MMR on the G2/M cell cycle checkpoint are also possible (15-25-26). In our study, conversely, we observed no subsets where MMR expression was higher in the short term survivors. Rather, our observations that MMR expression is lower in the short term GBM survivors irrespective of MGMT and p53 expression suggest that MMR is functioning through another mechanism, most likely, as explained above, through DNA lesion sensing and initiation of the apoptosis signaling pathways. Although MGMT can repair the cytotxic damage caused by alkylating agents, yet resistance to these agents can still be shown in tumors with low level of MGMT and MMR-deficiency ( see discussion under MGMT) (27).

Our results confirmed the prognostic value of hMLH1 previously reported by other investigators. The MMR gene hMLH1 maintains genomic integrity by mediating the activation of cell cycle checkpoints and apoptosis. It may be active in nitrosurea lesion repair, predicts the clinical response of malignant astrocytomas to nitrosureas, recognizes the O-6-methylguanine DNA adducts formed by methylation agents (procarbazine and temozolomide), and potently triggers the apoptotic pathway (15-25-28-29). In its absence, neither temozolomide nor BCNU can activate apoptosis (23). Potential models, with emphasis upon hMLH1, need to be further investigated for possible interactions between DNA repair factors, their pathways, the lesions that are recognized and repaired by them, and their relevance to the choice of chemotherapy.

5.2. MGMT

The most important initial DNA lesion induced by BCNU appears to be a chloroethyl adduct on the 0-6 position of guanine. This lesion can be removed by MGMT, the DNA repair factor that has received the most attention in GBM patients to date. The O-6 adduct is capable of activating apoptosis, and this signaling is initiated by MMR, possibly through recognition of a mismatched base incorporated into the complimentary DNA strand (23). Thus, it is of interest that our study failed to show a difference in expression of MGMT between the short and long surviving groups (Table 4). While this may have been due, at least in part, to statistical power, it is possible that the effect of MGMT is complex and related to the functioning of other factors. It is known that tumors with low levels MGMT can still show resistance to the cytotoxic effects of alkylating agents suggesting other mechanisms of resistance (30). Some of those mechanisms include: MMR deficiency, p53 mutations, overexpression of anti-apoptotic proteins (Bcl-2 or Bcl-X1), or active BER pathway (27-31). As has been reported by others an important interaction between MGMT and MMR does exist in patients receiving temozolomide (30). Several prior studies have suggested that MGMT expression does correlate with response and survival in patients with gliomas (8-9-10). However, these studies have included patients with grade 3 and 4 gliomas and sometimes low grade tumors. Among these studies, only Jaeckle, et al. has presented a subset analysis by tumor grade in patients receiving nitrosoureas. These authors found that MGMT levels appeared to be considerably more predictive of survival in patients with anaplastic astrocytoma (median survivals 8 vs. 29 months for high and low MGMT) than GBM (7 vs. 12 months).

Studies examining MGMT promoter methylation also present some difficulties in interpretation. A negative correlation between methylation (presumably associated with lower MGMT levels) and survival was found in a population of patients with grade 3 and 4 tumors, and in patients with low grade astrocytomas (9-32). However, in patients with GBM receiving predominantly nitrosoureas as chemotherapy, one study found no correlation (33). Conversely, in GBM patients receiving temozolomide, a strong relationship was shown (11-34). In sum, these results suggest that pre-treatment MGMT status may have a predictive value for patients with GBM receiving nitrosoureas and maybe more important in patients receiving O-6 guanine methylating agents (34). The recently completed international GBM trial (EORTC/RTOG 0525) will help in further clarifying this issue.

5.3. p53

Inactivation of p53 has been associated with both increased and decreased sensitivity to nitrosoureas in vitro (16-25). Likewise, p53 has been evaluated in a number of previous studies as an individual prognostic factor for patients with GBM with inconsistent results (35). These complex variations may indicate that p53 is important but only in the context of intact relevant signaling pathways. p53 expression is usually difficult to detect by IHC unless the molecule has been stabilized by mutation or cellular stress such as chemotherapy exposure. In these instances, p53 binding by the E-3 ubiquitin ligase, MDM2, is prevented. Therefore, high p53 expression by IHC in unstressed cells has been suggested as a marker for missense mutated, and presumably non-functional, p53. In gliomas, as in other tumors, the concordance between high IHC expression and mutation is about 70-80% (36-37).

Although not a significant prognostic factor individually in our study, p53 appears to have important interactions with MMR and ERCC1. As with other DNA repair factors, p53 has a complex and incompletely understood relationship with MMR, likely involving other molecules such as ATM/ATR (38-39). MMR and p53 participate in nitrosourea sensitivity pathways in complex ways that are dependent and independent of each other and yet to be fully elucidated (23). MMR/DNA interactions can result in phosphorylation of p53 on serine residues 15 and 392 with resultant stabilization (40). In addition, at least one pathway of MMR related apoptosis signaling goes through p53 (41). And recent data from our group and others have presented evidence that p53 may be a transcription factor for hMSH2, possibly participating in a feedback loop (42).

5.4. NER

NER factors ERCC1 and ERCC4 recognize and repair bulky DNA adducts (see Table 2). ERCC1 functions in the 5'-DNA nicking step of NER and may serve a similar role in HRR. Deficiencies of ERCC1 and ERCC4 seem to be most predictive of increased cyclophosphamide sensitivity in a panel of CHO excision repair mutant cell lines (43). Unfortunately, there is little information regarding these two repair pathways in the repair of nitrosourea induced DNA adducts. One group has suggested that expression of ERCC2, a component of NER, correlates in vitro with resistance to BCNU (14). Alternatively, one group proposed a model of DNA interstrand cross-link repair that involves the ERCC1 and ERCC4 driver through a mechanism of recombinational repair (44). The relationship of ERCC1/ERCC4 to survival will need to be investigated in series larger than ours. The question of the relative contribution of HRR to nitrosourea sensitivity is also complicated by the observation that HRR is also important in the repair of double-strand DNA breaks induced by radiation a treatment also received by our patients (45).

Limitations in this study include its retrospective nature and its small sample size. Retrospective studies have the flaw of not being able to adequately offer a satisfactory accounting or explanation for selection bias. The small sample is partly inherent in the fact that patients with GBMs have a short median survival time so it is a challenge to obtain pathology specimens on an adequate number of long-term survivors. This study was a retrospective analysis and, therefore, was not designed to take statistical power into consideration. It needs to be emphasized that just because some associations were not found between the DNA repair factors and length of survival does not mean that these associations do not exist. A prospective study needs to be done to insure enough statistical power to find these associations if they truly exist.


Presented as "Select Abstract" at the Thirteenth Annual Conference of the Society of Neuro-oncology, Las Vegas, Nevada, November, 2008. We acknowledge with gratitude the technical assistance of Janet Hansen and Bashar Dabbas, M.D and the contributions of Dr Clyde Ford to the original study design. Supported by RTOG U10CA21661, CCOP U10CA37422, and Stat U10CA32115 grants from NCI and in part by a generous grant from Intermountain Healthcare Deseret Foundation (AC). No actual or potential conflicts of interest exist with regard to this publication.


1. DN Louis, H Ohgaki, OD Wiestler, WK Cavenee, PC Burger, A Jouvet, B W Scheithauer, P Kleihues : The 2007 WHO classification of tumors of the central nervous system. Lyon, France, IARC Press, (2007)

2. CBTRUS 2008 statistical report: primary brain tumors in the United States, 1998-2002: Central Brain Tumor Registry of the United States, http://www.cbtrus.org (2000-2004)

3. HA Fine, KB Dear, JS Loeffler, PM Black, GP Canellos: Meta-analysis of radiation therapy with and without adjuvant chemotherapy for malignant gliomas in adults. Cancer 71, 2585-2597 (1993)

4. LA Stewart: Chemotherapy in adult high-grade glioma: a systematic review and meta-analysis of individual patient data from 12 randomized trials. Lancet 359, 1011-1018 (2002)

5. R Stupp, WP Mason, MJ van den Bent, M Weller, B Fisher, MJ Taphoorn, K Belanger, AA Brandes, C Marosi, U Bogdahn, J Curschmann, RC Janzer, SK Ludwin, T Gorlia, A Allgeier, D Lacombe, JG Cairncross, E Eisenhauer, RO Mirimanoff: Eirpoean Organization for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups; National Cancer Institute of Canada Clinical Trials Groups. Radiotherapy plus concomitant and ajuvant temozolomide for glioblastoma. N Engl J Med 352:987-96 (2005)

6. WJ Curran Jr, CB Scott, J Horton, JS Nelson, AS Weinstein, AJ Fischbach, CH Chnag, M Rotman, SO Asbell, RE Krisch: Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group malignant glioma trials. J. Natl. Cancer Inst 85, 704-710 (1993)

7. S Madhusudan, MR Middleton : The emerging role of DNA repair proteins a predictive, prognostic and therapeutic targets in cancer. Cancer Treat.Rev 31:603-617 (2005)

8. M Belanich, M Pastor, T Randall, D Guerra, J Kibitel, L Alas, B Li, M Citron, P Wasserman, A White, H Eyre, K Jaeckle, S Schulman, D Rector, M Prados, S Coons, W Shapiro, D Yarosh: 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)

9. M Esteller, J Garcia-Foncillas, E Andion, SN Goodman,OF Hidalgo, V Vanaclocha, SB Baylin, JG Herman: Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. New Eng. J. Med. 343, 1350-1354 (2000)

10. KA Jaeckle, HJ Eyre, JJ Townsend, S Schulman, HM Knudson, M Belanich, DB Yarosh, SI Bearman, DJ Giroux, SC Schold: Correlation of tumor O6 methylguanine-DNA methyltransferase levels with survival of malignant astrocytoma patients treated with bis-chloroethylnitrosourea: A Southwest Oncology Group Study. J. Clin. Oncol. 16, 3310-3315 (1998)

11. ME Hegi, AC Diserens, S Godard, PY Dietrich, L Regli, S Ostermann, P Otten, G Van Melle, N de Tribolet, R Stupp et al: Clinical trial substantiates the predictive value of O-6-methylguanine-DNA methyltransferase promoter methylation in glioblastoma patients treated with temozolomide. Clin. Cancer Res. 10:1871-1874 (2004)

12. WJ Bodell, K Tokuda, DB Ludlum: Differences in DNA alkylation products formed in sensitive and resistant human glioma cells treated with N-(2-chloroethyl)-N-nitrosourea. Cancer Res.48:4489-4492 (1998)

13. BP Engelward, A Dreslin, J Christensen, D Huszar, C Kurahara, L Samson: Repair-deficient 3-methyladenine-DNA glycosylase homozygous mutant mouse cells have increased sensitivity to alkylation-induced chromosome damage and cell killing. EMBO. J. 15:945-952 (1996)
PMid:8631315    PMCid:450292

14. ZP Chen, A McQuillan, G Mohr, LC Panasci: Excision repair cross-complementing rodent repair deficiency gene 2 expression and chloroethylnitrosourea resistance in human glioma cell lines. Neurosurg. 42:1112-1119 (1998)

15. G Aquilina, S Ceccotti, S Martinelli, R Hampson, M Bignami: M. N-(2-chloroethyl)-N'-cyclohexyl-N-nitrosourea sensitivity in mismatch repair-defective human cells. Cancer Res. 58:135-141 (1998)

16. GW Xu, CL Nutt, MC Zlatescu, M Keeney, I Chin-Yee, JG Cairncross: Inactivation of p53 sensitizes U87MG glioma cells to 1,3-bis(2-chloroethyl)-1-nitrosourea. Cancer Res. 61:4155-4159 (2001)

17. IU De Silva, PJ McHugh, PH Clingen, JA Hartley: Defining the roles of nucleotide excision repair and recombination in the repair of DNA interstrand cross-links in mammalian cells. Mol. Cell. Biol. 20:7980-7990 (2000)
PMid:11027268    PMCid:86408

18. M Milanes-Yearsley, ME Hammond, TF Pajak, JS Cooper, C Chang, T Griffin, D Nelson, G Laramore, M Pilepich: Tissue micro-array: a cost and time-effective method for correlative studies by regional and national cancer study groups. Mod Pathol 15(12):1366-73 (2002)

19. A Pollack, M DeSilvio, LY Khor, R Li, TI Al-Saleem, ME Hammond, V Venkatesan, CA Lawton, M Roach 3rd, WU Shipley, GE Hanks, HM Sandler: Ki-67 staining is a strong predictor of distant metastasis and mortality for men with prostate cancer treated with radiotherapy plus androgen deprivation: Radiation Therapy Oncology Group Trial 92-02.J Clin Oncol. 1; 22(11):2133-40 (Jun 2004)

20. A Chakravarti, K Heydon, CL Wu, E Hammond, A Pollack, M Roach, H Wolkov, P Okunieff, J Cox, J Fontanesi, R Abrams, M Pilepich, W Shipley: Radiation Therapy Oncology Group. Loss of p16 expression is of prognostic significance in locally advanced prostate cancer: an analysis from the Radiation Therapy Oncology Group protocol.86-10. J Clin Oncol. 1;21(17):3328-34 (Sep 2003)

21. S Madhusudan, ID Hickson: DNA repair inhibition: a selective tumor targeting strategy. Trends in Molecular Medicine 11(11):503-511 (2005)

22. S D'Atri, L Tentori, PM Lacal, G Graziani, E Pagani, E Benincasa, G Zambruno, E Bonmassar, J Jiricny: Involvement of the mismatch repair system in temozolomide-induced apoptosis. Mol Pharmacol 54:334-341 (1998)

23. MJ Hickman, LD Samson: Role of DNA mismatch repair and p53 signaling induction of apoptosis by alkylating agents. Proc. Natl. Acad. Sci. USA 96:10764-10769 (1999)

24. P Taverna, L Liu, AJ Hanson, A Monks, SL Gerson, et al: Characterization of MLH1 and MSH2 DNA mismatch repair proteins in cell lines of the NCI anticancer drug screen. Cancer Chemother Pharmacol 46:507-16 (2000)

25. G Aquilina, S Ceccotti, S Martinelli, S Soddu, M Crescenzi, P Branch, P Karran, M Bignami: Mismatch repair and p53 independently affect sensitivity to N-(2- chloroethyl)-N'-cyclohexyl-N-nitrosourea. Clin. Cancer Res. 6:671-680 (2000)

26. Y Hirose, M Katayama, D Stokoe, DA Haas-Kogan, MS Berger, RO Pieper: The p38 mitogen-activated protein kinase pathway links the DNA mismatch repair system to the G2 checkpoint and to resistance to chemotherapeutic DNA-methylating agents. Mol. Cell Biol. 23:8306-8315 (2003)
PMid:14585987    PMCid:262371

27. L Liu, SL Gerson : Targeted modulation of MGMT : clinical implications. Clin Cancer Res 12 :328-31(2006)

28. T Fukushima, Y Katayama, T Watanabe, A Yoshino, A Ogino, T Ohta, C Komine: Promotor hypermethylation of mismatch repair gene hMLH1 predicts the clinical response of malignant astrocytomas to nitrosourea. Clin Cancer Res 11:1539-1544 (2005)

29. D Fink, S Aebi, SB Howell: The role of DNA mismatch repair in drug resistance. Clin Cancer Res 4:1-6 (1998)

30. HS Friedman, RE McLendon, T Kerby, M Dugan, SH Bigner, AJ Henry, DM Ashley, J Krischer, S Lovell, K Rasheed, F Marchev, AJ Seman, I Cokgor, J Rich, E Stewart, OM Colvin, JM Provenzale, DD Bigner, MM Haglund, AH Friedman, PL Modrich: J Clin Oncol. 16(12):3851-3857 (1998 Dec)

31. JN Sarkaria, GJ Kitange, D James, R Plummer, H Calvert, M Weller, W Wick: Mechanisms of chemoresistance to alkylating agents in malignant glioma. Clin Cancer Res. 14:2900-2908 (2008)
PMid:18483356    PMCid:2430468

32. C Komine, T Watanabe, Y Katayama, A Yoshino, T Yokoyama, T Fukushima: Promoter hypermethylation of the DNA repair gene 06-methylguanine-DNA methytransferase is an independent predictor of shortened progression free survival in patients with low-grade astrocytomas. Brain Pathol. 13:176-184 (2003)

33. H Kobayashi, N Ishii, J Ikeda, Y Sawamura, Y. Iwasaki: Methylation profiling of the DNA repair genes in glioblastoma multiforme and its clinical relevance. Proc. Amer. Soc. Clin. Oncol. 23, 125 (2004)

34. ME Hegi, AC Diserens, T Gorlia, MF Hamou, N de Tribolet, M Weller, JM Kros, JA Hainfellner, W Mason, L Mariani, JE Bromberg, P Hau, RO Mirimanoff, JG Cairncross, RC Janzer, R Stupp: MGMT gene silencing and benefit from temozolomide in glioblastoma. N. Engl J Med 352:997-1003 (2005)

35. C Nieder, S Petersen, C Petersen, H Thames: The challenge of p53 as prognostic and predictive factor in gliomas. Cancer Treat. Rev. 26:67-73 (2000)

36. EW Newcomb, WJ Madonia, S Pisharody, FF Lang, M Koslow, DC Miller: A correlative study of p53 protein alteration and p53 gene mutation in glioblastoma multiforme. Brain Pathol. 3:229-235 (1993)

37. H Koga, S Zhang, T Kumanishi, K Washiyama, T Ichikawa, R Tanaka, J Mukawa: Analysis of p53 gene mutations in low- and high-grade astrocytomas by polymerase chain reaction-assisted single-strand conformation polymorphism and immunohistochemistry. Acta. Neuropathol 87:225-232 (1994)

38. Y Luo, FT Lin, WC Lin: ATM-mediated stabilization of hMut L DNA mismatch repair proteins augments p53 activation during DNA damage. Mol. Cell Biol. 24:6430-6444 (2004)
PMid:15226443    PMCid:434232

39. S Caporali, S Falcinelli, G Starace: DNA damage induced by temozolomide signals to both ATM and ATR: role of the mismatch repair system. Mol. Pharmacol. 66:478-491 (2004)

40. DR Duckett, SM Bronstein, Y Taya, P Modrich: hMutSalpha-and hMutLalpha-dependent phosphorylation of p53 in response to DNA methylator damage. Proc. Natl. Acad. Sci. USA 96:12384-12388 (1999)

41. GM Li: The role of mismatch repair in DNA damage-induced apoptosis. Oncol. Res. 11:393-400 (1999)

42. CT Warnick, B Dabbas, CD Ford, KA Strait: Identification of a p53 response element in the promoter region of the hMSH2 gene, required for expression in A2780 ovarian cancer cells. J. Biol. Chem. 276:27363-27270 (2001)

43. BS Andersson, T Sadeghi, MJ Siciliano, R Legerski, D Murray: Nucleotide excision repair genes as determinants of cellular sensitivity to cyclophosphamide analogs. Cancer. Chemother. Pharmacol. 38:406-416 (1996)

44. T Bessho, D Mu, A Sancar: Initiation of DNA interstrand cross-link repair in humans: the nucleotide excision repair system makes dual incisions 5" to the cross-linked base and removes a 22-28-nucleotide-long damage-free strand. Mol. Cell Biol. 17:6822-6830 (1997)
PMid:9372913    PMCid:232538

45. A Pastink, JCJ Eeken, PHM Lohman: Genomic integrity and the repair of double-strand DNA breaks. Mut. Res. 480:37-50 (2001)

Key Words: GBM, DNA repair factors, hMLH1, RPA Class, Long-Term Survivor

Send correspondence to: Ali Choucair, Intermountain Medical Center Neuroscience Clinic, 5171 So. Cottonwood Street, Suite 810, Murray, Utah 84107, Tel: 801-507-9825, Fax: 801-507-9841, E-mail:akiahc@gmail.com