Tumour and serum MGMT promoter methylation and protein expression in glioblastoma patients.

INTRODUCTION: Methylation of the promoter of the MGMT gene and MGMT protein expression are recognized as predictive markers for response to alkylating chemotherapy in glioblastoma (GB). MATERIAL AND METHODS: We have assessed MGMT methylation with the methylation-specific polymerase chain reaction (MSP) in tumor samples from 70 GB patients and in serum samples from 37 of these patients. We have also assessed MGMT protein expression by immunohistochemical (IHC) analysis in tissue samples from 63 of these patients. RESULTS: We found concordance between MGMT methylation status in tissue and serum (Cohen's Kappa = 0.586; p<0.0001). MSP for detection of non-methylated MGMT promoter in serum showed a sensitivity of 95.4% and a specificity of 60%, while the IHC methylation test showed a low specificity (8.9%). Patients whose MGMT promoter was methylated in tissue attained longer progression-free and overall survival. In the multivariate analysis, serum MGMT promoter methylation emerged as an independent factor for longer progression-free and overall survival. CONCLUSION: Serum-based MGMT methylation analysis offers a promising alternative to tumor-based MGMT analysis in cases where tissue samples are unavailable.

Evaluation of O6-methylguanine-DNA methyltransferase by immunohistochemistry: best clinical and research practices.

O6-Methyguanine-DNA methyltransferase (MGMT) repairs DNA damage and acts as a tumor suppressor in normal cells by preventing DNA mutations. Several antibodies against MGMT are used for immunohistochemical assessment of this marker and no universal standard is adopted. We evaluated the immunohistochemical expression of MGMT with 5 commercially available primary antibodies in 59 invasive breast carcinomas. A tissue microarray was constructed using 59 invasive breast carcinoma samples. Five primary antibodies against MGMT were used for immunohistochemistry, including clones MT3.1, SPM287, and MT23.2. Heat-induced antigen retrieval and polymer-based immunohistochemistry were performed. Stains were analyzed by microscopy and automated digital slide technology. qRT-PCR was performed for all tumors. Clone SPM287 had the highest sensitivity (p<0.001), and clone MT3.1 had the lowest sensitivity (p<0.001). Clone MT23.2 generated higher levels of cytoplasmic staining, which was not observed with the other antibodies (p<0.001). Fifty-six samples (94.9%) showed hypoexpression of MGMT compared with normal breast, as measured by qRT-PCR (p<0.001). Only clone SPM287 correlated significantly with the qRT-PCR results (p=0.027). Antibody clone SPM287 is the most sensitive and specific antibody for the immunohistochemical evaluation of MGMT, rendering it a reliable and effective reagent for research and clinical practice in breast cancer.

Correspondence between radioactive and functional methods in the quality control of DNA restriction and modifying enzymes.

We evaluated the use of two radiolabeled lambda DNA/Hpa II substrates to detect 5'-->3', 3'-->5' single and double stranded DNA dependent exonuclease and phosphatase activities found as contaminants in restriction and modifying enzyme preparations. Looking for the meaning of the radioactive assays results in a real cloning experience, we performed a cloning simulation assay using the same conditions established for the radioactive assay (enzyme units and pmols of DNA ends). As a result, we found that for degradation percentages of the radioactive DNA substrate per enzyme unit below 0.5, the false positives in the cloning stimulation assay were less than 5%. This conditions could ensure a good performance of the enzyme preparations for cloning experiments. Finally, we described the use of the radiolabeled [gamma 33P] ATP lambda Hpa II DNA substrate to detect 5'-->3' single stranded DNA dependent exonuclease and phosphatase contaminating activities in some critical steps of the purification process of the restriction enzyme Kpn I.