Prognostic value of O‐6‐methylguanine–DNA methyltransferase (MGMT) protein expression in glioblastoma excluding nontumour cells from the analysis

    loading  Checking for direct PDF access through Ovid


The prognosis in patients with glioblastoma multiforme (GBM) remains poor despite current standard therapy consisting of surgery, radiotherapy, and concomitant and adjuvant temozolomide (TMZ) 1. This regimen was introduced by Stupp et al. 2, and in the same patients Hegi et al. 3 showed that epigenetic silencing of the O‐6‐methylguanine–DNA methyltransferase (MGMT) gene is associated with increased overall survival (OS) in patients receiving TMZ.
Studies 4 have emphasised the importance of MGMT status when choosing treatment. These studies used a methylation‐specific polymerase chain reaction (MSP) assay to determine MGMT promoter status. This qualitative method is difficult and time consuming, and other detection methods have been proposed 6. However, the use of different methods provides different results 7. Noting that MGMT repairs mutagenic DNA lesions and that the MGMT protein level therefore is supposed to be of major importance, recent studies 12 suggested that information regarding both MGMT protein expression level and MGMT promoter methylation status optimises the prediction of the response to TMZ.
GBMs are characterised by a rich tumour microenvironment 14, and it has been reported that tumour‐associated microglia and macrophages constitute 30% of surgically removed tissue 15. These cells express the MGMT protein, thereby complicating the interpretation of MGMT immunohistochemical (IHC) staining. In a study by Burke et al. 20 an immunohistochemical double‐staining was performed followed by visual separation of tumorous and nontumorous elements. Based on the median, patients were divided into MGMT protein positive or MGMT protein negative. The results were then compared to MGMT methylation status obtained by methylation‐specific multiplex ligation‐dependent probe amplification (MS‐MLPA). The authors showed good correlation between the two different assays, indicating that differentiation of MGMT expression in tumour and nontumour nuclei are clinically important. Based on our previous studies 21, we developed a double immunofluorescence assay identifying MGMT protein and nontumour cells giving us the opportunity to exclude nontumour cells from our analysis. Moreover; in order to avoid inter‐ and intra‐observer variability we used automated image acquisition followed by an automated quantitative approach, which has been reported previously by our group 21.
Based on our automated quantitative measurements of MGMT protein expression, the aim of this study was to evaluate the prognostic value of MGMT protein expression using a double immunofluorescence staining and a quantitative estimation of the MGMT protein level. The method allowed exclusion of MGMT protein expression in nontumour cells. We hypothesised that evaluation of MGMT in tumour cells only would strengthen the prognostic importance of MGMT. Furthermore, the prognostic importance of combining information on MGMT protein level and MGMT promoter methylation status was evaluated. In order to validate our results in an independent cohort we used tissue from GBM patients included in a collaborative Nordic Study (NS), where patients with WHO grade 3 and 4 gliomas were treated with radiotherapy and different combinations of TMZ.
    loading  Loading Related Articles