O6-Methylguanine-DNA methyltransferase (MGMT) promoter hypermethylation has recently emerged as a powerful determinant of chemotherapy sensitivity in glioblastomas. To adapt such an important epigenetic biomarker to routine application in the clinical setting, we validated the conventionally used methylation-specific polymerase chain reaction (MSP) assay for its relevance in the determination of MGMT methylation status. MGMT promoter hypermethylation analysis employing MSP was performed on 25 primary glioblastoma samples and 7 cell lines, and compared with the more robust direct promoter sequencing that profiled the methylation status of 27 CpG sites within the MGMT promoter. In addition, the MGMT expression at the protein level was evaluated in the primary tumor samples using immunohistochemistry and in the cell lines using Western blotting analysis. Our MSP analyses yielded reproducible results, which were identical to the bisulfite sequencing data in all except one primary tumor that was negative on MSP. A poor correlation existed between the immunohistochemical staining results and the methylation status of the MGMT promoter in primary glioblastoma samples. Neither MSP-MGMT methylation nor immunohistochemical MGMT expression had prognostic implications in this small and non-uniform group of patients. In all of the cell lines with loss of MGMT expression, signals of methylated DNA were detected by MSP. Our data support the feasibility and reliability of MSP analysis, which could be routinely implemented in the diagnostic setting.