Prognostic significance of MGMT promoter methylation in glioblastomas treated with temozolomide and radiotherapy
Carol Walker1, Julie Dunn1, Atik Baborie3, Farida Alam2, Kathyrn Joyce1, Daniel Crooks3, Ross Sibson1, David Husband2, Aditya Shenoy2, Andrew Brodbelt3, Helen Wong2, Lakis Liloglou1, Brian Haylock2
1School of Cancer Studies, Liverpool, UK, 2Clatterbridge Centre for Oncology, Bebington, UK, 3Walton centre for Neurology and Neurosurgery, Liverpool, UK
Proffered paper presentation
Background
Epigenetic
silencing of O6-methylguanine-DNA-methyltransferase (MGMT) promoter
in pre-therapy diagnostic samples is associated with prolonged survival of
glioblastoma patients treated with alkylating agents, but most clinical studies
use the non-quantitative methylation-specific PCR assay to determine
methylation status. In this study we report outcome data in a series of
glioblastomas treated with temozolomide and radiotherapy using pyrosequencing
to provide quantitative data at individual CpG sites within the MGMT promoter.
Methods
Details of therapy and outcome were collated for 109 newly-diagnosed
glioblastoma patients treated with concurrent and adjuvant temozolomide and
radiotherapy in a single UK treatment centre from June 2004-October 2007. 61
patients were given CCNU at relapse. Pyrosequencing was used to investigate
MGMT promoter methylation in archival pretherapy tissue samples.
Results
Median overall survival (OS) following temozolomide and radiotherapy was 12.4
months with 2-year survival 17.9%. Methylation was an independent prognostic
factor associated with prolonged progression-free survival (PFS) and OS
(Log-rank unmethylated vs methylated: PFS P=0.0000; OS P=0.0000). Cases with
methylation >35% had the longest survival [median PFS 19.2, OS 26.2
(months), 2-year survival 59.7%]. Significant differences in PFS were seen
between those with intermediate or high methylation and unmethylated cases,
while cases with low, intermediate or high methylation all showed significantly
different OS. MGMT methylation at diagnosis was weakly associated with
survival following CCNU given at recurrence (Log-rank unmethylated vs
methylated P=0.033) and those with high methylation (>29%) had longer
survival from CCNU than unmethylated cases (Log-rank P=0.008).
Conclusion
MGMT methylation determined at diagnosis had greater prognostic significance
for temozolomide/radiotherapy as initial therapy than for CCNU given at
recurrence. Measurement of the extent of MGMT methylation in glioblastomas may
be used to provide additional prognostic stratification compared with
non-quantiative assays and cases with high methylation showed greatest clinical
benefit from therapy.
