Do we really know who has an MGMT methylated glioma? Results of an international survey regarding use of MGMT analyses for glioma.

Malmström A, Łysiak M, Kristensen BW, Hovey E, Henriksson R, Söderkvist P

Neurooncol Pract 7 (1) 68-76 [2020-01-00; online 2019-09-24]

Glioma O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status informs clinical decision making. Worldwide different methods and cutoff levels are used, which can lead to discordant methylation results. We conducted an international survey to clarify which methods are regularly used and why. We also explored opinions regarding international consensus on methods and cutoff. The survey had 152 respondents from 25 countries. MGMT methylation status is determined for all glioblastomas in 37% of laboratories. The most common methods are methylation-specific polymerase chain reaction (msPCR) (37%) and pyrosequencing (34%). A method is selected for simplicity (56%), cost-effectiveness (50%), and reproducibility of results (52%). For sequencing, the number of CpG sites analyzed varies from 1-3 up to more than 16. For 50% of laboratories, the company producing the kit determines which CpG sites are examined, whereas 33% select the sites themselves. Selection of cutoff is equally distributed among a cutoff defined in the literature, by the local laboratory, or by the outside laboratory performing the analysis. This cutoff varies, reported from 1% to 30%, and in 1 laboratory tumor is determined as methylated in case of 1 methylated CpG site of 17 analyzed. Some report tumors as unmethylated or weakly vs highly methylated. An international consensus on MGMT methylation method and cutoff is warranted by 66% and 76% of respondents, respectively. The method preferred would be msPCR (45%) or pyrosequencing (42%), whereas 18% suggest next-generation sequencing. Although analysis of MGMT methylation status is routine, there is controversy regarding laboratory methods and cutoff level. Most respondents favor development of international consensus guidelines.

Clinical Genomics Linköping

PubMed 32025325

DOI 10.1093/nop/npz039

Crossref 10.1093/nop/npz039

pii: npz039
pmc: PMC6993038