Digital PCR assessment of MGMT promoter methylation coupled with reduced protein expression optimises prediction of response to alkylating agents in metastatic colorectal cancer patients

Background O(6)-methylguanine-DNA-methyltransferase (MGMT) is a repair protein, and its deficiency makes tumours more susceptible to the cytotoxic effect of alkylating agents. Five clinical trials with temozolomide or dacarbazine have been performed in metastatic colorectal cancer (mCRC) with selection based on methyl-specific PCR (MSP) testing with modest results. We hypothesised that mitigated results are consequences of unspecific patient selection and that alternative methodologies for MGMT testing such as immunohistochemistry (IHC) and digital polymerase chain reaction (PCR) could enhance patient enrolment. Patients and methods Formalin-fixed paraffin embedded archival tumour tissue samples from four phase II studies of temozolomide or dacarbazine in MGMT MSP-positive mCRCs were analysed by IHC for MGMT protein expression and by methyl-BEAMing (MB) for percentage of promoter methylation. Pooled data were then retrospectively analysed according to objective response rate, progression-free survival (PFS) and overall survival (OS). Results One hundred and five patients were included in the study. Twelve had achieved partial response (PR) (11.4%), 24 stable disease (SD; 22.9%) and 69 progressive disease (PD; 65.7%). Patients with PR/SD had lower IHC scores and higher MB levels than those with PD. MGMT expression by IHC was negatively and MB levels positively associated with PFS (p < 0.001 and 0.004, respectively), but not with OS. By combining both assays, IHC low/MB high patients displayed an 87% reduction in the hazard of progression (p < 0.001) and a 77% in the hazard for death (p = 0.001). Conclusion In mCRC selected for MGMT deficiency by MSP, IHC and MB testing improve clinical outcome to alkylating agents. Their combination could enhance patient selection in this setting.