LAUSR

BACKGROUND The repair enzyme O6-methylguanine-DNA-methyltransferase (MGMT) is a validated predictor of benefit from temozolomide (TMZ) in glioblastoma. However, only 10% of patients with MGMT-methylated metastatic colorectal cancer (mCRC) respond to TMZ. METHODS Archived tumour samples (N = 41) from three phase II TMZ trials carried out in MGMT-methylated mCRC (assessed by methylation-specific polymerase chain reaction [PCR]) were stratified by MGMT status as assessed by three different methods: mass spectrometry, PCR/methyl-BEAMing and RNA-seq. The performance of each method was assessed in relation to overall response rate, progression-free survival (PFS) and overall survival (OS). RESULTS Overall, 9 of 41 patients responded to TMZ. Overall response rates were 50% (9/18), 50% (6/12) and 35% (8/23) among patients determined likely to respond to TMZ by mass spectrometry, methyl-BEAMing and RNA-seq, respectively. Low/negative MGMT protein expressors by mass spectrometry had longer PFS than high MGMT expressors (3.7 vs 1.8 months; HR = 0.50, P = 0.014). Results for OS were similar but statistically non-significant (8.7 vs. 7.4 months; HR = 0.55, P = 0.077). No significant association between survival and MGMT status by methyl-BEAMing or RNA-seq could be demonstrated as comparable subgroups survival could not be confirmed/excluded. Specifically, the association of high versus low methyl-BEAMing MGMT hypermethylation with survival was HR = 0.783, P = 0.46 for PFS and 0.591, P = 0.126 for OS, while association of low versus high RNA-seq MGMT level with survival was HR = 0.697, P = 0.159 for PFS and HR = 0.697, P = 0.266 for OS. CONCLUSIONS Quantitative proteomic analysis of MGMT may be useful for refining the selection of patients eligible for salvage treatment with single-agent TMZ.