Abstract Background: MGMT is a DNA repair protein which removes alkylating DNA adducts from the O6 position of guanine. Expression of MGMT is often silenced by promoter methylation in human cancers. MGMT methylation is a predictive marker for prolonged survival in glioblastoma patients treated with an alkylating agent, temozolomide. As MGMT methylation has been found in lung cancers, there is an increasing interest on the clinical utility of temozomolide in the treatment of human cancers. However, it is essential to use appropriate quantitative or semi-quantitative method methods to definitively establish the methylation status of the tumour. Methods: We critically assessed MGMT methylation status in 6 lung cancer cell lines and 56 lung tumours using three different methodologies. We first assessed the MGMT methylation pattern using methylation sensitive - high resolution melting (MS-HRM). The methylation status at each CpG dinucleotide was assessed bisulfite pyrosequencing of methylated clones. The level of MGMT methylation was quantified using quantitative methylation specific PCR. Results: MGMT methylation was found in 3 lung cancer cell lines by MS-HRM. The melting profiles of all methylated samples were indicative of heterogeneous methylation pattern by melting curve analysis. To examine the methylation status at each CpG sites of individual template, two MGMT methylated lung cell lines (H1666 and H69) were further tested by limiting dilution analysis and bisulfite pyrosequencing. The number and site of methylated CpG dinucleotides greatly varied in each template, confirming the heterogeneous methylation pattern in both cell lines. In 56 lung tumours, heterogeneous MGMT methylation was detected in seven samples (13%) by MS-HRM. The level of MGMT methylation was then estimated. 17 lung tumours, including the 7 MS-HRM positives and 10 additional tumours, were positive. However, the methylated level in all of the methylated samples was low, ranging from below 1% (12 samples) and up to 12%. Conclusions: The level of MGMT promoter in lung cancer is difficult to estimate. Ideally clonal analysis should be used to estimate the proportion of methylated alleles. Alternatively, methylation profiling using MS-HRM followed by pyrosequencing can be used to identify tumours showing significant levels of methylation. If MGMT methylation is found only in a small proportion of tumour cells, it is unlikely to be a useful target for therapy. Overcalling of MGMT methylated tumours may provide the explanation for the lack of survival benefit with temozolomide treatment in MGMT-methylated lung cancer patients in a recent phase II clinical trial (NCT00423150). This indicates that incorporation of immunohistochemistry for the MGMT protein should also be part of the assessment of the MGMT status of lung cancer. Citation Format: Hongdo Do, Paul Mitchell, Alexander Dobrovic. Is loss of MGMT a therapeutic target in lung cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 383. doi:10.1158/1538-7445.AM2014-383