MGMT Promoter Methylation as a Target In Metastatic Colorectal Cancer: Rapid Turnover and Use of Folates Alter its Study—Response

We agree with Ottaiano and Caraglia that the heterogeneity and dynamic evolution of MGMT status might cause temozolomide resistance related to MGMT reexpression/MGMT demethylation in patients with metastatic colorectal cancer. In phase II trials (1–5) and pooled analyses (6–8) of temozolomide-based therapy in patients with pretreated metastatic colorectal cancer, MGMT status was assessed in archival tumor tissue and may have undergone spatial and temporal variations in individual patients. Consistently, in the trial reported by Amatu and colleagues (3), a baseline pre-temozolomide biopsy was performed and the concordance with initial diagnostic samples in terms of MGMT methylation, assessed by methylBEAMing, was only 27%, probably accounting for the lower-than-expected response rate (3%).

We recently performed an in silico analysis of primary colorectal cancers included in The Cancer Genome Atlas (TCGA; ref. 9). Even if, as expected, MGMT methylation was found in 32% of cases, absence of residual MGMT enzyme expression was shown only in 27% of them. Therefore, a small subset (<10%) of colorectal cancers may be characterized by homogeneous and strong MGMT silencing, that is ideally derived by MGMT biallelic hypermethylation in the vast majority of cancer cells. Not surprisingly, evidence of residual immunohistochemical MGMT expression was invariably associated with lack of any benefit from temozolomide-based treatments in several retrospective analyses (2, 4, 5). Therefore, the simultaneous adoption of two assays such as immunohistochemistry and quantitative assessment of MGMT methylation may further refine the molecular selection of patients eligible for temozolomide-based treatment and may be associated with increased response rate to such strategies. Prospective evaluation of more restrictive molecular criteria is currently ongoing thanks to the MAYA trial (NCT03832621), which is aimed at exploiting single-agent temozolomide as a priming treatment to induce hypermutated status as reported previously (10) and sensitize microsatellite-stable tumors to the anti-PD-1 and anti-CTLA-4 combination of nivolumab plus low-dose ipilimumab.

Finally, we agree that MGMT methylation may be affected by several factors, including circulating folates, whose levels could have been temporarily increased by the administration of folinic acid only in the FOLFIRI arm. This topic warrants further clinical investigations, ideally with real-time monitoring of MGMT methylation status by means of serial liquid biopsies. However, we are still far from elucidating the mechanisms of temozolomide resistance. Thus, from a clinical perspective, the investigation of multiple and complex DNA damage response pathways may be even more crucial (9).

See the original Letter to the Editor, p. 3493