MGMT promoter methylation in 1p19q-intact gliomas

Abstract Objective Standard-of-care for 1p19q-intact anaplastic gliomas is defined by the international randomized phase III CATNON trial, which found an overall survival (OS) benefit for adjuvant temozolomide (TMZ) when added to radiotherapy. Paradoxically, TMZ did not appear to benefit patients with IDH-wildtype gliomas, regardless of MGMT promoter status. The authors concluded that well-powered prospective study on the clinical efficacy of TMZ for patients with IDH-wildtype anaplastic gliomas (meeting criteria for glioblastoma) is warranted. Given that the prognostic and predictive role of MGMT status for grade 2–3 gliomas is unresolved, we determined the effect of MGMT status on OS in patients with 1p19q-intact gliomas in the National Cancer Database (NCDB). Methods We queried the NCDB from 2018–2019 for patients with IDH-wildtype or -mutant astrocytomas who received chemotherapy with follow-up through 2022. The Kaplan-Meier method and Cox proportional hazards regressions models were used to determine the association of MGMT with OS. Results We identified 1,514 patients who were newly diagnosed with IDH-wildtype (n = 802, 33% methylated) or -mutant astrocytomas (n = 712, 48% methylated) and received chemotherapy during initial management. An unmethylated promoter was associated with poorer survival in patients with IDH-wildtype (3-year OS 34% [95%CI 29–39%] vs. 46% [95%CI 39–54%], p < .001, adjusted HR 1.53 [95%CI 1.24–1.89]) but not IDH-mutant astrocytomas (3-year OS 79% [95%CI 74–84%] vs. 80% [95%CI 75–86%], p = .81, HR 1.04 [95%CI 0.73–1.50]). Conclusions This ancillary analysis supports adjuvant TMZ as standard-of-care for anaplastic astrocytomas (IDH-mutant and 1p19q-intact), irrespective of MGMT status. Determining the optimal strategy for diffuse gliomas that are IDH-wildtype will be particularly important. MGMT promoter methylation should be considered as a stratification factor in future clinical trials for these patients.


Introduction
Standard-of-care for 1p19q-intact anaplastic gliomas is de ned by the international randomized phase III CATNON trial, which found an overall survival (OS) bene t for adjuvant temozolomide (TMZ) when added to radiotherapy. 1 The interim analysis did not show a bene t for the concurrent phase of TMZ. 2 In the initial trial design, isocitrate dehydrogenase (IDH1/2) mutational status was not collected and patients were recruited based solely on the absence of 1p19q codeletion, which is observed in approximately 97% of histologically diagnosed (historical) anaplastic astrocytomas and 26% of anaplastic oligodendrogliomas. 3Post-hoc analyses of the trial found the bene t of adjuvant TMZ to be limited to the 444 patients with IDH-mutant tumors.No bene t was observed in a subset of 159 patients with tumors that met the WHO 2021 molecular criteria for glioblastoma, IDH-wildtype. 46 -methylguanine-DNA methyltransferase (MGMT) promoter methylation was included as a strati cation factor in the CATNON trial, given its predictive role for bene t from TMZ in glioblastoma. Pomoter methylation was prognostic in patients with IDH-wildtype but not IDH-mutant tumors.4,5 However, to the surprise of the investigators, MGMT status did not predict bene t from TMZ in patients with IDH-wildtype tumors.They concluded that a well-powered prospective study on the clinical e cacy of TMZ for patients with anaplastic gliomas who meet the contemporary de nition of glioblastoma is warranted.Consistent with CATNON, we also found that MGMT status is prognostic in IDH-wildtype but not IDH-mutant, 1p19qintact gliomas in a pooled analysis of three prospective cohorts.6 MGMT status was not an independent predictor of OS on multivariable analysis in either study, though sample sizes in both our study and the CATNON study were limited.Therefore, the prognostic and predictive role of MGMT status for grade 2-3 gliomas, particularly IDH-wildtype tumors, is unresolved.
To address this uncertainty, we determined the effect of MGMT status on OS in patients with 1p19qintact gliomas in the National Cancer Database (NCDB) from 2018-2019 with follow-up through 2022.The NCDB accounts for approximately 70% of invasive cancers in the United States, so it is unlikely that inadequate power would lead to ambiguous results. 7
9][10][11][12][13][14][15] All statistical analyses were conducted using the RStudio software Version 1.4.1106(RStudio, Inc., Boston, Massachusetts).The Kaplan-Meier method with the log-rank test and adjusted Cox proportional hazards regressions models were used to determine the association of MGMT status with OS.Variables that were statistically signi cant on univariable analysis were included in the multivariable model.Schoenfeld's test of weighted residuals was utilized to assess proportional hazard assumption in the Cox model.All analyses were performed at the .05signi cance level based on two-sided statistical testing.

Discussion
This ancillary analysis supports adjuvant TMZ as standard-of-care for anaplastic astrocytomas (IDHmutant and 1p19q-intact), irrespective of MGMT status.Determining the optimal strategy for anaplastic astrocytomas that are IDH-wildtype (including those that meet the molecular criteria for glioblastoma) will be particularly important.The largest randomized trials (CATNON, RTOG 9402, EORTC 26951, RTOG 9802), paradoxically, have not demonstrated an improvement in OS with alkylating chemotherapy for patients with IDH-wildtype grade 2-3 gliomas, though these were post-hoc analyses with modest sample sizes.1,2,16-18 Based on our results, MGMT promoter methylation should be considered as a strati cation factor in future clinical trials of patients with IDH-wildtype gliomas.
Limitations of our study include retrospective analysis with relatively short follow-up.To clearly determine if MGMT promoter methylation may be a prognostic or predictive biomarker in IDH-mutant tumors, longer follow-up is required in both our dataset and the CATNON trial, as approximately 80% and 66% of patients were still alive at the database lock.However, we are not aware of any prospective studies that support a role for MGMT status in IDH-mutant astrocytomas.In contrast, emerging data suggest a role for MGMT as a biomarker in 1p19q-codeleted oligodendrogliomas. 6,7The method of testing for MGMT promoter methylation is not reported in the NCDB, and results and cutoff levels may vary depending on the speci c assay.Considering these limitations, our study is the largest on the subject and our results are consistent with other literature.Collectively, these data support a role for MGMT testing in IDH-wildtype gliomas and IDH-mutant/1p19q-codeleted oligodendrogliomas but not IDH-mutant/1p19-intact astrocytomas.Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Figures
Figures

Table 1
Iwamoto has obtained grants or contracts through Columbia from Merck, Bristol Myers Squibb, Roche, Sapience, Novocure, Celldex, Tocagen, Forma, Celldex, and Northwest Biotherapeutics; is in consulting agreements with Novocure, Regeneron, Tocagen, Alexion Pharmaceuticals, Abbvie, Guidepoint Global, Merck, Kiyatec, PPD, Massive Bio, Medtronic, MimiVax, Gennao Bio, and Xcures; has two US provisional patent applications (62/739,617 and 63/062,805) through Columbia University; received support for meetings and travel from Roche and Oncoceutics; and participates on advisory boards of Mimivax and Northwest Biotherapeutics.Dr. Neugut has consulted for Otsuka Pharmaceuticals, GlaxoSmithKline, Organon, Value Analytics, Merck, and United Biosource Corp.He has received grant support from Otsuka and is on the medical advisory board of EHE Intl.Dr. Yu receives speaking and consulting fees from Re eXion Medical, Boston Scienti c, and P zer/Myovant and is an investor in Modi Bio.Dr. Cheng reported receiving grants from Janssen and travel funding from Caris outside the submitted work.Dr. Wang reports personal fees and non-nancial support from AbbVie, personal fees from Cancer Panels, personal fees from Doximity, personal fees and non-nancial support from Elekta, personal fees and non-nancial support from Merck, personal fees and non-nancial support from Novocure, personal fees and non-nancial support from RTOG Foundation, personal fees from Wolters Kluwer, grants and non-nancial support from Genentech, grants and non-nancial support from Varian, personal fees from Iylon Precision Oncology, outside the submitted work.Funding/Support: This study was supported by National Cancer Institute grant P30 CA13696 and by the generous unrestricted gift from Barry Neustein and Poly ex Inc. to the lung cancer research program in Radiation Oncology at Columbia University.No other outside grant support was received for this study.