<i>MGMT</i> promoter methylation and survival following chemotherapy for WHO grade 4 IDH-mutant astrocytoma

MGMT promoter methylation is an established prognostic and predictive biomarker in patients with IDH-wildtype glioblastoma.1–3 However, World Health Organization (WHO) central nervous system (CNS) grade 4 IDH-mutant astrocytomas—unlike glioblastoma—uncommonly exhibit monosomy 10 (harboring MGMT), therefore the effect of MGMT promoter methylation on response to alkylating chemotherapy is unclear. Using US registry data, we previously defined the epidemiology of WHO grade 4 IDH-mutant astrocytomas.4,5 Here we investigate the outcomes associated with MGMT promoter methylation and chemotherapy among WHO grade 4 IDH-mutant astrocytoma patients in the United States. Patients with WHO grade 4 IDH-mutant astrocytomas from 2018 to 2021 were identified from the National Cancer Database (NCDB). Starting in 2018, cancer registries implemented new site-specific data items for MGMT promoter methylation, WHO grade, and “Brain Molecular Markers” (including IDH status).4–6 The NCDB only reports treatment data from the first-line setting. Chemotherapy is reported as none, single-agent, or multi-agent, without drug details. Because first-line temozolomide is recommended by National Comprehensive Cancer Network (NCCN), Society for Neuro-Oncology (SNO), and European Association of Neuro-Oncology (EANO) guidelines in this population and other single-drug regimens are rare for this indication, we assumed that first-line single-agent chemotherapy represented temozolomide.1–3 The relationship between MGMT promoter status, single-agent chemotherapy, and OS was examined using multivariable Cox regression, adjusting for patient age, radiotherapy, and extent of resection (EOR). The NCDB is publicly available and only contains deidentified data, which qualifies for human subjects research exemption. A total of 916 patients with WHO grade 4 IDH-mutant astrocytoma met inclusion criteria. The median age at diagnosis was 43 years (interquartile range [IQR] = 34–58) and 42.3% were female. MGMT promoter was unmethylated in 35.3% (n = 323) and methylated in 64.7% (n = 593). As first-line treatment, 87.4% received radiotherapy and/or chemotherapy. Among chemotherapy, 95.4% was a single agent (ie, likely temozolomide). Radiotherapy and single-agent chemotherapy were initiated at a median of 42 (IQR = 33–54) and 39 (IQR = 31–55) days, respectively, after diagnosis. Overall survival (OS) data were available for 650 patients, of whom 44.3% had died with a median follow-up of 23.6 months (IQR = 13.0–33.4). OS analyses were restricted to patients with ≥1 month of follow-up and at least a tissue biopsy of tumor (n = 546). Among the 475 patients that received first-line single-agent chemotherapy, the median OS was 27.9 months for MGMT promoter unmethylated patients (95% CI = 23.0–42.7) and not reached (NR) for methylated patients (95% CI = NR–NR; log-rank P < .001; Figure 1). In multivariable Cox regression adjusted for patient age, EOR, and radiotherapy, MGMT promoter methylation remained associated with improved OS in patients receiving single-agent chemotherapy compared with unmethylated patients (hazard ratio [HR] = 0.54, 95% CI = 0.40–0.72; P < .001). Overall survival associated with MGMT promoter methylation and temozolomide among patients with WHO grade 4 IDH-mutant astrocytoma. Kaplan–Meier OS estimates with 95% confidence intervals (light gray shading) and at-risk table for patients with WHO grade 4 IDH-mutant astrocytoma who received first-line single-agent chemotherapy (ie, likely temozolomide, TMZ, black lines) versus no first-line chemotherapy (gray lines), stratified by MGMT promoter status (methylated, solid lines, unmethylated, dashed lines). Patients had at least tissue biopsy and follow-up of ≥1 month. For the 71 patients who did not have first-line chemotherapy, median OS was 11.5 months for MGMT promoter unmethylated patients (95% CI = 3.8–15.5) and 11.9 months for MGMT promoter methylated patients (95% CI = 6.5–32.3; log-rank P = .20). There was no significant difference observed in the multivariable regression model (methylated HR = 0.57 vs. unmethylated, 95% CI = 0.32–1.03, P = .06). Although in univariable analysis temozolomide displayed improved OS among MGMT promoter unmethylated patients compared with no chemotherapy (HR = 0.35, 95% CI = 0.22–0.55, P < .001), a significant difference was not observed in the multivariable Cox regression adjusted for age, EOR, and radiotherapy (HR = 0.74, 95% CI = 0.38–1.45, P = .38). Our findings suggest that MGMT promoter methylation is predictive of improved OS in WHO grade 4 IDH-mutant astrocytoma patients who receive first-line temozolomide. The highest OS was observed among patients who had MGMT promoter methylation and first-line temozolomide. Although we did not observe a significant OS difference by MGMT status among patients who did not receive first-line chemotherapy, the small size of this group may limit the power to detect a prognostic effect. Notable limitations of the NCDB included a lack of details regarding chemotherapy type and subsequent lines of treatments, leading to the study’s assumption that first-line single-agent chemotherapy likely represented temozolomide in this population. However, given the consistency of standard-of-care treatment plans in clinical practice and the limited survival benefit from current therapies for tumor recurrence, these factors are unlikely to have influenced the results of this study. Although additional data will be needed to prospectively confirm these findings, our results are highly relevant when considering clinical trial design in patients with WHO grade 4 IDH-mutant astrocytoma, where the paradigm may shift away from IDH-wild-type glioblastoma-based approaches and where the influence of MGMT promoter methylation may need to be factored in. No funding was received to support this study. The authors declare no conflicts of interest related to the content of this manuscript. In accordance with the NCDB data use agreement, data are available by application to the NCDB. Not considered human subjects research. This study was performed in accordance with the Declaration of Helsinki. The NCDB Participant User Files contain deidentified national data for which consenting was not applicable.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.