https://orcid.org/0000-0002-3296-8652
It has been established that glioblastoma (GBM) survival differs by sex, with females having a significant survival advantage.1 Another prognostic factor associated with GBM survival is O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation, associated with sensitivity to alkylating chemotherapy, such as temozolomide (TMZ), and improved survival.2,3MGMT methylation status impacts treatment patterns for GBM. Studies suggest that patients with unmethylated tumors should receive only radiotherapy.4 Here, we investigate, utilizing a large real-world data set, the impact of MGMT methylation status on sex-specific survival for different GBM treatment patterns.
Despite aggressive multimodal treatments, GBM remains uniformly lethal. Survival is largely dictated by the extent of surgical resection and response to standard-of-care radiation and TMZ. MGMT promoter methylation status has been shown to have prognostic and predictive value.2,4 It has been postulated that enhanced MGMT methylation may predispose an individual to better responses to alkylating treatments and radiotherapy. To analyze the impact of sex, MGMT promoter methylation, and treatment modality on survival, we analyzed 2108 adult (>25 years old) individuals with GBM as determined by a combination of annotated histology and IDH wild-type status from the CARIS Lifesciences data set from 2013 to 2021 (61.5% male, 38.5% female). MGMT promoter methylation analysis was performed by pyrosequencing.5 Samples with ≥7% and <9% methylation were considered to be equivocal or gray zone results.5 Real-world overall survival and treatment information was obtained from insurance claims data from payers and calculated from start of any type of GBM treatment to last contact6; uninsured patients were not included. Individuals were categorized by MGMT promoter methylation status, TMZ treatment at any dosage/time period, and radiation treatment at any dosage. Treatment information was obtained from the insurance claims data. Date of first treatment was determined by the first insurance treatment of interest claims, independent of the tissue collection date.6 Kaplan–Meier survival curves were evaluated to assess survival differences by sex, stratified by 3 different treatment regimens (TMZ alone, radiation alone, and TMZ plus radiation). Corresponding log-rank test P-values are reported (P < 0.05 as significant). All analyses were performed using the Caris CODEai data platform. Limitations with the analytical tools within the platform prevented multivariate analyses, and therefore all results are presented stratified by MGMT promoter methylation and treatment status. Additional stratifications, including age at diagnosis, were out of this analysis scope.
Within each treatment modality, median survival estimates were generally higher among individuals with MGMT promoter methylation compared to unmethylated individuals (Figure 1). This is consistent with other studies that have demonstrated that overall survival is better in individuals with MGMT promoter methylation compared to individuals that are unmethylated.7 Similar results were observed here in individuals treated with standard of care. Previous studies also demonstrated that there is no difference in survival, by sex, in individuals receiving standard of care who have unmethylated MGMT. This was observed here, as there were no significant survival differences observed by sex within any treatment group among unmethylated individuals (Figure 1A, C, and E). This larger study does show a direction of lower survival in male survival among the TMZ-only (Figure 1C) and radiation-only (Figure 1E) treatment groups; while this did not reach significance, it may be worth examining in a larger study group. There were no observed sex differences in survival within GBM individuals with MGMT promoter methylation receiving either TMZ and radiation (Figure 1B) or TMZ alone (Figure 1D). A notable survival difference was observed among MGMT-methylated individuals, with female individuals having significantly lower survival than males (Figure 1F; 13.4 m vs. 24.5 m, respectively, log-rank P = 0.01).
Sex-specific Kaplan–Meier survival curves for (A) MGMT promoter-unmethylated glioblastoma (GBM) cases treated with temozolomide (TMZ) and radiation, (B) MGMT promoter-methylated GBM cases treated with TMZ and radiation, (C) MGMT promoter-unmethylated GBM cases treated with TMZ only, (D) MGMT promoter-methylated GBM cases treated with TMZ only, (E) MGMT promoter-unmethylated GBM cases treated with radiation only, and (F) MGMT promoter-methylated GBM cases treated with radiation only. P-values of <.05 were determined to be statistically significant.
MGMT promoter methylation has been implicated as a prognostic biomarker even in the absence of TMZ treatment, suggesting that MGMT promoter methylation may be a predictor for overall treatment responsiveness in general.8 This study, however, did not stratify based upon sex while other studies, due to the rarity of GBM, were carried out in small data sets. In this larger study set here, the observed male:female survival difference among individuals who were treated with only radiation is notable. While not significant, median survival was higher for females among unmethylated patients who received radiation alone (Figure 1E), in contrast to the significant survival difference observed for males when assessing MGMT-methylated individuals treated with radiation alone (Figure 1F). Female GBM individuals with MGMT promoter methylation treated with radiation alone had a median survival that was almost half that of male GBM individuals with MGMT promoter methylation treated with radiation alone. Methylated females who received only radiation appear to have little to no survival benefit relative to their unmethylated counterparts, unlike the other treatment groups in this study. Current clinical recommendations suggest the use of concurrent TMZ and radiation for treatment of MGMT-methylated GBM, and the results here may suggest that this course of treatment is particularly important for females.2 Further research into this would be strengthened by the inclusion of other prognostic factors for GBM, such as extent of surgical resection, and Karnofsky Performance status which were unavailable in this data set. Due to the unavailability of these data, we cannot rule out that potential differences in clinical outcomes may be driven by covariates not captured in this analysis.
Previous research suggested MGMT promoter methylation impacts survival of individuals with GBM receiving standard-of-care treatment (TMZ and concurrent radiation) in a sex-dependent manner.9 We did not observe a significant survival difference, by sex, among individuals who received only TMZ or TMZ and concurrent radiation in this larger study. The prominent sex difference observed among individuals with MGMT promoter methylation who received only radiation treatment may suggest that there is a potential sex bias on the impact of MGMT promoter methylation and survival prognosis post treatment. This potential sex bias may be related not to the established role in sensitivity to alkylating chemotherapy, but rather to underlying mechanisms of tumor biology that remain to be elucidated. MGMT promoter methylation has been correlated with increased occurrence with K-Ras mutation in female colorectal cancer individuals,10 it remains to be determined if similar pathways are involved with GBM.
Funding
The research performed by G.C., K.A.W., M.D., and J.S.B.-S. was provided by the Division of Cancer Epidemiology and Genetics (DCEG) of the National Cancer Institute (NCI). G.C. and K.A.W. are full-time contractors of the NCI. M.D. is a full-time research fellow of DCEG at NCI. J.S.B.-S. is a full-time employee of NCI and is supported through intramural funds.
Conflict of interest statement
The authors have no conflict of interest to declare. J.S.B.-S. is a full-time employee of the NIH/NCI. G.C. and K.A.W. are full-time contractors of the NIH/NCI. M.D. is a full-time postdoctoral fellow of the NIH/NCI.
Authorship statement
G.C.: conceptualization, investigation, formal analysis, visualization, writing original draft and editing; K.A.W.: funding acquisition, investigation, writing original draft and editing, project administration, resources; M.D.: conceptualization, investigation, methodology, visualization, writing original draft and editing, supervision; J.X.: data curation, formal analysis, investigation, methodology, reviewing final draft; J.S.B.-S.: conceptualization, investigation, project administration, resources, supervision, reviewing final draft.
Data availability
Deidentified data were analyzed and summarized. Additional data can be requested from CARIS Life Sciences.
