Metformin exposure after glioblastoma diagnosis and mortality: A large population-based study.

Background: Glioblastoma (GBM) is an aggressive primary brain malignancy with limited therapeutic options and poor survival outcomes. Recent real-world studies and early-phase clinical trials evaluating metformin have yielded mixed findings. This study assessed the association between post-diagnosis metformin exposure and all-cause mortality in a large population-based cohort of older adults with GBM. Methods: A retrospective cohort study was conducted utilizing the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. Adults aged ≥66 years diagnosed with GBM between 2007 and 2018 were included, excluding those with other malignancies or diagnoses made by autopsy or death certificate. Metformin exposure was modeled as a time-varying covariate, defined by the first prescription filled after diagnosis. Cox models with time-dependent covariates estimated adjusted time-varying hazard ratios (HRs) for mortality in the overall cohort and selected subgroups. Results: Among 16,069 eligible patients, 2,850 (17.7%) were exposed to metformin after diagnosis. Over half (55.8%) died within six months. In the full cohort, metformin exposure was not associated with mortality during the first three months (HR = 1.02; 95% CI: 0.91-1.14) but was linked to increased mortality between 3-9 months (HR = 1.22; 95% CI: 1.10-1.34) and beyond nine months (HR = 1.20; 95% CI: 1.09-1.14). A similar pattern was consistent across subgroup analyses. Conclusions: In this large cohort of older adults with GBM, metformin exposure after diagnosis was not associated with reduced mortality and was linked to increased mortality beyond the initial three months. These findings underscore the need for prospective studies and randomized controlled trials to clarify metformin's role in GBM outcomes.