Abstract
Background: Despite advances in glioblastoma (GBM) management, median overall survival (mOS) remains poor, and multi-modal disparities persist. We sought to evaluate trends in GBM treatment and survival outcomes from 2005-2020, with a focus on sociodemographic and geographic disparities. Methods: We conducted a retrospective US-based cohort study using the National Cancer Database (NCDB), stratifying study period into four intervals (2005-2008, 2009-2012, 2013-2016, and 2017-2020). Logistic regression was used to identified predictors of receipt of combination surgery, radiation, and chemotherapy (Sx+RT+Chemo). Kaplan-Meier and multivariable Cox proportional hazards approaches were used to assess mOS. Results: A total of 111,955 adults with GBM were included. From 2005-2008 to 2017-2020, mOS increased from 7.8 to 9.5 months, with geographically unequal gains in survival across the US. In multivariable logistic regression model adjusting for known confounders, combined Sx+RT+Chemo was less likely to be received by female patients (OR 0.90, 95% CI 0.88-0.92) vs. male, non-White patients (OR 0.90, 95% CI 0.86-0.94) vs. White, patients treated at community hospitals (OR: 0.78, 95% CI 0.76-0.80) vs. academic centers, publicly-insured patients (OR 0.74, 95% CI 0.71-0.76) or uninsured patients (OR 0.54, 95% CI 0.50-0.58) vs. privately-insured, and patients living in the South (OR 0.88, 95% CI 0.85-0.91), Midwest (OR 0.83, 95% CI 0.80-0.86), and West (OR 0.85, 95% CI 0.81-0.88) compared to the Northeast. In multivariable Cox regression, significantly poorer survival was seen amongst non-metropolitan patients, community-based hospital patients, and publicly-insured and uninsured patients (vs. privately-insured), despite adjusting for prognostic factors. Conclusions: Only modest improvement in mOS of GBM patients has occurred across 2005-2020, with persistent disparities linked to sociodemographic and structural factors, whose redressal warrants multi-pronged efforts.