Cancer health disparities affecting glioblastoma survival in Ontario, Canada: a population-based study

Background and objectives — The role of cancer health disparities in a non–screen-detectable, high-mortality cancer such as glioblastoma (GBM) remains uncertain. We investigated the association between socioeconomic factors and overall survivorship and receipt of adjuvant treatment among patients with GBM.

Methods — This population-based cohort study used linked provincial administrative databases in Ontario, Canada, and was conducted among adult patients with GBM who underwent surgery from 2006 to 2022. Our exposures were neighborhood-based material deprivation index, immigration status, and residence rurality index. The primary outcomes were overall survival (OS) and receipt of adjuvant treatment within 6 months. Secondary outcomes were cancer-specific survival (CSS) and the number of emergency department visits in the year preceding surgery, used as a proxy for potential management delay. Kaplan-Meier survival analysis and multivariable Cox proportional hazard models were used to assess OS and CSS while logistic regression and Poisson regression were applied for receipt of adjuvant treatment and emergency department visits, respectively.

Results — Among 7,232 patients included in the cohort, patients residing in the most deprived neighborhoods had worse overall survival (median: 228 days vs 294 days, adjusted hazard ratio [aHR] 1.167, 95% CI 1.078–1.229, p < 0.001) and cancer-specific survival (aHR: 1.216, 95% CI 1.112–1.329, p < 0.001) compared with patients in least deprived areas. They also had 26% decreased adjusted odds of receiving adjuvant treatment (77.9% vs 83.0% in the most vs least deprived neighborhoods; adjusted odds ratio [aOR]: 0.741, 95% CI 0.608–0.902, p = 0.003) compared with patients in least deprived neighborhoods. Immigration status was significantly associated with OS and CSS, shown to be longer among remote immigrants (aHR for OS: 0.875, 95% CI 0.794–0.964, p = 0.007; aHR for CSS: 0.816, 95% CI 0.730–0.912, p < 0.001) and recent immigrants (aHR for OS: 0.746 95% CI 0.566–0.982, p = 0.037; aHR for CSS: 0.705, 95% CI 0.517–0.961, p = 0.027) compared with nonimmigrants. Higher material resource dimension index (p = 0.015) and rural residence (p < 0.001) were also associated with more ED visits before surgery.

Discussion — These findings suggest that cancer health disparities persist for GBM in Ontario, Canada, despite a single-payer, publicly financed system that guarantees universal health coverage of cancer services. This highlights the need for targeted interventions to address disparities and improve outcomes in marginalized populations.