Comparative Analysis of Survival Outcomes and Prognostic Factors of Supratentorial versus Cerebellar Glioblastoma in the Elderly: Does Location Really Matter?

Abstract

Cerebellar glioblastomas (cGBMs) are rare tumors that are uncommon in the elderly. In this study, we compare survival outcomes and identify prognostic factors of cGBM compared with the supratentorial (stGBM) counterpart in the elderly. Data from the SEER 18 registries were used to identify patients with a glioblastoma (GBM) diagnosis between 2000 and 2016. The log-rank method and a multivariable Cox proportional hazards regression model were used for analysis. Among 110 elderly patients with cGBM, the median age was 74 years (interquartile range [IQR], 69-79 years), 39% were female and 83% were white. Of these patients, 32% underwent gross total resection, 73% radiotherapy, and 39% chemotherapy. Multivariable analysis of the unmatched and matched cohort showed that tumor location was not associated with survival; in the unmatched cohort, insurance status (hazard ratio [HR], 0.11; IQR, 0.02-0.49; P= 0.004), gross total resection (HR, 0.53; IQR, 0.30-0.91; P= 0.022), and radiotherapy (HR, 0.33; IQR, 0.18-0.61; P < 0.0001) were associated with better survival. Patients with cGBM and stGBM undergoing radiotherapy (7 months vs. 2 months; P < 0.001) and chemotherapy (10 months vs. 3 months; P < 0.0001) had improved survival. Long-term mortality was lower for cGBM in the elderly at 24 months compared with the stGBM cohort (P= 0.007). In our study, elderly patients with cGBM and stGBM have similar outcomes in overall survival, and those undergoing maximal resection with adjuvant therapies, independent of tumor location, have improved outcomes. Thus, aggressive treatment should be encouraged for cGBM in geriatric patients to confer the same survival benefits seen in stGBM. Single-institutional and multi-institutional studies to identify patient-level prognostic factors are warranted to triage the best surgical candidates.