Population-based Study Determining Predictors of Cancer-Specific Mortality and Survival in Pediatric High-grade Brainstem Glioma

BACKGROUND: Pediatric high-grade brainstem gliomas are aggressive tumors with dismal prognoses. Large-scale studies are needed to further characterize these tumors and determine factors influencing cancer-specific mortality and survival at varying time points. METHODS: We used the SEER (Surveillance Epidemiology and End Results) database to conduct a population-based study of pediatric patients with histologically confirmed anaplastic astrocytoma or glioblastoma tumors located within the brainstem. Multivariate analyses incorporating patient demographics, tumor characteristics, and treatments were used to determine predictors of cancer-specific mortality and survival at 6 months, 9 months, 1 year, and 2 years. RESULTS: We included 154 patients from the SEER database: 72 patients with anaplastic astrocytoma (47%) and 82 (53%) with glioblastoma. Median survival for the entire cohort was 10.0 months. Glioblastoma histology, developmental stage, and large tumor size were significantly associated with cancer-specific mortality. Six-month, 9-month, 1-year, and 2-year survival was 75%, 57%, 42%, and 20%, respectively. Glioblastoma histology was associated with worsened survival at 6 months (odds ratio [OR], 0.19; P = 0.0081), 9 months (OR, 0.18; P < 0.001), 1 year (OR, 0.19; P < 0.001), and 2 years (OR, 0.14; P = 0.0055). Radiation therapy was associated with improved survival at 6 (OR, 8.53; P = 0.0012) and 9 months (OR, 3.58; P = 0.035) but not at 1 or 2 years. Radiation therapy was associated with improved survival in glioblastoma (9.0 vs. 3.0 months; P < 0.001). CONCLUSIONS: This population-based study showed that glioblastoma histology is associated with a poor prognosis in pediatric patients with high-grade brainstem gliomas. Regardless of histology, radiation therapy improved survival at 6 and 9 months but not long-term.