Background
Gliomas constitute a wide spectrum of neuroepithelial tumors arising from glia or supporting cells (astrocytes, oligodendrocytes, and ependymocytes) of the central nervous system (CNS)1,2. Gliomas are responsible for nearly 24% of all primary brain and CNS neoplasms and vastly differ in histology and behavior ranging from benign ependymomal tumors to the most aggressive and lethal grade IV glioblastoma multiforme3. Specifically, brainstem gliomas are diverse groups of neoplasms that primarily affect children and include low-grade focal brainstem gliomas, as well as high-grade diffuse intrinsic pontine gliomas (DIPG)4,5. Although 80% of gliomas arise within the pons as DIPGs, the remaining low-grade gliomas are located within the midbrain, dorsal medulla, or the cervico-medullary junction. The aforementioned locations of these neoplasms pose therapeutic challenges, and hence, may negatively affect treatment outcomes.
Radiotherapy, chemotherapy, or combined treatment modalities are the standard therapeutic options for gliomas6. Previous studies have reported failure of chemotherapy in treating DIPGs due to the lack of intra-tumor penetration7,8. Recently, the unraveling of the genetic landscape of DIPGs together with identification of the K27M mutation (mutation in both histones H3.1 and H3.3) has improved the understanding of the pathogenesis of gliomas and identification of novel targeted therapies9,10.
Despite a wealth of information on cancer staging, prediction of survival and treatment strategies, little is known about the determinants of cancer-specific survival in childhood brain stem glioma. The Surveillance, Epidemiology, and End Results (SEER) database of survival data from population-based cancer registries encompass ~28% of the American population11,12. The aim of the current study was to characterize a comprehensive, accurate and useful prognostic model using a population-based SEER analysis to predict survival of pediatric cases of stem gliomas.