Abstract
INTRODUCTION: Comprehensive, genome-wide profiling and next-generation sequencing based studies have dramatically improved our understanding of pediatric brain tumor biology in recent years. However, the vast majority of these studies rely on the assumption that single biopsies are representative of all areas within a tumor. Intratumor heterogeneity comprises a common phenomenon previously described in glioblastoma multiforme and other tumors. Highly disparate genetic profiles of spatially separated areas within the same tumor may preclude development of personalized, molecularly targeted therapies based on single tumor biopsies. MATERIAL AND METHODS: To assess the degree of intratumor heterogeneity, we conducted multiregion whole exome sequencing, high-resolution DNA copy number analysis (Cytoscan HD) and DNA methylation profiling (Infinium HumanMethylation450 BeadChip) on over 25 distinct pediatric and adult brain tumors with a median of six biopsies per tumor (range 4-9). Histological entities comprised ATRT (n = 2), DIPG (n = 2), ependymoma (n = 1), glioblastoma (n = 10), medulloblastoma (n = 10), and medulloepithelioma (n = 1). We elucidated the degree of intratumor heterogeneity and subgroup affiliation using integrated genomics and unsupervised hierarchical clustering algorithms. RESULTS: Epigenetic signatures were highly similar in multiregion biopsies from a single tumor. However, we identified up to 250,000 CpG dinucleotides that were differentially methylated when determining the intertumor heterogeneity of DNA methylation patterns even within disease subgroups. In addition, pediatric brain tumors displayed highly similar focal and broad DNA copy number alterations compared to their adult counterparts. Multiregion sequencing further reinforced the relatively higher degree of intratumor homogeneity in pediatric brain tumors. Lastly, we showed that subgroup affiliation was stable in all multiregion biopsies in medulloblastomas and glioblastomas. CONCLUSIONS: Our results reveal that single biopsies are representative of the tumor genomics landscape in pediatric brain tumors and that DNA methylation based subgrouping is geographically stable.