LGG-13. RESOLVING TRANSCRIPTIONAL PROFILES IN BRAF-REARRANGED PILOCYTIC ASTROCYTOMA USING SINGLE CELL RNA SEQUENCING

Pilocytic astrocytoma (PAs) are pediatric low grade gliomas that frequently harbor oncogenic KIAA1549-BRAF fusions and exhibit low rates of other somatic genetic alterations. Previous transcriptomic profiling of bulk PA tissues sought to uncover gene expression programs that distinguish PA from other tumor types and which may underlie the unique biological features of PA. However, bulk PA transcriptomic data is often confounded by gene signatures thought to be contributed by normal, non-tumor cells. Also, analysis of bulk PA tissue provides incomplete information on whether distinct cell populations exist among the tumor cells. To resolve the cell types contributing to PA transcriptional profiles and to identify gene expression programs that distinguish PA tumor cells from those of higher-grade tumors, we performed RNA sequencing of >1000 single cells (scRNA-seq) collected from six PAs. To confidently distinguish tumor cells from non-tumor cells, we sorted cells by glial progenitor marker A2B5 status and also profiled KIAA1549-BRAF fusion status for every cell. This scRNA-seq approach revealed unique tumor and non-tumor cell populations within PAs. We identified tumor cell populations expressing oligodendrocyte precursor and radial glia gene signatures. Non-tumor cells consisted of populations expressing microglia, T lymphocyte, and endothelial cell gene signatures which accounted for many of the expression patterns seen in bulk PA transcriptional data. In summary, scRNA-seq has revealed the cellular ecosystem and population structure of PA.