EPCO-58. SINGLE-CELL PROFILING REVEALS CELLULAR DIVERSITY AND A STEM-LIKE COMPARTMENT IN SUBEPENDYMOMA

Abstract Subependymoma (SE) is a rare, WHO grade 1 central nervous system tumor that typically arises in adults. Despite its benign classification, SE often develops near eloquent brain regions—most commonly the floor of the fourth ventricle—where even small lesions can cause severe neurological symptoms. As a result, surgical resection is frequently pursued for symptom relief, though these procedures carry substantial risk and can result in lasting complications. Diagnosis of SE is currently relies on histopathologic criteria; however it often exhibits morphological overlap with other glial tumors such as ependymoma, complicating clinical decision-making. Due to its rarity and largely stable genetic profile, SE remains poorly characterized at the molecular level. Here, we performed proof-of-principle single-cell RNA-sequencing on 7,377 cells from a supratentorial SE tumor resected from a 26 year old male to demonstrate that high-resolution single-cell profiling is feasible for SE, even in the context of its low cellularity and dense extracellular matrix. We identified a diverse cellular landscape, comprising ependymal, microglial, astrocyte-like, and oligodendrocyte-like glial populations, along with a small fraction of cells expressing lymphocyte markers. In addition, we observed a prominent population with transcriptional profiles consistent with a neoplastic phenotype, characterized by activation of gene programs involved in ribosome biogenesis, cell cycle regulation, DNA repair, and mitochondrial metabolism. These neoplastic cells also exhibited upregulation of Wnt signaling and stemness-associated pathways, suggesting they reside in a stem-like state. This represents the first single-cell transcriptomic characterization of SE and provides foundational insight into its cellular and molecular architecture. Ongoing work will expand this analysis across a larger cohort to define conserved tumor-specific features and assess potential intratumoral heterogeneity within SE.