Multimodal analysis of pediatric pilocytic astrocytomas reveals tumor location-associated cellular and transcriptional heterogeneity

Background: Pilocytic astrocytomas (PAs) are the most common pediatric central nervous system tumors, which present with limited genetic but significant clinical heterogeneity. Current treatment strategies are partly effective, but tumors often progress, and patients experience long-term side effects, highlighting the need for additional novel therapeutic approaches. A promising alternative approach could be targeting the tumor immune microenvironment (TIME); however, a comprehensive overview of the TIME of PAs across different anatomical tumor locations is currently lacking. The aim of this study was to comprehensively characterize the cellular and transcriptional landscape of pediatric PAs. Methods: We employed a multimodal, integrative approach using single-nucleus and bulk RNA-sequencing alongside high-dimensional immunofluorescence imaging and flow cytometry to elucidate the transcriptional landscape and cellular composition of pediatric PAs across tumor locations. Results: Our analysis uncovered neoplastic cell transcriptional heterogeneity reflective of aberrant glial differentiation. Moreover, we provided a detailed characterization of the TIME, revealing pro-inflammatory brain-resident microglia and abundant activated monocyte-derived macrophages and T cells. Deconvolution of bulk RNA-sequencing data revealed variation in tumor and TIME composition across anatomical tumor locations. Suprasellar tumors additionally exhibited increased expression of immune-related genes compared to tumors arising in other anatomical locations. Conclusions: Together, our multimodal in-depth characterization provides detailed insights into the transcriptional and cellular heterogeneity of pediatric PAs across distinct anatomical tumor locations, which could aid in the development of novel (immuno)therapeutic approaches.