Abstract
BACKGROUND: Pediatric high-grade gliomas (HGGs) present a therapeutic challenge due to their aggressive behavior and limited treatment options. Traditional treatments often fall short in addressing the genetic complexity and aggressiveness of pediatric HGGs, urging the search for and validation of novel genetic dependencies as potential therapeutic targets. METHODS: Loss of function CRISPR screens enable the systematic investigation of gene function by selectively disrupting gene expression, offering invaluable insights into the genetic dependencies fuelling tumor growth and survival. Herein, we conducted genome-scale, loss of function CRISPR/Cas12 screens across genetically-defined subtypes of pediatric HGG, including diffuse midline gliomas (n=30 cell lines). RESULTS: Distinct genetic dependencies exclusive to pediatric high-grade gliomas (pHGG) are uncovered compared to their adult counterparts, particularly within druggable pathways. These dependencies are associated with specific pathways such as cellular metabolism, epigenetics, and tissue development. Notably, histone-altered subtypes of pHGG exhibit dependency on distinct gene sets for growth. Furthermore, machine learning techniques are applied to correlate various -omics features (RNA, DNA, epigenomics, proteomics) with genetic dependencies, facilitating the identification of biomarkers predictive of therapeutic response. CONCLUSION: Our work provides a comprehensive genetic dependency map of pHGG. We expect this data to serve as a springboard for enriching our biological understanding of pHGG, ultimately improving clinical outcomes and quality of life.