Biol-14. from non-malignant to malignant: generating and targeting diffuse midline glioma in a novel human pontine organoid model

Abstract Diffuse Midline Glioma (DMG) is a rare and aggressive pediatric cancer with no chance of survival, demonstrating a critical need for therapy development. Predictive preclinical models that take into account the unique developmental background and anatomical environment of DMG, are so far lacking. We developed a new regionally patterned brain organoid protocol based on a timely regulated sequence of morphogens. The patterned organoids give rise to pontine and medulla oblongata identity, as confirmed by transcriptomic profiling and immunofluorescence imaging. As the tissue matures, the complexity inherently takes shape and gives rise to pons-specific neurons and glial cells, offering the first human in vitro model for this region of the brain. Introducing H3K27M, including its partner mutations P53 and PDGFRA-D842V via electroporation forms de novo H3K27M diffuse midline glioma with a similar genetic make-up as found in patients. Indeed, contrary to common cell lines or PDX models, single cell sequencing and 3D imaging reveal that Organoid-DMG resembles WHO-defined primary patient cancer, including its diverse cell population heterogeneity and invasive nature. Utilizing barcode-based lineage tracing allows us to unravel and track the cancerous developmental evolution. Thus, we can now model early emerging populations and identify their transcriptomic signatures, which is conceptually impossible to conduct with patient-derived tumor tissue. In summary, we generated a bona fide DMG in vitro model which will give insights into tumorigenesis of DMG and ultimately contribute to pre-clinical drug- and immunotherapy development.