TMOD-07. DEVELOPMENT AND CHARACTERIZATION OF A SPINAL CORD DIFFUSE MIDLINE GLIOMA MODEL

Abstract Diffuse midline gliomas (DMGs) are very aggressive central nervous system tumors that occur mainly in children and adolescents. Most DMGs develop in the brainstem, but they can also be found in the thalamus and the spinal cord, the latter comprising 4% of all DMGs. Despite advances in current treatments, the outcome for DMG patients remains extremely poor. The success of immunotherapy in other tumors encourages its translation to DMGs, for which the understanding of the tumor immune microenvironment (TIME) is imperative. The field counts with several murine DMG models in the pontine and thalamic locations. However, there is a paucity of spinal cord models. Thus, in this work, we have generated an orthotopic syngeneic model of H3K27M spinal cord DMG by injecting the tumor cells directly into the spinal cord under the T2 vertebra. We monitored tumor growth by bioluminescence and PET using 18F-FDG and 11C-methionine. As the tumor grew, we observed human-like symptoms in the mice, such as hemiplegia and hemiparesis. TIME characterization showed that this tumor model is mainly infiltrated by macrophages and almost devoid of lymphocytes. We are currently delving into the phenotype of these populations by scRNAseq and spatial transcriptomics. In addition, we are comparing our findings with human samples (n=15) to assess whether our model mimics human disease and can serve as a bonafide model of this tumor location. As a proof of concept, treatment of spinal DMG-bearing mice with a TIM-3 antagonist antibody, previously described as effective in pontine DMG-bearing mice, increased the median overall survival. We showed by microPET imaging that anti-TIM-3 administration into the brain ventricles diffuses to the tumor. These results open a path for the characterization of the spinal cord DMG TIME and for testing preclinical treatments in this new model.