Identifying and Targeting the Cell-to-Cell Interactions Driving Diffuse Intrinsic Pontine Glioma Growth After GD2-CAR T Cell Treatment

Diffuse intrinsic pontine glioma (DIPG) and other H3K27M mutated diffuse midline gliomas are universally fatal pediatric brain cancers, with a median survival of just 11 months. Emerging immunotherapies offer new promise, as early results from a clinical trial using CAR T cells targeting GD2 have demonstrated clinical and radiographic improvements, including one long-term survivor. Yet, responses are heterogeneous, and most patients still relapse. A major barrier to effective therapy is the profoundly immunosuppressive, pro-tumorigenic microenvironment, marked by scarce adaptive immune infiltration and abundant myeloid cells. Neuronal signaling and distinct myeloid cell states are known to promote tumor growth; however, the cell–cell interactions that establish this immunosuppressive niche and drive tumor growth and therapy resistance remain poorly understood. 

Advances in single-cell and spatial transcriptomics now allow unprecedented dissection of the tumor microenvironment at unprecedented resolution, offering a powerful opportunity to uncover the mechanisms driving treatment failure. Furthermore, a comprehensive repository of patient-derived DIPG cultures and orthotopic xenograft models developed in our lab enables functional testing of candidate cell–cell interactions in vitro and in vivo, bridging discovery and translational validation. The central hypothesis of this proposal is that signaling between neural, immune, and tumor cells generates spatially organized immunosuppressive microniches that promote both tumor progression and resistance to CAR T therapy. To test this hypothesis, we will leverage unique access to clinical biopsy material from patients treated with or without GD2 CAR T cells, together with our robust in vitro and in vivo models. The specific aims are: 1: Define the transcriptional programs and cellular states linked to DIPG progression after GD2-CAR T therapy using single-cell RNA sequencing of tumor biopsies. 2: Identify spatially organized cell–cell interactions that drive treatment-resistant microniches using spatial transcriptomics of DIPG tumor sections. 3: Functionally validate and target candidate cell–cell interactions to overcome tumor progression and therapy resistance using patient-derived DIPG cultures and orthotopic xenograft models. 

This work will advance fundamental understanding of tumor–immune–neural interactions in DIPG, identify novel therapeutic targets, and provide a pipeline for rational design of more effective immunotherapies. The intellectual merit of the project lies in its integrated, multi-modal approach combining rare human tissues, high-resolution spatial and single-cell mapping, and functional validation. Broader impacts include the potential to improve survival for children with DIPG, inform immunotherapy strategies for other pediatric and adult brain tumors, and provide mechanistic insights into how neural–tumor–immune ecosystems influence therapy outcomes more generally.