Experimental models of pediatric brain tumors. Establishment, immunophenotyping and clinical implications

Abstract: Brain tumors are the most common solid tumors in children. Current treatment protocols fail in 25% of patients and are associated with significant long-term adverse effects in survivors. Experimental models of these tumors are scarce and will be crucial for the development of more efficient treatment strategies, including molecular targeting and immunotherapy. In this thesis, I describe the establishment and characterization of novel in vitro and in vivo models of pediatric brain tumors. I initially define a standardized protocol for establishment of patient-derived cell cultures, based on the concept of serum-free monolayer culturing. In addition, I describe the generation of an orthotopic xenograft model of a high-risk Group 3 medulloblastoma (MB-LU-181) by cerebellar inoculation of low-passage tumor cells. The newly established experimental models were phenotyped alongside patient samples, with emphasis on inflammatory mediators that could serve as future targets for therapeutic intervention. Tumor markers, cytokine signatures and components of the COX-2/mPGES-1/PGE2 pathway were generally preserved following propagation of tumor cells in vitro and in vivo, demonstrating the biological faithfulness of the models. CD24 was identified as a clinically and experimentally useful immunomarker for medulloblastoma cells, but additional detailed studies are needed to determine the prerequisites for targeted treatment. PTGS2 (COX-2) and VEGFA were overexpressed in Group 3 medulloblastoma compared to other medulloblastoma subgroups; COX-2 was further evaluated as a therapeutic target in an immunocompetent high-grade glioma model, where simultaneous administration of COX-2 inhibitors and GM-CSF based immunotherapy cured >60% of tumor-bearing mice. I finally performed a systemic immune characterization of children with brain tumors. Multiplex analysis of preoperative plasma samples identified patient groups with distinct cytokine profiles, which could have important implications for the development and clinical implementation of immunotherapies. In brief, this thesis presents novel experimental models that recapitulate the phenotype of pediatric brain tumors and will serve as tools for future studies of tumor biology and preclinical drug evaluation. The results also implicate a role for immune intervention and monitoring in the treatment of children with brain tumors.