Brain tumors have become the leading cause of cancer-related death in children. An important hurdle to scientific and clinical progress in the field has been the limited availability of preclinical tumor models. Historically, few pediatric brain tumor cell lines have been established and these often poorly recapitulate the phenotypes of the original tumors. In recent years, the Children’s Brain Tumor Network (CBTN) has accelerated the development of patient-derived cell lines and xenografts, offering these resources to the community through open-source access. While these models are extremely valuable, their development process can be lengthy and result in clonally selected lines which presents a challenge for studying complex tumor biology. To address the need for three-dimensional tissue culture, our group in conjunction with CBTN, utilized organoid culture from fresh tissue specimens obtained directly from surgical resection of various pediatric brain tumor histologies. This resulted in the development and banking of over 30 organoid models, which included ependymoma, high-grade glioma, medulloblastoma, atypical teratoid-rhabdoid tumor, diffuse midline glioma, and low-grade glioma diagnoses. Tissue was processed within an hour post extraction and cultured with universal media composition for each diagnosis. Organoid growth was observed within 2-3 weeks of initiation and continued for up to three months before banking. Banked organoids established growth upon return to culture. Phenotypic analysis revealed organoid cell composition that represented clinical histology. Importantly, organoids returned to culture post-banking demonstrated similar cell composition to those in the original culture, indicating their utility for subsequent preclinical testing. Here we provide a simple and efficient workflow for the generation and characterization of three-dimensional tumor organoids generated from fresh surgical pediatric brain tumor tissue. The platform has the potential to accelerate investigations into tumor biology and empower a diverse array of translational studies for the pediatric brain tumor field.

Pediatric brain tumor preclinical field suffered for years from the lack of in vitro and in vivo models. With the explosion of novel therapy approaches for solid and brain tumors, including the immunotherapies it is essential to maximize the access to preclinical models for preclinical specificity, efficacy as well and safety. One of the many ways the Children’s Brain Tumor Network (CBTN) accelerates the pediatric brain tumor research and discovery is through support of the tumor model development program. This program focuses on the generation, characterization, and distribution of diverse models to investigators worldwide provided free of charge. Here we present the resource platform with over 150 cell lines, organoids and patient derived xenografts (PDX) developed and/or propagated at D3b at CHOP on behalf of CBTN. This platform maximizes the tumor tissue use to generate a combination of cell line, organoids and/or xenograft models grown in animals. In recent years, consortium supported over 40 requests for cells lines used in basic biology and translational studies internationally. Current efforts focusing also on supporting large-scale data generation and testing through its collaborative model (Childhood Cancer Model Atlas, Procan, National Center for Advancing Translational Sciences) to maximize the molecular information available for each tumor model essential in preclinical screenings. The generated and returned to consortia data are bound with the deidentified patient clinical information and genomic data and freely available through Kid’s First Data, Cavatica and PedcBio portals. These efforts have already attracted interest from pharma stakeholders previously not observed in pediatric brain environment. This open-source repository model is an example of a unique research partnership supported by patients and their families and built with one mission to bring fast change to kids suffering from brain tumors.

Brain tumors are the leading cause of cancer-related death in children. An important hurdle to scientific and clinical progress in the field has been the limited availability of preclinical tumor models. In recent years, the Children’s Brain Tumor Network (CBTN) has accelerated the development of patient-derived cell lines and xenografts, offering these resources to the community through open-source access. While these models are extremely valuable, their development process can be lengthy and result in clonally selected lines which presents a challenge for studying complex tumor biology. To address the need for three-dimensional tissue culture, our group in conjunction with CBTN, generates organoids from fresh tissue specimens obtained directly from surgical resection. This effort resulted in numerous organoid models generated for high-grade glioma (HGG), medulloblastoma, ependymoma, atypical teratoid-rhabdoid tumor, and low-grade glioma tumors. The tissue was processed within an hour post-extraction and cultured in growth factor-free media. Organoid formation and growth were observed as early as in 2 weeks post initiation and continued for up to three months. Banked organoids established growth upon return to culture and resembled similar cell composition to those in the original culture. This provides an excellent preclinical testing model for precision medicine-driven trials. To represent this, we utilized CAR-T testing approach and treated GD2-positive HGG organoids with GD2 CAR-T cells and control CD19 CAR-T cells. 72h exposure to GD2 CAR-T induced morphological changes and significant increase in dead cells within organoids, which was not observed in CD19 CAR-T treated counterparts (p<0.05). Here we provide a simple and efficient workflow for the generation and testing of three-dimensional tumor organoid models generated from fresh surgical tissue. The platform can accelerate not only tumor biology explorations but also empower a diverse array of precision medicine-driven testing for pediatric brain tumors.