OTHR-12. LOCATION-RELEVANT GROWTH FACTOR DEPENDENCE ON THE FORMATION OF PEDIATRIC BRAIN TUMOR ORGANOIDS FROM PDOX MODELS AND PATIENT SAMPLES

Abstract

Abstract Brain cancer constitutes the most common type of childhood malignancies and the leading cause of death in children. Despite recent advances in diagnostic methods, surgical techniques as well as chemo- and radio-therapeutic strategies; the survival rates in many types of malignant pediatric brain tumors remain dismal, with <25% of children with high grade glioma surviving five years after initial diagnosis and nearly all patients with diffuse intrinsic pontine glioma succumbing to the disease within 9-12 months. Moreover, even among brain tumor survivors, many patients are left with long-term cognitive and/or neuroendocrine sequalae. Therefore, more efficacious therapeutic modalities that can improve the patient’s quality of life and increase their survival are urgently needed. 3D organoid cultures have arisen as clinically relevant and molecularly accurate in vitro brain tumor model systems for preclinical drug testing. These models have demonstrated to faithfully recapitulate the genomic and phenotypic complexities of the original tumor and thereby, can reliably predict the patient response to treatment. Recognizing the impact of the tumor microenvironment and the brain developmental stage on brain tumor growth, we developed a novel protocol to evaluate the impact of tumor location-related niche factors on the development of brain tumoral organoids. Utilizing different growth factor combinations that are unique/selective to different regions of brain development, we were able to develop 39 PDOX-derived organoids from a total of 40 PDOX models tested (success rate: 97.5%) and 23 patient-derived organoids from a total of 29 patient samples tested (success rate: 79%). Furthermore, cellular characterization studies revealed the replication of an heterogenous cell population. Altogether, our study identifies differences on growth factor dependence for the formation of organoids from brain tumors of different locations and demonstrates that PDOX models are valuable tools for the development and optimization of protocols for organoid model establishment.