STEM-11. INTERROGATING INTRATUMORAL TRANSCRIPTOMIC HETEROGENEITY AND PLASTICITY AS RESISTANCE MECHANISMS IN GLIOBLASTOMA

Intratumoral heterogeneity is one of the major barriers in the treatment of many tumor types. In glioblastoma (GBM) – the most aggressive brain tumor - survival of patients following standard therapy has stagnated to an average of 14 months, largely due to the inherent heterogeneity and phenotypic plasticity. Patient-derived orthotopic xenografts (PDOX) are important tools for preclinical cancer studies. We have shown that they recapitulate histological and molecular features of patient tumors, thus producing more clinically relevant study outcomes when compared to other preclinical models. To understand the impact of treatment on intratumoral heterogeneity and interactions of tumor cells with the microenvironment, we applied single-cell RNA-seq in GBM PDOX models generated from naive and treatment exposed patient tumors. Here we show that PDOX models recapitulate all the major cell types and transcriptional programs reported in GBM patient samples. We demonstrate that GBM tumor cells occur in distinct phenotypic states dictated by microenvironmental cues and treatment-induced pressure. Non-neoplastic cells present within xenografts undergo transcriptomic adaptation similar to the GBM microenvironment in patients. Analysis of treated PDOXs and longitudinal PDOX models derived from patients prior and after treatment allows us to investigate short-term and stable long-term transcriptomic adaptations. We aim to reveal key molecular regulators enhancing tumor cell plasticity and resistance, that could be targeted in combinatory treatments. Our GBM PDOX models provide a platform for high-quality pre-clinical research and an improved clinical relevance in the development and testing of novel therapeutics.