P18.06.B ASSESSING TEMPORAL METABOLIC EVOLUTION OF GLIOBLASTOMA IN TUMOR MICROENVIRONMENT USING THE COFM XENOGRAFT IN VIVO MODEL.

Abstract BACKGROUND Glioblastomas (GBM) are constructs that demonstrate a high degree of adaptability, capable of altering their characteristics to suit the prevailing conditions and thereby ensure their own survival. Recent studies have demonstrated that tumors not just undergo morphological changes over time, but their metabolic processes also exhibit variability depending on their status. In this study we assessed dynamics of the metabolome in the tumor microenvironment (TME) by collecting interstitial fluid (ISF), with Open Flow Microperfusion (cOFM), cerebrospinal fluid (CSF), and blood from human GBM-bearing animals at several time points along the tumor progression timeline. The samples were then compared to those collected from non-GBM-bearing animals (non-GBM controls) in order to enhance the focus on the metabolome in TME and to identify new potential metabolic biomarkers or biomarker profiles that reflect the local situation of TME. MATERIAL AND METHODS We applied the atraumatic access to human GBM xenograft with cOFM that allows sampling of metabolites inside the TME with functional blood-brain barrier using U87 cells. Cancer cells were implanted orthotopically into the striatum region of immunodeficient mouse model (Athymic nude) by using the previously implanted cOFM. Tumor growth was monitored with magnetic-resonance-imaging. On days 7, 10, and 15 post-implantation, ISF from TME was collected with cOFM. CSF and brain tissue were collected from the same animals at terminal time point. Blood plasma samples were taken at matching time points. The collected samples were analyzed using a LC-HRMS metabolomics platform. Metabolomics data from GBM-bearing animals were analyzed and compared with that from non-GBM controls. RESULTS Tumor progression curve indicated a typical development of a U87 tumor. Moreover, histology revealed that even multiple collections from TME did not affect the tumor morphology. The metabolome from the GBM-bearing animals showed distinct metabolic patterns when compared to that from the non-GBM controls. Furthermore, the metabolome within the GBM group demonstrated distinguishing patterns at the different collection time points along the tumor progression trajectory. This metabolic adaption was visible in particularly in energy metabolism and inflammatory pathway. CONCLUSION This in vivo xenograft study not just demonstrated the unique capabilities of the cOFM technology investigating evolution of brain tumors, but also nicely showed the ability of GBM tumors adapting their metabolism with respect to their current needs during tumor progression. These results indicate potential therapeutic windows and highlight the importance of the time point for the GBM treatment.