Abstract
In a multi-disciplinary project, patient derived glioblastoma stem-cell cultures (GSCs) have been established and characterized extensively with the goal of applying computational efforts to integrate results from high-throughput screens of drugs and RNAi with genomic and transcriptional profiling to predict the most successful therapy for individual glioma patients. The main goal of this study was to develop a platform for testing the tumor-initiating capacity of GSCs in mouse brain and to label the cells with GFP-luciferase to enable non-invasive quantification of tumor growth by in vivo bioluminescence imaging. Adherently grown GFP-luciferase labeled glioblastoma stem-cell cultures were dissociated and injected stereotactically into immunodeficient mice. Tumor growth was monitored by IVIS imaging for up to 40 weeks and brains were collected for histopathological and immunohistochemical stainings. Automatic quantification and growth pattern analysis of tumor cells in brain sections was set up based on human cell specific staining and a CellProfiler Analyst’s machine learning classifier with a manual observer correlation of 0.86. GSC xenograft tumors with a wide range of histopathological features and biological behaviors recapitulating high grade astrocytoma was confirmed in mice injected with 14 of the 29 glioblastoma cell cultures (48%). Glial lineage markers, such as Sox2, GFAP, and Olig2, were expressed both in patient tumors and patient derived xenografts. Individual glioblastoma cell cultures were either characterized by formation of condensed tumors (8/14, 57%) or diffuse infiltrative growth through the brain parenchyma (6/14, 43%). In summary, we present a valuable mouse model for preclinical studies of glioma. Integrative analysis of molecular profiles with growth pattern data has the potential to unravel genes that distinguish tumors with more infiltrative growth. Our biobank of luciferase labeled GSCs may be useful for longitudinal monitoring of tumor growth dynamics in the setting of therapies and provides a foundation for testing the effect of predicted drug vulnerabilites in orthotopic xenografts mouse models.
Citation Format: Cecilia Krona, Soumi Kundu, Karl Holmberg-Olausson, Riasat Islam, Rashmi Ramachandra, Ludmila Elfineh, Sven Nelander. In vivo modeling of high grade glioma for oncology drug development [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4813. doi:10.1158/1538-7445.AM2017-4813