Development of a novel preclinical immunocompetent model of glioblastoma that reflects tumor heterogeneity 2260765

E Eliese Moelker (Duke Univ) A Alexandra Hoyt-Miggelbrink (Duke University) S Shannon Wallace (Duke University) V Vatsal Jain P Pamela Norberg (Duke University School of Medicine) K Katayoun Ayasoufi (Duke University School of Medicine)

Abstract

Abstract Introduction Glioblastoma is an aggressive, lethal brain tumor with limited treatments and poor patient survival. Several murine tumor lines are used as models in preclinical studies, each with their own strengths and weaknesses, but little has been published about which most closely reflects human tumors in respect to cell composition, immune microenvironment and disease progression. Tumor heterogeneity is often given as a reason for high reoccurrence rate in GBM, but most preclinical work uses only one cell type at a time. A systematic, cross-species comparison of each model cell type and a mixture of all three to human tumor is critical to identify which model is best used preclinically for development of promising immunotherapies. Methods C57BL/6 mice were intracranially implanted with GL261, SB28, CT2A, and a combination of the three and monitored until humane endpoint. MRI, bulk RNAseq, and flow cytometry on brain and immune organs was performed. Additionally, tumor-bearing mouse brains and resected human GBM tissue were fixed and imaged by MRI. Results Differences in survival, tumor invasion, volume, gene expression profiles, and immune landscape were observed across both murine and human tumors. Bulk RNAseq results suggest model-specific patterns in highly expressed genes. Top genes for each murine tumor type were: immune evasion, including SERPINA3 (GL261), fibrosis programming, including collagen (CT2A), and cell adhesion and innate immunity, including CD44 and IFITm2 (SB28). Conclusion To our knowledge, we have made the first preclinical GBM model to truly capture tumor and immune heterogeneity. Differences in murine GBM tumor models and correlating aspects of human tumors offer promising insights for refining preclinical modeling strategies. Our approach serves as a more representative platform, improving translational relevance of available murine models for better immunotherapy development to improve outcomes for patients and their families affected by this devastating disease. Funding Source n/a Topic Categories Tumor Immunology: Cellular Responses and Tumor Microevironment (TIME)

Article Details

Volume / Issue Vol. 215, Issue Supplement_1
Published August 01, 2026
ISSN 0022-1767
Publisher American Association of Immunologists

Authors (6)

E

Eliese Moelker

Duke Univ

A

Alexandra Hoyt-Miggelbrink

Duke University

S

Shannon Wallace

Duke University

V

Vatsal Jain

P

Pamela Norberg

Duke University School of Medicine

K

Katayoun Ayasoufi

Duke University School of Medicine