Abstract B092: A preclinical ex vivo model for glioblastoma captures patient heterogeneity in drug response

Abstract Introduction: Glioblastoma multiforme (GBM) is an aggressive malignancy characterized by rapid progression, heterogeneity, and resistance to conventional therapies. Traditional in vitro and animal models often fail to replicate the complex tumor microenvironment and cellular diversity of human GBM, limiting their utility in effective drug testing and translation to clinical application. To address these limitations, we developed a 3D Ex Vivo Patient Tissue (EVPT) assay for GBM which uses patient-derived tumor tissue. This approach maintains the tumor heterogeneity and cytoarchitecture of the tumor, enabling more clinically relevant assessment of therapeutic candidates. In this study, the EVPT model was used to test novel tumor-specific therapies, highlighting potential breakthroughs in the treatment of GBM. Methods: 10 patient tumor tissues from GBM patients were procured from commercial tissue providers and processed within 24 hours of dissection to preserve the TME. Fresh, minimally processed tumor material was embedded in a protein-rich hydrogel and exposed to novel peptide-drug conjugates (PDC): a peptide carrier and a cytotoxic payload from Oncopeptides at various doses in 384-well format for 5-7 days. As controls, we included Staurosporine (maximum killing), Temozolomide (one of the few options and standard of care for GBM), as well as investigational drugs currently in clinical trials: QBS72S (Alkylating Agent) and Panobinostat (Histone Deacetylase (HDAC) Inhibitor). Phenotypic screening was conducted using our proprietary automated high-content imaging (HCI) platform. Additionally, sample characterization was performed on fixed tissue samples by the IHC, validating GBM tumor identity, proliferation and TME presence. Results: All patient tissues responded strongly to Staurosporine, while exhibiting resistance to Temozolomide. QBS72S and Panobinostat showed sample-dependent killing activity but were outperformed by the PDC compounds, demonstrating over a five-fold increase in effectiveness. PDC compounds elicited strong to mild killing responses, reflecting the heterogeneity in the samples' reactions (>10-fold difference). Notably, the lead PDC compound demonstrated strong cytotoxic activity, effectively killing tumor cells at highest dose in 9 out of 10 samples. Conclusion: Our results demonstrate the novel PDCs' great potential over standard of care Temozolomide, highlighting its promise as a more effective therapeutic strategy for GBM. The EVPT platform effectively aided in lead compound selection, with a clear assay window between less and more potent compounds. Moreover, drug response heterogeneity was accurately captured between patients. This represents a major advancement in preclinical drug testing and glioblastoma therapy development, enhancing the likelihood of successful translation to clinical outcomes. Ultimately, this approach promises to improve treatment strategies and patient outcomes in the long term. Citation Format: Ezgi Kaya Aksoy, Ulrica Westermark, Kristin Hammer, Luca Gandini, Jolie Flach, Emma Spanjaard, Stefan Svensson. Gelius, Marrit Putker, Nataliia Beztsinna. A preclinical ex vivo model for glioblastoma captures patient heterogeneity in drug response [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2025 Oct 22-26; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2025;24(10 Suppl):Abstract nr B092.