Abstract 6275: Leveraging the tumor microenvironment (TME) atlas to identify the best PDX-model for pre-clinical testing of TME targeting agents

Abstract Tumor-infiltrating immune cells play a critical role in tumor progression, metastasis, and resistance to anti-cancer therapies. This is particularly evident in the context of tumor evasion from immune surveillance, which is driven by tumor microenvironment (TME) modulation, ultimately impairing therapeutic efficacy. Consequently, accurately predicting the composition of immune populations within the TME holds great promise for uncovering the complex interplay between immune cells and tumors, as well as for the selection of the optimal pre-clinical model to test new TME-targeted therapeutics. The relevance of experimental settings in preclinical immunotherapy studies is often undermined by the lack of comprehensive TME characterization in the cancer models selected for pre-clinical testing. To address this gap, Champions Oncology generated the TME Atlas. Champions Oncology’s preclinical program is built on Patient-Derived Xenograft (PDX), a highly accurate and clinically relevant platform for preclinical research. The PDX bank encloses 1500 models, characterized with transcriptomic, proteomic, and patient treatment history data to guide the model’s selection mimicking the future clinical patient cohorts. On top of this, the TME Atlas will provide an additional layer of characterization to refine the model selection based on TME composition, including immune cells and fibroblasts. The TME Atlas derived from the transcriptomic data from Champions Oncology PDX bank and enabled the prediction of TME composition patterns in each specific model. To profile the TME, we applied different computational methods to Champions Oncology’s proprietary datasets. The computational tool xCell was identified as the optimal strategy to perform cell-type enrichment analysis for 64 immune and stromal cell types. The implementation of xCell enabled comprehensive profiling of immune infiltration across the entire PDX bank. This detailed characterization of the TME unlocks the selection of optimal models and experimental settings to test TME-engaging drugs. Histological analyses further validated the computational findings enclosed in the TME Atlas. To validate the computational data, we engrafted selected PDX models in humanized mice and assessed TME composition in tumors through Immunohistochemistry (IHC). IHC analyses validated the profile predicted within Champions Oncology TME Atlas. This immune atlas enables the selection of the most suitable PDX models, and humanized hosts to optimize preclinical studies targeting immune and stromal components of the TME. This approach ensures clinically translatable results, that will advance the development of innovative TME targeting strategies. Citation Format: Mara Gilardi, Gilad Silberberg, Kurtis Cruz, Stefano Cairo, Marianna Zipeto, Michael Ritchie. Leveraging the tumor microenvironment (TME) atlas to identify the best PDX-model for pre-clinical testing of TME targeting agents [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6275.