Abstract 7281: Predicting clinical checkpoint inhibitor responses with TumorGraft3D PBMC co-culture

Abstract Although immunotherapy-based therapeutics such as checkpoint inhibitors (CPI) have revolutionized cancer therapy, a lot needs to be done in the field. In preclinical settings, ex vivo models still face significant challenges in translating the results into clinical responses. Traditional approaches, including cell-lines, are limited by their lack of genetic heterogeneity. Similarly, 2D systems fail to capture spatial complexity, significantly reducing their translational relevance. Clinical data on CPI are not available for commercially available cell lines since they were developed before CPI treatments were approved. These limitations hinder the accurate modeling of CPI efficacy and the prediction of therapeutic responses. To address these issues, Champions Oncology leverages a tumor bank of 1500 patient-derived xenograft (PDX) models across solid tumors and hematological malignancies offering 3D co-culture assays specifically tailored for different mechanisms of action. To provide a clinical translational platform that retains human tumor heterogeneity, we developed Champions TumorGraft3D (CTG3D) low-passage 3D PDX-organoid models. The use of proprietary, indication-specific media enabled CTG3D to preserve the genetic, proteomic, and phenotypic diversity of the original tumors. Additionally, their 3D architecture accurately models the cell-cell interactions that are crucial for mimicking immune cell engagement. The in-depth clinical annotation of our bank, which includes detailed patients’ clinical history, enables this system to provide predictive insights into the CPI efficacy. In this project, we assessed the translational value of our CTG3D co-culture system, based on its ability to recapitulate the clinical response in patients. We selected models derived from patients displaying resistance or sensitivity to CPI, by leveraging the data collected in Champions Oncology Tumor Bank. Then, we optimized the CTG3D-PBMCs co-culture settings including media and cell ratios. High throughput flow cytometry was used to interrogate the model. We profiled the co-culture with a custom-designed panel allowing for cancer cell killing and immune phenotyping readout in parallel. Beyond immune activation, we were able to determine the correlation of the data obtained from the co-culture with the clinical records of the original patients. In summary, we developed a cutting-edge ex vivo CTG3D PBMC co-culture platform allowing for dynamic evaluation of immune engagement and checkpoint inhibitor responses, therefore providing valuable translational datasets. By overcoming the limitations of traditional assays, CTG3D co-culture systems represent a transformative step forward in preclinical immunotherapy research and the development of personalized cancer treatments. Citation Format: Mara Gilardi, BanuPriya Sridharan, Veena Jagannathan, Brandon Walling, Stefano Cairo, Marianna Zipeto, Michael Ritchie. Predicting clinical checkpoint inhibitor responses with TumorGraft3D PBMC co-culture [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 7281.