Modeling immunotherapies in live 3D human cancer tissue bioreactors

Abstract

Background: Cancer immunotherapies have shown remarkable efficacy in advanced malignancies, yet many patients remain unresponsive. This variability, along with concerns about adverse effects and healthcare costs, highlights the need for predictive biomarkers and physiologically relevant cancer models to forecast individual treatment responses. Existing systems inadequately recapitulate the human tumor microenvironment (TME), which is essential for understanding immune–tumor interactions and treatment efficacy. Here, we developed an ex vivo 3D human tissue culture model that preserves the native TME for functional immunotherapy testing. Such a short-term culture platform also supports functional precision medicine by enabling rapid ex vivo assessment of therapeutic responses to guide clinical decisions. Methods: Fresh, intact human lymph node (LN) tissue pieces were cultured in optimized perfusion bioreactors for three days, during which CAR T cell therapies and antibody-based treatments were administered. Post-culture analyses were performed using flow cytometry, histology, and multiplexed fluorescence microscopy. Results: The bioreactor system significantly improved tissue viability compared to traditional plate cultures. Novel CAR T cells with enhanced PI3K signaling exhibited superior tissue infiltration but showed comparable cytotoxicity to conventional CAR T cells. Pembrolizumab, a PD-1 inhibitor, significantly reduced lymphoma and melanoma cell viability without affecting benign LN tissues. Conclusions: This optimized bioreactor culture system provides a robust platform for evaluating immunotherapy efficacy within a physiologically relevant TME. It offers valuable potential for advancing personalized treatment strategies, accelerating the understanding of immunotherapy mechanisms, and improving clinical outcomes.