Micro-organospheres retain patient tumor microenvironment for precision immuno-oncology.

2573 Background: Current patient-derived organoid (PDO) models are largely devoid of immune components. We developed a precision microfluidic and membrane platform to generate patient-derived micro-organospheres (MOS) that retain tumor-resident immune and stromal components for personalized immuno-oncology (IO) assays. Methods: MOS were generated from lung, kidney, and colorectal cancer patients. The composition and function of patient tumor-resident immune cells in MOS were characterized by flow cytometry, single-cell RNA-seq, antibody staining, and TCR-seq. High-content and longitudinal imaging with AI analyses were used to quantify tumor cell death and immune cell dynamics inside MOS in response to IO therapies including checkpoint inhibitors, T cell bispecific antibodies, and adoptive tumor infiltrating lymphocyte (TIL) therapies, followed by single-cell analyses. Results: Tumor and stromal cells quickly form tissue niches inside MOS to sustain the viability and function of encapsulated immune cells. MOS derived from lung and kidney cancer patients respond to Nivolumab, indicated by the Annexin V apoptosis marker. ESK1* (TCB antibody targeting HLA-A2/WT1) induces killing in eight lung tumor patients derived MOS. We further developed a MOS T-cell potency assay as autologous TILs or PBMC efficiently infiltrate MOS from lung, kidney and CRC patients and induce tumor cell apoptosis. Adjunctive therapies combining TCB with TILs enhanced the potency of adoptive cell therapy against lung tumor. Based on the data, three ongoing and upcoming personalize IO clinical trials will further validate the ability of the MOS assay to predict patient response to TCB, checkpoint combinations, and adoptive cell therapy. Conclusions: MOS provide a rapid and scalable personalized platform for developing and testing IO therapies such as checkpoint inhibitors, bispecific antibodies, and T cell therapies on patient tumor models that still retain the original tumor microenvironments.