Combinatorial Engineering of Multicellular Synthetic Immunotherapy Systems

Project Summary/Abstract: Immune cells—T cells, natural killer (NK) cells, or macrophages—engineered to express synthetic receptors have emerged as powerful anti-cancer treatments, but suffer from several shortcomings including short immune cell persistence, immunosuppression by tumor associated macrophages, and failure to recognize evolving tumors. Some of these problems might be overcome by co-engineering two receptors, chimeric antigen receptors (CARs) and synthetic cytokine receptors (SCRs), to program immune cells with optimized anti-tumor properties. A more ambitious strategy to improve cell therapies is to draw inspiration from the natural immune system, which simultaneously deploys multiple immune cell types to mount an immune response, harnessing the unique strengths of each cell type. This proposal outlines a research program in my lab to create new cell therapy technologies that (1) co-optimize the signaling domains of CARs and SCRs to achieve enhanced anti- tumor immune cell function, (2) co-engineer T cells, NK cells, and macrophages to synergize as a synthetic immune system against cancer, and (3) train machine learning models to predict and understand how synthetic receptors encode the functions of these engineered immune cells. We have developed a platform to rapidly build hundreds to thousands of synthetic signaling proteins and screen them in primary human immune cells in pooled or arrayed contexts. In previous work and pilot studies, we have shown that this platform enables generation of receptors with diverse effects on immune cell phenotypes such as tumor killing and immune cell state, and that library screening data can be used to train machine learning models for rational cell therapy design. Our lab will adapt this platform to optimize pairs of synthetic receptors that enhance immune cell survival, resistance to immune suppression, and tumor killing. We will use synthetic receptors that encode diverse phenotypes to create the first synthetic immune systems in which T cells, NK cells, and macrophages synergize to effectively target cancers. Screening and analysis of synthetic receptor libraries will reveal how crosstalk between two engineered signaling domains encodes cell functions, and how engineered immune cells interact in the context of a synthetic anti-cancer immune system. Neural networks and dynamical models trained on the data will uncover design rules that aid in development of next-generation cell therapies. This work will expand the scales on which we engineer immune cells, enabling us to co-engineer multiple receptors to work together within a cell, and to engineer multiple cell types to work together as a cell therapy.