Abstract A031: Engineering antigen density sensors for T-cell immunotherapy

Abstract

T-cells can be redirected to kill tumor cells via synthetic T-cell receptors known as chimeric antigen receptors (CARs); this approach is becoming a highly promising therapeutic strategy for cancer treatment. Current CAR T-cells, while effective at killing cells expressing the target antigen, fail to discriminate between cells expressing high and low levels of common antigens. Potential target antigens are often also found on normal cells at lower expression levels, and CAR T-cells get activated even by low amounts of target antigen, resulting in potentially deadly toxic effects. We are approaching this problem of antigen density discrimination by using common biochemical mechanisms for cooperative recognition to design new synthetic T-cell receptors. CAR T-cells are highly sensitive to even low densities of the target antigen. Current strategies for generating CAR T-cells consist of selecting antibodies with high affinity since previous studies have shown that the CAR T-cell activity is inversely correlated with the antibody’s affinity. However, these CARs are unable to discriminate between cancer and normal cells based on antigen density. Recent studies suggest that lowering the receptor affinity on a CAR T-cell could increase its selectivity against targeT-cells with different antigen densities. Nevertheless, lowering the affinity is unlikely to provide a sharp density threshold of activation. Instead for a sharp threshold transition from off to on, one would require cooperativity and nonlinear recognition. We have designed a two-step recognition-activation circuit that involves two receptors. In this circuit an initial recognition event alters the potency of a subsequent response. We combined a weak receptor that turns on the activity of a high-affinity receptor that fully activates the cell. For this circuits we have made used of a new class of modular receptors called synthetic notch receptors (SynNotch). SynNotch receptors use antibody-based domains to recognize a target antigen, but when activated, they control transcription via a cleaved transcriptional domain. We have designed and expressed several synNotch/CAR circuits and have tested in vitro their ability to achieve antigen density sensing. These circuits use a synNotch receptor to control the expression of a CAR—all recognizing the same Her2 ligand. The constructs are introduced into T-cells via standard lentiviral infection, and sorted for receptor expression using flow cytometry (anti-myc stain). We have screened candidate receptors for expression and activation response in the human Jurkat T-cell line and human primary CD4+ and CD8+ T-cells from several donors, using standard flow-based assays (e.g. CD69 Activation, T-cell proliferation) and cytokine assays (e.g., IL-2, INF-γ). We are now taking the best circuit candidate and proceeding in mouse models. We have first characterized the growth curves of tumors made by engineered targeT-cells to measure the baseline tumor growth and then I have tested either conventional CAR T-cells or my new engineered T-cells in bilateral tumors into NSG immunocompromised mice, in which one side contains a low-density Her2 K562 tumor, and the other side will contain a higher-density Her2 K562 tumor. The engineered T-cells are injected into the tail vein, and allowed to traffic freely to either tumor. We are monitoring tumor size by caliper over the course of 4 weeks, and tracking the differential clearance of the high- and low-density tumors. We have confirmed that when using a fixed affinity standard CAR T-cell, then cells kill either high- or low-expression-level Her2 tumors with equal efficiency. I have begun parallel mouse experiments with the two receptor circuits that show improved antigen density discrimination in vitro. While still in progress, preliminary results look promising. Citation Format: Rogelio A. Hernandez-Lopez, Wendell A. Lim, Wei Yu. Engineering antigen density sensors for T-cell immunotherapy [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr A031.