Abstract PL01-05: Next-generation CAR T cells designed to overcome tumor resistance

Abstract

Abstract Personalized cell therapy using chimeric antigen receptor (CAR) modified T cells has rapidly evolved from a boutique treatment under study at a handful of academic centers to the first global, commercialized cell and gene therapy. Despite significant and ongoing challenges regarding cost and scalability, the impressive and sustained antitumor effects observed in some patients with otherwise incurable B cell malignancies lead to the prediction that applications for this novel class of therapeutics will increase in the coming years. Further optimism derives from the rapid pace of scientific and technological advances occurring in cellular engineering, which are yielding increasingly sophisticated “next generation” CAR T cell therapeutics, endowed with greater tumor specificity and control and designed to overcome tumor resistance. Current concepts regarding mechanisms of tumor resistance to CAR T cells largely implicate tumor evasion or T cell failure, and these are typically mutually exclusive. In B-ALL, tumor evasion most often derives from selection of variants lacking the targeted epitope (CD19neg relapse following CD19-CAR therapy) or expressing reduced levels of the epitope that are below the threshold for CAR recognition (CD22lo relapse following CD22-CAR therapy)[1, 2]. Antigen neg and antigen lo relapse are anticipated to be increasingly problematic as CARs are applied to solid tumors and AML, diseases which display significant target antigen heterogeneity. Several engineering approaches are available to address this problem, including multispecific CARs. We are currently testing a bivalent, CD19/22-CAR in clinical trials in children and adults with B cell malignancies[1], and this agent demonstrates acceptable safety and significant clinical activity. We, and others, have also generated tri- and quad-specific CARs and anticipate increasing importance of such higher order therapeutics in expanding the application of CAR T cells beyond B cell malignancies. The second major mechanism of resistance to CAR T cells is T cell exhaustion, driven by excessive or prolonged CAR signaling in response to high tumor burdens and/or tonic signaling as a result of antigen-independent clustering of the CAR receptor[3]. Functional, phenotypic and transcriptomic profiling of “exhaustion” in CAR T cells reveals it to be essentially identical to exhaustion found in T cells within the tumor microenvironment or those exposed to chronic viral infection. We have therefore utilized a reductionist, in vitro model of a tonic signaling CAR to study the biology of human T cell exhaustion, as a first step toward engineering “exhaustion-resistant” CAR T cells. Retroviral transduction of healthy human T cells with a tonically signaling CAR, rapidly induces phenotypic, functional, transcriptomic and epigenomic hallmarks of T cell exhaustion. Whole genome interrogation using ATAC-Seq reveals that the most differentially accessible chromatin regions in exhausted vs. non-exhausted T cells are those bound by AP1/IRF transcription factors and RNA-Seq confirms overexpression of both the canonical “activating” AP1 transcription factors (TFs), Fos and cJun, as well AP1/IRF TFs, which have been associated with inhibitory transcriptional profiles (JunB, BATF3, ATF3, and IRF4). We hypothesized that an imbalance between classical vs inhibitory TFs might drive dysfunction in exhausted T cells. Consistent with this, overexpression of cJun (cJun OE) in tonically signaling CAR T cells reduced expression of inhibitory receptors, increased IL2 production, and deletion of IRF4 or BAT3 replicated this effect, and cJun OE also enhances long-term proliferative capacity of CAR T cells bearing receptors that do not tonically signal. In vivo, cJun OE CAR T cells demonstrate enhanced efficacy in multiple xenograft tumor models and enhanced persistence. Molecular mapping reveals that the bZIP domain of cJun for exhaustion prevention, whereas phosphorylation sites involved in transactivation and the DNA binding domains is dispensable, which suggests that the major mechanism of action of cJun OE is to compete for binding, and therefore disrupt inhibitory AP1/IRF complexes. cJun OE also enhanced CAR recognition of antigen lo targets, consistent with a model wherein cJun OE amplifies otherwise subthreshold levels of CAR signaling. These data provide new insights into the biology of human T cell exhaustion, highlighting an imbalance between “activating” versus “inhibitory” AP1/IRF complexes in the fundamental pathophysiology of this state, and demonstrate that CAR T cells engineered for “exhaustion resistance” mediate impressively increased potency in preclinical models. 1. Fry, T.J., et al., CD22-targeted CAR T cells induce remission in B-ALL that is naive or resistant to CD19-targeted CAR immunotherapy. Nat Med, 2018. 24(1): p. 20-28. 2. Walker, A.J., et al., Tumor Antigen and Receptor Densities Regulate Efficacy of a Chimeric Antigen Receptor Targeting Anaplastic Lymphoma Kinase. Mol Ther, 2017. 25(9): p. 2189-2201. 3. Long, A.H., et al., 4-1BB costimulation ameliorates T cell exhaustion induced by tonic signaling of chimeric antigen receptors. Nat Med, 2015. 21(6): p. 581-90. Citation Format: Crystal L. Mackall. Next-generation CAR T cells designed to overcome tumor resistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr PL01-05.