Highlight of 2023: CAR T cells driving precision therapy for autoimmune disease

CAR T cell therapy is showing remarkable results in autoimmune disease with treatment-refractory patients showing durable drug-free remission. Here, we highlight five key papers from 2023 that are driving the development of CAR T cells to improve precision, safety, efficacy and accessibility for the treatment of autoantibody-associated autoimmune diseases. Autoimmune diseases account for a significant burden of chronic incurable disease, affecting approximately 1 in 10 individuals.1 The treatment and ongoing management of patients is difficult, with standard care involving immunosuppressive agents that confer significant side-effects. The emergence of therapeutic monoclonal antibodies (mAbs) and molecular inhibitors, that specifically target B cells, cytokine and complement pathways have improved outcomes. However, many patients remain refractory to treatment and experience ongoing severe pathology. Chimeric antigen receptor (CAR) T cells are a promising therapeutic approach that have revolutionized the treatment of hematological malignancies and are showing remarkable early results in autoimmune disease. CAR T cells are engineered T cells, often patient derived, that have genetically altered receptors conferring the antigen specificity of an antibody linked to intracellular signaling domains that activate cytotoxicity. The most common CAR T cell target is CD19, which has been harnessed recently in autoantibody (B cell)-associated autoimmune diseases. CD19-CAR T cells were administered to five patients with treatment-refractory systemic lupus erythematosus (SLE).2 The patients have remained in drug-free remission for at least 2 years.3 These remarkable results have sparked further investigations for other autoantibody-associated autoimmune diseases and the refinement of CAR T cells that can specifically target autoreactive immune cells. This Research Highlight will discuss five key studies from 20234-8 that are driving the development and translation of precision cellular therapy for autoimmune disease. While CD19-CAR T cell therapies have shown promising results in SLE, another strategy for autoantibody-mediated autoimmune diseases is to directly target antibody-secreting cells. Qin et al.4 used CAR T cells targeting the B cell maturation antigen (BCMA), which is expressed predominantly by plasma cells and some activated B cells and plasmacytoid dendritic cells. Anti-BCMA CAR T cells were administered to 12 patients with neuromyelitis optica spectrum disorder (NMOSD), an autoimmune disease caused by anti-aquaporin 4 (AQP4) autoantibodies damaging the optical nerve and spinal cord. Improvements in vision were reported in half of the patients, while 8 (67%) showed improvement in their disabilities, including walking without assistance. No serious cases of cytokine-release syndrome (CRS) were encountered. During a median follow-up of 5.5 months, 11/12 (92%) of patients remained in remission with undetectable anti-AQP4. Interestingly, patients with concomitant autoimmune diseases, Sjögren's disease (SjD) and rheumatoid arthritis, also demonstrated marked improvements in subjective and objective measures of these diseases including decreased SSA/Ro autoantibodies for the former. While these data represent a promising new approach for treating autoimmune disease, these findings also provide insights into the cellular origins of autoantibodies. Therapeutic anti-CD20 mAbs do not reliably reduce the serum titers of anti-AQP4 or anti-SSA/Ro IgG in NMSOD and SjD, respectively.9, 10 These findings imply that anti-AQP4 or anti-SSA/Ro autoantibodies could derive from long-lived CD20-negative plasma cells. Therefore, anti-BCMA CAR T cells may represent a promising approach for autoimmune diseases where pathogenic autoantibodies are derived from CD20-negative plasma cells. Given that autoimmune diseases are highly heterogenous and pathogenic autoantibodies likely derive from multiple sources, studies are currently underway to use combined targeting of CD19 and BCMA CAR T cells.11, 12 However, these treatments will require caution, as seven (58%) of the patients treated with anti-BCMA CAR T cells developed infections and all patients developed a low-grade CRS.4 Given the increased risk of infection from depleting total B cells or plasma cells, the next steps are to specifically target the rare population of B cells/plasma cells that are responsible for producing pathogenic autoantibody. Chimeric autoantibody receptor (CAAR) T cells have been developed to target B cells bearing specific autoantibodies. One of the earliest examples was CAAR T cells targeting the pemphigus vulgaris autoantigen, desmoglein 3,13 which has progressed to clinical trials.14 In 2023, Oh et al.5 engineered CAAR T cells targeting muscle-specific tyrosine kinase (MuSK) autoantibodies in myasthenia gravis, which disrupt neuromuscular junction signaling. Xenograft mouse models of Nalm-6 B cells expressing anti-MuSK autoantibodies isolated from patients, confirmed specificity and cytotoxicity of MuSK-CAAR T cells. Moreover, in an experimental mouse model of myasthenia gravis, MuSK-CAAR T cells reduced serum anti-MuSK autoantibodies one week post-infusion without altering total IgG levels or total B cell counts. One of the concerns for CAAR T cells is that autoantibodies present in the patient serum could block their activity. In vitro experiments with anti-MuSK expressing Nalm-6 cells, indicated minor inhibition of CAAR T cell cytotoxicity in the presence of myasthenia gravis patient IgG. However, this was overcome by increasing the co-incubation time and the numbers of CAAR T cells. Chimeric autoantibody receptor T cells have also been engineered for the treatment of Graves' disease caused by autoantibodies targeting thyroid-stimulating hormone receptor (TSHR). Duan et al.6 manufactured CAAR T cells, from human peripheral blood mononuclear cells (PBMC), expressing the first extracellular domain of TSHR fused with CD8 transmembrane region, 4-1BB and CD3ζ. In preclinical models of Graves' disease, B-NDG mice (severely immunodeficient mice) injected with a hybridoma producing anti-TSHR autoantibody develop biochemical hyperthyroidism and weight loss. The infusion of human TSHR-CAAR T cells reversed these pathologies. A potential issue for CAAR T cells that express a hormone receptor is that cytotoxic activity could be directed to cells making TSHR-binding hormones instead of cells making anti-TSHR autoantibodies. Importantly, circulating hormone levels were not altered in mice receiving TSHR-CAAR T cells and TSHR-CAAR T cells did not exhibit activation or cytotoxicity in the presence of pituitary cells in vitro. Preclinical studies of both MuSK and TSHR CAAR T cells show promise for myasthenia gravis and Graves' disease, respectively, and are both progressing to clinical trials in humans. However, there are some limitations for CAAR T cells. Firstly, this approach requires prior knowledge about pathogenic autoantibodies and extensive efforts mapping disease relevant epitopes. It also assumes that elimination of this autoantibody would significantly ameliorate disease, which is difficult to study in pre-clinical models that may not accurately reflect complex human disease. Secondly, plasma cells that have lost their surface immunoglobulin expression will not be targeted by this approach. Autoantibody responses that are resistant to anti-CD20 mAb may be produced from long-lived plasma cells that have down-regulated surface immunoglobulin receptors, and may not be the best candidates for CAAR T cells. Another precision cellular approach currently under investigation is to engineer autoantigen specific regulatory T cells (Tregs) to suppress effector responses in patient tissues.15, 16 A potential limitation for this approach is the stability/plasticity of the CAR-modified Treg cell product, where autoantigen-specific CAR Tregs could convert into T effectors and initiate more tissue damage. To address this issue, Henschel et al.7 overexpressed FOXP3 in human PBMC-derived CAR Tregs, by fusing FOXP3 cDNA to the CD28 CD3ζ signaling domain. CAR Tregs overexpressing FOXP3 maintained a stable Treg, even in inflammatory conditions and in the absence of IL-2. FOXP3 CAR Tregs also showed significantly enhanced ability to inhibit allogenic T effector proliferation compared with control CAR Tregs. Moreover, forced expression of FOXP3 converted CAR T effector cells to a Treg phenotype with suppressor activity. Traditionally, CAR T cells are engineered using DNA viral vectors, which require patients to undergo lymphodepleting chemotherapy and continued monitoring for toxicity after infusion. Therefore, CAR T cell treatment of autoimmune disease is currently limited to those with severe and refractory symptoms. To increase accessibility of CAR T cell therapy, outside a hospital setting, Granit et al.8 tested the feasibility of producing CAR T cells using RNA. RNA-based CAR T cells have several advantages over DNA, including transient expression of CAR molecules and no viral vectors, meaning they can be administered repeatedly in an outpatient setting without the need for lymphodepletion. To evaluate the safety and activity of RNA CAR T cell therapy, autologous anti-BCMA RNA CAR T cells were administered to 14 patients with myasthenia gravis in an outpatient clinic. None of the patients received chemotherapy but were continued on a background of prednisone and/or steroid-sparing immunosuppression. Patients received regular infusions and collectively demonstrated decreases in clinical severity indices, improved function and quality of life. Importantly, patients did not demonstrate any decreases in total IgG and had an acceptable side-effect profile with no serious cases of CRS or neurotoxicity encountered. These encouraging early findings pave the way for RNA CAR T cells, as a potentially safer option that can be more broadly adopted in autoimmune disease than viral vector engineered CAR T cells. Overall, 2023 represented a significant year for the development of precision cellular therapies in autoimmune disease. In particular, the ongoing development of CAR T cells that can specifically eliminate pathogenic autoreactive cells and appear to be superior compared with current B cell-depleting regimens (Figure 1). Approaches to increase accessibility and tolerability of CAR T cells, that reduce costs and requirements for specialized facilities, should be a focus in the future. Moreover, research efforts uncovering fundamental knowledge on the source of pathogenic autoantibodies and the cause of severe symptoms17 will be imperative to making more precise cellular therapy to improve patient outcomes. AL is supported by a National Health and Medical Research Council (NHMRC) Postgraduate scholarship. JR is supported by a NHMRC Investigator Grant. Adrian YS Lee: Conceptualization; writing – original draft; writing – review and editing. Joanne H Reed: Conceptualization; writing – original draft; writing – review and editing. The authors declare no conflicts of interest.