Featured Student: Gustavo Rodrigues de Moraes

Gustavo Rodrigues de Moraes is a second-year medical student at Kansas City University College of Osteopathic Medicine. His interests include dermatology, with a particular focus on autoimmune diseases with cutaneous manifestations and the intersection of dermatological conditions and mental health. Gustavo is also passionate about providing care to LGBTQ+ and underserved populations. In his free time, he enjoys spending time with his dogs, watching shows on Netflix, mixology, and traveling whenever possible.

Introduction

Autoimmune diseases are a heterogeneous group of disorders that are characterised by a breach of immune tolerance. The sensitisation against antigens leads to the formation of autoreactive T cells and B cells, as well as autoantibodies that trigger organ damage. Autoimmune diseases include rheumatoid arthritis, systemic lupus erythematosus, type 1 diabetes, pemphigus, and multiple sclerosis. In all autoimmune diseases, autoreactive B-cell clones and autoantibodies directed against a patient’s own antigens are formed long before the onset of clinical symptoms. However, not all chronic inflammatory diseases are autoimmune diseases. For instance, diseases like psoriasis, Crohn’s disease, or ankylosing spondylitis have no B-cell-mediated pathophysiology but clinically present as chronic inflammation. This group of diseases usually genetically associated with major histocompatibility class I alleles, whereas autoimmune diseases are often associated with specific MHC class-II alleles.In this paper, we focus on the treatment of autoimmune diseases by using chimeric antigen receptor (CAR) T cells directed against B cells. Autoimmune diseases are also susceptible to the therapeutic targeting of B cells by monoclonal antibodies. However, to some extent, results on the treatment of autoimmune disease by monoclonal B-cell- depleting antibodies were disappointing, as some trials failed their clinical endpoints and induction of drug-free remission or cure has not been shown. CAR T cells, on the other hand, might have curative potential, at least in malignant diseases, as implied by data from clinical trials.

Current management of autoimmune diseases Autoimmune diseases are usually not chronic and drive organ damage when no therapeutic intervention is

initiated. Autoimmune T-cell and B-cell clones trigger disease and autoantibodies by inductions of autoantibodies that act as effectors driving target cell lysis (such as in autoimmune haemolytic anaemia), mediating cell activation (like in Graves’ disease), disturbing cell communication (like in pemphigus), or leading to immune complex formation and inflammation (like in systemic lupus erythematosus). Controlling altered T-cell and B-cell function is of the utmost importance in the treatment of autoimmune disease. Conventional immune suppressants (eg, azathioprine, inhibiting purine synthesis or mycophenolate blocking the enzyme inosine-5’-monophosphate dehydrogenase and depleting guanosine nucleotides) and calcineurin inhibitors (eg, ciclosporin A tacrolimus and voclosporin, which downregulate the production of T-cell-derived cytokines through binding to cytoplasmic receptors cyclophilin and FK binding protein) are widely used for the treatment of autoimmune disease. In addition, targeted drugs that block T-cell migration to inflammatory sites, such as natalizumab and vedolizumab, blocking integrin a4 and a4b7, respectively, and those modulating the sphingosine-1 phosphate receptor, have been developed to treat autoimmune diseases. These agents often sufficiently attenuate the inflammatory process, but treatment with the need to be given continuously over years, or even lifelong. Even when remission is reached, recurrence of disease often happens when immunosuppression is discontinued. Furthermore, suboptimal control of autoimmune disease by immune suppressants requires additional use of glucocorticoids to dampen inflammation, which is associated with considerable side-effects. Selectively interfering with B-cell activation and autoantibody production in autoimmune diseases is a promising approach. The breakthrough work by Edwards and colleagues in 2004 showed that B-cell depletion with rituximab—a monoclonal antibody directed against CD20 antigen—showed efficacy in the treatment of rheumatoid arthritis. Despite this success of rituximab, a substantial proportion of patients do not show sufficient improvement of disease activity of rheumatoid arthritis. Some studies on the use of rituximab in autoimmune diseases, including a randomised controlled trial in systemic lupus erythematosus, showed no clinically or statistically significant effect in ameliorating the disease, suggesting escape mechanisms that prevent eradication of autoimmunity in patients with an autoimmune disease. Also, the disease often relapses after cessation of rituximab, necessitating repeated application to sufficiently control disease. This observation also applies to treatments that inhibit B-cell activation, such as belimumab, a monoclonal antibody that blocks the B-cell activating factor, which is used in the treatment of systemic lupus erythematosus. Furthermore, long-term use of B-cell depleting antibodies is associated with low serum immunoglobulin G levels, which increases the risk of infections.

Limitations of current B-cell targeting approaches and the concept of deep B-cell depletion in autoimmunity

Control of autoimmune diseases and even a potential

immunological reset in autoimmunity might require deep B-cell depletion. Rituximab mediates effects via antibody-dependent cytotoxicity. However, treatment does not always fully eliminate circulating B cells, which has been observed to be associated with the level of therapeutic efficacy in systemic lupus erythematosus. Circulating B cells are the most susceptible to antibody-mediated B-cell depletion, as effector cells (ie, monocytes and natural killer cells) and complement factors are readily available.



However, effector cells that mediate antibody-induced cellular toxicity might not always be present in the tissues, and antibody concentrations are eventually lower (eg, because of uptake through marginal zone macrophages), which could contribute to insufficient target cell clearance. In the case of systemic lupus erythematosus, complement factors are consumed, and there is evidence of lower phagocytic activity of

myeloid cells. In accordance with this concept, several studies have shown that memory B cells escape depletion by rituximab. For instance, in the abdominal lymph nodes, numbers of CD19+ B cells were not reduced after rituximab treatment, when the drug was administered to prevent rejection of kidney transplants. Furthermore, B cells have been detected in the synovial membrane of patients with rheumatoid arthritis upon rituximab treatment, while peripheral B cells were depleted. Finally, in tonsils of patients with systemic lupus erythematosus, memory B cells can reside in niches after rituximab treatment, which indicates that tissue B cells can escape antibody-mediated targeting.

These data clearly show that depletion of tissue-resident B cells poses a greater challenge compared with depletion of their circulating counterparts, suggesting that memory B cells in the tissues might have a greater resistance to depletion. This lack of complete depletion of B cells within the tissues might explain the inconsistent results of rituximab therapy in systemic lupus erythematosus. Also, negative studies showed in post-hoc analyses that patients with complete peripheral B-cell depletion had better responses to treatment than those with only partial depletion. In accordance, newer anti-CD20 antibodies with higher antibody-dependent cellular cytotoxicity activity, such as the humanised ocrelizumab and, in particular, the glycoengineered IgG1 type 2 obinutuzumab, show clinical efficacy in systemic lupus erythematosus. However, in the case of ocrelizumab, treatment of systemic lupus erythematosus was associated with an increased risk of severe infections. Although the degree of B-cell depletion might be better with obinutuzumab and ocrelizumab than with rituximab, the need to recruit functional effector cells to the site of B cells, or absence of complement factors might still limit efficacy and allow autoreactive B-cell populations to survive in their respective niches. Another open question is the ideal target antigen. For example, CD20 antigen is not expressed on plasmablasts and plasma cells. A high cell proportion of plasmablasts are found in post-rituximab systemic lupus erythematosus flares. Targeted elimination of CD38+ plasma cells through the monoclonal anti-CD38 antibody daratumumab has shown efficacy in patients with refractory systemic lupus erythematosus.

Principle of targeting B-cell-derived malignant cells with CAR T cells

Researchers have been trying to develop strategies to reset autoimmune disease in the way that leads to abrogation of disease by deeply resetting the immune system and allowing patients to permanently stop immunosuppressive drugs. Such concepts of rebooting the immune system in autoimmune disease have been previously attempted by autologous haematopoietic stem cell transplantation (auto-HSCT) following high- dose chemotherapy. However, significant toxic effects have restricted the use of auto-HSCT in systemic lupus erythematosus. Cell-based therapy has undergone a fast evolution over the years thanks to the development of genetically engineered receptors, such as CARs. The term chimeric is based on the different origins of the individual CAR components: an extracellular antigen recognition domain derived from antibodies, a transmembrane domain, and an intracellular activation domain derived from T cells. The CAR-encoding DNA can be transferred into ex-vivo immune cells, such as T cells, to generate CAR T cells (figure 2). Upon infusion of CAR T cell into the host the cells recognise the antigen, become activated, and destroy the target. In malignancies, such as B-cell-derived acute lymphoblastic leukaemia or non-Hodgkin lymphoma, it is of utmost importance to eliminate malignant clones to prevent relapse. To achieve this goal, targeting of hard-to-reach, tissue-resident tumour cells is a prerequisite. CAR T cells represent a suitable tool for that thanks to the T cells’ natural ability to infiltrate tissues, their high-affinity specific target binding, and their anti-tumor effector functions. Engagement of the CAR expressed on the plasma membrane of T cells, or alternatively of natural killer cells, by the target antigen leads to their activation and subsequent killing of tumour cells. Surface molecules restricted to B-cell lineage, such as CD19, CD20, and CD22, serve as the tumour-associated antigens. These antigens are recognised by the extracellular domain of the CAR that represents a single-chain variable fragment of an antibody specific to the respective antigen. The cytoplasmic part of CAR contains signalling (CD3 zeta-chain of the T-cell receptor) and co-stimulatory (4–1BB or CD28) domains to ensure proper expansion and activation of CAR T cells, as well as target cell killing. CAR T cells expand and survive in the tissues, where they are exposed to their target antigens.Another key factor for the expansion and persistence of CAR T cells is the lymphodepleting chemotherapy (usually cyclophosphamide and fludarabine) before CAR T-cell infusion. The resulting lymphopenia leads to a compensatory homoeostatic proliferation of CAR T cells and the preferential formation of a memory phenotype. Under ideal circumstances, CAR-expressing cells can survive for many years in the tissues. CAR T cells have become a powerful tool in cancer therapy, leading to a deep and sustained eradication of target antigen-expressing cells. The most advanced CAR-based treatments are those directed against malignant B cells, allowing long-lasting remissions in up to 50% of the patients. In fact, all CD19 CAR T-cell products approved by the US Food and Drug Administration (ie, axicabtagene ciloleucel/axi-cel, tisagenlecleucel/tisa-cel, lisocabtagene maraleucel/liso-cel, and brexucabtagene autoleucel/brexu-cell) are directed against B-cell-derived malignancies, such as B-cell-derived acute lymphoblastic leukaemia, large B-cell lymphoma, follicular lymphoma, and mantle cell lymphoma. For treatment of refractory and relapsed diseases CAR T cells are superior in inducing remission over the standard treatments, and lead to long-term remissions (up to 10 years), indicating CAR T cells’ curative therapeutic potential. In consequence, CD19 CAR T cells have become an established tool in haematology practice. In the context of malignant diseases, the risk of immune escape with reappearance of the tumour and toxic effects are among the strongest limitations of CAR T-cell therapy.

Cytokine release syndrome is of particular concern as a toxic effect of CAR T-cell therapy. Mild cytokine release syndrome manifests as fever, headache, arthralgia, and myalgia, but can also lead to hypotension and even cytotoxic shock in severe cases. Cytokine release syndrome rates, including milder forms of the condition, range from 42% to 93% across all therapeutic cell products, and life-threatening and lethal events have occurred. Mechanistically, cytokine release syndrome is mediated by CAR T-cell activation upon target cell engagement and the release of proinflammatory mediators, such as IL-6. Another typical side-effect is the immune effector cell-associated neurotoxicity syndrome, which can manifest as fine motor impairment leading to dysgraphia and speech alterations. Headache, confusion, seizures, and behavioural changes have also been reported in conjunction with immune effector cell- associated neurotoxicity syndrome. The pathophysiology of this condition is not well understood, but there is evidence that endothelial activation and disruption of the blood-brain barrier are involved. The key risk factor for both cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome is a high target cell (tumour) burden. Treatment of cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome is based on antipyretics, glucocorticoids, and IL-6 receptor blockade with the monoclonal antibody tocilizumab.

Autoimmune diseases that might benefit from CAR T-cell-based therapy

Autoimmune diseases can potentially be treated with CAR T cells that recognise B-cell-specific surface molecules. However, the pathophysiological role of B cells in individual autoimmune diseases is not entirely clear apart from the fact that autoantibodies are formed. B-cell activation and autoantibody formation might just represent an epiphenomenon of underlying T-cell-mediated autoimmunity, rather than being the causal factor in triggering disease. In such cases, even complete B-cell depletion might not be an effective treatment, as remaining autoreactive T cells would still sustain the disease. Furthermore, it is not entirely clear whether B-cell-derived plasmablasts or long-lived plasma cells are the main source of autoantibodies in individual autoimmune diseases. As plasma cells have different antigens from B cells and plasmablasts, they might escape treatment with B-cell-directed CAR T cells and therefore still maintain the autoimmune disease process. Finally, clinical symptoms in autoimmune diseases might not only be the result of ongoing inflammation, but to some extent also result from permanent organ damage limiting the reversibility of the symptoms even after successful CAR T-cell treatment. Examples for this include kidney damage in systemic lupus erythematosus, lung fibrosis in systemic sclerosis, muscle atrophy in myositis, secretory gland fibrosis in Sjögren’s syndrome, neuronal damage in multiple sclerosis, and disappearance of β cells in the pancreatic islets in diabetes. For this reason, it might be particularly important to use CAR T cells as early as possible in the treatment of autoimmune diseases, in order not to miss the window of opportunity for reversibility of the symptoms and to minimise the risk of permanent organ damage. Clinical CAR T-cell treatment in cancer is currently limited to autologous cells (extracted from the same patient), which require manufacturing of an individual advanced therapy medicinal product. Off-the-shelf allogeneic CAR T-cell products (cells extracted from the same patient) are not yet commercially available but are yielding encouraging results in ongoing clinical trials. The manufacturing of autologous CAR T cells requires leukapheresis of sufficient numbers of functional lymphocytes, cell transfection using viral vectors, and in-vitro expansion of CAR T cells following strict quality controls in each patient. Correct identification of patients who might benefit the most from such complex intervention and who are in the highest need of CAR T cells is of utmost importance.

Evidence for efficacy of CAR T-cell therapy in autoimmune diseases

CAR T-cell therapy for the treatment of autoimmune diseases is promising. In the last 2 years, substantial progress has been achieved in making the therapy available to patients CAR T-cell therapy to patients with autoimmune diseases. Landmark work in pre-clinical models has strongly supported the use of CAR T cells targeting B cells in autoimmune diseases, showing that CD19 CAR T cells abrogate disease-specific B-cell autoimmunity and organ (ie, kidney) inflammation in murine models of systemic lupus erythematosus. These findings have led to the concept that deep cell-based depletion of (autoreactive) B cells might induce remission of systemic lupus erythematosus and potentially also other autoimmune diseases. Rolling out CAR T-cell treatment of patients with autoimmune diseases has been challenging for several reasons. First, patients with autoimmune diseases, particularly those with severe manifestations of the diseases, have usually been exposed to glucocorticoids and other drugs, such as mycophenolate or cyclophosphamide, that have an effect on T-cell quantity and quality. Therefore, retrieving sufficient amounts of functional T cells and expanding them in patients with autoimmune diseases might be impossible, especially because lymphopenia is not uncommon in autoimmune diseases. Second, the T-cell compartment in autoimmune disease patients contains autoreactive T-cell clones, which might also expand during the manufacturing of CAR T cells. In fact, reinfusion of CAR T cells that also carry autoreactive T-cell receptors could aggravate autoimmune diseases. Finally, cytokine release syndrome resulting from activated CAR T cells, which in turn activate macrophages and other immune cells, is a serious complication for cancer patients receiving CAR T cells. Risk factors for cytokine release syndrome include elevated inflammatory parameters, such as C-reactive protein, which is also seen in active autoimmune diseases. Similarly, immune effector cell-associated neurotoxicity syndrome is more prevalent in patients with previous CNS dysfunction, and many autoimmune diseases display CNS involvement. Such events, especially if they are severe, might prevent applicability of CAR T cells in the treatment of non-malignant diseases and autoimmune diseases. In autoimmune diseases, CD19-directed CAR T cells were used for the first time in 2021, in the treatment of a woman aged 20 years with severe treatment-refractory systemic lupus erythematosus.55 The approach showed that CAR T-cell production from patients with autoimmune diseases is feasible, allowing a stable transfection of T cells with the CAR construct and generation of sufficient numbers of CAR T cells to be applied for therapy. CAR T-cell infusion was well tolerated by the patient and did not lead to any high-grade toxic effects. This first CAR T-cell-based approach in autoimmune disease treatment led to fast depletion of B cells in vivo, which was accompanied by a rapid and robust expansion of CAR T cells in peripheral blood. With respect to the manifestations of the autoimmune disease, complete clinical remission, including cessation of proteinuria, was reached after 3 months and seroconversion of disease-associated antibodies against double-stranded DNA was observed. Furthermore, all immunosuppressive agents, including glucocorticoids, could be successfully discontinued in this patient without signs of disease relapse for up to 18 months. A deeper analysis of the effects of CD19 CAR T-cell therapy in systemic lupus erythematosus was done in a group of five patients with severe treatment resistance. This study showed stable and reproducible production of CD19 CAR T cells from the peripheral blood, despite pre-treatment with T-cell targeting agents, such as mycophenolate and glucocorticoids, as part of systemic lupus erythematosus therapy. These agents do not seem to impair the production of CAR T cells when readily stopped (in the case of mycophenolate) or tapered to lower doses (in the case of glucocorticoids). Furthermore, the temporal dynamics of the expansion of CAR T cells in vivo were similar among these treated patients, with a peak in circulating CAR T cells around 1 week after administration. Clinical manifestations of systemic lupus erythematosus rapidly ceased in all five patients, reaching a state of remission according to DORIS criteria. This remission was accompanied by seroconversion of several systemic lupus erythematosus-associated antibodies, such as those against double-stranded DNA, nucleosomes, and Smith antigen, but long-standing vaccination responses (ie, those against measles, rubella, and tetanus) were mostly stable or only slightly decreased. This observation indicated that at least some autoantibody species seem to stem from the memory B cell and plasmablast compartment and not from the CD19-negative plasma cells. Vice versa, antibody responses against vaccines that the patients received before CAR T-cell therapy seem to primarily stem from the long-lived CD19-negative plasma cell compartment, which should not be affected by CD19 CAR T cells. The finding is also important given that humoral vaccination responses are seriously impaired during the B-cell aplasia phase after CD19 CAR T-cell therapy, highlighting the importance of previous vaccinations and the stability of anti-vaccine antibody titres after CAR T-cell treatment. Another key finding of the impact of CAR T-cell therapy in these patients with systemic lupus erythematosus was that no long-term B-cell aplasia was observed as B cells reconstituted around 100 days after the infusion. Analysis of these B cells showed that memory B cells and plasmablasts virtually disappeared, and the newly emerging B cells in the circulation revealed a naive phenotype with expression of IgM and IgD heavy chains, indicating a pre-class switched naive B-cell phenotype. The B-cell recurrence after CAR-T cell therapy was not associated with disease recurrence, as all patients remained in treatment-free remission and did not produce double-stranded DNA autoantibodies. This finding is important, as it has been shown that after antibody-mediated B-cell depletion with rituximab, the rising levels of B-cell-activating factor can perpetuate autoreactive B cells and plasmablasts, stimulate T follicular helper cells, and trigger flares of systemic lupus erythematosus.60 Tolerability of CAR T-cell therapy in this small group of patients with systemic lupus erythematosus seems good, as no high-grade cytokine release syndrome, neurotoxicity, haematotoxicity, or infectious complications occurred. Hence, the key lessons from these first patients with autoimmune disease treated with CAR T cells are that the procedure is feasible, well tolerated, and allows induction of longer treatment-free remissions with clear indications of a rebooting of the B-cell compartment. As these results were gathered from a small group of patients, the risk of infections in conjunction with conditioning therapy and B-cell depletion by CAR T cells remains to be determined. In addition to patients with systemic lupus erythematosus, a patient with severe multidrug-resistant dermatomyositis (anti-synthetase syndrome) has been treated with CD19 CAR T cells.61 The patient was in drug- free remission after infusion with CD19 CAR T cells. Deep abrogation of B cells was followed by normalisation of concentrations of creatine kinase, seroconversion of anti-Jo-1 autoantibodies, regression of muscle and lung inflammation, and complete regain of physical function. The observed changes suggest that the effect of CD19 CAR T-cell therapy might not be limited to systemic lupus erythematosus but might also work in several different B-cell-mediated and autoantibody-driven human autoimmune diseases.

Required length of CAR T-cell activity and duration of response in autoimmune diseases

The length of expected drug-free remission after CAR T-cell therapy for an autoimmune disease remains to be determined. Disease-free observational periods extend to up to 2 years for some patients with systemic lupus erythematosus, with patients being in the state of full B-cell reconstitution for the majority of this period. Hence, it might well be that some patients receiving CAR T-cell therapy can be cured of autoimmune diseases; however, longer follow-up is necessary. New B cells mature in response to infections and vaccinations, and it will be important to know whether this maturation process is not accompanied by the formation of autoreactive B-cell clones, autoantibodies, and eventually the recurrence of the disease. The genetic links (eg, HLA- DR2 and DR3 alleles) for systemic lupus erythematosus are still in place. Furthermore, it is not known how CAR T-cell treatment affects the aberrant T-cell responses in systemic lupus erythematosus, particularly the type I interferon signature associated with autoimmune diseases.62 If the autoimmune disease does not recur for years, however, CAR T-cell therapy has restored an important immune checkpoint that could be associated with the permanent absence of symptoms and even a cure of the underlying disease. Such a scenario would push the use of CAR T-cell treatment as an early intervention strategy in selected patients with severe forms of autoimmune diseases. In the context of treatment of B-cell malignancies, CAR T cells must have the potential to survive in the body for many years. Whether such long survival of CAR T cells is possible and necessary in autoimmune diseases’ treatment is unclear. Consistent and stable repopulation of patients with an autoimmune disease with B cells following CAR T-cell treatment speaks against the CAR T-cells’ long-term persistence, either suggesting contraction, activation-induced cell death, or exhaustion. The underlying mechanism remains unclear, however, unlike patients with cancer, patients with autoimmune diseases have received less cytotoxic agents with the corresponding effects on stem cells and immune cells. Thus, it is conceivable that the reconstructing cells (after lymphodepletion) more successfully compete for niches (ie, bone marrow or lymphoid tissues) than the substantially pre-damaged cells of patients with cancer. Another straightforward explanation could be the rapid elimination of cells carrying the target antigen (ie, CD19), as the B-cell load is many times lower in autoimmune disease than in patients with tumours. Hence, loss of the target antigen could trigger early contraction of the CAR T-cell population. The value of long-term persistence of functional CAR T cells in autoimmune diseases appears questionable as the time-limited deep B-cell depletion was shown to be sufficient to fully abrogate disease activity and reboot the B-cell system. Long-term B-cell depletion might not be desirable in autoimmune diseases, as B cells provide humoral protection and are involved in immunological homoeostasis (eg, as regulatory B cells). If poor CAR T-cell persistence leads to autoimmune disease relapse, retreatment with ready-to-use CAR T cells, which had been cryopreserved during the initial production round, could be considered. A similar strategy is also exploited in B-cell-derived malignancies.64 Although not observed to date, relapse of autoimmune disease might occur with the disappearance or exhaustion of CAR T cells. Relapses of B-cell lymphoma have been described several years after CAR T-cell treatment and autologous stem cell transplantation.65 Hence, long-term observation of patients with autoimmune disease receiving CAR T-cell therapy is of the utmost importance.

Challenges regarding safety and cost-effectiveness

Although mortality is increased in patients with autoimmune diseases, it is substantially lower than in patients with relapsed or refractory B-cell malignancies, which inevitably cause death of the affected patient if they do not respond to treatment. Hence, safety considerations for CAR T-cell therapy in autoimmune diseases are different from considerations in cancer therapy. Life-threatening or even fatal cytokine release syndrome and neurotoxicity (ie, immune effector cell-associated neurotoxicity syndrome) is not acceptable in autoimmune disease patients. Therefore, a low rate of higher-grade side effects, which also include haematotoxicity, hypogammaglobulinemia, and infectious complications, is of utmost importance for making the use of CAR T cells an acceptable treatment option in the future. In the small number of patients with autoimmune diseases who have been treated with CAR T cells, good tolerability of the treatment was observed, with no cytokine release syndrome greater than grade 1 (fever) and no immune effector cell-associated neurotoxicity syndrome. Although this low toxicity might be based on the much lower target engagement of CAR T cells in autoimmune diseases than in B-cell malignancies, the effects of standard treatments for cytokine release syndrome, including glucocorticoids and the IL-6 receptor antibody tocilizumab, must be carefully assessed in patients with autoimmune disease. The use of glucocorticoids has not shown any negative effects on CAR T-cell function in patients with cancer, and it remains to be shown whether this is also true for patients with autoimmune diseases. In addition, there is the possibility of exacerbating T-cell-mediated autoimmunity by co-expanding autoimmune T-cell clones from patients with autoimmune disease during the process of CAR T-cell manufacturing. However, the chance of such a scenario is low, as CAR T cells are likely to quickly become exhausted by simultaneous T-cell receptor and CAR engagement, which has also been shown for donor-derived alloreactive T cells, which are unable to trigger graft-versus-host disease.71 Given the high cost of CAR T cells manufacturing, the therapy might initially be available only to patients with particularly severe forms of autoimmune disease. However, since CAR T cells have allowed discontinuation of all other immunosuppressive agents in patients with systemic lupus erythematosus, the treatment of which is associated with substantial cumulative drug costs (up to US$100 000 per year), CAR T cells could actually reach their break-even point over a few years. Since this type of cellular therapy holds promise to achieve a long-term, if not permanent, reset of the immune system in autoimmune diseases, CAR T cells might be cost-effective for autoimmune disease treatment.

Future strategies for CAR T-cell therapies against autoimmune diseases

The feasibility of CAR T-cell treatment in autoimmune diseases will strongly depend on a multidisciplinary collaboration between different specialists (ie, immunologists and hematologists, together with colleagues specialized in rheumatology, nephrology, and neurology). Highly specialised centres will have a key role in the selection of suitable patients with autoimmune diseases, the planning and execution of individualised therapeutic concepts, and the prevention and management of treatment-related toxic effects.72 If the safety data for CAR T cells in autoimmune diseases documented so far are confirmed, studies in the outpatient setting, similar to the approaches in malignant diseases, would be the next step. The best procedure to treat an autoimmune disease with CAR T cells is unknown. Clinical tolerability and efficacy data for CAR T-cell therapy in autoimmune disease are so far confined to one construct targeting CD19 and one containing a 4–1BB co-stimulatory domain. CD19 seems to be a promising target, as it is highly specific for the B-cell lineage and is expressed widely across different B-cell differentiation stages, including plasmablasts and potentially a small portion of plasma cells.74 CD20 and CD22 also represent interesting antigens for CAR T cells in systemic lupus erythematosus and other autoimmune diseases. However, it is unclear how CD20 and CD22 would be a better target than CD19, as their expression overlaps with CD19 but is low or absent in plasmablasts and plasma cells. Some patients with an autoimmune disease might also respond to CAR T cells targeting plasma cells. CAR T cells directed against CD38 and B-cell maturation antigen (BCMA), both of which are antigens expressed on plasmablasts and plasma cells, have been developed for the treatment of malignant plasma cell disease (multiple myeloma). CAR T cells directed against CD38 or BCMA could also be interesting for application in autoimmune diseases after the first evidence for the efficacy of CD38 antibodies in systemic lupus  rythematosus.25 Stable antibody titres against vaccinations that the patients had before CART-cell therapy and only a mild decrease in immunoglobulin concentration suggest that most of the plasma cell compartment remains intact upon treatment with CD19 CAR T cells, which might be a major advantage, as the CD19-directed approach was sufficient to eliminate double-stranded DNA and nucleosome antibodies. However, long-lived plasma cells in the bone marrow, which contribute to antibody and probably to autoantibody production, are usually CD19-negative and would, therefore, escape elimination by CD19 CAR T cells. Some forms of systemic lupus erythematosus and other autoimmune diseases might show a higher dependence on plasma cells and, therefore, might benefit more from plasma cell targeting rather than deep B-cell (and plasmablast) depletion. In these patients, using BCMA-targeted CAR T cells or CD38-targeted CAR T cells might be beneficial. Newer constructs, such as CAR T cells that simultaneously target B cells and plasmablasts through CD19 and plasma cells through CD38, might be interesting in the treatment of autoimmune diseases.81 The safety of CD38-targeted and BCMA-targeted CAR T cells in autoimmune disease treatment is unknown, and specific features for BCMA-targeted CAR T cells regarding safety in hematological malignancies such as neurocognitive and hypokinetic disorders82 and lung changes need to be considered. Apart from the target-binding domains, other features of the CAR vector might also affect the therapeutic efficacy of CAR T-cell treatment in autoimmune diseases. First-generation CAR T cells without co-stimulatory domains did not proliferate sufficiently in vivo and had poor lasting efficacy.84 In the currently approved second-generation CD19 CAR T cells, CD28 or 4–1BB co-stimulatory domains are used. Which domain has better suitability for the treatment of malignancies and autoimmune diseases has not been determined. CD28-driven CARs are considered to elicit faster and more intense downstream signalling, promoting differentiation into effector memory CAR T cells, than 4–1BB-driven CARs.85 However, CD28-driven tonic signaling favors early exhaustion.86 In contrast, 4–1BB-driven CARs lead to slower and more persistent signalling, skewing T cells towards a central memory phenotype.87 Some studies have reported that CD28-costimulated CAR T cells might have higher rates of cytokine release syndrome, but this observation is not reproducible. In fact, a comprehensive review has not supported consistent differences between 4–1BB-containing and CD28-containing CARs with respect to efficacy and safety in the treatment of B-cell-derived malignancies. A short-term but very deep depletion of the B-cell compartment seems to be sufficient for an immunological reset, as autoimmune disease does not recur despite the reconstitution of B cells and the halting of immune-suppressive drug therapy. Thus, construct modifications for the prolongation of CAR T-cell persistence might not be crucial for the treatment of autoimmune diseases. However, in cases of early recurrence of autoimmune diseases, such modifications could be necessary, such as the use of third-generation CARs that combine 4–1BB and CD28 co-stimulation to improve the robust signaling that is important for better longevity and effector functions of CAR T cells. Another important component of CAR T-cell treatment is the preparatory conditioning regimen, which for most patients is composed of a short course of cyclophosphamide and fludarabine. The current doses of cyclophosphamide (usually a cumulative dose of about 1 g/m²) used in conjunction with CAR T-cell treatment are lower than the cumulative doses usually applied for treating autoimmune diseases and in some systemic lupus erythematosus patients receiving CAR treatment had been non-responsive. Data on fludarabine in autoimmune diseases, and particularly on systemic lupus erythematosus, are anecdotal and suggest that the medication showed substantial toxicity if used over longer periods of time. Overall, conditioning chemotherapy might have some short-term effects in autoimmune diseases as it reduces immune cell count (including B cells) for approximately 1 week. However, lymphodepletion might not be responsible for the long-term effects of CAR T-cell therapy in autoimmune diseases, including sustained B-cell depletion, disappearance of autoantibodies, rebooting of the B-cell compartment, and resolution of disease symptoms. The effect is regarded as necessary to empty niches for CAR T cells (eg, in the bone marrow and in secondary lymphoid organs), which allows them to better proliferate and survive.

Conclusions

CAR T cells have been successfully introduced into the treatment of autoimmune diseases. This approach is unique not only because it is based on complex manufacturing of a personalized genetically modified autologous cell product but also because it is con conceptualized as a single-shot intervention to induce long-standing drug-free remission; this ambitious approach could herald a new era of autoimmune disease treatment, transforming the current principle of long-term immunosuppression into a strategy that induces an immune reset with no need for further treatment. Further studies addressing the potential of CAR T-cells’ applications for the treatment of autoimmune diseases are underway and will shed more light on the potential of this treatment approach.

Citation

Schett G, Mackensen A, Mougiakakos D. CAR T-cell therapy in autoimmune diseases. Lancet. 2023 Nov 25;402(10416):2034-2044. doi: 10.1016/S0140-6736(23)01126-1.

Epub 2023 Sep 22. PMID: 37748491.