The first clinical findings showing that engineered CAR-T cells can deplete pathogenic B cells in progressive multiple sclerosis were published in March 2024. These early results, released by Kyverna Therapeutics, added momentum to an already effervescent field. In June, a phase 1 study in patients with systemic lupus erythematosus (SLE) presented by iCell Gene Therapeutics at the European Alliance of Associations for Rheumatology Congress in Vienna showed that a CAR-T therapy eliminated all autoantibodies in 11 out of 12 treated patients, with those whose disease responded remaining clear of disease and medication-free for up to 4.5 years. Emboldened by the safety profile and results in other early-stage clinical studies, several companies and academic centers are launching trials to test this new approach to autoimmune disorders (Table 1). Although interim data from patients with lupus nephritis presented in June by Kyverna have been mixed, the hope of a sustained therapeutic option is stoking excitement over this new chapter for CAR-T cell therapies.

Table 1 Selected CAR-T cells in clinical trials for autoimmune disorders
Neurons in multiple sclerosis are damaged by autoantibodies and inflammation. Credit: Stocktrek Images, Inc. / Alamy Stock Photo

CAR-T therapies are living drugs. To make them, T cells are generally removed from a person and genetically engineered to produce chimeric antigen receptors (CARs) that recognize a specific target. Once reinfused, they seek and destroy their target. So far, CAR-T cells have proved successful in treating blood cancers by destroying pathogenic B cells causing leukemias. Because B cells also drive autoimmune conditions, wiping out B cells has potential for treating these diseases too.

Multiple sclerosis is being closely watched. Kyverna, in collaboration with researchers from the University Medical Center Hamburg-Eppendorf in Germany, has tested whether autologous CAR-T cells could safely eradicate B cells in the central nervous system. Their product KYV-101 was used as part of a compassionate use program for patients whose disease failed to respond to Ocrevus (ocrelizumab), an antibody therapy that depletes B cells in the periphery but not in brain tissue. “KYV-101 really shows a differentiated impact in this disease compared with the traditional monoclonals,” says James Chung, Kyverna’s chief medical officer.

Chung says that the CAR construct in KYV-101 is “pretty well suited for autoimmune disease.” Its CD19 antibody domain is fully human to reduce immunogenicity, plus other domains are optimized for safety. There was no evidence of neurotoxicity of KYV-101 in the two participants, despite it reaching the central nervous system. To gather more data, Kyverna is teaming up with Stanford University researchers on a US-based phase 1 trial and conducting its own phase 2 trial in the United States and Europe.

To destroy disease-causing B cells, CAR-T cells target CD19 or B cell maturation antigen (BCMA) on B cells. These targets can also be hit by monoclonal antibodies such as Rituxan (rituximab) which is indeed one of the mainstay treatments for autoimmune disorders. But antibody therapies need to be taken chronically and put patients at risk of serious infection. Companies are instead striving for therapies that allow healthy B cells to bounce back.

Cartesian Therapeutics is designing CAR-T cells that are only transiently expressed in patients. Its approach uses mRNA transfection to drive expression of BCMA-directed CARs in patient-derived cells. Once these cells are infused into the body, they divide, mRNA is degraded and CAR expression is lost. This strategy means that there is no need to create room for the incoming T cells, and therefore patients do not need grueling lymphodepletion and hospital stays before infusion. And unlike with permanently modified cells, there is opportunity to control dosing and potentially reduce side effects. “We’ve overcome some of those hurdles that traditionally have been problematic to bringing CAR-T therapies to broader patient groups,” says Chris Jewell, Cartesian’s CSO.

The company’s mRNA CAR-T Descartes-08 is in phase 2 trials for SLE and the muscular weakness disorder myasthenia gravis. It is the first CAR-T therapy in any indication to undergo a randomized placebo-controlled trial, with a crossover design to ensure all patients get therapy. Results of an open-label trial in myasthenia gravis showed that Descartes-08 reduces disease severity and levels of circulating autoantibodies, and preprint findings suggest these effects can be durable, being maintained for one year in five of seven people given six doses one week apart.

These early results with Descartes-08 add to tantalizing evidence that CAR-T cells can send autoimmune disorders into remission, working better than B-cell-targeted monoclonal antibodies. CAR-T cells can get into places that are out of reach for antibody therapies, such as lymphoid organs and inflamed tissues. “This is the reason why we are having such good results with CAR-T cells in autoimmune disorders,” says Raffaella Greco, a hematologist at the San Raffaele Hospital in Milan, Italy, who is part of a multidisciplinary team investigating cell therapies for autoimmune disorders. There are also fewer dysfunctional B cells and a lower disease burden in autoimmune disease compared with cancer. “It’s a completely different setting in terms of the target,” says Greco. This difference in target setting may in part explain why CAR-T cells in autoimmune disorders appear so far devoid of severe side effects such as the cytokine release syndrome and neurotoxicity seen in cancer settings.

A clue to the durable effects of CAR-T cells in autoimmune disease comes from a compassionate use program led by investigators at the University of Erlangen in Germany. They looked at B cell levels in 15 patients with one of three autoimmune disorders who were in drug-free remission up to two years after a one-time dose of CD19 CAR-T cells. After CAR-T infusion, the patients’ B cells were rapidly eliminated. CAR-T cell levels then declined quickly. When the eliminated B cells were replaced by the body, around 3.5 months later, the new B cells had a naive phenotype. The CAR-T cells had reset the B cell immune system, wiping it clean of disease.

For Haig Aghajanian, co-founder and vice president of research at Capstan Therapeutics, this study was important evidence that validated the company’s transient CAR approach. “It doesn’t make sense to have long-term B-cell ablation,” he says, given the immune system reset. This contrasts with cancer, where long-term persistence of CAR-T cells is likely to be needed. Capstan, a spin-out of the University of Pennsylvania, is developing an in vivo approach to CAR-T cells in which nanoparticle-delivered mRNA prompts the body to make its own CAR-T cells.

So far, the in vivo CAR -T production strategy has worked in a mouse model of fibrosis. The company also successfully generated CD19-directed CAR-T that ablated B cells in a humanized mouse. “T cells turn into CAR-T cells in the body without any need for lymphodepletion, taking any cells out, or ever handling cells,” says Aghajanian. The strategy aims to remove some of the liabilities of ex vivo approaches.

The company plans to take its lead asset, CPTX2309, an in vivo-generated CAR that is fully human and clinically validated, into autoimmunity trials using a recent $175 million financing round that included Johnson & Johnson. Aghajanian says the company will learn from emerging data on other CAR-T therapies before choosing indications.

Yet any CAR-T therapy directed at B-cell antigens such as CD19 or BCMA will deplete all B cells expressing that target, leaving patients at risk of infection until B cells return. “It’s usually just a tiny fraction of the B cells that are misbehaving, and yet we’re just wiping them all out,” says Aimee Payne, a dermatologist and researcher at Columbia University. Payne’s lab designed CAR-engineered T cells, dubbed chimeric auto-antibody receptor (CAAR) T cells, that destroy only the B cells that produce autoantibodies. “The goal is to have the lasting efficacy of regular CAR-T cells without the side effects of global immunosuppression,” she says.

CAAR constructs express a protein antigen of interest, and when introduced ex vivo into a patient’s T cells via a lentiviral vector, the CAAR-T cells target only B cells that express autoantibodies against the antigen. Preclinical studies provided proof of concept of the CAAR-T cell approach in mouse models of pemphigus vulgaris, a blistering skin disorder mediated by autoantibodies against desmoglein 3, and a form of myasthenia gravis mediated by autoantibodies to muscle-specific tyrosine kinase (MuSK) that disrupts neuromuscular junction signaling. To take the approach into clinical trials, Payne co-founded Cabaletta Bio and now acts as a scientific advisor to the company. Cabaletta is running phase 1 trials of CAAR-Ts in mucosal pemphigus vulgaris and MuSK myasthenia gravis, with and without lymphodepletion.

In engineering their CAR-T cells, Sonoma Biotherapeutics is focusing on a different type of immune cell: regulatory T (Treg) cells. Treg cells suppress the immune system to restore homeostasis, whereas the effector T cells modified in most CAR-T therapies focus on cell destruction.

Sonoma is building on clinical studies showing that the use of Treg cells to prevent transplant rejection and autoimmune disorders is safe, albeit with limited efficacy. The aim is to boost the cells’ prowess by directing Treg cells to an inflammation hotspot with a CAR. The company’s lead asset, SBT-77-7101, has a vector-transformed CAR that recognizes citrullinated proteins. These modified proteins are present at inflamed sites such as the joints of people with rheumatoid arthritis and the skin nodules of people with hidradenitis suppurativa, where they induce autoantibodies. The therapy is in phase 1 trials for both these disorders, without lymphodepletion.

By acting on several cell types, including T cells, B cells and macrophages, and potentially promoting tissue repair, SBT-77-7101 should “quiet down the whole system … have a field effect in the tissue,” says Joseph Arron, Sonoma’s chief scientific officer. Other preclinical CAR-Treg programs include those from GentiBio and AstraZeneca-backed Quell Therapeutics in both inflammatory bowel disease and type 1 diabetes.

Off-the shelf CAR-T cells offer another opportunity to democratize CAR-T cells, making them more convenient and accessible and potentially treating hundreds or thousands of people from one batch of cells. Companies such as CRISPR Therapeutics and Caribou Biosciences are aiming to start clinical trials of allogenic CAR-T cells in autoimmune disorders soon. Nkarta is looking to a different immune cell type: CAR natural killer (NK) cells. These immune cells survey the body, recognizing abnormal cells without having seen an antigen before or needing to expand. “Those properties allow us to build a more off-the-shelf therapy,” says David Shook, Nkarta’s Chief Medical Officer. The company sees NK cells as a good fit with autoimmune disorders, having recently announced it was shifting focus from cancer to autoimmune disorders. “In indications where there’s a premium on safety and a premium on accessibility, CAR NK cells are really the appropriate cell there,” says Shook.

A clinical trial in lupus nephritis of Nkarta’s NKX019, allogeneic vector-transformed NK cells engineered with a humanized CD19-directed CAR, will start soon, using single-agent cyclophosphamide lymphodepletion. Meanwhile, Takeda announced in May that its allogeneic CAR-NK cell therapy TAK-007 will pivot from blood cancer to autoimmune disorders, noting that it has a favorable safety profile and off-the-shelf manufacture. A clinical trial in lupus nephritis is expected to start next year.

Right now, patients recruited to clinical trials have severe disease that is refractory to multiple therapies. Using CAR-T and CAR NK cells in milder disease is an alluring possibility, provided their long-term safety profile proves favorable and accessibility barriers can be knocked down. “People are already talking about diseases like multiple sclerosis or systemic sclerosis where you know the disease will progress,” says Chung. “Why don’t we go earlier so we could stop that process from occurring?”