‘Good First Try’: In Vivo CAR T for Neuro Autoimmune Disease

A single intravenous infusion of an experimental therapy that trains the immune system to attack B cells involved in multiple sclerosis (MS) and other neurologic autoimmune disorders was associated with decreased disability and improvement in motor and cognitive function, a small early-stage trial showed.Called JY231, the therapy is an off-the-shelf lentiviral vector that directs the immune system to produce chimeric antigen receptor T cells (CAR T) directly in the body.While CAR T-cell therapy has been approved for nearly a decade for oncologic disorders, the current regimen requires clinicians to collect a patient’s T cells through leukapheresis, ship them to a specialized facility for weeks of genetic engineering, and then reinfuse the modified cells, usually following lymphodepleting chemotherapy.In the phase 1 trial, all 16 patients who received JY231 produced CAR T cells in vivo, without the need for cell manufacturing or chemotherapy.After 6 months of treatment, patients with MS, myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD), generalized myasthenia gravis, and idiopathic inflammatory myopathies showed B-cell depletion and improvements on a range of disease-specific clinical measures.“These findings provide proof-of-concept that in vivo CAR T-cell generation through lentiviral vector engineering is associated with manageable side effects and may be effective for treating refractory neurologic autoimmune disorders,” lead investigator Yu‑Hang Cheng, PhD, Tongji Hospital of Huazhong University of Science and Technology, Wuhan, China, and colleagues wrote.“By circumventing the need for complex ex vivo cell manufacturing and toxic lymphodepletion, JY231 has the potential to transform CAR T-cell therapy into an accessible, scalable molecular intervention capable of crossing anatomical barriers to eliminate target cells,” they added.The study was published on September 3 in The New England Journal of Medicine.An Off-the-Shelf ApproachPatients with progressive MS have few effective treatment options. Ocrelizumab, the only drug approved for primary progressive MS, offers modest slowing of disability. For patients with refractory MOGAD or generalized myasthenia gravis who have failed rituximab and other immunosuppressants, the options narrow further.CAR T-cell therapy has offered a potential solution. Early trials in lupus and other autoimmune conditions have shown prolonged treatment-free remission after a single course, as previously reported by Medscape Medical News.JY231 offers a simpler alternative to conventional CAR T-cell therapy. As described in the study, the lentiviral vector carries not only a CD3-targeting fragment on its surface that directs it to T cells in the bloodstream but also a co-stimulatory molecule (CD137 ligand) that activates those T cells on contact.Once inside, the vector inserts the genetic instructions for a CAR that recognizes CD19, a protein found on the surface of B cells, directly into the T cell’s genome. In all four conditions enrolled in the trial, B cells produce the autoantibodies that damage nerve and muscle tissue, and in MS they also sustain inflammation in the central nervous system by presenting antigens to autoreactive T cells and secreting proinflammatory cytokines.By programming a patient’s T cells to recognize CD19, the treatment turns the patient’s own immune system against the cells driving the disease, seeking out CD19-positive B cells without the need for a manufacturing step outside the patient’s body, investigators noted.Safety and Key FindingsInvestigators used a 3 + 3 dose-escalation framework with two dosing regimens in the open-label trial. Patients who first underwent lymphodepletion with fludarabine plus cyclophosphamide started at a lower dose of JY231, and those who did not receive lymphodepletion received a dose 10 times higher.To find the highest tolerated dose, investigators administered a starting dose to three patients. Depending on the presence of dose-limiting toxicities, additional patients were added at that same dose or at an increased dose, continuing through two more dose levels to identify the maximum tolerated dose.All patients were premedicated with acetaminophen and diphenhydramine before infusion. The primary endpoint was safety and tolerability. Secondary endpoints included preliminary efficacy and the pharmacokinetic characteristics of both the vector and the CAR T cells it generated. Patients were monitored during a 28-day observation period, with follow-up planned through 24 months.At 6 months of follow-up, the seven patients with progressive MS showed decreased scores on the Expanded Disability Status Scale (EDSS) — indicating reduced disability. Cerebrospinal fluid ( CSF) neurofilament light chain, a marker of ongoing nerve damage, also declined. All seven had no new lesions on MRI.The three patients with MOGAD showed decreases in serum MOG-specific autoantibody titers.Patients with generalized myasthenia gravis (n = 3) had reductions on the Quantitative Myasthenia Gravis and Myasthenia Gravis Activities of Daily Living scales, correlating with decreased anti-acetylcholine receptor antibodies. The three patients with inflammatory myopathies showed increased Manual Muscle Testing 8 scores, decreased serum creatine kinase, and reduced muscle inflammation on imaging.The vector achieved higher copy numbers in CSF than in peripheral blood in the patients with MS and MOGAD, a finding consistent with the investigators’ assertion that JY231 is capable of crossing anatomical barriers to reach target cells in the central nervous system, which conventional anti-CD20 antibodies such as rituximab and ocrelizumab can’t do.No cases of immune effector cell-associated neurotoxicity syndrome were observed. Grade 1 cytokine release syndrome occurred in 11 of 16 patients (69%), resolving within 2 weeks. Three patients had reversible grade 3 or 4 neutropenia. No infections were reported. Four patients developed low immunoglobulin G levels during later follow-up, though levels showed recovery.Study limitations include the lack of a control group and clinical assessments that were performed by the treating investigators without blinding. Additionally, half the cohort had near-zero B-cell counts at baseline from prior rituximab treatment, so they were already substantially B-cell depleted before the infusion.Michael Levy, MD, PhD, associate professor of neurology and research director of the Division of Neuroimmunology and Neuroinfectious Disease at Massachusetts General Hospital in Boston, praised the engineering behind JY231, calling it the first study he had seen of a true lentiviral vector approach that transforms cells in vivo and directs them to function as intended.“If it’s all correct and true, then it would be a true breakthrough on the engineering side of things,” he told Medscape Medical News.Whether the treatment achieved what the investigators call a “systemic immunologic reset,” in which B cells repopulate as entirely naive cells with no trace of the original disease-associated clonotypes, was less clear, he said. Residual memory B cells were visible in flow cytometry data, and if a comparable result from an autologous CAR T-cell trial showed that same flow cytometry data, it would likely be considered a failure, Levy added.The CSF and bone marrow results were more promising. Rituximab does not reach either compartment, and the investigators found no B cells in either. “You don’t see that with rituximab because rituximab doesn’t get there,” Levy said. “Maybe these T cells do.” The EDSS improvements in progressive MS were harder to interpret, he said, given that disability at that disease stage is driven largely by neurodegeneration rather than active inflammation. “An MS patient isn’t going to get this much better no matter what you do to them,” Levy said.Levy said the study was best understood as a proof of concept for a delivery method. “The engineering still needs to be worked out. There’s no best way to do this yet,” he said. “I think this adds to that possible list of possible ways to do this.”The trial was supported by grants from the National Natural Science Foundation of China and the Shenzhen Science and Technology Major Project. The vector was manufactured by Shenzhen Genocury Biotech, which employed two coauthors. Full disclosures are available at NEJM.org. Levy reported receiving personal compensation for advisory roles from Alexion, Horizon, Genentech/Roche, UCB, Sanofi, and Mitsubishi and research grants through Massachusetts General Hospital from Alexion, Horizon, Genentech, Roche, UCB, and Sanofi.