First human trial of injectable CAR-T therapy shows early promise against multiple sclerosis

A research team in China has published the first human data on "in vivo" CAR-T cell therapy for neurologic autoimmune disease, an approach that reprograms a patient's immune cells to fight disease from inside the bloodstream rather than in a laboratory. All 16 participants in the small trial, most with treatment-resistant multiple sclerosis, showed complete depletion of the targeted immune cells after a single dose.

The trial, led by neurologist Dai-Shi Tian at Tongji Hospital, part of Huazhong University of Science and Technology in Wuhan, was published as a research letter in the New England Journal of Medicine this month. Sixteen patients with refractory neurologic autoimmune conditions, including seven with progressive multiple sclerosis and others with myasthenia gravis, received a single intravenous dose of an engineered lentiviral vector built by the Shenzhen biotech firm Genocury.

Skipping the lab entirely

Conventional CAR-T therapy, which has transformed treatment of B-cell blood cancers and shown recent promise against severe autoimmune diseases such as lupus, requires extracting a patient's T cells, genetically engineering them outside the body over several weeks, and infusing them back after chemotherapy conditioning. German researchers at Friedrich-Alexander University Erlangen-Nuremberg were among the first to show, in a series of case reports beginning in 2021, that ex vivo CAR-T could drive lupus into sustained remission, work that helped spur interest in extending the approach to other autoimmune diseases, including multiple sclerosis. But the lab-manufacturing step remains slow and expensive, typically taking weeks per patient and requiring specialized cell-processing facilities that few hospitals have.

In vivo CAR-T instead delivers a targeted vector directly into the bloodstream that reprograms T cells to attack CD19-marked B cells while they remain inside the patient — no extraction, no lab manufacturing, no chemotherapy beforehand. The vector used in the Tongji Hospital trial was engineered to seek out T cells specifically and deliver genetic instructions telling them to recognize and destroy CD19-positive B cells, the same cell population implicated in driving the autoimmune attacks seen in progressive MS and myasthenia gravis. Myasthenia gravis, one of the conditions represented in the trial, causes the immune system to attack the connection between nerves and muscles, producing weakness that can affect breathing and swallowing in severe cases; like progressive MS, it has historically been managed with broad immune suppression rather than a targeted cellular therapy.

In the trial, all 16 patients achieved complete B-cell depletion, with preliminary clinical improvement observed over roughly six months of follow-up. Side effects were described as manageable, and researchers reported no cases of immune effector cell-associated neurotoxicity syndrome, a potentially fatal complication sometimes seen with standard CAR-T treatment.

"A big achievement," with caveats

Independent researchers described the results as an important proof of concept while cautioning against reading too much into a small, uncontrolled trial. Bing Du, an immunologist at East China Normal University in Shanghai who was not involved in the study, called achieving in vivo CAR-T for any disease "a big achievement," adding that the results were "comparable with those for ex vivo CAR-T-cell therapy."

"The responses are promising signals, but they are not yet definitive evidence of efficacy or permanent restoration of immune tolerance," Dai-Shi Tian, the trial's lead investigator, said.

David Simon, a clinician-researcher at Charité–Universitätsmedizin Berlin, called the study "a very exciting proof-of-concept," a characterization echoed in a Nature News analysis of the findings published the same week. Du also flagged a longer-term concern: because the therapy relies on a lentiviral vector that inserts genetic material into a patient's own cells, researchers say monitoring for cancer risk from that insertion should continue for years — Du suggested up to a decade — even as short-term results look encouraging.

Who stands to benefit

The patient population most directly affected is people with autoimmune neurologic diseases who have exhausted standard treatments, particularly progressive multiple sclerosis, for which options remain limited compared with the relapsing-remitting form of the disease. If the in vivo approach holds up in larger trials, specialists say its biggest advantage may be accessibility: removing the need for specialized cell-manufacturing facilities could make CAR-T-style treatment feasible in hospitals and even countries that lack the infrastructure for conventional ex vivo therapy, sharply lowering per-patient cost. A related trial from the same investigator group targeting neurological autoimmune disease with a CD19/BCMA construct is registered with the U.S. National Library of Medicine's ClinicalTrials.gov database, suggesting the Tongji Hospital team is already expanding the approach. Biotech investors tracking in vivo gene-therapy platforms, including Genocury, are also watching closely, since a viable injectable alternative to lab-manufactured CAR-T could reshape the competitive landscape against companies pursuing the conventional ex vivo route.

What happens next

Tian's team said it plans larger, disease-specific follow-up trials to test whether the initial results hold up in bigger patient groups and over longer periods, likely separating progressive MS, myasthenia gravis and other conditions into distinct trial arms rather than pooling them as this first-in-human study did. Outside researchers, writing in specialist forums including an analysis by MS specialist Gavin Giovannoni, said the central open questions are durability of the immune effect and the long-term insertional-mutagenesis risk from the lentiviral vector, both of which only years of follow-up can answer. A separate critique published by Clinical Trial Vanguard argued that the trial's small, single-arm design means the encouraging headline numbers should be treated as hypothesis-generating rather than conclusive, and that a randomized, placebo-controlled trial will be needed before regulators anywhere are likely to consider approval.