When CAR-T Crosses into Autoimmune Diseases.

In the early autumn of 2026, the most attention-grabbing news in the cell therapy field came in the form of two pause announcements.

Novartis temporarily halted its CD19-targeted CAR-T candidate rapcabtagene autoleucel (rap-cel) in multiple autoimmune disease trials after reporting three patient deaths. The deaths were attributed to immune effector cell-associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS), a known but severe systemic inflammatory side effect of CAR-T therapy.

At almost the same time, Bristol-Myers Squibb also voluntarily paused its autoimmunity studies of zolacabtagene autoleucel (zola-cel), a CD19-targeted autologous CAR-T, due to "transient and reversible inflammatory events" observed.

Unlike in oncology, autoimmune disease patients are not typically in immediate danger of death. They may have suffered from their disease for years, but many still live with it. In this patient population, even one treatment-related death could alter the entire risk-benefit calculation for the field.

This is not the first time CAR-T has faced safety scrutiny. But when the target population shifts from blood cancers to chronic autoimmune diseases, where exactly should we set the tolerance for risk?

Chapter 1: Where to Draw the Risk Boundary?

According to Novartis's disclosure, the paused rap-cel trials covered multiple indications including systemic lupus erythematosus/lupus nephritis, systemic sclerosis, myasthenia gravis, and multiple sclerosis. All three deaths were caused by severe IEC-HS complications.

IEC-HS is a rare but extremely dangerous adverse event in CAR-T therapy. When genetically modified T cells rapidly expand and activate in the patient's body, they may trigger an abnormal systemic immune response, leading to macrophage hyperactivation and multi-organ damage.

Novartis emphasized that IEC-HS is a "recognized" side effect of CAR-T and that the deaths would not affect rap-cel's development in oncology. But this statement did little to quell the debate.

The core of the controversy is this: autoimmune diseases are not cancer. Dr. Robert Kruse, Medical Director of Apheresis at the University of Texas Medical Branch, believes that for autoimmune diseases, any risk of death is unacceptable because patients are not immediately dying from their disease.

Oncologists often have different tolerances: cancer patients are typically in worse physical condition with rapidly progressing disease, and they accept higher risks. In autoimmune diseases, many patients are young and relatively healthy overall, with safety expectations closer to those of chronic disease management.

But the other side of the coin is that autoimmune diseases are not without fatal consequences. Lupus nephritis can lead to end-stage renal disease, multiple sclerosis can cause disability, and myasthenia gravis crises can be fatal. Long-term high-dose immunosuppressive therapy itself carries risks of infection, malignancy, and cardiovascular disease. Therefore, CAR-T safety assessment should not simply benchmark against "zero deaths," but should be weighed against the cumulative risk of living with chronic disease.

Some industry professionals told TXY that CAR-T therapy initially faced cytokine release syndrome (CRS) and neurotoxicity without effective management tools, but interventions like tocilizumab later emerged and significantly improved CRS management. Companies are already researching drugs to prevent and treat IEC-HS — a side effect that likely comes with efficacy — and it may eventually be manageable, provided the industry is willing to invest resources in solving it rather than abandoning the approach.

Kruse expressed a similar view: for patients whose quality of life is severely affected by their disease, the prospect of durable remission and freedom from long-term immunosuppressive therapy after a single infusion holds immense value.

"This is the most exciting thing in biotech right now," Kruse said. "I don't think the research should stop."

The currently available information is insufficient to determine the exact cause of the three IEC-HS cases. One notable clue is that both Novartis's rap-cel and BMS's zola-cel use rapid manufacturing platforms.

William Blair analysts speculate that rapid production may lead to enhanced cell expansion, increasing toxicity risk. Dr. Bruce Levine, a University of Pennsylvania professor and CAR-T pioneer, also noted that these trials use more potent versions of CAR-T, and that for certain patient categories, lower doses may be appropriate. If this speculation holds, the problem is not the CAR-T platform itself, but the fine-tuning of dosing and manufacturing processes.

Chapter 2: The Second Leap

In hematologic malignancies, CD19-targeted CAR-T is effective because it completely eliminates B cells expressing CD19, including malignant B cells. In autoimmune diseases, the pathological process similarly originates from self-reactive B cells and their differentiated plasma cells. These cells produce autoantibodies that attack self-tissues, driving inflammation and tissue damage.

Thus, in theory, deep B-cell depletion should be able to cut off the disease at its source — but the goal is even more ambitious than in oncology.

In cancer, the goal of CAR-T is to kill tumor cells. In autoimmune diseases, the goal is "immune reset." Researchers hope that after deep B-cell depletion, the immune system can rebuild from primitive precursor cells, inducing durable, even drug-free, remission.

Early data provide preliminary support for this hypothesis.

A study from Friedrich-Alexander-Universität Erlangen-Nürnberg in Germany showed that 15 patients with three different severe autoimmune diseases — systemic lupus erythematosus, inflammatory myositis, and systemic sclerosis — achieved drug-free remission after CD19 CAR-T therapy.

A report published in the New England Journal of Medicine in February 2024 stated that no patients experienced moderate or severe CRS or immune effector cell-associated neurotoxicity syndrome; only mild adverse events such as mild infections were observed.

These results suggest that in autoimmune patients, CAR-T side effects may be tolerable, and responses may persist even after CAR-T cells disappear and B cells reconstitute.

It is precisely on this promise that autoimmunity has become one of the most active directions in CAR-T research. According to a 2025 estimate, more than 119 registered clinical trials cover lupus, myasthenia gravis, multiple sclerosis, pemphigus vulgaris, and other B-cell-mediated autoimmune diseases.

The parallel exploration of different technology platforms indicates that the industry has not yet reached consensus on the "optimal CAR-T modality for autoimmune diseases." But the trend is toward more controllable and precise immune reset tools, rather than the "maximum firepower" approach used in oncology.

However, the very goal of "immune reset" is itself coming under scrutiny. Some clinical researchers have asked: when we pursue "cure," are we chasing "B-cell depletion" on a lab report, or "returning patients to normal life"? If these two goals do not fully align, is it necessary to insist on the former to achieve the latter?

Another perspective draws a comparison: in oncology, medicine has accepted "living with tumors." In autoimmunity, should we instead demand a "format and reinstall" approach to immune clearance? Is this logic reasonable?

A more realistic goal might be to push pathogenic clones below the tissue damage threshold, allowing patients to maintain low disease activity and organ preservation over the long term, rather than striving for complete eradication.

At the same time, some industry professionals told TXY that these patients have extremely poor quality of life, and that long-term immunosuppression-induced organ damage is itself a major morbidity factor. Their desire for "true cure or long-term drug-free remission" is real. From this perspective, "immune reset" remains a highly attractive and promising direction in autoimmunity.

Chapter 3: The Triple Barriers to Accessibility

Even if safety and efficacy issues are ultimately resolved, the widespread application of CAR-T in autoimmune diseases faces another major barrier: accessibility.

According to a U.S. claims analysis of commercially insured patients with B-cell lymphoma, the median cost of the CAR-T product alone exceeds $400,000, not including hospitalization, preconditioning, toxicity management, and long-term follow-up.

For oncology patients, this price can be partially justified by the "life-saving" logic with payers. But for autoimmune patients — especially young, working-age individuals whose disease is not immediately life-threatening — persuading payers to cover such high upfront costs requires a completely different economic rationale.

At a minimum, payers will need to see long-term follow-up evidence that remission is durable and that long-term organ damage is indeed reduced. Before then, the payment arrangement for CAR-T in autoimmune diseases may require innovative models such as installment payments, outcomes-based pricing, or pay-per-year-of-remission.

Equally challenging as cost is the clinical infrastructure gap. CAR-T cannot be delivered by a single physician or department. In oncology, early programs required coordination across hematology, cell therapy, apheresis, pharmacy, neurology, and intensive care around a single treatment.

Expanding into autoimmune diseases retains this complexity while adding cross-specialty clinical co-management. Cell therapy teams understand lymphodepletion, product infusion, and acute toxicity management; rheumatologists or neurologists understand how to manage the underlying disease, interpret disease-specific symptoms, and monitor for relapse. Neither side possesses all the expertise needed to independently manage patients.

One widely discussed solution is the hub-and-spoke model. Large academic centers serve as hubs, responsible for complex patient evaluation, product administration, cryopreservation, and management of acute severe toxicities. Smaller clinics or community infusion centers serve as spokes, handling patient identification, initial testing, T-cell collection, product infusion, and long-term follow-up.

Historically, infusion therapies for multiple sclerosis have driven the construction of specialized infusion centers. CAR-T will likely follow a similar path. As safety profiles simplify and preconditioning chemotherapy is potentially eliminated, infusions may gradually shift to outpatient settings.

Chapter 4: The Next Stop

The pauses by Novartis and BMS have forced the industry to slow down and more carefully examine which patients are best suited for this powerful but not risk-free therapy, and how to make the therapy itself safer and more accessible.

From a longer perspective, CAR-T is just the beginning of a paradigm shift in autoimmune treatment. The field is moving from the old mindset of "immune suppression" to a new framework of "precise reprogramming." Beyond CAR-T, next-generation modalities such as in vivo therapy, gene editing, and off-the-shelf cell therapies may demonstrate greater scalability.

In vivo CAR-T is particularly noteworthy. This approach uses targeted delivery vectors to directly engineer T cells inside the patient's body, bypassing the logistics and supply chain of ex vivo manufacturing. Theoretically, it could be administered in outpatient settings, even in regions lacking cell therapy facilities. Eli Lilly's $2.4 billion acquisition of Orna Therapeutics was driven by the potential of this direction.

Of course, in vivo therapy is not without risk. Some industry observers told TXY that in vivo CAR-T is essentially a gene therapy whose long-term safety has not been fully validated. Using it directly as a first-line indication in autoimmunity may carry elevated risk. A more rational path might be to first select disease categories with the widest safety tolerance window, establish a foothold, and then gradually expand.

Gene editing and off-the-shelf cell therapies are also progressing. But regardless of how the technology evolves, one point worth反复 considering is that what autoimmunity may need is not just a "CAR-T product," but a "CAR-T mindset" — the therapeutic concept of targeting and eliminating specific pathogenic immune cell subsets to achieve immune homeostasis reset.

This concept can be realized through cell products, bispecific antibodies, ADCs, or even small molecules. If the industry fixes its gaze on a single product class, it may miss a broader space for therapeutic innovation.

A decade ago, CAR-T rewrote the prognosis of hematologic malignancies. Today, it is learning how to treat patients who are not imminently dying but are lifelong prisoners of autoimmune attack. The difficulty of this transition is no less than the original breakthrough from scratch.

The pauses by Novartis and BMS may be precisely a necessary deep breath in this long journey.

References:

1. UPDATED: Novartis halts autoimmune CAR-T trials after 3 deaths, as BMS also pauses studies; Fierce Biotech

2. Novartis pauses CAR T studies due to 3 deaths; BMS follows suit after safety events; BioSpace

3. Opinion: CAR Ts will only treat autoimmune diseases if we can scale delivery; BioSpace