To date, the remarkable effectiveness of chimeric antigen receptor (CAR) T-cell therapy has been largely limited to hematologic malignancies. Blood cancer patients facing a terminal diagnosis are often willing to undergo a burdensome treatment regimen and tolerate significant side effects in exchange for a chance at long-term remission. Nevertheless, broad adoption faces significant challenges including the incredibly high cost of CAR T-cell therapy, the complex manufacturing process requiring significant specialized facilities and potential delays in treatment.
However, scientific advancements are converging to enable in vivo generation of CAR-T cells with the potential of improving their safety, efficacy, accessibility and timeliness. Furthermore, with in vivo CAR-T cells, other conditions such as autoimmune conditions become more feasible and promising targets. To fully appreciate why these in vivo approaches are so transformative, it helps to first understand how the operational challenges of current CAR T-cell therapies have slowed broader adoption.
Ex vivo CAR T cell therapy: a compelling mechanism with high operational costs
Today’s approved CAR T-cell therapies represent a transformative medical breakthrough for the treatment of relapsed or refractory B-cell driven cancers. Treatment with T cells synthetically engineered to target a B-cell-specific surface protein, often CD19, clears the immune system of B cells that drive disease. The immune system can then be reconstituted with naive B cells produced by the bone marrow. While the clinical reality is complicated, CAR T-cell therapy has the potential to be curative, and has led to the long-term remission of B-cell-driven cancers that would otherwise have been terminal.
CAR T’s therapeutic mechanism translates intuitively to autoimmune diseases with high B-cell involvement, including lupus and multiple-sclerosis where they eliminate a patient’s B cells that are producing autoreactive antibodies. This initiated an immune reset leading to repopulation of the immune system with naïve, non-autoreactive B cells. However, hurdles of safety, cost and logistical burden have dampened this otherwise promising area of therapeutic development.
To manufacture an ex vivo CAR T-cell therapy, a patient’s T cells are collected, engineered in a laboratory to express CARs, expanded in culture for two to four weeks, activated and quality-tested before being shipped back to the patient for administration. The process is lengthy; costing upwards of half a million dollars. Furthermore, CAR T-cell therapy involves significant supplemental treatments to prepare the patient to receive the T cells and subsequently, to manage side effects; and requires an infrastructure that exists at only a handful of academic medical centers worldwide.
While this therapeutic approach has been challenging to develop and commercialize for cancer, it faces even more severe headwinds for autoimmune conditions, in which patient populations have high unmet need, but can manage their condition long-term with preexisting immunosuppressive treatments. For many of these patients, the potential for a CAR T-cell therapy to produce long-term remission is compelling. However, the burdens of treatment, or participation in a clinical trial, are impractical.
Technological advancements converge to enable in vivo CAR T
A confluence of technological advancements enabling targeted in vivo delivery are beginning to release the brakes on the development of CAR T-cell therapies for autoimmune disease, where the patient population’s risk-benefit calculus favors interventions that are precise, reversible, and calibrated.
To engineer a CAR T cell to treat autoimmunity in vivo, T cells must receive instructions, ideally as mRNA, that temporarily allows them to produce B-cell-targeting CARs. This requires a very specialized delivery vehicle that can encapsulate fragile, charged nucleic acid cargo and facilitate entry into cells.
From a mechanistic standpoint, the optimal delivery vehicle for in vivo CAR T-cell therapies must also be able to specifically target T cells. To reach these T cells, it must avoid provoking an immune response, and minimize removal from the blood by the liver. The instructions must be transient to prevent long-term or permanent elimination of B cells from patients. And, because both therapeutic potency and the severity of side effects are hard to predict, the therapy should allow for controlled, repeat dosing. Finally, from an operational standpoint, the optimal delivery vehicle should be cost-effective to manufacture and store.
Modified lipid nanoparticles (LNP) have proven to be well suited to these mechanistic and operational requirements. First used to deliver mRNA vaccines, the infrastructure for LNP-based nucleic acid therapies are already in place, and LNP have demonstrated the ability to repeatedly deliver transient mRNA instructions to cells, while enabling either minimal or potent immune activation, depending on the specific application.
However, when administered systemically, conventional LNP lack natural T-cell specificity, and are rapidly sequestered by the liver, which limits the maximum tolerated dose and reduces access to the target cells. While LNP in their native form are suboptimal for in vivo CAR T applications, they can be modified to overcome these limitations through the addition of protein-targeting molecules and modifications to minimize liver uptake.
For instance, LNP can be conjugated with monoclonal T-cell antibodies to improve their specificity for T cells. However, the relatively large size of antibodies makes this method of targeting relatively crude, introducing instability and potentially triggering immune responses. More recently, antigen mimetics including designed ankyrin repeat proteins (DARPins), have emerged as a promising alternative to antibodies for LNP targeting. DARPins are an order of magnitude smaller than monoclonal antibodies, lack reactive disulfide bonds, and maintain high affinity for target proteins. In head-to-head mouse models, DARPin-LNP showed significantly higher target cell binding, uptake, and mRNA expression in CD8+ T cells compared to their antibody-conjugated counterparts, as well as superior safety.
DARPin-conjugated LNP can also be modified to avoid uptake by the liver through alteration of their lipid composition. This is important for improved T-cell specific uptake and toxicity reduction, enabling more potent dose-effects and improving redosability. In mouse-model preclinical studies, LNP modified in this way increased their plasma half-life from 25 minutes to over 2 hours, yielded an 11-fold enhancement in specific T-cell uptake, a 7.5-fold increase in payload expression in target T cells and 10-fold reduction in off-target liver expression.
Finally, LNP may be compatible with a manufacturing method that separates the production of LNP and their cargo and recombines them at the point of care. This method of therapeutic manufacturing could remove long-standing operational and infrastructural barriers to the delivery of personalized nucleic acid therapeutics, lower costs, increase access and improve the feasibility of personalized, single-batch therapeutics. For therapeutic developers and patients who have thus far been dissuaded from pursuing CAR T-cell therapy for autoimmune conditions, a more cost-effective method of manufacturing and delivery could be transformational.
Technology changes, but the goal remains the same
As technological and scientific breakthroughs converge, investment in CAR T-cell therapies for applications beyond oncology is gaining momentum. Already, researchers have established clinical proof of concept for in vivo CAR T-cell therapy in oncology and in autoimmune disease. As therapeutic developers are energized by the advancements in the field, they must carefully weigh the choice of delivery vehicle. It determines not only how effectively a therapy reaches its target, but also the balance between efficacy, safety and patient burden. Delivery platforms that offer precise targeting, repeat dosing and superior safety, while remaining scalable and cost-effective, are increasingly well positioned to meet these demands. Selecting the right delivery approach will be key to translating innovation into real-world impact.
