In Vivo CAR-T Arrives: Early Signals from a New Generation of Cell Therapies

Key Takeaways

• Early in vivo CAR-T clinical data show that T cells can be genetically engineered directly inside the body, producing MRD-negative responses in small multiple myeloma cohorts and signaling a potential breakthrough in next-generation immunotherapy.

• Lentiviral in vivo CAR-T platforms such as ESO-T01 and KLN-1010 demonstrate biological activity without traditional ex vivo CAR-T manufacturing or preparative chemotherapy, highlighting a possible shift in cell therapy production models.

• Strategic acquisitions by AstraZeneca and AbbVie underscore accelerating industry investment in in vivo CAR-T and targeted LNP-based approaches.

• Nonviral lipid nanoparticle (LNP) CAR-T technologies broaden the field by enabling mRNA-based in vivo T-cell engineering, reinforcing the growing importance of scalable, off-the-shelf cell therapy solutions.

• Although early findings are promising, long-term safety, durability, and real-world clinical performance remain critical questions that will determine whether in vivo CAR-T can transform future cancer and autoimmune treatment landscapes.

After more than a decade of rapid advances, the field of cellular immunotherapy may be approaching another pivotal transition point. Researchers and developers are beginning to test whether chimeric antigen receptor (CAR-T) cells — the engineered immune cells that have transformed treatment options for several hematologic malignancies — can be created directly inside a patient’s body rather than manufactured externally. Although still in early clinical stages, this emerging in vivo paradigm has generated significant attention because it could reshape how patients access and receive cellular therapies.

Cancer immunotherapy’s impact has been recognized for years: Science named it the 2013 Breakthrough of the Year, and CAR-T cell therapy was included among the approaches highlighted in that package. In its conventional form, CAR-T relies on genetically engineering a patient’s own T cells to better recognize and attack malignant cells. That process, while highly effective for many patients, is operationally demanding. Autologous CAR-T manufacturing remains complex and typically requires several weeks, with reported vein-to-vein timelines ranging from roughly two to six weeks depending on the product and process.

Given this logistical burden and the pressures it places on patients, facilities, and the broader healthcare system, it is not surprising that researchers are exploring a fundamentally different approach. The concept of generating CAR-T cells in vivo offers a potential path to simplify or even bypass the bottlenecks inherent to traditional manufacturing. These early clinical experiments, though still small and preliminary, suggest that in vivo engineering could become a meaningful alternative strategy as the field continues to evolve.

Although CAR-T therapies have delivered transformative outcomes for many patients, the economic and operational realities surrounding their use have driven intense interest in more streamlined alternatives. The earliest commercial products entered the market with strikingly high list prices: tisagenlecleucel (Kymriah) at $475,000 and axicabtagene ciloleucel (Yescarta) at $373,000. Commentators have further noted that the total cost of CAR-T treatment can reach half a million dollars in some cases.

These price points — combined with the multi-week, resource-intensive manufacturing process — have reinforced the need for approaches that are faster, less complex, and potentially more scalable. While conventional CAR-T therapy remains a cornerstone of modern hematologic oncology, the logistical and financial demands associated with ex vivo engineering create strong incentives to explore models that could reduce these burdens.

This motivation helps explain the growing attention around in vivo CAR-T strategies now entering early-phase clinical testing. Two of the most visible examples are ESO-T01 from EsoBiotec and KLN-1010 from Kelonia Therapeutics, both designed to generate CAR-T cells directly within patients and both being evaluated in relapsed or refractory multiple myeloma. Their early findings have become focal points for discussions about what the next generation of cellular immunotherapy might look like.

Among the earliest in vivo CAR-T candidates to reach human testing is ESO-T01, developed by EsoBiotec. ESO-T01 is an immune-shielded lentiviral vector engineered to specifically reprogram T lymphocytes in vivo into BCMA-targeted CAR-T cells, enabling the therapeutic cell product to be generated directly within the patient. This design represents one of the first attempts to translate lentiviral gene delivery — long used in ex vivo CAR-T manufacturing — into an in vivo strategy intended to reduce procedural complexity.

Initial clinical experience with ESO-T01 has been reported in an investigator-initiated study involving a patient with relapsed or refractory multiple myeloma in China. According to EsoBiotec, the patient received ESO-T01 without prior lymphodepletion, a notable departure from conventional CAR-T protocols. By day 28, several encouraging biological signals were observed: minimal residual disease (MRD) became undetectable in the bone marrow, the patient’s elevated free light chain levels normalized, and no significant adverse events were reported. While these findings reflect only a single case, they establish feasibility for lentiviral in vivo reprogramming in patients.

Subsequent secondary analyses have contextualized this early clinical experience within a small four-patient cohort. The ESO-T01 effort represents the first clinical report of in vivo CAR-T therapy in patients with relapsed or refractory multiple myeloma, involving treatment of four individuals. An independent summary notes differentiated responses across the group: two stringent complete responses and two partial responses, with MRD negativity achieved in the bone marrow by day 28 in these patients. These early-phase findings, although preliminary and limited in scale, underscore the biological activity of the in vivo approach.

For developers and manufacturers, one of the most consequential aspects of the ESO-T01 platform lies in its potential economic impact. EsoBiotec estimates that their in vivo vector approach could achieve a cost of goods an order of magnitude lower than ex vivo CAR-T manufacturing.

If validated in larger studies, such a shift could have meaningful implications for scalability and affordability — long-standing constraints in the cellular therapy field.

While ESO-T01 is one of the first in vivo CAR-T efforts to generate clinical data in multiple myeloma, KLN-1010, developed by Kelonia Therapeutics, showcases a related but distinct lentiviral platform., KLN-1010 is a replication-incompetent, self-inactivating lentiviral vector that is retargeted to CD3+ T cells and encodes a fully human anti-BCMA CAR for use in relapsed or refractory multiple myeloma.

This design, similar in conceptual spirit to ESO-T01, aims to deliver the CAR construct directly to T cells in vivo so that the therapeutic cell product is generated inside the patient rather than in an external manufacturing facility.

Kelonia’s in vivo BCMA program entered the spotlight with a late-breaking presentation of early data from the phase I inMMyCAR study (KLN-1010) at the 2025 American Society of Hematology (ASH) annual meeting. According to Kelonia, early results showed MRD-negative responses in all treated patients, accompanied by robust CAR-T cell expansion in the absence of preparative chemotherapy and the presence of persistent memory CAR-T cells in all patients evaluated.

Early findings show that KLN-1010 produced initial MRD-negative responses with persistent CAR-T expansion in four patients with relapsed or refractory multiple myeloma.

Taken together, these early data points position KLN-1010 as one of the leading examples of lentiviral in vivo CAR-T in the clinic. Although the sample size remains very small and follow-up is still short, the convergence of MRD negativity, in vivo expansion without preparative chemotherapy, and persistence of memory CAR-T cells has made this program a focal point for discussions about the feasibility of the in vivo model in multiple myeloma.

Even at this early stage, the competitive landscape surrounding in vivo CAR-T technologies is already beginning to consolidate, with major pharmaceutical companies moving decisively to secure strategic positions. One clear example is AstraZeneca’s acquisition of EsoBiotec, the developer of the ESO-T01 in vivo lentiviral platform. In parallel, AbbVie has signed a definitive agreement to acquire Capstan Therapeutics for $2.1 billion, a deal that reflects growing interest in alternatives to lentiviral delivery.

Capstan’s lead asset CPTX2309 exemplifies a distinct nonviral approach: a targeted lipid nanoparticle (tLNP) designed to deliver mRNA encoding an anti-CD19 CAR directly to CD8-expressing cytotoxic T cells in vivo. Unlike traditional CAR-T modalities, CPTX2309 is described as functioning without the need for lymphodepletion or ex vivo manufacturing, relying instead on the transient expression of the CAR to deplete pathogenic B cells and potentially “reset” aberrant immune activity in autoimmune disease..

Beyond this single program, the scientific foundations for nonviral in vivo CAR-T strategies continue to strengthen. A recent study demonstrated that LNPs can deliver mRNA encoding chimeric antigen receptors to immune cells in vivo, validating the broader concept that immune cells can be genetically reprogrammed within the body using LNP-mediated RNA delivery.

Together, these developments illustrate how rapidly the in vivo CAR-T field is diversifying. Lentiviral systems like ESO-T01 and KLN-1010 are advancing in the clinic, while nonviral platforms such as Capstan’s CPTX2309 and the LNP strategies emerging in preclinical research are attracting significant investment. The momentum from both scientific progress and industry consolidation suggests that in vivo engineering is becoming a central focus of the next generation of cell therapy innovation.

The early clinical signals now emerging from both lentiviral and nonviral in vivo CAR-T programs point toward a potentially broader therapeutic modality rather than a collection of isolated experimental efforts. The fact that multiple technological strategies — ranging from targeted lentiviral vectors to lipid nanoparticle RNA delivery — are advancing in parallel suggests that in vivo engineering has become a meaningful new axis for cell-therapy innovation.

At the same time, these developments highlight the importance of rigorous long-term evaluation. Safety, durability of response, and the stability of in vivo–generated CAR-T populations over time remain open questions, and the currently available sources do not yet offer extended follow-up. As with any first-in-human platform, long-term risks will need to be assessed methodically and across larger, more diverse patient populations.

If future data reinforce these early findings, in vivo CAR-T approaches could eventually reshape the manufacturing and access landscape, potentially reducing logistical bottlenecks and decentralizing the processes that currently constrain the availability of autologous cell therapies. The initial reports of MRD-negative responses in small four-patient cohorts within both lentiviral programs offer a compelling biological proof-of-concept, but the modest sample sizes underscore how early this field remains. The coming years will determine whether these encouraging signals translate into a clinically and commercially sustainable new model for cellular immunotherapy.

The emergence of early clinical data, combined with decisive strategic moves by major pharmaceutical companies and the rapid expansion of both viral and nonviral engineering platforms, signals that in vivo CAR-T therapy has progressed from a conceptual ambition to a tangible frontier in cell therapy development. The convergence of these elements reflects growing confidence that direct in-body programming of T cells may ultimately offer a viable alternative to today’s complex, resource-intensive manufacturing paradigm.

At the same time, the field remains in its infancy. The encouraging signals reported so far come from very small cohorts, with short follow-up and many unanswered questions about durability, safety, and long-term biological behavior. As larger studies unfold and longer-term outcomes become available, the true clinical and commercial trajectory of in vivo CAR-T will come into clearer focus.

For now, the momentum is unmistakable: early results, accelerating investment, and technological diversification are pushing the field forward quickly. But translating this momentum into lasting therapeutic impact will require sustained rigor, careful monitoring, and continued innovation as developers navigate the transition from first-in-human milestones to broader clinical reality.