In Vivo CAR Engineering and the Next Phase of Cellular Immunotherapy

Background: Why Delivery Has Become the Next Barrier

Chimeric antigen receptor (CAR) T-cell therapy has fundamentally reshaped the treatment paradigm for hematologic malignancies, delivering durable remissions in diseases once defined by therapeutic resistance.¹ Yet, despite this success, the field remains constrained by the structural limitations of its own manufacturing paradigm. Current CAR T-cell platforms depend on ex vivo engineering—leukapheresis, genetic modification, expansion, and reinfusion—a process that is not only logistically complex but biologically restrictive. Time to treatment remains a critical vulnerability in aggressive malignancies, while variability in T-cell fitness across patients introduces inconsistency in product quality and therapeutic response. These constraints are not peripheral; they define the outer limits of scalability, accessibility, and ultimately impact.

Against this backdrop, emerging strategies in in vivo CAR engineering suggest that the next major advance in cellular therapy may not come from refining the CAR construct itself, but from fundamentally rethinking how these therapies are delivered.² Rather than extracting and reprogramming immune cells externally, these approaches seek to transform circulating immune cells directly within the patient through targeted delivery of genetic payloads. In doing so, they collapse the distinction between drug and manufacturing process, reframing cellular therapy as a programmable biologic system rather than a static product.

In Vivo Programming: Replacing Manufacturing With Biology

The conceptual shift introduced by in vivo CAR engineering is as significant as the original development of CAR T-cell therapy itself. Using lipid nanoparticles or viral vectors with defined tropism, these platforms deliver CAR constructs directly to immune cells in circulation, most commonly CD8-positive T cells, enabling in situ genetic modification and expansion.³ This approach bypasses the need for centralized manufacturing infrastructure and eliminates the delays inherent to ex vivo processing, allowing for near-immediate initiation of therapeutic activity following administration.

Importantly, the biology of in vivo–generated CAR T cells appears to differ meaningfully from their ex vivo–expanded counterparts. Without the artificial stimulation and expansion processes that can drive terminal differentiation, these cells retain a more stem-like phenotype, characterized by improved proliferative capacity and persistence.³ This distinction is not merely theoretical; persistence has emerged as a central determinant of long-term response durability in cellular immunotherapy, and strategies that preserve early memory states may offer a meaningful advantage in sustaining antitumor activity.

What the Data Suggest

Early preclinical and translational data demonstrate that in vivo–engineered CAR T cells can achieve robust expansion and antitumor activity across multiple models.⁴ Following vector administration, CAR expression is detectable within days, with expansion kinetics that parallel or exceed those observed with conventional products.⁵ Functional analyses indicate preserved cytotoxicity and cytokine production, suggesting that in situ programming does not compromise effector function.

Equally significant is the potential for dose modulation. Unlike ex vivo CAR T-cell therapy, where dosing is constrained to a single infusion of a fixed cellular product, in vivo approaches introduce the possibility of repeat administration and titration based on response. This flexibility may prove particularly important in solid tumors, where dynamic adaptation to tumor burden and microenvironmental pressures is likely required for sustained control.

Control and Safety: The Central Tension

The principal challenge facing in vivo CAR engineering is not feasibility, but control. Ex vivo systems allow for precise characterization of the cellular product prior to infusion, including phenotype, viability, and transduction efficiency. In contrast, in vivo delivery introduces variability in transduction rates, cell subset targeting, and expansion magnitude, raising important safety considerations.

To address this, next-generation platforms are incorporating multiple layers of regulation, including cell-type–specific promoters, transient expression systems, and nonviral delivery technologies designed to improve targeting fidelity.⁶ These innovations aim to balance the scalability advantages of in vivo programming with the precision required for safe clinical application. Whether this balance can be achieved consistently across diverse patient populations remains an open question and will be central to the clinical translation of these approaches.

Integrative Perspective: From Engineering Cells to Programming Systems

Viewed within the broader trajectory of the field, in vivo CAR engineering represents a natural progression in the evolution of cellular therapy. The first generation of CAR T-cell therapies established proof of concept, demonstrating that engineered immune cells could achieve meaningful clinical responses.¹ Subsequent efforts have focused on optimizing cellular biology—addressing exhaustion, improving metabolic fitness, and refining receptor design.²^,³ The current shift toward in vivo programming reflects a third phase, centered not on the cell itself, but on the system through which it is generated and deployed.

This transition mirrors broader trends in oncology, where therapeutic development has increasingly moved toward dynamic, adaptive systems capable of responding to tumor evolution in real time. In this context, in vivo CAR platforms can be understood not simply as a new delivery method, but as a step toward programmable immunotherapy—an approach in which therapeutic activity is initiated, modulated, and sustained within the patient.

Clinical and Translational Implications

The implications of this paradigm shift are substantial. By eliminating the need for individualized manufacturing, in vivo CAR engineering has the potential to expand access to cellular therapy beyond specialized centers, addressing one of the most significant barriers to global implementation. At the same time, the ability to modulate dosing and potentially combine with other systemic therapies introduces new opportunities for integration into existing treatment frameworks.

In solid tumors, where trafficking, persistence, and microenvironmental suppression have limited the efficacy of conventional CAR T-cell approaches, in vivo programming may offer distinct advantages. The capacity for continuous generation and expansion of effector cells within the patient could enhance tumor infiltration and sustain activity in hostile microenvironments, although this remains to be demonstrated in clinical settings.

Open Questions

Despite its promise, several critical questions remain. The durability of in vivo–generated CAR T-cell responses relative to established ex vivo platforms is not yet defined. The predictability of expansion across heterogeneous patient populations, particularly in heavily pretreated settings, requires further study. Long-term safety, including the risk of off-target transduction and uncontrolled immune activation, must be carefully evaluated. Finally, the degree to which targeting specificity can be refined to minimize off-tumor effects will determine the ultimate applicability of this approach across tumor types.

What Comes Next

The path forward will depend on the successful translation of these platforms into early-phase clinical trials, with particular emphasis on safety, dose optimization, and biomarker-driven monitoring. Integration with advances in receptor design and metabolic engineering may further enhance the efficacy of in vivo approaches, enabling the development of next-generation cellular therapies that address multiple dimensions of tumor resistance simultaneously.

Bottom Line

In vivo CAR T-cell engineering represents a fundamental rethinking of how cellular therapies are produced and delivered. By shifting the site of manufacturing from the laboratory to the patient, this approach addresses key limitations of current platforms while introducing new opportunities for adaptability and scale. Whether it can match or exceed the durability and safety of established therapies remains to be seen, but its potential to redefine the framework of cellular immunotherapy is unmistakable.

References

1. Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma. N Engl J Med. 2017;377(26):2531–2544. doi:10.1056/NEJMoa1707447
2. Majzner RG, Mackall CL. Tumor antigen escape from CAR T-cell therapy. Cancer Discov. 2018;8(10):1219–1226. doi:10.1158/2159-8290.CD-18-0442
3. Guedan S, Ruella M, June CH. Emerging cellular therapies for cancer. Annu Rev Immunol. 2019;37:145–171. doi:10.1146/annurev-immunol-042718-041407
4. Munir M, Moore M, Tiberti S, et al. Metabolically supercharged NK cells engineered with adenoviral E4ORF-1. Cancer Res. 2026;86(suppl 7):5197. doi:10.1158/1538-7445.AM2026-5197
5. Tanyi JL, Haas A, O’Hara M, et al. Initial results of a first in human dose-escalation study of KIR-CAR in patients with advanced mesothelin-expressing solid tumors. Cancer Res. 2026;86(suppl 8):CT104. doi:10.1158/1538-7445.am2026-ct104
6. Zhang Z, Ho Y-J, Fang X, et al. A convergent uPAR-positive tumor ecosystem creates broad vulnerability to CAR T cell therapy. Cell. 2026. doi:10.1016/j.cell.2026.03.002