Next-Generation CAR T in Oncology: Ex Vivo Evidence and In Vivo Promise

Cell and gene therapy (CGT) continues to redefine oncology, creating new avenues for treating complex cancers. Among these innovations, chimeric antigen receptor (CAR) T-cell therapies have delivered remarkable results in hematologic malignancies; however the full promise is tempered by access barriers, complex manufacturing logistics and varied real-world performance.

As the field advances toward next-generation approaches, two areas stand out: ex vivo CAR T programs entering randomized trials against currently approved CAR T  treatments and in vivo CAR T platforms that perform T-cell engineering inside the patient’s body.

This exciting scientific movement also raises critical questions for oncology drug developers and others involved in clinical trial design. For example, how should first-in-human and proof-of-concept studies for in vivo CAR Ts account for unique pharmacology and safety profiles? Or what evidence thresholds will ex vivo CAR Ts need to meet as comparator expectations rise?

In this article, we examine advances in both modalities and discuss strategic development considerations that can reduce uncertainty and accelerate decision-making and widen access in a space that consistently needs additional treatment options. 

Ex vivo CAR T: the state of play and strategic design considerations

Ex vivo CAR T therapy has proven efficacy in hematologic malignancies, particularly B-cell leukemias and lymphomas, with multiple treatments approved by the US Food and Drug Administration (FDA).

Commercially approved products have continued to move up to earlier lines of therapy, while the next generation of CAR T includes novel off-the-shelf, allogeneic products as well as dual-target CARs, such as CD19/CD20 or CD19/CD22, designed to minimize the risk of relapse secondary to antigen escape. Clinical trials indicate the promise of dual-target CARs for the treatment of blood cancers, helping to keep patients in remission longer.

CAR T therapy for solid tumors continues to lag in part due to tumor microenvironments that can weaken or block immune cells. To overcome this, scientists are using checkpoint inhibition, which helps remove the block or hold that tumors put on immune cells, and chemokine receptor engineering, which helps CAR T cells follow targeted chemical signals to infiltrate tumors.

Through these innovations, clinical trial sponsors aim to unlock next-generation CAR T designs’ full potential across a wider range of cancers and go beyond oncology, exploring autoimmune disease and more.

Facing a higher bar: ensuring patient access and updating trial frameworks

With multiple approved products and at earlier lines of therapy, the next-generation ex vivo CAR T clinical trials in heme malignancy now face a higher bar, where the following need to be heavily considered early in clinical development: 

• Comparator arm strategy – Regulators and health technology assessment authorities are increasingly expecting randomized evidence against the standard-of-care. In these cases, trial sponsors can prioritize head-to-head or add-on design approaches. When direct comparison is impractical, sponsors may consider rigorously designed external control arms with high-quality real-world data. Understanding the global CAR T landscape regarding approval, reimbursement and uptake for the desired indication and line of therapy is critical. For example, if randomizing to commercial CAR T, the study footprint should be limited to countries where CAR T is readily available and prescribed. If looking for CAR T naïve patients at later lines of therapy, selecting countries with limited usage can bolster enrollment.   
• Completeness of endpoint reporting – Beyond progression-free survival and objective response rates, trial programs must incorporate minimal residual disease (MRD) negativity where applicable, time-to-next-treatment and patient-reported outcomes to align with reimbursement and health technology assessments.
• Manufacturing metrics – Vein-to-vein timelines, product success rates and exposure to the bridging therapy received between T-cell collection and downstream CAR T infusion should be transparent for regulators and payers. Expanded access programs for out-of-spec products must also be considered. 

For effective regulatory and payer engagement, it is vital that trial sponsors integrate commercial access planning early in trial design stages.   

In vivo CAR Ts: excitement and uncertainty 

In vivo CAR T introduces a paradigm shift by engineering CAR T cells directly within the patient, eliminating leukapheresis, initial blood cell collection from patients, and ex vivo manipulation and the associated time lag.

With no apheresis and no personalized manufacturing, operational scalability is a major draw for in vivo CAR T design. These platforms avoid chemotherapy-based lymphodepletion, a barrier for frail patients, and reduce the need for repeated treatment quality checks that can delay progress.

In this shift, two in vivo platforms have entered first-in-human trials, primarily in B-cell malignancies, with encouraging early data outcomes:

• Engineered viral vectors, such as lentiviral or gamma-retroviral systems, are gene therapies that deliver genetic material needed to produce the desired CAR, directly to T cells and/or other immune cells inside the body. This approach genetically modifies the T cells and aims to deliver durable, long-lasting CAR expression akin to traditional ex vivo CAR T therapy. Like ex vivo CAR T, these products require long-term monitoring for insertional mutagenesis per FDA gene therapy guidance.

• Targeted RNA–lipid nanoparticle (LNP) systems deliver linear or circular RNA encoding CAR constructs via targeted LNPs, resulting in transient CAR expression. This method allows controlled dosing schedules and treatment re-administration to potentially offer a safer, more flexible alternative, with potential for tumor control in preclinical phases and B-cell depletion in non-human primate models. Establishing the duration of response or persistence necessary for durable remission are critical for these assets. Given that there is no genetic modification, insertional mutagenesis is not a concern. However, there is the potential for a unique safety profile that includes LNP toxicity, immunogenicity and hypersensitivity reactions, in addition to standard CAR T complications like cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome.  

Early‑phase in vivo clinical design: measure more, earlier and longer

Unlike ex vivo CAR T trials, which hinge on individualized manufacturing and long vein-to-vein times, in vivo studies resemble conventional drug trials, potentially offering streamlined enrollment, faster dosing and broader eligibility. However, unique complexities should be noted early in trial design, including:

• Endpoint considerations, such as progression-free survival, MRD negativity, and translational biomarkers (e.g., circulating CAR T cells and circulating tumor DNA), to capture early pharmacodynamic signals.
• Dose and related schedule for the RNA therapeutics, as well as first-in-human development plan (i.e., oncology indication vs autoimmune disease and consideration for healthy volunteer approach more in line with traditional RNA therapeutics development).  
• In vivo CAR platforms do not require sites with cell therapy capabilities (e.g., apheresis). This has the potential to dramatically improve accessibility in the long term. For early phase studies, as side effect profile is still being established, it is recommended to identify sites with expertise in managing CAR T-specific toxicities.
• Robust safety monitoring that includes considerations for traditional CAR T-specific toxicity but is also guided by the platform (e.g., LNP toxicity, immunogenicity, etc). 
• Although there is a high interest in in vivo CAR T platforms, it can be challenging to find CAR T-naïve patients in later lines of therapy. Creative trial design and country selection strategies can ensure timely dose finding and expansion. 

To address regulatory concerns and requirements, trial sponsors are increasingly using adaptive trial designs to accommodate small patient populations and evolving endpoints, which also need to account for extended follow-up protocols for long-term monitoring of delayed adverse events.

Teach Me in 10: Accelerating Cell & Gene Therapy Development 2026

In a series of rapid, expert-led sessions, this event will spotlight the tools, technologies and strategies driving faster, smarter progress from early research through to scalable manufacturing in cell and gene therapy.

View Online Event

Advertisement

Regulators will also demand rigorous biodistribution studies to confirm vector targeting and minimize off-target effects alongside strict Chemistry, Manufacturing and Controls for vector consistency and potency. This is particularly important where RNA-based in vivo CAR T platforms are evaluated in healthy volunteers. 

Where can the promise of CAR Ts go?

Looking ahead, both modalities are advancing toward solid tumors and combination regimens and also expanding beyond oncology into non-malignant, B-cell-mediated diseases.

Ex vivo CAR T trials are exploring armored CARs and checkpoint inhibitor combinations, while in vivo approaches aim to integrate multi-receptor CARs and localized delivery systems to overcome tumor microenvironment barriers. These programs face the need to provide additional validation, demanding robust comparator design and endpoints that demonstrate differentiation.