CRISPR-Based Delivery System Generates CAR T Cells In Vivo

A two-vector delivery system generates tumor-targeting CAR T cells directly inside the body, achieving complete responses in mouse cancer models

October 13, 2026
3 min

The Medicine Maker

Overview

Researchers developed a two-vector CRISPR–Cas9 system that inserts a CAR transgene into the TRAC locus of T cells following direct administration in preclinical models. The approach generated CAR-expressing cells and produced antitumor responses in mouse models of B-cell acute lymphoblastic leukemia, multiple myeloma, and sarcoma; safety and translation remain unestablished.

Background

Conventional CAR T-cell therapy requires immune cells to be collected, genetically modified outside the body, and reinfused, a process described in the source as complex, time-consuming, and expensive. In vivo CAR T-cell engineering aims to generate therapeutic cells by delivering genetic material directly to endogenous T cells. The reported system combines CRISPR-mediated targeted insertion with delivery vehicles designed to transport editing machinery and a DNA template. The study remains preclinical, and the authors identify limitations in assessing immunogenicity, biodistribution, and safety using immunodeficient mouse models.

Data Highlights

Model or assessment	Reported result
Humanized mouse models	CAR-expressing cells represented up to 19.7% of splenic T cells; B cells were completely depleted.
B-cell acute lymphoblastic leukemia	One injection produced complete responses in 18 of 20 animals across four human immune-cell donors.
Direct comparison with in vivo lentiviral CAR T-cell engineering	Complete tumor control occurred in all six mice treated with the CRISPR-based approach; the source reports greater antitumor activity than the comparator.
Multiple myeloma	All eight treated mice achieved complete responses. After tumor rechallenge, three of four animals maintained disease control.
Sarcoma	Complete responses occurred in five of six animals from one donor and three of eight from another.

Key Findings

The platform pairs enveloped delivery vehicles carrying CRISPR–Cas9 ribonucleoproteins with AAV-delivered DNA templates.
The CAR transgene was inserted into the T-cell receptor alpha constant (TRAC) locus, enabling expression regulated by endogenous cellular machinery rather than random genomic integration.
The engineered AAV-hT7 variant was reported to resist pre-existing neutralizing antibodies and show enhanced uptake by human T cells; CD3-targeted EDVs were used to direct editing toward T cells.
In humanized mouse models, CAR-expressing cells reached up to 19.7% of splenic T cells, alongside complete B-cell depletion.
Complete responses were reported in mouse models of B-cell acute lymphoblastic leukemia and multiple myeloma, with antitumor activity also observed in sarcoma models.
The authors state that immunodeficient mouse models limit evaluation of immunogenicity, biodistribution, and safety, and that further studies in larger, immunocompetent animals, including nonhuman primates, are needed before clinical translation.

Clinical Implications

The findings describe a preclinical alternative to ex vivo CAR T-cell manufacturing, but they do not establish clinical safety, efficacy, or feasibility in patients. The study authors call for further evaluation in larger, immunocompetent animal models before clinical translation.

Conclusion

This two-vector system generated targeted CAR T cells in vivo and showed antitumor activity across several mouse models. Its clinical relevance remains undetermined pending further safety and translational studies.