Engineering Choices in Modern Cell Therapy

First-generation CAR-T therapies followed one blueprint. T cells were drawn from a patient’s blood, modified to carry a chimeric antigen receptor (CAR) using a lentiviral vector, expanded in culture, and returned to circulation. The approach has produced durable remissions in blood cancers and underlies every FDA-approved CAR-T product. The same workflow also accounts for the field’s persistent constraints—manufacturing is slow and expensive, efficacy against solid tumors is limited, and toxicity risks persist.¹,²

Newer strategies address these issues. CAR genes can be inserted at defined positions in the genome rather than at random; CRISPR-based editing can target multiple loci at once; synthetic receptors can be designed to require multiple antigens for activation (improving tumor specificity and reducing off-target effects);³ and some platforms generate CAR-T cells inside the patient rather than in a manufacturing facility.

Ultimately, every new cell therapy comes down to how genetic cargo is delivered, how the genome is edited, and how the cell is programmed to recognize and respond to cancer. The choice of starting cell type shapes all three. The sections that follow examine each option, the trade-offs, and how they map to specific cancers.

Delivery: Viral vectors and their alternatives

Lentiviral and gammaretroviral vectors are the standard delivery method for every FDA-approved CAR-T product, inserting the CAR gene into the genome for long-lasting expression.¹ Their main limitation is semi-random integration: insertion sites vary across the cell population, producing inconsistent CAR expression, and only a fraction of cells reach the level needed for strong antitumor activity. Random integration also carries a small but documented genotoxicity risk.⁴

Site-specific integration solves both problems. The most studied site is the T cell receptor alpha constant (TRAC) locus, which offers two coupled advantages. Inserting the CAR gene at TRAC links its expression to the cell’s native T cell receptor promoter, producing uniform CAR levels in every engineered cell. The same integration event disrupts the TCR. In allogeneic CAR-T products (donor-derived rather than patient-derived), TCR removal prevents graft-versus-host disease (GVHD). It also reduces the chronic signaling that drives T cell exhaustion.⁴

Non-viral delivery has historically lagged behind viral vectors in integration efficiency and durability. The Sleeping Beauty transposon system has changed that, using a transposase enzyme to insert the CAR gene into the T cell genome. CARCIK-CD19 combines a CD19 CAR with cytokine-induced killer (CIK) cells, a donor-derived population with innate tumor-killing capacity, engineered using Sleeping Beauty. In Phase I/II trials in patients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) relapsed after stem cell transplantation, 83% achieved complete remission and 89% of those had no detectable cancer cells.⁵ Lower manufacturing cost continues to drive its development.

In vivo CAR-T generation removes ex vivo manufacturing entirely. In a 2025 preclinical study, targeted lipid nanoparticles (LNPs) delivered anti-CD19 CAR mRNA directly to CD8 T cells. The resulting CAR-T cells controlled tumors in humanized mice and depleted B cells durably in primates.⁶

Two in vivo engineering strategies are gaining traction. Capstan Therapeutics uses targeted LNPs to deliver CAR mRNA preferentially to CD8 T cells, producing transient CAR expression without altering the genome. CPTX2309 carries an anti-CD19 CAR; a second version, CPTX2506, carries a CAR targeting B-cell maturation antigen (BCMA), which is highly expressed on multiple myeloma cells, and is in preclinical development. Kelonia Therapeutics' KLN-1010, by contrast, uses a targeted lentiviral vector that engineers T cells in vivo with stable, integrated CAR expression.

Updated Phase I data from the inMMyCAR trial, reported at the 2026 ASCO Annual Meeting, showed a 100% overall response rate and undetectable residual disease in bone marrow at one month in all evaluable patients (n=18) with relapsed/refractory multiple myeloma.⁷ AbbVie’s completed acquisition of Capstan and Eli Lilly’s pending acquisition of Kelonia both signal growing industry confidence in in vivo CAR-T approaches.

Azalea Therapeutics’ preclinical platform uses two T cell-targeted vectors for in vivo, site-specific integration. One delivers CRISPR-Cas9 to cut the TRAC locus; the other delivers the CAR gene as the template for insertion. A single intravenous dose produced therapeutic CAR-T cell levels in humanized mouse models of both hematologic and solid cancers.⁸

CRISPR and beyond: From single edits to multiplex programs

CRISPR-Cas9 is the standard tool for editing allogeneic CAR-T cells. It is used to knock out the T cell receptor (preventing GVHD) and human leukocyte antigen (HLA) proteins, which could otherwise trigger host rejection of donor cells. Base editors offer an alternative to nuclease cutting, making single-letter DNA changes without creating double-strand breaks. This reduces the genotoxicity and chromosomal rearrangements that can result from multiple simultaneous CRISPR cuts.

Zugocaptagene geleucel (zugo-cel) illustrates how these editing tools combine in practice. The product uses TRAC locus insertion for site-specific CAR expression and multiple gene knockouts to evade host rejection and improve antitumor activity. December 2025 Phase I/II data reported a 90% overall response rate and 70% complete response rate in 10 patients with relapsed/refractory large B-cell lymphoma.⁹

Solid tumors require more extensive editing because their microenvironment imposes multiple immunosuppressive barriers. A 2025 study described donor-derived CAR-T cells engineered with six edits, using both CRISPR base editors and a CRISPR nuclease. In addition to standard rejection- and GVHD-prevention knockouts, the strategy disabled three tumor-imposed immunosuppressive pathways (PD-1 checkpoint signaling, TGF-β signaling, and adenosine receptors). This improved antitumor activity in mouse models of human lung cancer, suggesting that disabling multiple immunosuppressive pathways simultaneously may be necessary for solid tumor CAR-T to succeed.¹⁰

Engineered receptors and synthetic circuits

A standard CAR pairs an antigen-binding region with costimulatory signaling components that activate the T cell when the target antigen is recognized. Newer designs replace the CAR with a T cell receptor (TCR), which recognizes peptide fragments displayed on the cell surface by HLA molecules, including fragments of proteins originating inside the tumor cell. This broadens the range of targetable antigens but restricts eligibility to patients whose HLA type matches the TCR’s specificity.

The August 2024 FDA approval of afamitresgene autoleucel (Tecelra) for synovial sarcoma, the first TCR-T approval for a solid tumor, illustrates both the promise and the constraint: the therapy is limited to HLA-A*02-positive patients.

Synthetic biology adds logic gates that require specific antigen combinations before the CAR activates. Synthetic receptors like SynNotch enable AND-gating, where the CAR activates only when both a primary and secondary tumor antigen are present, sparing healthy cells that express only one. A2 Biotherapeutics’ Tmod system uses AND-NOT logic, activating only when a tumor antigen is present and HLA-A*02 is absent, a pattern seen in patients who carry this allele but whose tumors have lost it.

Safety switches provide a final layer, allowing the engineered cells to be eliminated if toxicity becomes severe, though this also ends the therapy.

Cell sources beyond autologous T cells

The starting cell type shapes scalability, persistence, and safety, and no single type solves all three. The 2024 FDA approval of lifileucel (Amtagvi) for advanced melanoma showed that even minimally engineered tumor-infiltrating lymphocytes (TILs) can deliver clinical benefit in solid tumors, a setting where CAR-T has long struggled.¹¹

Other cell types address different limitations. CAR-engineered natural killer (NK) cells reduce the need for patient-matched donors and retain the NK cell’s natural ability to recognize and kill abnormal cells, independent of the CAR. A cord-blood CAR-NK product engineered to co-express IL-15 for improved persistence achieved a 48.6% overall response rate at 100 days in 37 patients with relapsed/refractory B-cell malignancies, with no severe cytokine release syndrome.¹² Induced pluripotent stem cell (iPSC)-derived approaches take scalability further still, enabling clonal, off-the-shelf manufacturing for both CAR-T and CAR-NK products, with multiple programs now entering clinical testing.

Getting these cells into solid tumors (and keeping them active there) is one of the hardest problems in cell therapy. CAR-engineered natural killer T (NKT) cells, which combine T cell specificity with the rapid tumor-sensing ability of innate immune cells, may disrupt the immunosuppressive tumor-associated macrophages that shield tumors from attack. CAR-macrophages take a different approach: naturally recruited into tumors, they gain direct access to the microenvironment and reprogram it from within. A first-in-class CAR-macrophage targeting HER2 (overexpressed in breast, gastric, and other solid tumors) achieved stable disease in 44% (4 of 9) of patients with the highest HER2 expression in Phase I.¹³

Conclusion

Hematologic malignancies are well-served by patient-specific CAR-T, with Sleeping Beauty transposons offering lower-cost alternatives. Solid tumors require multiplex editing and synthetic logic gates matched to the immunosuppressive tumor microenvironment. Allogeneic cell therapies (CAR-NK, CAR-NKT, iPSC-derived) address scalability. In vivo CAR-T generation, already in early clinical trials, may eventually eliminate ex vivo manufacturing entirely. The field has moved from one approach to many, each designed for a specific cancer biology and, increasingly, a specific patient.

References

1. Abdo L, Batista-Silva LR, Bonamino MH. Cost-effective strategies for CAR-T cell therapy manufacturing. Mol Ther Oncol. 2025;33(2):200980. 

2. Matos LN, Costa MS, Serra M, Costa MHG. Are we there yet? The road to faster and more efficient CAR T cell manufacturing. J Biol Eng. 2026;20(1):77.

3. Shirzadian M, Moori S, Rabbani R, Rahbarizadeh F. SynNotch CAR-T cell, when synthetic biology and immunology meet again. Front Immunol. 2025;16:1545270. 

4. Eyquem J, Mansilla-Soto J, Giavridis T, et al. Targeting a CAR to the TRAC locus with CRISPR/Cas9 enhances tumour rejection. Nature. 2017;543(7643):113-117. 

5. Lussana F, Magnani CF, Galimberti S, et al. Donor-derived CARCIK-CD19 cells engineered with Sleeping Beauty transposon in acute lymphoblastic leukemia relapsed after allogeneic transplantation. Blood Cancer J. 2025;15(1):54. 

6. Hunter TL, Bao Y, Zhang Y, et al. In vivo CAR T cell generation to treat cancer and autoimmune disease. Science. 2025;388(6753):1311-1317. 

7. Kelonia Therapeutics. Kelonia Therapeutics presents updated first-in-human data from Phase 1 inMMyCAR study of KLN-1010 in vivo BCMA CAR-T therapy at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting. Press release. May 31, 2026. Accessed June 17, 2026. 

8. Nyberg WA, Bernard PL, Ngo W, et al. In vivo site-specific engineering to reprogram T cells. Nature. 2026;652(8110):712-721. 

9. CRISPR Therapeutics. CRISPR Therapeutics provides broad update on zugocaptagene geleucel (zugo-cel; formerly CTX112) in autoimmune diseases and hematologic malignancies. Press release. December 22, 2025. Accessed June 17, 2026. 

10. Murray R, Chowdhury MR, Botticello-Romero NR, et al. Multiplex gene-editing strategy to engineer allogeneic EGFR-targeting CAR T-cells with improved efficacy against solid tumors. Nat Commun. 2025;16(1):11593. 

11. Medina T, Chesney JA, Kluger HM, et al. Long-term efficacy and safety of lifileucel tumor-infiltrating lymphocyte cell therapy in patients with advanced melanoma: a 5-year analysis of the C-144-01 study. J Clin Oncol. 2025;43(33):3565-3572. 

12. Marin D, Li Y, Basar R, et al. Safety, efficacy and determinants of response of allogeneic CD19-specific CAR-NK cells in CD19+ B cell tumors: a phase 1/2 trial. Nat Med. 2024;30(3):772-784.

13. Reiss KA, Angelos MG, Dees EC, et al. CAR-macrophage therapy for HER2-overexpressing advanced solid tumors: a phase 1 trial. Nat Med. 2025;31(4):1171-1182.