CAR NK cell therapy — the engineering of natural killer (NK) cells with chimeric antigen receptors (CARs) — has emerged as one of the most compelling off-the-shelf alternatives to autologous CAR-T in modern cancer immunotherapy. By grafting antigen-targeting specificity onto the innate cytotoxic machinery of NK cells, researchers have unlocked a manufacturing paradigm that could make cell therapy accessible to far more patients, at lower cost, and with a markedly improved safety profile. This article reviews the biology, manufacturing, clinical landscape, and research tools that define CAR-NK cell therapy in 2025.
What Is CAR-NK Cell Therapy?
Natural killer cells are innate lymphocytes that patrol the body for stressed, infected, or malignant cells — and destroy them without requiring prior sensitization or MHC-restricted antigen presentation. NK cells express a suite of germline-encoded activating receptors — including NKp30, NKp44, NKp46, and NKG2D — that recognize ligands upregulated on tumor and virally infected cells. Inhibitory receptors, primarily killer-cell immunoglobulin-like receptors (KIRs), survey HLA class I expression on target cells; loss of HLA-I ("missing self") lifts inhibition and triggers killing. NK cells also engage in antibody-dependent cellular cytotoxicity (ADCC) via the Fc receptor CD16 (FcγRIII), enabling them to redirect killing toward any target coated with IgG antibody.
Engineered natural killer cell therapy — more precisely, natural killer cell therapy augmented with a CAR construct — superimposes antigen-specific recognition on top of this broad innate killing capacity. The CAR construct consists of an extracellular single-chain variable fragment (scFv) targeting a tumor-associated antigen, a hinge and transmembrane domain, and intracellular signaling modules (commonly CD3ζ plus NK-optimized costimulatory domains such as 2B4 or DAP10). When the scFv binds its target, the intracellular domain triggers calcium flux, granule polarization, and perforin/granzyme-mediated cytotoxicity — layered on top of any simultaneous innate killing signals the NK cell receives.
NK cells used in CAR-NK manufacturing are sourced from three main platforms: peripheral blood NK cells isolated from healthy donors, umbilical cord blood NK cells expanded ex vivo, and iPSC-derived NK cells generated from induced pluripotent stem cells. Each source offers distinct trade-offs in scalability, phenotypic consistency, and allogeneic safety profile — discussed further in the manufacturing section below.
CAR-NK Cells vs. CAR-T Cells: Key Differences
The defining advantage of CAR-NK over CAR-T lies in allogenicity. Because NK cells do not mediate graft-versus-host disease (GvHD) — they lack the clonotypic T-cell receptor responsible for alloreactive tissue destruction — donor-derived NK cells can be infused into HLA-mismatched recipients without the immunosuppressive conditioning required for allogeneic CAR-T. This makes allogeneic cell therapy genuinely feasible with NK cells: a single manufacturing run from one healthy donor can yield doses for many patients, enabling true "off-the-shelf" supply chains analogous to a conventional drug product.
The landmark clinical demonstration of this principle was the cord blood CAR-NK trial published by Liu et al. in the New England Journal of Medicine (2020) (NCT03056339), which showed that allogeneic CD19-directed CAR-NK cells derived from cord blood could be infused without GvHD, without cytokine release syndrome, and with clinical responses in patients with relapsed/refractory CLL and NHL — establishing the safety and feasibility of the allogeneic NK cell therapy approach.
| Feature | CAR-NK cells | CAR-T cells (allogeneic) |
|---|---|---|
| GvHD risk | Absent — NK cells lack clonotypic TCR | Present even with HLA matching; requires prophylaxis |
| Allogenicity (off-the-shelf) | High — donor NK cells can be used in any recipient | Limited — alloreactive TCR must be knocked out (TALEN/CRISPR) |
| Cytokine release syndrome risk | Low — NK cells produce limited IL-6 and TNF | High — IFN-γ, IL-6, TNF storms documented |
| In vivo persistence | Short (days–weeks without cytokine support) | Longer (weeks–months); memory T-cell subsets persist |
| Manufacturing scalability | High — cord blood and iPSC platforms yield large batches | Moderate — leukapheresis, gene editing, quality steps |
| Off-the-shelf potential | Strong — no HLA-matching or patient leukapheresis required | Requires TCR/HLA KO engineering for true allogeneic use |
| Current clinical stage | Phase I/II; >100 registered trials as of 2025 | Multiple approved products (axi-cel, tisa-cel, liso-cel, ide-cel) |
The absence of GvHD is the pivotal property that makes CAR-NK cells suitable for allogeneic manufacturing at scale. Whereas allogeneic CAR-T programs must engineer out the endogenous TCR (and often HLA-I) using gene-editing tools, CAR-NK products require no such step — the innate TCR-free biology of NK cells resolves the problem by default. This simplification in the manufacturing process translates directly to lower per-dose cost and shorter release timelines.
NK Cells and Cancer: Mechanism of Action
Understanding the therapeutic potential of CAR-NK cells requires a grounding in how NK cells kill tumors through multiple overlapping mechanisms — a redundancy that is both a biological strength and a manufacturing design consideration.
Missing-self recognition. Normal healthy cells display HLA class I molecules that engage inhibitory KIRs on NK cells, preventing activation. Malignant cells frequently downregulate HLA-I to evade cytotoxic T lymphocytes — but this same escape strategy renders them vulnerable to NK cells, whose inhibitory tone is removed when the HLA-I ligand is absent. This is the "missing self" hypothesis articulated by Vivier et al. (Science, 2011).
Activating receptor engagement. Tumor cells upregulate stress ligands — MICA, MICB, and the UL16-binding proteins (ULBPs) — that are recognized by NKG2D on the NK cell surface. NKG2D signaling, together with inputs from NKp30, NKp44, and NKp46, tips the activating/inhibitory balance toward killing when the cumulative activating signal exceeds the inhibitory threshold. Caligiuri (Blood, 2008) provides a comprehensive overview of human NK cell biology and these activation pathways.
ADCC via CD16. NK cells express FcγRIII (CD16), which binds the Fc region of IgG antibodies coating tumor cells. Antibody-dependent cellular cytotoxicity (ADCC) is therefore a major effector mechanism in NK cells — and, critically, one that CAR engineering augments rather than replaces. A CAR-NK cell retains full CD16-mediated ADCC activity while gaining the additional antigen-specific targeting layer of the CAR scFv.
How CAR augments tumor recognition. The CAR construct allows NK cells to recognize tumor antigens independently of NKG2D or missing-self signals — making CAR-NK cells effective even against tumors that have not upregulated stress ligands or downregulated HLA-I. The result is a cell with orthogonal killing modalities: innate receptor-driven, ADCC-driven, and CAR-driven cytotoxicity operating in parallel.
Engineering and Manufacturing CAR-NK Cells
CAR-NK manufacturing follows a sequence of NK cell source selection, ex vivo expansion, CAR gene delivery, quality release testing, and cryopreservation.
NK cell sources and expansion. The three donor platforms differ in their expansion characteristics and resulting phenotype. Peripheral blood NK cells from healthy donors are phenotypically mature, express high CD16, and are potent ADCC effectors — but primary isolation yields are low, requiring substantial ex vivo expansion. Cord blood NK cells offer higher initial yields and excellent expansion potential but are less mature. iPSC-derived NK cells (exemplified by Fate Therapeutics' FT596, currently in Phase I/II clinical trials as of 2025 — not yet approved) offer the most scalable and phenotypically homogeneous platform, with the ability to engineer additional modifications at the stem-cell stage before differentiation.
Ex vivo expansion relies on either feeder cell co-culture or feeder-free cytokine-driven protocols. The most widely used feeder system employs irradiated or growth-arrested K562 cells — a myeloid leukemia line engineered to express membrane-bound IL-21, 4-1BBL, and other co-stimulatory ligands — which drive robust NK cell proliferation. BioHippo's Growth-Arrested NK Feeder Cells (BHC18200328) are a ready-to-use K562-based feeder system optimized for this purpose. Cytokine support — particularly IL-2 and IL-15 — is used alongside feeder cells and in feeder-free protocols to sustain NK viability and promote memory-like differentiation.
CAR gene delivery. The CAR construct is introduced into NK cells by one of three main methods: lentiviral vector transduction (stable genomic integration, used in the landmark Liu et al. 2020 NEJM trial with a retroviral vector), retroviral transduction, or mRNA electroporation (transient expression, no integration). Lentiviral vectors offer the most durable transgene expression and are the preferred platform for clinical-grade CAR-NK manufacturing; retroviral and mRNA approaches each offer manufacturing simplicity trade-offs. Integration site diversity and CAR copy number uniformity are critical quality attributes for GMP release.
Cryopreservation. Off-the-shelf viability depends on cryopreservation protocols that preserve post-thaw cytotoxic function. NK cells are more sensitive to freeze-thaw stress than T cells, making optimized cryomedia formulations and controlled-rate freezing protocols critical to product quality. BioHippo's Expanded Human Peripheral Blood NK Cells, Frozen (BHC18200327) are quality-controlled for post-thaw viability and cytotoxic function and serve as a validated starting material for downstream CAR engineering research.
CAR-NK Cell Therapy Clinical Trials: Where the Field Stands in 2025
CAR NK cell therapy has moved decisively from preclinical proof-of-concept into clinical investigation over the past five years, with the trial registry growing from approximately 32 registered studies in 2022 (as counted in systematic reviews such as Moscarelli et al., Transplant Cell Ther., 2022) to more than 100 trials registered on ClinicalTrials.gov globally as of 2025.
Anchor clinical data. The pivotal published clinical study remains Liu et al. (NEJM, 2020) (NCT03056339): 11 patients with relapsed/refractory CD19+ CLL or NHL received cord blood-derived, CD19-CAR-NK cells co-expressing IL-15 and an inducible caspase-9 safety switch. No GvHD, no cytokine release syndrome, and no neurotoxicity were observed. Eight of eleven patients responded, with complete remission in seven — the first peer-reviewed clinical evidence of allogeneic CAR-NK efficacy and safety in humans.
Target landscape. The initial focus on hematologic malignancies has broadened substantially. Current registered trials target CD19 (B-cell lymphoma, CLL), CD33 (AML), CD7 (T-cell malignancies), NKG2D ligands (AML, MDS), BCMA (multiple myeloma), EGFR, and HER2. The first published CAR NK-92 trial targeted CD33 in AML (Tang et al., Am J Cancer Res, 2018). CD138-targeting CAR-NK cells have demonstrated preclinical efficacy in multiple myeloma (Jiang et al., Mol Oncol, 2014).
Solid tumor expansion. A significant development in the 2023–2025 period has been the expansion of CAR-NK trials into solid tumors — historically more resistant to cell therapy due to the immunosuppressive tumor microenvironment. Glioblastoma (GBM), ovarian cancer, and colorectal cancer are the most active solid tumor indications in current trials. Strategies to improve solid tumor infiltration include engineering chemokine receptor expression (e.g., CXCR4 for bone marrow homing, CXCR2 for tumor trafficking) and arming CAR-NK cells with cytokine payloads (IL-15, IL-21) to sustain in vivo persistence.
Memory-like CAR-NK cells represent a clinically important subpopulation. Cytokine-induced memory-like (CIML) NK cells — generated by brief IL-12/IL-15/IL-18 stimulation — exhibit enhanced and durable antitumor responses. Romee et al. (Sci Transl Med, 2016) showed CIML NK cells induce complete remission in AML patients. Gang et al. (Blood, 2020) subsequently demonstrated that CAR-modified CIML NK cells exhibit enhanced degranulation and specific killing of NK-resistant lymphomas compared to conventional CAR-NK cells — an important advance toward overcoming tumor resistance.
Regulatory status. As of 2025, no CAR-NK cell therapy product has received FDA approval. Multiple Investigational New Drug (IND) applications have been authorized, and the field is progressing through Phase I/II safety and dose-finding studies. FDA approval will require demonstrated efficacy in randomized controlled trials — an outcome the field anticipates within the coming decade as Phase II data mature.
BioHippo's CAR-NK Research Reagents
BioHippo supplies a focused portfolio of primary NK cells, feeder systems, and viral vector tools that directly support CAR-NK cell research programs:
- Expanded Human Peripheral Blood NK Cells, Frozen (BHC18200327) — Pre-expanded, cryopreserved human NK cells from peripheral blood; quality-controlled for post-thaw viability and cytotoxic function. Suitable as starting material for CAR transduction or as effector cells in cytotoxicity and ADCC assays.
- Growth-Arrested NK Feeder Cells (BHC18200328) — K562-based feeder cells optimized for ex vivo NK cell expansion. Growth-arrested format eliminates the need for irradiation in most research settings; supports 100–1,000-fold NK expansion in conjunction with cytokine supplementation.
- Human CD56+ NK Cells (BHC16406253) — Primary human NK cells isolated from peripheral blood by positive CD56 selection; suitable for phenotyping, functional assays, and ex vivo CAR transduction experiments.
- NK Cell Expansion Kit (Cat #: BHE18200001) — Complete expansion system combining K562-based feeder cells (BHC18200328) and NK Cell Basal Medium (Cat #: BHM18200001) for two-to-three-log (100–1,000×) NK cell expansion. Note: BHE18200001 and BHM18200001 are catalog products available directly from BioHippo — contact support@biohippo.com for ordering.
- NKp44 and NKG2D ELISA Kits — Quantitative assay kits for measuring shed NKp44 and NKG2D receptor levels in cell culture supernatants; useful for monitoring NK cell activation state in CAR-NK expansion and cytotoxicity experiments.
- Lentiviral Vectors for CAR Construct Delivery — High-titer lentiviral vectors for stable CAR transgene integration into NK cells, including custom CAR virus services. Contact support@biohippo.com to discuss custom CAR lentivirus manufacturing.
Frequently Asked Questions
What is the difference between CAR-NK and CAR-T cells?
NK cells do not require donor-recipient MHC matching and do not cause graft-versus-host disease, making allogeneic (off-the-shelf) manufacturing genuinely feasible; autologous CAR-T cells carry significant GvHD risk when used allogeneically and require patient-matched production or extensive gene editing to remove the endogenous TCR. CAR-NK cells also produce substantially less IL-6 and TNF, resulting in a lower cytokine release syndrome (CRS) risk profile. The trade-off is shorter in vivo persistence for NK cells compared to memory T-cell subsets, which is an active area of engineering improvement.
Are CAR-NK cells FDA approved?
As of 2025, no CAR-NK cell therapy product has received FDA approval. Multiple IND-authorized Phase I/II clinical trials are ongoing for hematologic malignancies and solid tumors. The field is progressing toward Phase II randomized studies; FDA approval is anticipated to follow as efficacy data mature in the coming years.
What are the sources of NK cells used in CAR-NK manufacturing?
The three main donor sources are peripheral blood NK cells, umbilical cord blood NK cells, and iPSC-derived NK cells; each offers distinct trade-offs in scalability, phenotypic consistency, and manufacturing complexity. Peripheral blood NK cells are phenotypically mature and potent ADCC effectors but are low-yield. Cord blood NK cells expand robustly but are phenotypically immature. iPSC-NK cells (e.g., Fate Therapeutics FT596 platform, currently in Phase I/II trials) offer the most scalable and genetically uniform supply but require stem cell differentiation expertise.
How are CAR-NK cells manufactured?
Manufacturing involves: (1) NK cell isolation from the chosen donor source; (2) ex vivo expansion using feeder cells (K562-based systems) and cytokines (IL-2, IL-15); (3) introduction of the CAR construct by lentiviral or retroviral transduction, or mRNA electroporation; (4) quality release testing (viability, CAR expression, sterility, cytotoxicity); and (5) cryopreservation for off-the-shelf storage and distribution. A detailed methods review is provided by Rezvani et al. (Mol Ther, 2017).
What cancers are CAR-NK cells being tested against?
CAR-NK trials are most advanced in hematologic malignancies including B-cell lymphoma, chronic lymphocytic leukemia (CLL), acute myeloid leukemia (AML), and multiple myeloma. Solid tumor trials targeting glioblastoma, ovarian cancer, and colorectal cancer have expanded significantly in 2023–2025, representing the current frontier of CAR-NK clinical development.
How many CAR-NK clinical trials are currently registered?
As of 2025, over 100 CAR-NK trials are registered on ClinicalTrials.gov globally, up from the approximately 32 counted in 2022 systematic reviews (Moscarelli et al., 2022). The majority are Phase I/II safety and dose-finding studies, with a significant proportion conducted in China. Targets span CD19, CD33, CD7, BCMA, NKG2D ligands, EGFR, HER2, and others.
References:
- Eshhar Z, Waks T, Gross G, Schindler DG. Specific activation and targeting of cytotoxic lymphocytes through chimeric single chains. Proc Natl Acad Sci. 1993;90(2):720–724. doi:10.1073/pnas.90.2.720
- Vivier E, Raulet DH, Moretta A, et al. Innate or adaptive immunity? The example of natural killer cells. Science. 2011;331(6013):44–49. doi:10.1126/science.1198687
- Caligiuri MA. Human natural killer cells. Blood. 2008;112(3):461–469. doi:10.1182/blood-2008-09-077438
- Sadelain M, Brentjens R, Rivière I. The basic principles of chimeric antigen receptor design. Cancer Discov. 2013;3(4):388–398. doi:10.1158/2159-8290.CD-12-0548
- Romee R, Rosario M, Berrien-Elliott MM, et al. Cytokine-induced memory-like natural killer cells exhibit enhanced responses against myeloid leukemia. Sci Transl Med. 2016;8(357):357ra123. doi:10.1126/scitranslmed.aaf2341
- Rezvani K, Rouce R, Liu E, Shpall E. Engineering Natural Killer Cells for Cancer Immunotherapy. Mol Ther. 2017;25(8):1769–1781. doi:10.1016/j.ymthe.2017.06.012
- Liu E, Marin D, Banerjee P, et al. Use of CAR-Transduced Natural Killer Cells in CD19-Positive Lymphoid Tumors. N Engl J Med. 2020;382(6):545–553. doi:10.1056/NEJMoa1910607
- Gang M, Marin ND, Wong P, et al. CAR-modified memory-like NK cells exhibit potent responses to NK-resistant lymphomas. Blood. 2020;136(20):2308–2318. doi:10.1182/blood.2020006619
- Jiang H, Zhang W, Shang P, et al. Transfection of chimeric anti-CD138 gene enhances natural killer cell activation and killing of multiple myeloma cells. Mol Oncol. 2014;8(2):297–310. doi:10.1016/j.molonc.2013.12.001
- MacKay M, Afshinnekoo E, Rub J, et al. The therapeutic landscape for cells engineered with chimeric antigen receptors. Nat Biotechnol. 2020;38(2):233–244. doi:10.1038/s41587-019-0329-2
- Tang X, Yang L, Li Z, et al. First-in-man clinical trial of CAR NK-92 cells: safety test of CD33-CAR NK-92 cells in patients with relapsed and refractory AML. Am J Cancer Res. 2018;8(6):1083–1089. PMID: 30018879.
- Moscarelli J, Zahavi D, Maynard R, Weiner LM. The Next Generation of Cellular Immunotherapy: Chimeric Antigen Receptor-Natural Killer Cells. Transplant Cell Ther. 2022;28(10):650–656. doi:10.1016/j.jtct.2022.06.025
