ADC Reviews

What is CAR-NK therapy


Chimeric antigen receptor (CARs)-based T-cell therapy has shown great success against blood cancers; however, potential lethal toxicity (eg, cytokine release syndrome) and high cost limit the clinical application of CAR-T cells Some shortcomings need to be overcome.

Natural killer (NK) cells are the focus of current immunology research, and so far, the use of NK cells to treat malignant tumors has achieved certain success. Recent advances in NK cell receptor biology have dramatically changed our understanding of how NK cells recognize and kill tumor cells. CAR-NK cells are strong candidates for cancer retargeting therapy due to their unique recognition mechanism, strong cytotoxicity and clinical safety. Furthermore, NK cells reduce the risk of allogeneic reactions and can be used as “off-the-shelf products”.

Although the current booming CAR-NK therapy still faces many challenges and obstacles, its huge potential has shown broad prospects in the field of tumor immunotherapy.

NK cell biology

NK cells are the first subtype of innate lymphocytes (ILCs) to be identified and can exert multiple effector functions on virus-infected and/or transformed cells, primarily cell killing and production of proinflammatory cytokines. NK cells and other ILC family members (ILC1s, ILC2s, and ILC3s) are derived from the same lymphoid progenitors as B and T cells. The cytotoxic activity of NK cells makes them functionally most similar to CD8+ T cells, whereas the cytokine production pattern of the ILC1, ILC2, and ILC3 populations corresponds these cells to the TH1, TH2, and TH17 subsets of CD4+ T cells, respectively.

The two most typical subsets of NK cells are the CD56brightCD16- and CD56dimCD16+ populations. CD56bright cells are less abundant in peripheral blood (90% of NK cells in circulation are CD56dim), whereas NK cells in tissues are predominantly CD56bright. CD56bright NK cells are potent cytokine producers and are less cytotoxic unless stimulated by pro-inflammatory cytokines such as IL-15. In contrast, the CD56dim NK cell population mediates serial killing of infected and malignant cells, primarily through the exocytosis of pre-assembled cytolytic granules containing granzyme B and perforin at immune synapses, ultimately inducing target cells apoptosis.

Activating and inhibitory receptors for NK cells

Unlike B cells and T cells, NK cells do not express somatically rearranged antigen receptors, but rather a random combination of activating and inhibitory receptors. The balance of stimulatory and inhibitory signals through these various receptors generates a response or tolerance to target cells. MHC-I (major histocompatibility complex class I) antigen-specific inhibitory receptors closely regulate NK cell-mediated cytotoxicity and lymphokine production. Inhibitory signaling by MHC-I-specific receptors is essential for hematopoietic cells to avoid destruction by NK cells. This concept is called “lost self” and was originally proposed by Ljunggren and Karre. This MHC-I recognition inhibitory receptor forms three families of NK cell surface receptors, namely KIRs (killer cell immunoglobulin-like receptors), LIRs (leukocyte immunoglobulin-like receptors), and NKG2A (natural killer) Cell family 2 A).


KIRs, members of the immunoglobulin superfamily, are type I transmembrane molecules that recognize classical human leukocyte antigens A, B, and C (HLA-class Ia). LIRs, also known as ILTs (immunoglobulin-like transcripts), form a second group of receptors that, in addition to HLA class Ia, mainly recognize non-classical HLA-G (class Ib) molecules. LIRs and KIRs belong to the same Ig superfamily. NKG2A is a member of the NKG2 family, including A, B, C, D, E, F and H, and dimerizes with CD94 to form the NKG2A/CD94 receptor. It belongs to the C-type lectin family of receptors and recognizes non-classical HLA-E class I molecules as its ligands.

The killing effect of NK cells requires not only the detection of MHC-I molecules on transformed cells through inhibitory receptors, but also the activation of NK cells through activating receptors. Natural cytotoxicity receptors (NCRs) are a group of natural killer cell surface activating receptors including NKp46, NKp30 and NKp44. These receptors, along with NKG2D and DNAM-1 (DNAX accessory molecule-1), recognize ligands expressed on the surface of virally infected or malignantly transformed cells. Some co-receptors (2B4, NKp80, NTB-A, and CD59) are also expressed, and they only function when they bind to other activating receptors. CD16 (or FcγRIII) is also an activating receptor, mainly expressed by the CD56dim NK cell subset, critical for antibody-dependent cellular cytotoxicity (ADCC) of IgG-coated target cells.


Like CAR-T cells, CAR-NK cells have extracellular, transmembrane and intracellular signaling domains. NK cells increase their cytotoxic capacity and cytokine production through two other co-stimulatory molecules, NKG2D and CD244 (2B4). Therefore, it has stronger tumor-specific targeting and cytotoxicity than CAR-T cells. CAR-NK cell therapy may become an alternative to CAR-T therapy in the future because CAR-NK cells have the following unique features that surpass CAR-T.


First, allogeneic NK cells are quite safe for adoptive cell therapy (ACT) because they do not usually mediate the development of GVHD. In addition, NK cells secrete only a small amount of IFN-γ and GM-CSF, but do not produce IL-1 and IL-6 that initiate CRS. Second, in addition to recognizing tumor surface antigens to inhibit cancer cells by single-chain antibodies, NK cells can also inhibit cancer cells by recognizing various ligands through a variety of receptors, such as natural cytotoxicity receptors (NKp46, NKp44, and NKp30), NKG2D and DNAM-1 (CD226). Finally, NK cells are abundant in clinical samples and can be generated from peripheral blood (PB), umbilical cord blood (UCB), human embryonic stem cells (HESC), induced pluripotent stem cells (IPSC) and even the NK-92 cell line.

Clinical study of CAR-NK cell therapy

Research on CAR-NK therapy is currently in its infancy, and the number of clinical studies is increasing year by year. In addition, in terms of targets studied, the CD19 antigen was most common in hematological tumors. The most widely advanced CAR-NK therapy in solid tumors includes targets such as tumor-associated antigens HER2, MUC1, and PMSA.

Currently, several ongoing clinical trials are investigating the safety and efficacy of CAR-NK cells in the treatment of hematological and solid tumors, and these trials are listed in the table below.


Challenges of CAR-NK Cell Therapy

low persistence

The lack of in vivo persistence of infused cells in the absence of cytokine support is one of the major drawbacks of adoptive NK cell therapy. While it may be safer, it also limits the effect of NK cell immunotherapy. Exogenous cytokines have been shown to increase the proliferation and durability of adoptive NK cells; however, they may also lead to undesirable side effects, including the growth of suppressive immune subpopulations such as Tregs.

Metastases to the desired tumor site

Rapid homing to the tumor bed is critical for adoptive cell therapy efficacy and is governed by complex interactions between NK cells and chemokines released by tumor cells. However, the efficiency of NK cells homing to tumor sites has been controversial, thus prompting continuous efforts to improve them.

Several researchers have investigated various engineering approaches to improve NK cell homing. For example, NK cells were electroporated with mRNA encoding the chemokine receptor CCR7 to increase migration to lymph nodes expressing the chemokine CCL19. NK cells transduced with a viral vector encoding CXCR2 exhibited better motility against renal cell carcinoma tumors expressing cognate ligands such as CXCL1, CXCL2, CXCL5, CXCL6 and CXCL8. To improve the success of NK cell immunotherapy in patients with solid tumors, several novel techniques for promoting NK cell transport to tumor sites have been investigated in mouse models; however, the effectiveness of these approaches needs to be validated in clinical trials.

Immunosuppressive tumor microenvironment

TME includes immunosuppressive molecules, immunosuppressive cells, and an unfavorable environment that hinders the function of immune cells, which is the main obstacle to CAR-NK cell therapy. TGF-β; adenosine; indoleamine 2,3-dioxygenase (IDO) and prostaglandin E2 (PGE2) are immunosuppressive cytokines and metabolites found in the TME that impair NK cell activity. Treg cells; regulatory B cells; myeloid-derived suppressor cells; tumor-associated macrophages (TAM); platelets; Immunosuppressive.

Therefore, researchers are working to develop CAR-NK cells that can prevent certain immunosuppressive effects. For example, by using CRISPR/Cas9 technology to knock out the TGF-βR2 gene or block the high-affinity A2A adenosine receptor on NK cells. Another important way in which TME leads to NK cell depletion is immune checkpoint molecules. To overcome this problem, genome editing was used to eliminate checkpoint components of NK cells to improve their function.

Low transduction efficiency of lentivirus

Lentivirus-based transduction systems are one of the most commonly used methods for intracellular gene modification and delivery. However, due to their natural properties, NK cells are resistant to lentiviruses, which makes lentivirus-based transduction a challenge. To improve viral transduction, various chemicals have been used, for example, protamine sulfate or dextran can be used to remove the charge on the cell membrane.

Future Prospects of CAR-NK Cell Therapy

Identify target antigens

The most critical step in the design of CARs is the identification of highly consistently expressed target tumor antigens. Most tumor-associated antigens (TAAs) are also expressed by some healthy cells, thus inevitably bringing about a “targeting non-tumor” effect. Furthermore, the expression of these TAAs may vary widely among single-cell clones of the same tumor. To overcome this problem, bispecific CARs have been designed that can target multiple antigens simultaneously.

This can be achieved in a number of ways, for example, by injecting different CAR-NK cells targeting different antigens at the same time; or by designing a CAR that recognizes multiple antigens, which can be achieved through a “tandem CAR”, where two combine Dots are attached to individual molecules to increase the efficiency of immune synapses. In addition, multiple CARs can be generated on the same immune cell by using a single vector called a “bicistronic CAR.”

Increase NK cell activity

Several immune checkpoints regulate and inhibit NK cell activity. These immune checkpoints act as “natural brakes” to prevent autoimmune diseases or immunopathological conditions caused by overactivation. Genetic deletion or blockade of these checkpoints can help CAR-NK cells remain overactive and shed cancer and metastases more quickly.

For example, a new NK-92 cell line was designed as a CAR targeting PD-L1, ER-retained IL-2, and high-affinity CD16, termed PD-L1-targeted haNK (t-haNK). Exciting preclinical data demonstrate that these cells have specific antitumor effects against 15 tumor cell lines in vitro and strong antitumor effects against triple-negative breast, bladder, and lung cancers in vivo.

Another important strategy to enhance the activity of CAR-NK cells is the regulation of tumor metabolism, but this strategy has not received the attention it deserves. Under hypoxic conditions, adenosine is produced by the metabolism of ATP by CD39 and CD73, which are involved in immune evasion, prevent NK cell trafficking to tumor sites, and prevent NK cell maturation. NKG2D-engineered CAR-NK cells showed good efficacy in the treatment of lung cancer after anti-CD73 antibody inhibition.

Overcoming the immunosuppressive microenvironment

Tumors have multiple immunosuppressive factors, such as TGF-β, IL-10, PD-1, or arginase. There are several ways to reduce the inhibitory effect of TGF-beta. For example, a combination of a TGF-β kinase inhibitor and NK cells was found to restore NK cell cytotoxicity and preserve NKG2D and CD16 expression. Furthermore, the use of hybrid CARs with extracellular TGF-β receptor domains has been found to be quite successful in enhancing the antitumor potential of NK-92 cells. The cytotoxic activity of NK cells has been enhanced by knocking out SMAD3 in solid tumors.

Improve security

Important approaches to improve the safety of CAR-NK cell-based therapy may include modifying the CAR structure by adding suicide genes or developing bispecific CAR molecules to better target tumor-specific antigens.

CAR-NK cells can target tumors in both CAR-dependent and CAR-independent ways; thus, this property of NK cells can be exploited to exert enhanced tumor killing and to develop non-signaling CARs. These non-signaling CARs lack direct killing signals, but can enhance NK cell-specific killing by promoting NK cell residency and adhesion on target cells. Another interesting strategy is to design CAR-NKs that can modulate the TME, which are named “armored” CAR-NK cells. These very specialized CAR-NK cells express several exogenous genes that can modulate the local TME to prevent any deleterious effects.

Improve accessibility

To overcome the accessibility of CAR-NK cells in solid tumors, several approaches can be used, including topical, intraperitoneal, and focused ultrasound-guided drug delivery. For example, pleural injections were found to be highly effective in orthotopic models mimicking human pleural malignancies, with functional durations even longer than those obtained with intravenous injections. Local administration of CAR immune cells may also help reduce therapeutic doses.


Overall, advances and advances in the field of NK cell immunobiology have laid the foundation for better and more novel immunotherapies, and the excellent antitumor effects of NK cells have made them the focus of cellular immunotherapy. CAR-NK cell therapy is a promising field of clinical research. Compared with CAR-T cells, CAR-NK cells have their own unique advantages, but still face some challenges. These challenges include cell persistence, overcoming the immunosuppressive microenvironment, and transduction efficiency. It is believed that solving these problems, based on the excellent anti-tumor lineage of NK cells, is very likely to bring new breakthroughs to tumor treatment under the arm of CAR modification.

1. CAR-NK Cells: From Natural Basis to Design for Kill. Front Immunol. 2021;12:707542