Weighing the Promise and Trade-offs of CAR NK Cell Therapy for Cancer

Natural killer (NK) cell therapies, including chimeric antigen receptor (CAR)–engineered NK cells, are emerging as a potential off-the-shelf alternative to autologous CAR T-cell therapy in hematologic malignancies. One example is FT596, an induced pluripotent stem cell (iPSC)–derived, CD19-directed CAR NK cell therapy. In a first-in-human phase 1 trial (NCT04245722), 10 of 86 patients with relapsed/refractory B-cell lymphoma (BCL) who received FT596 experienced grade 1 or 2 cytokine release syndrome (CRS), and no neurotoxicity or graft-vs-host disease was reported.

At the Big Ten Cancer Research Consortium (Big Ten CRC) Summit, CancerNetwork® spoke with Veronika Bachanova, MD, PhD, about the biology, safety, and clinical potential of NK cell therapies. Bachanova is a professor of medicine in the Division of Hematology, Oncology and Transplantation at the University of Minnesota and co-leader of the Transplant and Cellular Therapy Program at the Masonic Cancer Center.

Bachanova began by explaining how the CAR-dependent and CAR-independent killing mechanisms of NK cells set them apart from CAR T cells. She then discussed the trade-off between their favorable safety profile and limited persistence, along with the rationale for using NK cell therapy after progression on CAR T-cell therapy. Next, she identified myeloid malignancies as the area where NK cells may have the greatest near-term impact. She concluded by urging colleagues to stay committed to a field whose complex biology has slowed its progress.

CancerNetwork: Given the CAR-dependent and CAR-independent mechanisms of tumor killing for NK cells, how significant is that dual mechanism in practice, and does it translate into different response patterns compared with CAR T-cell therapy?

Bachanova: That is a critical question for understanding the difference between how NK cells and T cells work as antitumor effectors. As a field, our thinking on the mechanisms of NK cell killing has evolved. Initially, most of the adoptive NK cell trials used allogeneic NK cells because autologous NK cells, particularly those that are unmanipulated, are inhibited by the patient’s own HLA [human leukocyte antigens]. That makes autologous NK cells essentially inactive in terms of antitumor effect.

That is the main reason the field pivoted to allogeneic NK cells from the very beginning. Allogeneic NK cells, usually from related donors, have a higher frequency of cells that are educated and not inhibited, so they are alloreactive against the patient’s tumor cells.

This is based on the expression of inhibitory and activating receptors on NK cells, and their ligands on the patient’s cells, typically HLA molecules, which either inhibit or engage NK cell signaling. There is complicated science around this, but the bottom line is that, unlike CAR T cells or T cells, NK cells are typically taken from an allogeneic donor. That is one big difference. No. 2, allogeneic NK cells are more likely to be alloreactive, to detect the tumor, and to eliminate it.

No. 3, allogeneic NK cells are more amenable to genetic manipulation, particularly if they are derived from cord blood or from iPSC lines. Those manipulations make the NK cells more targeted, and that is the CAR NK cell. These cells are particularly interesting, and the manipulation can be very comprehensive and quite advanced. There can be up to 8 manipulations to make CAR NK cells specific for 1, 2, or 3 targets, or to help them engage better with antibodies.

Some NK cell products express a noncleavable, high-affinity CD16, so any combination with an antibody makes them better effectors. They can also be made less visible to the patient’s immune system, which reduces their detection and removal.

The advantage of NK cells is that they can be active through multiple mechanisms, combining the CAR with autonomous NK cell function. That autonomous function is based on the expression of activating ligands, which develop during malignant transformation and are already present on cancer cells. It is a complicated field, and a lot is still evolving.

What trade-offs have you seen in durability of response when considering the safety profiles of CAR NK cell therapies?

The safety profile is one of the biggest advantages of CAR NK cells, and that is exciting for the field because NK cells do not seem to trigger significant cytokine release syndrome and neurotoxicity. The incidence of these immune toxicities is low, or really absent, for several of the products. That is a huge advantage of NK cells.

The trade-off is that they may not be persistent. We are still learning whether the immediate presence of NK cells for maybe 2 to 3 weeks is enough to deliver an antitumor response. The use of NK cells could also be different from the use of CAR T cells. As opposed to “one-and-done,” we are learning that NK cells may need repeated infusions, either several days in a row or over a couple of cycles, to really expand their persistence and induce durable and deeper remissions.

In terms of the durability of responses, there is precedent from phase 1 and phase 2 studies that some responses are durable. That is perhaps a reflection of the depth of response while the cells are there. They are not persistent long term; this is not a living drug. NK cells are not likely to be detectable after several months of treatment, but they are certainly very effective and active soon after infusion to elicit that antitumor response.

For patients with lymphoma who have already progressed on CAR T-cell therapy, is there a rationale for sequencing to a CAR NK or allogeneic NK cell approach?

There are no approved NK cell products; everything is experimental, and clinical trials are ongoing. That said, most trials allow prior CAR T-cell therapy in their eligibility criteria.

I am encouraged by some of the data already published, particularly with iPSC-derived NK cells. Investigators have used these products in phase 1 and phase 2 studies in patients previously treated with CAR T-cell therapy, and in a number of those patients, we saw responses. I think the response rate in the study by Ghobadi et al with an iPSC-derived NK cell product was [54%].¹ Responses can certainly be observed, and some of them will be durable.

The mechanism of action is probably different from that of CAR T cells, which is why sequencing makes clinical sense. It is certainly an opportunity. The NK cell field is rapidly evolving, and combinations with antibodies are another opportunity for the field to expand, particularly in multiple myeloma. But there is a level of speculation in my responses here.

Looking at the full landscape of cellular therapy in hematologic malignancies, where do you see NK cell approaches having the most near-term impact: in lymphoma, myeloma, or elsewhere?

The development of agents in lymphoma and multiple myeloma has been accelerated tremendously by CAR T cells and bispecific antibodies. One of the areas where hematologic malignancies are really struggling is myeloid malignancies, particularly [acute myeloid leukemia (AML)] and [myelodysplastic syndromes (MDS)]. Those are diseases where NK cell killing is particularly interesting and appealing. I still hope that one of the first areas where NK cells can make strides is in myeloid malignancies because NK cell cytotoxicity is powerful in AML.

Studies are still ongoing. I think the sequencing of NK cells with chemotherapy may point to using them in patients with residual disease, which may be more amenable to NK cell killing, as opposed to patients with refractory disease.

The other area of interest is in combination or sequencing with allogeneic stem cell transplantation. We still use allogeneic transplantation as the most effective curative therapy for myeloid malignancies. But I think there are biological opportunities to combine it with cellular therapy in a way that enhances the graft-vs-leukemia effect, and NK cells in particular could maximize it.

What do you hope your colleagues take away from this conversation?

I hope that people will look at NK cells as an ongoing, exciting frontier. The biology of NK cells is so complicated that it turns off a lot of people because it is not straightforward to understand, and that is exactly why the pace of progress has had ups and downs. It requires persistence and real focus to get where we need to go.

Those of us who work with NK cells have no doubt that they are incredibly powerful at killing cancer cells. How to translate that into the clinic is not straightforward. I am part of the NK cell community where we continue to be dedicated and committed to bringing a clinical product to patients because the biology clearly shows that the potential exists.

References

Ghobadi A, Bachanova V, Patel K, et al. Induced pluripotent stem-cell-derived CD19-directed chimeric antigen receptor natural killer cells in B-cell lymphoma: a phase 1, first-in-human trial. Lancet. 2025;405(10473):127-136. doi:10.1016/S0140-6736(24)02462-0.