Good afternoon, everyone, and thank you for joining our second annual Cell Therapy virtual conference. My name is Emily Bodnar, and I'm an equity research analyst at H.C. Wainwright. I'm pleased to introduce Paul Hastings, President and Chief Executive Officer, and David Shook, Chief Medical Officer of Nkarta. David will start by going over some of the new translational data and background on the company's autoimmune space with their lead asset NKX-019, and then we'll have a few minutes for Q&A. So if anyone has any questions, feel free to submit through the Q&A/chat bar. And David, you could get started whenever you're ready.
Well, thank you, Emily, for the introduction and opportunity to present and discuss our platform today, where there's obviously been some considerable excitement related to cell therapy and autoimmune disease, particularly with some academic data on the effectiveness of CD19-directed CAR-T. And we feel that that success is driven not only by modality of cell therapy but also target. It's known that in autoimmune diseases like lupus and lupus nephritis, that that autoimmune disease is driven by autoantibody-producing cells, and particularly the plasmablasts. Plasmablasts notably express high levels of CD19 and low levels or no expression of CD20 or BAFF, which are the targets of some agents that have limited success in lupus and lupus nephritis. And we think that by targeting CD19, cell therapy allows for not only the elimination of those plasmablasts but the induction of an immunologic reset through a depletion of CD19 cells.
And obviously, CAR-T has been around for a long time now in the treatment of B-cell malignancies, and that experience is important to draw on, but that experience differs greatly from the experience in the autoimmune disease setting thus far, most notably in CAR-T persistence and B-cell suppression. Historically, with CAR-T and B-cell malignancies, there's been rapid expansion of B-cells in response to large amounts of CD19-targeted antigen, whereas in autoimmune disease, that expansion is more muted and results in a shorter term of persistence and not the long-lived persistence seen in B-cell malignancies. That equates to large differences in B-cell suppression, and in autoimmune disease, we see recovery of B-cells as soon as 45-50 days, and that varies greatly with the B-cell suppression in oncology, where the median is about 18 or 19 months.
What that importantly demonstrated was that there is not, in autoimmune disease, no link between cell persistence, B-cell suppression, and long-term durable response. Because while long-term B-cell suppression is necessary or correlative to a response in malignancy, as you can see in the Schett data of the follow-up presented at ASH this year, despite having a short-term B-cell suppression, some of these responses last in excess of two years. So when we sought to replicate these data, we first focused on B-cell killing. And what we showed earlier this year was that in our patients with Non-Hodgkin's lymphoma treated with NKX-019, we see rapid and deep B-cell suppression of both malignant and non-malignant B-cells by the completion of a single cycle.
In addition to our clinical evidence, we took NKX-019 and mixed it or tested it against the B-cells from patients with various autoimmune diseases, including systemic lupus erythematosus, scleroderma, myositis, and myasthenia gravis, some other diseases for which CD19 CAR-T is being considered or in active use, and showed a very effective targeting and killing of those B-cells from patients. So after we demonstrated that those cells could be killed, we then focused on what happens following B-cell reconstitution because, as I mentioned, that really seems to be the mechanism by which CD19 CAR-T has been effective. And the data here is looking at the sequencing by PCR of the B-cell heavy chain. And in the bottom left panel, you see data from the Schett group in the initial publication in Nature Medicine in 2022, where they showed, at baseline, patients have a mixture of mature and naive B-cells.
Taking back to that original slide I showed about B-cell ontogeny, and that after treatment, that reset occurs and you're left with a predominance of naive B-cells, which are unable to produce long-term autoantibody. Of those 8 patients, we had 5 patients who were in remission and we had adequate B-cell samples to test and showed a similar reset to a naive phenotype. One of those patients, patient 4, we had extended data on and showed that, similar to the Schett data, the patient went from a baseline mixture of mature and immature B-cells to a predominance of immature naive B-cells after treatment. Then, on par with what we would expect from normal B-cell development, including after allogeneic transplant, you see recovery of normal B-cell subsets at the 6, 9, and 12-point time check.
Looking at those same 5 patients using a different assay but same color scheme, we did a deeper single-cell RNA evaluation to make sure that that naive phenotype was there beyond just simple surface expression. What we showed is, by transcriptomic signature, that these B-cells indeed were all naive, including some purely naive B-cells and those that have intermediates, which are B-cells which are naive but beginning to express some of the more mature transcripts. This is certainly comparable to what was seen in the autologous CAR-T data. Moving on, now that we know that we could kill B-cells effectively and have them recover in a naive way, in our NHL study, we sought to make this autoimmune disease study beginning with Lupus Nephritis. What I'm showing here on the upper left-hand panel is the totality of our PK experience.
In NKX-019 for NHL, patients are treated with 3 doses of cells on days 0, 7, and 14 following fludarabine and cyclophosphamide conditioning. Importantly, in this study, we're moving to cyclophosphamide alone. We're doing that because fludarabine has considerable short-term and long-term risks, including cytopenias, infections, and secondary MDS or AML, as well as this medicine and dose of cyclophosphamide at 1 gram per meter squared is the same dose used by rheumatologists currently for the management of lupus nephritis and other autoimmune diseases. That allows our investigators to use a drug that they're comfortable with and not asking patients to take on a new risk.
In addition, we think there's a possible regulatory advantage of using the totality of historical data using 1 gram per meter squared and potentially leverage that in a single-arm or reduced-size registration study and not worry about contribution of component. So, as we said, we're changing conditioning regimens from flu cy to cy alone. I want to go through some data that gives us confidence that that's going to be an effective LD regimen. First and foremost, fludarabine immune suppression tends to be delayed. The peak effect is about 13 days after administration or about day 10 or so after cell infusion. By that time, most of PK exposure has occurred with NK cells. In T cells, as I showed you before with the Schett data, that occurs around day 10.
Here, it happens ex vivo in manufacturing, so we never see a higher peak than we do on day zero. And that allows us to front-load cells and move that PK curve to the left. And in the oncology study, we've moved from a 0, 7, and 14 model to a 0, 3, and 7 model dosing. But also, we've allowed investigators to omit Fludarabine when they have concerns about potential toxicity. Sometimes patients come with pre-existing cytopenias or infections, and Fludarabine, like it would do in Lupus Nephritis, poses that additional risk. So we're allowing physicians, if they feel it appropriate, to omit Fludarabine in those subsequent cycles. So sometimes a patient will get a if you respond, you can get additional cycles. So if a patient achieves a complete response, they can get an additional cycle to consolidate that response.
Or if they get a partial response, they can get a second cycle to try and deepen that response. What we're showing here is the first two patients that have undergone that approach of getting the first cycle with Fludarabine and Cyclophosphamide and the second cycle with Cyclophosphamide alone. What you'll see here is that there's comparable PK, including CMAX, of flu cy versus Cyclophosphamide alone in patient one and patient two. Unfortunately, in patient two, we missed some PK time points because of some confusion with the sites, but we do have the totality of the data with cycle two. We were encouraged and validated that our engineering and differential approach would allow Cyclophosphamide-only monotherapy in the Lupus Nephritis study, and these data give us more comfort with that.
So when we show these data and we discuss these data, and we often get a question, "If cyclophosphamide alone works, why do autologous CAR-T companies use or CAR-T products use fludarabine as well?" And the answer is that these are autologous cells, and despite no immune barrier, this immunosuppressant is necessary. And it's really because that immunosuppressive effect is not what's driving the need for fludarabine. It's a cytokine profile. Patients that receive fludarabine as a part of their condition get a massive increase in IL-15 peak in the blood. That IL-15 peak allows expansion of CAR-T, and that expansion drives response. So what happens is you get a direct correlation between the amount of cytokines, particularly IL-15, and the possibility of response.
Now, inherent to engineering of all Nkarta cells is a molecule called membrane-bound IL-15, where we've tethered the IL-15 molecule to the surface of the cells. It's introduced at the same time as the CAR through a bicistronic vector and allows cells to live in culture in the absence of immune pressure for weeks to months without any stimulation whatsoever. And importantly, in our clinical data, we've shown that our responses, unlike CAR-T, are independent of the IL-15 in serum. In addition, we think it's important to highlight that NK cells are actually superior to CD19 CAR-T cells at killing CD19-bearing targets. And that sensitivity is only increased as CD19 is downregulated. This is important on the oncology side when it comes to downregulation of CD19 for immune evasion or antigen escape, but also in B-cell autoimmunity because CD19 expression, as I shared, is variable and not binary.
So we think that this sensitivity allows us to get some of those cells that maybe express low levels of CD19, like the early naive B cells or even the long-lived plasma cells, to allow a deeper immune reset in autoimmune disease. This I'll highlight here, the B-cell malignancies. We've been able to drive remissions in B-cell malignancies, but these are far more difficult targets than autoimmune disease. And so we feel that part of the reason for the dramatic success in targeting CD19 is that normal B cells are not malignant B cells. They don't have antigen escape. They don't have the tumor microenvironment or infrastructure to concern about or prevent trafficking and infiltration. And the amount of disease burden is far lower. And that low antigen burden favors NK cells because T cells are necessarily a one-to-one therapy as you infuse a subtherapeutic dose.
I showed how T cells expand in response to the amount of antigen that's there. So a low number of cells, you'll get a responsive low number of T cells. Whereas in NK cells, if you have a low number of target cells, we're infusing the same 1 billion-1.5 billion NK cells regardless of antigen burden. So that will potentially increase our in vivo ET ratio. And we've seen that, and you saw it in our data, that sometimes actually subsequent cycles, when there's a lower burden, increases our PK exposure. Then finally, I want to cover NK cell trafficking. NK cells are an evolutionary link between the innate and the adaptive immune system and, for that reason, have the ability to traffic to any tissue in the body, including those immunologic privileged sites like the reproductive tract or the brain.
In addition, that trafficking only increases under inflammatory conditions like autoimmune diseases and allows NK cells to traffic easily between different tissues and organs. What we're showing here is our animal model that supported our IND and showed that NKX-019, like normal NK cells, can traffic to any tissue in the body. These are mouse data. In addition, we've shown previously and cataloged the extent of disease in our various B-cell malignancies and showed effective clearance of malignant B cells from not only blood, lymph nodes, and bone marrow, but also secondary lymphoid tissue like liver and spleen and target tissue like the kidneys. We think that those malignant B cells, despite, as I said, being a higher bar than normal B cells, offer an excellent proxy for the biodistribution of B cells. In summary, I'll just mention NK cells reach peak activity at infusion.
That allows maximum effect without in vivo expansion and facilitates our different LD approach using cy alone. They're cleared by host immunity, which means that long-lasting B-cell aplasia, as shown and like CAR-T, is not required for response. And there's no risk of T cell malignancies because these cells are rapidly cleared. And we feel that NK cells have a superior safety and accessibility advantage without having to deal with on-site infrastructure for manufacturing and administration and low risk of expansion-related toxicities like CRS and ICANS. So for that reason, we've started the study that I alluded to early, which is NKX-019 for autoimmune diseases, beginning with lupus nephritis with a single dose of cyclophosphamide as LD followed by NKX-019 on days 0, 7, and 14, looking obviously at safety and tolerability as well as PK, renal function, and the clearance of autoantibodies.
We've guided to first patient dosed here in the first half of this year and are actively evaluating the applicability in other autoimmune diseases. With that, I'll stop sharing and happy to cover any of that or additional questions. Thank you.
Great. Thanks so much, David, for the great update. Question on the peak expansion. So with CAR-T, it looks like that occurs around day 10, whereas, as you explained with CAR-NK therapy, that seems to occur right away. Would you expect that to translate to any differences in efficacy? I guess what role does that kind of play into potential effects?
Yeah. And we think the data so far, the clinical data so far shows that it's depth of initial depletion and maximum exposure. So we feel that that immediate activity should translate to a deep, meaningful reset without the need for in vivo expansion. Our ability to control the dose not only allows us to manage toxicity through the avoidance of those expansion-related toxicities, but it also allows us to intensify dose right away. And T cells, although they've been effective, depend on that in vivo expansion. So you dose it at a subtherapeutic dose and sort of allow the body to manage what that effective dose is. So we expect that dosing strategy to have superior safety and equivalent activity.
Are you confident that starting from cycle 1 with cyclophosphamide alone would have the same PK profile as how you showed it did in cycle 2?
We are. So we're looking for an immunologic window, sort of immune suppression window. We're not looking for a window of cytopenia. And in fact, we're trying to avoid it. And so yes, we think that the immune suppression that's afforded by high-dose cyclophosphamide, that same immune suppression that's led it to be a standard of care in lupus nephritis and other autoimmune diseases, creates that sort of flattening or a limited suppression window. And by doing that in cycle one, should allow patients to avoid that cytopenia. So yes, we're confident that similar PK exposure will occur in cycle one. And we think the activity is really driven by the drug. And that's why we think PK exposure is the best correlate.
Okay. That makes sense. How does the phase 1 study design relate to the Dr. Schett study in terms of endpoints and any specific things you're evaluating?
Sure. So there's always going to be some variability in inclusion and exclusion criteria and how the studies are executed. Certainly, most notably is the omission of fludarabine. It's the same dose of cyclophosphamide, but omission of fludarabine. In terms of patient population study, I think everybody's studying a very similar patient population. However, given the fact that the LD regimen of cyclophosphamide alone may allow an earlier entry point of patients, and the feedback we've heard from our investigators is that as they go to dose patients in the CAR-T cell studies, and we know that that enrollment has been challenging, part of that challenge is identifying patients for whom they think they've sort of failed all available therapies, including high-dose cyclophosphamide.
A message we have heard several times is if we're going to give flu cy followed by CAR-T, we want them to have seen cy alone first because that's kind of a component of that. Whereas in our study, we've heard, "Well, I wouldn't necessarily need to have seen cy alone because I'm not introducing really additional risk." So despite having similar enrollment criteria, we may see a slightly different patient population. We think that's a good thing that we don't feel like patients this has obviously been a revolutionary opportunity. Despite the sort of compelling rationale, people are not lining up, and centers are not enrolling as rapidly as one might expect given the compelling data. Part of that is just finding that right window and asking the patients to take on additional risk. So I know you asked about inclusion and exclusion criteria.
Just sometimes there's what's on paper versus what as somebody who still enrolls patients on clinical trials, you've got to make sure that the study is right for the patient, but also that the patient is right for the study. Paul, I don't know if you have anything to add to that?
I have nothing more to add to that.
Okay.
I think two of the potential, I guess, benefits of CAR-NK therapy, particularly on the allogeneic side, are potential to retreat patients and also potential for outpatient treatment. So how do you kind of think about those two benefits and if you're planning to evaluate those in your study?
Yeah, certainly. Yeah. We think those are great opportunities. And I think I'll start with the outpatient administration, the off-the-shelf product. In the oncology side, this is given fully as an outpatient or LD and cell infusion totally as an outpatient. And we think that for a therapy to become truly accessible to patients, that outpatient administration, you've got to get there. And so that safety profile of NK cells has really allowed that. And in terms of retreatment, we really think that's a real keen advantage here of NK cells and not as it has been posed as, "Well, you need to retreat with NK cells." And we've shown reliably every patient that has relapsed with NK cells in this 019 study, we've been able to put every one of them back into remission with retreatment. And we feel like these are patients in these autoimmune diseases.
They're in their 20s or early in life developing autoimmune disease. And these are not bone marrow transplants. There's a possibility patients will relapse even after CAR-T because the B cells are recovering, and these patients are not; they're retaining the ability to develop lupus. And should those patients relapse 5 years from now, 10 years from now, the idea of exposing them to fludarabine again, that risk becomes considerably higher of saying that serial exposure to genotoxic stress. So we think if that relapse were to occur, now, we have no reason to believe that NK cells will be any less effective than CAR-T. And we are looking for single-cycle CRs here.
However, if that, we think, inevitable relapse occurs in CAR-T or CAR-NK, that the ability to use single-agent Cyclophosphamide as is used right now, sort of monthly doses, that we think if you need to redose somebody a year or 2 years later, the ability to have an off-the-shelf product with significantly safer LD is a huge opportunity, and we think is a much better and more patient-centric profile than a custom-manufactured drug that requires a more toxic regimen.
That makes sense. And maybe as a last question to end, you mentioned you plan to dose the first patient the first half. Do you expect that you might have initial data sometime this year, and what might that look like?
I'll grab that one just as the last hurrah here. We have guided to FPI in the first half, and we're going to maintain that guidance. We feel we're well on the way to get there. Once we've dosed that first patient, to everything that Dave just spoke about, we also believe that the outpatient ability to administer these NK cells is going to have an advantage in our enrollment as well over the current autologous CAR-T therapies that are out there. We'll give an update on when we may see data after we've enrolled a patient or two through the mandatory FDA STAGR and cell therapy. We'll give that update as soon as we get that FPI in and have some experience with patients under our belt.
Okay. Great. Thank you so much, David and Paul. It was great speaking with you. And thanks, everyone, for tuning in. Have a great rest of your day.
Thank you.
Thanks, Emily.