Good afternoon, everyone, and thank you for joining me at the third day of the Annual Needham Healthcare Conference. It is my pleasure to have with me today, Ed Dulac, the Chief Financial Officer of Fate Therapeutics. As a reminder, any viewers who are watching through our conference portal are able to submit questions via the Ask a Question box underneath the video. Maybe a good place to start, just a bit of background, on Fate, the cell therapy company, but a slightly different one. So, Ed, if you can go ahead and provide a little bit of background.
Yeah, thanks for having us, Gil. So I think most people are likely familiar with cell therapy. It's been a concept that's been around for quite some time. As it is today, we often think about cell therapy in an autologous format, which is a fancy way of just saying it's a bespoke solution. We take a patient's T cells, we do some very simple engineering over a manufacturing process that's called roughly 30 days, and then we can infuse those T cells to cancer patients. Today, that has proven to be very effective, and it's done wonders clinically for patients that have certain types of lymphoma, leukemia, and multiple myeloma are sort of the three that come to mind. However, it's not without limitations. There are significant costs associated with that process, although it's one and done.
There's a significant manufacturing turnaround time, expense associated with that, and just some practical limitations in terms of scalability. So while these therapies have proven to be very efficacious, we have more of a supply problem today than a demand one. So the natural question of how can you bring cell therapy to more people? And a lot of folks have thought about an allogeneic approach or an off-the-shelf approach. Most companies, which is not Fate, will use a donor-derived pool of cells, which they can do engineering. That process is still fairly cumbersome. It's fairly expensive, and it results in a heterogeneous mixture of cells at the end of that process. There are numerous companies doing that in various stages of development. However, our approach is uniquely different. We use what we call induced pluripotent stem cell technology. So we can take a healthy donor cell.
We can go through what we call a dedifferentiation process, but push that cell back to a juvenile state. Then we can do as many edits as we'd like. We can do one, three, seven, 10. We do, though, go through a very careful and rigorous selection process. So we do engineering one time ever in the course of the product's history, but we then go through a single clonal selection process where we make sure these things are genetically and have a high degree of integrity. We do phenotypic screenings, and we select the cells that are not just perfectly edited, but also capable of killing a lot of different cell types preclinically. We then take that single clone and expand it a little bit to create a master cell bank. So that's the first important step of our process.
After we create that master cell bank, we never do engineering again. Now we can run a manufacturing process that takes roughly 30-40 days. In our case, we can create either T cells or NK cells that will fight a number of potential different cancer types and autoimmune disorders. We can do this at scale. We can do this at a cost point unrivaled in cell therapy. At the end of the day, what we're hoping to do is take the power of cell therapy and create a cell therapy product that can be scaled, that can be delivered cost-effectively, that can meet the patient demand, whether that's broadly in cancer or in autoimmune disorders.
So before we dive into individual discussion topics, just to give us a highlight of key events and catalysts that Fate expects this year in 2024.
I'd say there's a few. It'll be a fairly busy calendar year for us. We'll start in May. There's a conference called ASGCT, which is in the first part of May. I think that'll be the opportunity for us to share a little bit of an update on a program that we call FT819. This is a simple iPSC-derived cell therapy. It's a T cell. We have a CAR-T CD19, so it's more similar than different to a lot of the existing autologous or allogeneic approaches. And what we hope to show is an update on the oncology data that we've generated. We've dosed almost 40 patients across different types of lymphoma, leukemia, including CLL.
I think a lot of focus will be on the translational insights that we have from that phase I experiment as it relates to autoimmune disorders, which we can talk a little bit more about. That'll be an important update for us at ASGCT in May. Then I think most people are going to look towards the end of the year for three important clinical updates. The first one will be this product, FT819, our CAR T- cell, in what I'd call the first three to five patients in lupus. So I'm sure we'll talk about this in subsequent questions, but there's been an emergence of interest in autoimmune disorders, including in lupus. We have our own program there, and we should have the first few patients of experience by the end of this year. The second key program is one that we call FT522.
This is an NK cell. It has five different edits, including some very unique features that may allow us to move away from conditioning chemotherapy. But we have initiated and are already dosing patients in a phase I study in B-cell lymphoma. And that study has two arms. One is the traditional paradigm of let's give patients conditioning chemotherapy, standard Cy/Flu conditioning, followed by three doses of NK cells. And once we clear the first dose level, which we're starting at 300 million cells per dose, once we clear that first dose level, we open up or ungate the second arm, which is no Cy/Flu conditioning, no conditioning chemotherapy whatsoever, and just infusion of NK cells. So a lot of folks are looking at that as an early proof of concept at the end of the year.
Can we give a cell therapy without traditional chemotherapy conditioning? And does it yield results in this case in lymphoma? That will have implications for oncology, potentially autoimmune disorders as well. And then the last one, Gil, will be a program we call FT825. This is a 7-point edited T cell. It is in a collaboration with ONO Pharmaceutical, a collaboration that we've been benefiting from since 2018. And it targets HER2. It has a lot of different unique aspects of synthetic biology that may allow cell therapy to have an effect in solid tumor indications, which has historically had been very difficult. So again, another high-risk, high-reward opportunity. We should enroll the first patient here in the next couple of months. And the hope is we could share the initial data maybe at SITC in November.
That will be an important proof of concept for a more highly edited T cell and, you know, potential proof of concept in different solid tumors.
All right. I do want to spend a significant portion of our conversation today on autoimmunity. It sounds like you have more than one shot on goal here, as you walk through these programs. But maybe starting, at this point from some of the recent disclosures, it appears that Fate is prioritizing its iPSC-derived, you know, CD19-targeting CAR-Ts, FT819, for use in autoimmune diseases. So maybe you can provide some of the context as to the data as it evolved from lymphoma, which supported this kind of decision-making.
You know, by now, most people are familiar with the Nature Medicine paper back in September 2022, right? German investigator, five lupus patients. It sort of created this excitement.
It's up to eight. No longer.
There's a few more patients, and they've broadened out of lupus. But that's been the epicenter of what started this kind of excitement around cell therapy, particularly CAR T- cell therapy in potential autoimmune disorders. We, like others, have, you know, pursued that initial set data from September 2022. And I think, look, this whole field started with proof of concept in CD19 targeting CAR-T. So kind of first principles, we have a CD19 targeting CAR T- cell. Wouldn't it be great to see if we can get into lupus patients and have an effect? So as I look at what we've done in oncology, though, as I mentioned earlier at the outset, we have dosed about 40 patients. So we have a lot of clinical experiments with FT819. And while we haven't provided a broad update, again, we'll have a little bit more in May.
What we have seen to date is a typical CAR T- cell PK profile where we have a peak around day eight. We have persistence out to day 15, maybe a little bit longer in some patients. And so we do believe we have a CD19-directed CAR T- cell that's acting like a T cell therapy. So we've been encouraged by that profile. I would say, secondly, we have, like others, shown B- cell depletion. And I'm sure we'll talk about the effect of cells versus conditioning chemotherapy, but make no mistake about it, we have seen patients come on our study, either have very low levels, sometimes no levels of B cells, but varying levels of B cells at baseline that have been depleted through a combination of lymphoconditioning therapy as well as the administration of FT819.
I will put in a teaser right now because it'll maybe set up a question for later if you have one. But we have seen patients that have been unresponsive to side flu, interestingly, at baseline, giving conditioning chemotherapy. They don't respond well to side flu, but have had a dramatic reduction following FT819 infusion. So I do think there's value in not just conditioning chemotherapy, but certainly cell therapy that follows. And then I would say the third point that has gotten us at least interested in pursuing T cells is we've seen good tissue penetration. So I think it's one thing to see reduction of B cells in the plasma, in the blood. I think that's a relatively easy task. However, I do think you're going to have to get out of the blood and into certain tissues.
And we have translational data from our oncology program showing that we see FT819 in places like the bone marrow, certainly in lymph nodes. So I think there's reason to believe while our specific program may help replicate the early SLE experience that we referenced from the German group. And so we're going to explore that. But as you mentioned, we do have NK cells as well. We've been working with NK cells for quite some time. And I mentioned FT522, which we're also going to explore and probably file an IND in the middle of the year this year to potentially explore lupus or other autoimmune disorders.
A couple of things to unpack here. I do want to start with ASGCT just because it's top of mind and upcoming in May. I mean, the data is going to be in lymphoma. How should investors, you know, look at this data and kind of gauge, you know, potential translatability to autoimmune disease?
Yeah, it's super early days, right? I mean, drug development takes a long period of time. We've been working with, you know, this German data for a year and a half. What we've indicated publicly is that we cleared IND at the end of July of last year. We've been working diligently with investigators to understand the opportunities here. And we haven't disclosed publicly if we've dosed the first patient, but we feel very well equipped and prepared to do so in the next few weeks, couple of months. So that's kind of where we are with the program. As you think about the oncology data, look, autoimmune disorders is not oncology. And I think what we've learned from NK cells in particular is if you look at DLBCL, for example, tumor burden is very different, right? Much higher tumor burden on average, much higher proliferative disease.
And it takes probably a different type of cell, more likely a T cell, to attack that equation. But if you were to look at follicular lymphoma, still oncology, it's a slower moving disease, right? i.e., indolent. And we don't see often as high a tumor burden. So that E:T ratio that we talk about in terms of how many effector cells, whether that be T or NK versus target cells, is important. A T cell expands. We can, you know, infuse 300 million cells, for example, and that will expand to meet the needs of the demand in oncology. That may be important in autoimmune disorders, but it also may be less so. And that may also present an opportunity for NK cells where you don't have that expansion. But because you have what I'll call relatively bad acting, but normal B cell counts, it's an easier E:T ratio.
I do think there's safety and other considerations in autoimmune disorders that I think will be possible. As I look at the ASGCT data with our T cell program, I'll be looking for safety. You know, we have 819. What does the safety profile look like? We give less hospitalization than any other T cell alternative. It appears to be much safer. And then I alluded to some of the other things that we're looking for. What do we see in terms of B cells at baseline? What do we see after conditioning? What do we see with T- cell therapy? And, you know, what do we see in terms of plasma, beyond plasma cell penetration? Do we see these cells in different tissues?
I think that checks a number of boxes of things that should be important in autoimmune disorders, but we're going to have to run that experiment.
Right. And maybe stepping back a bit here and looking at this from the outside for a second. So we've had quite a few channel checks with experts. And it seems that when you're looking at standard lymphodepletion, this is likely to restrict the overall market, at least in SLE, to somewhere between 2,000-3,000 patients a year. We've talked about ways to address lymphodepletion, but just in that specific context of lymphodepletion, don't you feel there's still a bit too many companies going for a very small market?
Yeah. So I guess the way you've phrased the question, I'll say yes. If, you know, if cell therapy is going to have to be administered in autoimmune disorders like it is in cancer, I think the total adjustable market is smaller. You referenced 2,000-3,000 patients. I don't know what the number is exactly, but I would say a few things. One, we all need to start somewhere. I think the FDA has been very thoughtful about its approach. A lot of us are doing things that are very similar in terms of study design because of the limited data sets that's available. So if we're beginning and ending with side flu therapy in lupus, this is a much smaller opportunity than most investors and companies are expecting.
What I think will be the case and how we're thinking about this is that lupus with conditioning chemotherapy is a place to start, but at least two things have to happen. I think one, you have to get away from side flu. And that could be as simple as can you combine with something like cyclophosphamide, which is used as a standard treatment in various autoimmune disorders? That would be one option. So you're still giving some sort of conditioning therapy. You're still probably benefiting that cell therapy by the cytokines and the space that's created from giving conditioning chemotherapy. And then the other, I think, key tenet to this is how do you expand beyond lupus? And I think you've seen a number of companies and you'll see us do the same going outside of lupus into other B cell mediated disorders and potentially beyond that.
So I think people are going to start with lupus. I think they'll look to expand more broadly. And I also think they're going to look to expand outside of lupus. I would say generally, though, whether it's our platform and products or others, a winning profile is going to be different, right? Your average autoimmune patient or lupus patient is female. They're typically between 15 and 45. They have lives. They may be of reproductive age. I mean, they have very different considerations than an oncology patient who is significantly older with far fewer treatment options and far, far fewer life events to sort of consider. So I do think safety is going to matter. And I do think getting outside of the traditional academic centers where we're all starting is going to be super important.
These patients are going to be treated in the community more often than not. And so I don't know how to think about the total addressable market today. It will be as large as a therapy's ability to scale, to do it cost-effectively, and to do it in a way that's safe and treats these patients like autoimmune patients, not oncology patients.
I think we're in total agreement about, you know, the requirement of removing lymphodepletion. But, and this is, you know, something I wanted to allude to, and this is specific to allogeneic therapeutics, not just you. Allo rejection is a real issue. We've seen this in lymphoma very strongly. Cells are usually rejected before they expand. Unless you engineer a bunch of bells and whistles in order to help them avoid the host immune system. Isn't this a major concern when thinking about autoimmunity, especially in the context of no lymphodepletion?
Yeah, it could be. It's interesting because I don't know that our data suggests we see lots of rejection. I'm not sure we see any rejection, to be perfectly honest. I think the other way to look at this, Gil, is the other lens to think about is potency. And like us and like many, if not all, allogeneic players, there is a disruption to the T cell receptor, right, in the donor-derived cells. I do think there's a potency impact on that. And so one of the reasons you're seeing some companies choose to go higher conditioning is you just have less oomph, right? You're not using the natural signaling of a T cell through the disruption of the editing process, through the manufacturing of what these companies, including what we have done. So I don't know that it's all rejection. I do think that lymphoconditioning does create space.
It does create cytokines. So there's less competition for those cytokines. It does help adaptively transfer cells, kind of wake up and perform and do what they need to do. You mentioned an important point, and it's where I think one of our advantages as a platform using iPSCs is that we can engineer in a lot of features and functionality that may allow us to get away from conditioning, right? I mentioned cytokines a couple of times. You know our programs very well, at least on the NK cell side and eventually on the T cell side. We do have cytokine support, IL-15 in NK cells, IL-7 as we did with FT825 in a T cell. So we do build in that cytokine support. That exogenous source is less important.
As we may talk about, with at least our FT522 program, it's our first program in the clinic, at least, that we have what we call an ADR technology. You can think about this as another CAR. It targets 4-1BB. And it really allows a couple of things to happen. 4-1BB is upregulated on T cells, NK, other immune cells in a patient's immune system. So if we're not giving conditioning chemotherapy and there's activity from the host immune system, we can engage with those cells and eliminate those alloreactive T cells or NK cells, for example. It does, though, signal to our NK cell. It's almost a potentiation signal at the same time. So we're incorporating elements of synthetic biology that we believe will help offset or compensate the lack of or reduce lymphoconditioning.
So we will get to ADRs a little later, but there is one thing I've always maybe struggled with. I think there's this fundamental difference between T cells and NK cells as regards to their expansion and pharmacokinetics, to be honest. So because T cells require a host to expand in, that makes the lymphodepletion pretty important for an expanding cell population. NK cells being innate, are killers. You could probably flood the entire system with them. They don't really expand very much to begin with. So I think that's a fundamental difference. But like I said, we will talk about the ADR a little further down. And this is a question you guys have been asked before. So why not throw in an ADR in a T cell? And I think the answer is basically that. You probably still need the chemo anyways.
Yeah. I mean, it's interesting because I do think you'll see us incorporate ADR into a T cell. Whether it's that simple or not, I don't know. You know, I would say I want to be careful because as much as a home run could exist in the form of removing conditioning chemotherapy altogether, you have seen us and you know we're big believers in also just looking to get away from side flu and combine with other existing standards, right? So bendamustine is the one that we talk about most. KYMRIAH uses it. It's in the label.
I don't think there's anything magical about Cy/Flu, but I do think it's important to understand the treatment paradigms of different diseases and think about can we add a cell therapy on top of some existing standard and thereby benefit from some of the things that Cy/Flu does also can be done maybe by other chemotherapeutic agents or other modalities. So I think we're thinking about it both from the perspective of, hey, wouldn't it be great to just remove chemotherapy altogether? When you say the word chemotherapy, no one has a great reaction to it. And maybe we can do that and we're going to try for sure. But I also think it's equally valuable to combine with these existing standards like bendamustine and we have a lot of experience doing so.
And you can see from our data, I think there's data from Penn that show that it may actually be a little bit safer, but you still get the same cytokine support. You still create the host and space for T cells to do what they need to do. So I think there's different ways to attack it. I think the point being is I'm not sure there's anything magical about side flu. And I think we would all benefit in the industry to move away from that and try to meet patients the way they are. To the extent we can do that safely, it also allows us to maybe get out of these academic centers and get more into the community where most patients are treated.
Yep, totally agree. I do want to go back to something you mentioned earlier. What do you think is the contribution from the conditioning to the treatment effect in AID? And you mentioned you had a couple of patients who weren't even responsive to their B-cell lymphoma, right? Their B cells, their mutated B cells seem to be extra hardy. This is a common question that comes up a lot, right? It's in a healthy person versus someone with a lymphoma, but how much of a role is lymphodepleting even playing?
Yeah, it's hard to tell. I would say in lymphoma, at least, as you look at different data sets, most of these lymphoma patients come in with really low B-cell counts, right? And so when you administer side flu as a chemotherapy agent, look, it should kill aggressively proliferating cells, right? It should have an effect. So whether or not, though, that is sufficient to now in autoimmune disorders create this immune reset that we're all hoping is where we're all going to end up in this space. I don't think that's going to be the case. And I say that knowing that, as we mentioned earlier, cyclophosphamide is given in fairly relapsed refractory patients with lupus as a good example. It's infused monthly for up to six cycles. It has an effect, but something like the complete renal remission and CR rate is pretty low.
You look at the different data, it's maybe 20%-35%. So you're not getting a meaningful long-term durable effect in the vast majority of patients. And so why might that be the case? Well, it could be what we talked about earlier. You could reach the cells that are hiding in plain sight, right, in the plasma and elsewhere. Whether or not you're getting to some of these key secondary tertiary tissues, I think, is a critical question. rituximab, even as a non-chemotherapy agent, is used often in these patients. It depletes B cells really well in the plasma, but you don't get this long-term durable effect. You don't get this immune reset. And I think part of the hypothesis is you're not getting to some of these key tissues and eliminating the disease to enable that reset to occur. So there seems to be something else required.
I have no doubt lymphoconditioning is doing something, but in the vast majority of patients, it's not going to be doing enough. I do think there's going to be substantial benefit of incorporating a cell therapy, whether that be a T or NK cell, to get into those more difficult to reach tissues, make sure you have a deeper, more durable effect. The hope is that then allows the patient to become drug-free or certainly have a prolonged impact that doesn't require as much immunosuppressive therapy in the future.
Okay. So maybe spending a moment here on FT522. I mean, you've added the ADR technology and still, you know, for the average investor, when you say the word NK cell and lymphoma, they go, "Oh, that just doesn't work." Why is it going to be different this time?
So a few things. Look, NK cells, as you mentioned, are not T cells. So when you infuse 300 million cells or 1 billion cells, that's about it, right? You're only going to get a decline from there. So it's a totally different paradigm versus 300 million cells as a T cell. That's a much larger in vivo amount of T cells, right? And so as I think about the world of NK cells, we've had to do lots of exploration, right? We have a long history of NK cells, including discussions with the FDA having to go very conservative because it's a cell and it incorporates a CAR. We've had to do things like start at single doses, low doses. Those are never going to be effective.
Now, as we've gotten to higher doses in certain programs, we've seen dose-dependent PK, and I think we've seen some activity both in our data and in others. But I would say a few things. You know, when you're working in diseases like AML, really tough disease. DLBCL, really tough disease. For some of the reasons we talked about before, the amount of cells, the E:T ratio matters a lot. You're going to have to give billions and billions of NK cells potentially to have a therapeutic effect. So I do think disease matters. As I alluded to earlier, though, and if you look at a program like FT596, which is kind of the predecessor to 522 that we may talk a little bit more about, it's got, you know, three or four different features and functionality built into it.
And I think if you looked at our data in two different buckets, one being aggressive lymphoma, we definitely haven't met the mark, right? We've had low response rates, low CR, not a tremendously durable effect. However, if you look at more indolent disease like follicular lymphoma, we have seen profound response rates, including profound CR rates that have had long, durable effect. Again, that is a slower moving disease. And I think probably is a closer cousin, if you will, to autoimmune disorders where you don't have these super physiological levels of B cells. You have bad-acting B cells oftentimes, but they're not in crazy numbers. So I do think the E:T ratio for an NK cell in something like follicular autoimmune disorders is different. And I don't think you're going to need expansion to slow disease, eliminate disease, and have a profound effect.
I would also say, you know, NK cells have really short half-life. So we're big believers in multiple dosing. You'll see us do at least three doses like we have done historically. And the good news is with our program in 522, at least in oncology, we'd be able to start at 3 times 300 million cells per dose. So I think we're already at a dose level that makes sense. And to the extent that we could extrapolate that into autoimmunity, I think it gives us a much better starting point and a much more effective E:T ratio in something like lupus. So I wouldn't count out NK cells. I also do believe, and I think we've seen this in many data sets, the safety profile has been pretty exquisite, right? And so we've seen treatment effect, but without really any toxicity to speak of.
That profile plays pretty well outside of an academic center. So whether or not safety and efficacy in that balance is T cell versus a T cell equivalent, I do think NK cells are much safer to the extent you can have a pretty profound treatment and get into the community center. I think that could be a winning profile. But as I said earlier, we'll file an IND in the middle of the year for 522 and look to explore that end of this year and beyond.
I would wager to say that, again, the safety profile is directly associated with the pharmacokinetics of that drug, right? The lack of expansion of NK cells makes for a very different profile and tox. As to, and this is kind of a far-out question there, right? NK cells and FT522 probably has a pretty benign safety profile. Why not do a healthy volunteer study? No lymphodepletion. See what happens to their B cells.
Yeah. I mean, we've done preclinical work here. I mean, the fact that we have 522, a cleared IND, dosed a few patients, I mean, we're very close to having this proof of concept play out in oncology. And I think we'll have valuable human data, translational data from that program by the end of the year. So I just think, given where we are and how soon we'll be in possession of clinical data with 522, albeit lymphoma, I mean, it's probably our best guess to understand what that program is, our ability to move away from Cy/Flu. And I think our history, including programs like 516 and 596, give us really good benchmarks to be able to quickly ascertain whether or not moving away from conditioning is going to be a reality.
Okay. I do want to spend a moment on the oncology programs, maybe on the ONO program, just to remind us where it sits. And I mean, I'm assuming mostly looking at tumors with HER2 expression.
Yeah. So FT825, we filed and cleared the IND in 30 days in the fourth quarter of last year. We're in sort of a, just as a reminder, we're in a 50/50 Co-Co with ONO outside of Asia, so in Europe and across the U.S. And study is open. We have a couple of sites and we're screening patients and hope to enroll our first patient very soon. There is a unique aspect to many unique aspects to FT825. So it is a 7-point edited cell. It does have a CAR with a unique binder that ONO contributed against HER2. And so not surprisingly, we're going into various different tumor types with varying degrees of HER2 expression, including very low levels of HER2 expression.
The binder, which we'll hopefully talk about later this year, is novel and should allow us to have preferential activity against HER2 expression on tumors only while preserving any sort of normal HER2 expression on normal healthy tissue. But more to be said about that. We do have, though, incorporated into FT825 the CD16, the high-affinity non-cleavable CD16. So in addition to HER2, you can envision us doing something like a cetuximab combination in various different diseases to go after that. So it is HER2. It's a crowded space, but we're looking to find certain niches. But really, this is a seven-point edited cell. I think by my calculus, it's probably the most advanced, most highly edited cell therapy in human testing.
It's important, obviously, to ascertain the safety of that profile and begin to see if we have any early responses, whether it's HER2 or EGFR or other different targets that we can pursue with that program.
Yeah. The idea of having a multi-antigenic approach using a combination with antibodies has always been pretty appealing, especially in the solid tumor space where probably targeting a single antigen would be insufficient.
Yeah. And I'll just, yeah, no, I agree. And you know very well, in solid tumors, there's a lot more to be asked, right? Trafficking is important. We've included CXCR2 as a chemokine receptor there. The tumor suppressive microenvironment, we have a TGF-beta redirector to kind of help stimulate the cell while also warding off that tumor suppressive environment. We mentioned CD16. So we've got a variety of features that, at least on paper, on a slide, look really good as you think about the reason to believe in a multi-point edited cell. So it feels like a fair fight. Now, whether we have the right constellation, that's why we're running the phase l, and we'll see what the early data look like at the end of the year.
Great. Maybe a last one. This mostly relates kind of to the company's moat as it comes to your technology. Actually, creating some of these master clones isn't trivial at all. Differentiation, de-differentiation, it's actually also pretty expensive. How much of a moat does that provide, Fate, with respect to maybe others?
Quite a bit. I'm glad you brought this up. I mean, we've got IP as you would expect us to have. We have a lot of know-how. I'm pretty confident I could give you the formula of how to do an iPSC, and it would take a long time to figure. I'm sure you have better things to do. But the point being, to have your hands on this technology for the better part of seven or so years that we've been working on this is not trivial. And the process of generating a clone, and I think it's important because, look, when we edit iPSCs, I don't care if you edit donor cells, primary CAR T- cells from a patient, you're going to end up with a lot of edits in all the wrong places.
So our ability to go through and quantify, qualify the edits that we've done, go through that process, it is time-consuming. It's fairly expensive, but once it's done, it is done. And we take great pains to make sure we don't have a loss of master cell banks. We diversify the risk of where we keep master cell banks. So the beauty of the master cell bank is once it's created, it's done for the lifecycle of the product. We get a lot of leverage out of the master cell bank. So we could talk a little bit more about that, but the know-how and the IP is substantial.
One of the reasons we started with NK cells is, quite frankly, it was easier to produce an NK cell off an IPSC than a T cell, in part because of the T- cell receptor that we had to knock out. That's actually pretty important to differentiate. So we had to figure out a workaround, which we have done. So it's very hard for a company to come into this space with two or three years of experience and do this well and compete effectively.
Maybe touching on a last challenge there. Over many rounds of replication, there's an effect of genetic drift. How does the company deal with potential risk of your product having changes in it naturally over time?
Yeah. So again, a lot of experience here. We only have a certain number of passages to go from the single clone that we select to see the master cell bank. We're very thoughtful and very careful and very limited on how many passages we have. Once we create that master cell bank, though, we then will go to a stage called a CD34 intermediate, which takes about 10 days. There'll be a little bit of expansion there as well, but not much. And we actually get a lot of leverage off that. So we can bank a lot of those CD34 intermediates for later use. We don't always have to go back to the master cell bank. And so we can carry forward a smaller subset of those CD34s to create an NK or a T cell, particularly for phase l clinical development.
We don't need thousands and thousands of doses. So we create a lot of leverage there. But we go through painstaking detail at that master cell bank to genetically understand where the engineering is, phenotypic screening that we do. There's a lot of proprietary know-how in that process as well to make sure that we don't have that genetic drift consideration. So we feel really confident about our starting material is homogeneous. It's got the edits where they're supposed to be. We've gone through painstaking detail to get there. So it allows us to start with this very predictable, very consistent starting material. And at the back end, we do, again, a lot of phenotypic screening assays to make sure that the final product is, I wouldn't call it clonal, but very homogeneous relative to other alternative cell therapy approaches.
All right. Excellent, Ed. Again, it's always a pleasure speaking with you. Thank you for attending today.
Thanks, Gil. My pleasure. Thanks for having us.