Sana Biotechnology, Inc. (SANA) Earnings Call Transcript & Summary

All right. Good afternoon from the 40th Annual JPMorgan Healthcare Conference. My name is Cory Kasimov. I'm the senior large-cap biotech analyst. And it's my pleasure to introduce Sana Biotechnology and CEO, Steve Harr. Please note that following this presentation, we will move right into a Q&A session. [Operator Instructions] So with that, Steve, thanks for joining us today. Let me turn things over to you for an update.

Thank you, Cory, and thank you, everybody, for joining us today. We'll be making some forward-looking statements today, as you might imagine. So please do take a look at our 10-Q and 10-K. We spent a lot of time on the risk factors, so hope you'll find that useful. So Sana was founded with the belief that one of, if not the most important transformation that will occur in medicine over the next several decades, is the ability to modulate genes and use cells as medicines. And our goal is to build one of the leading and sustainable companies of that era. And if you take a step back, pretty much every disease and certainly the clinical manifestations of each disease are really caused by damage to or dysfunction to a cell. And our goal, our aspirations are relatively straightforward. One, we would like to be able to fix any cell that's damaged by either modulating the gene or gene expression; two, is we would like to be able to build cells from scratch to replace cells that are too far damaged or actually missing; and three, we want to do this all in a way that meaningfully expands access for patients. The company started out by trying to understand what are the fundamental challenges to reaching that aspiration, those aspirations. And really go after what I'd say are the most tractable challenges to getting there. And A few years into this, I'd say a few things. One, the strategy that we chose to go after is at least as relevant today as it was when we founded the company. The technologies we brought in are working. That's a ways from saying that they work. And we're really excited about where we are. We've built a pipeline that we think can begin to generate meaningful INDs beginning as early as this year. And we have the capability to go forward now, we think, with 2 to 3 per year. We're building expertise across really important areas: delivery of genetic material with virus-like particles and other things that we'll talk about; gene modification and gene editing; immunology, using the immune system both as a weapon and, importantly, hiding cells from immune rejection; disease in stem cell biology and the manufacturing sciences. And we have the balance sheet to really see this through with $866 million in the bank as of the end of the third quarter 2021. So you often hear companies talk about being a cell therapy company or a gene therapy company. And at the end of the day, what you're trying to do is engineer a cell. And we're relatively agnostic as to whether we do that inside the body or out. And importantly, by going after both of these areas, we've been able to build the core capabilities, which are highly overlapping in a competency and scale that we otherwise wouldn't have been able to. And the risk though in the near-term are actually relatively idiosyncratic. So as we've approached in vivo cell gene or people called cell therapy -- or sorry, gene therapy, basically what you're trying to do is deliver a payload to a cell that modulates the gene or gene expression. And you can, more or less, do anything you want to sell in a petri dish. And the real challenge has been delivery, which is what we went after. And our goal is relatively simple, not to be able to deliver any payload to any cell in a specific and repeatable way. And every time we do 1 of those 4 things, we actually open up a whole host of medicines for patients. And the company was founded really on specific technology that allowed us to do cell-specific delivery with any payload using things like virus-like particles or VLPs. As we made progress there, we also have been targeting the ex vivo cell engineering. And this is the side of the page where we get to go after the diseases that most of our loved ones will suffer from. And it's harder to be really clear. Because to make a medicine, if you distill it down to 4 things, you have to do all 4. And we want to be able to manufacture cells at scale that can graft, function and persist. And the field has been fraught with challenges on each of these autologous cells, who are complicated to scale, but allogeneic cells are using somebody else's cells, have a real problem with immune rejection. From the outset of the company, we felt that if we were able to overcome this immune rejection and transplant allogeneic cells and utilize stem cells as medicines, we would have a significant impact on the industry. So we really started the company on 2 core technologies. One was an ability for cell-specific delivery, and the other was a technology that we believe allows us to hide cells from immune rejection. And from that, we've been building up our capabilities and pipeline. So it's translated into, I think, a pretty robust set of programs moving forward. And a couple of things to know. These are just some of the assets in our pipeline. First and foremost, you see that we're pretty balanced between the in vivo and ex vivo cell engineering. Second is we're building meaningful capabilities and pipeline in oncology. And importantly, we're also looking above and beyond that. And third is, hopefully, you'll begin to see some of these enter with INDs this year. So how do we go about building this, and we'll talk about this in more detail going forward. The hypoimmune technology we talked about, since the beginning of transplant medicine, a real challenge has been that when you transplant my organ or my cells into you, your immune system will reject them. And the way the field has tried to grapple with this today has been heavy immunosuppression. But that's really limited the field. And as people began to understand the power or potential of stem cells to differentiate into different medicines, they realize that unless you could overcome allogeneic medicine, you really -- rejection, you wouldn't really get there. And when some of the founders of the field went to their immunology colleagues, they got this very simple set of advice, and that is to go figure out the paradox of pregnancy. And the paradox of pregnancy is that each of us is half mom and half dad. And the only reason we're on this Zoom call together today as our mothers didn't reject us as foreign. So they really set about trying to understand what was different about that maternal-fetal border. And you have to overcome 2 parts of the immune system. One is the adaptive immune system of B and T cells. And many of us figured out a long time ago how to grapple with that. You disrupt Class I and Class II MHC through gene editing or other mechanisms. Unfortunately, cancers and viruses [indiscernible] a long time ago, and we evolved something called the innate immune system, including things like natural killer cells to go after those cells, and they actually get killed very quickly, if you don't go after the -- and target the innate immune system. And what we found was that overexpression of CD47 is able to turn off both the natural killer cells and macrophages of the innate immune system. And this system is something that if we can get to work we want to put into IPS cells or pluripotent stem cells that we can then differentiate in all kinds of different cell types and transplant, if they're missing. So how are we doing? So first off, we got the system to work in a petri dish or in vitro, which is always important. But biologic systems are more complex. So the first place we went was in the mouse models. And we've done this both with the mouse cells with a mouse immune system as well as human cells with a humanized immune system. And what you see on the left-hand side of the page is that our gene-edited hypoimmune cells are able to avoid, evade being seen by T cells and antibodies and that's the top set of graphs. And on the bottom, what you see is that when you don't modify the cells and you put in for -- or allogeneic cells, the T cells and B cells recognize them and reject them. Through the right-hand side of the page what you see is that we're also able to prevent our cells from getting killed by these natural killer cells. So natural killer cells don't recognize on wild type of regular cells, unfortunately, the B and T cells will. If you just disrupt or knock out Class I and Class II, you can see that natural killer cells rapidly kill these cells. That's at middle panel on the right-hand side. And if you look at the far right, which I'm circling now, which is well the cells that have the gene modifications of our hypoimmune platform, you can see that they have evaded natural killer cells. And over here on the left, you see that they're not seen by antibodies and T cells. It's one thing to do things in mice. It's another thing to do them in nonhuman primates, which is the closest we'll get to the human primate. And sometimes a picture is worth a thousand words. And this is an experiment where we took 8 nonhuman primates and we injected our cells into the right leg at time 0 and follow them over time. And at 6 weeks in the left leg, we put in the other type of cells. If you got -- if the animal got hip cells or hyperimmune cells first in the right leg, they get unmodified cells in left leg and vice versa. And the goal was both to see what happens de novo and also when you have an ongoing immune response. What you see is for the duration of this study, in the right leg and the top panel that our cells lived and thrived, again a picture being worth it as you see those -- the color is metabolic activity. And it did that despite the fact that at week 6, in the left leg, we put in unmodified cells which were rapidly killed. So that gave us some confidence that "Hey, maybe our cells will live in the context of an ongoing immune response." In the bottom panel, what you see is if you inject unmodified cells, they're rapidly killed within a couple of weeks. And we can later put our hypoimmune cells in and then they will continue to thrive and not be seen, meaning that even if an animal has a preexisting immune response, our cells will live. So how -- what did that look like in terms of B and T cells? What you see -- this is what you see here. And to make a long story short, basically, B and T cells don't see our hypo immune cells, and they do react to normal allogeneic cell. And importantly, on the right-hand side, what you see is not only do not -- killer cells not see ourselves, but we also can have a built-in safety switch. And that is if we make an antibody to CD47, those cells will be killed. And that both proves our mechanism as well as provides maybe a bailout if something were to go wrong. This is the first and only time that we're aware of that anybody has shown an ability to transplant allogeneic cells in nonhuman primates with no immunosuppression. So the next step is really just to go and take this into humans. So where are we going? And we're going to start with 2 cell types: allogeneic T cells; and pancreatic islet cells. The allogeneic T cell, I believe both of these cell types are cells where the people have already transplanted cells, and that gives us an idea of how to go about this engraftment and persistent challenge. It's also an area where with the allogeneic T cells, we think we have a lot of the immune systems in our favor. The patient has cancer. They have to get conditioning chemotherapy, which will dampen the immune system for a period. And at the end of the day, we only need the cells to live for, I don't know, 3 to 6 months, something like that, which you'll get into. Pancreatic islets are kind of the opposite challenge where patients not only have an intact immune system, but they have an autoimmune response that is already killing the cells with type 1 diabetes. And that's where those monkey data are so important. Additionally, we need these cells to live for years and years for a patient to have the clinical benefit we want to see. So -- and then we have some programs on the right that are a bit further out. So why allogeneic T cells? And first and foremost, sometimes people say, there are a whole bunch of CAR T cell companies already out there. What do you guys offer? I will start with the fact that -- in the U.S. and Europe alone, approximately 100,000 patients a year die of lymphoma, leukemia, multiple myeloma. I think we all would acknowledge that the CAR T cells have had a tremendous positive impact. But in the history of humanity, fewer than 10,000 patients who have been treated with these therapies. And of those, many don't respond and a number of them don't -- who do respond, relapse. And my goal has been to turn lymphoma and leukemia and myeloma into Hepatitis C, diseases where the vast majority of patients are cured with a single therapy. So there are manufacturing and access challenges, there are safety challenges. And importantly, there are reasons why patients don't respond. Either the cell type isn't good enough. The cell goes away, probably due to the immune response and we think we can deal with those with our allogeneic platform. Or they lose CD19 in the case of B-cell malignancies we can get into in a minute. So with our allogeneic CAR T cell, there are a lot of different approaches out there when they're going to make an allogeneic T cell, you're going to put my cells into you as an example. You have to grapple with a few things. First off, my cells will try to kill you. That's called graft-versus-host disease. And the first thing that we will do in that is just -- that's been shown how to deal with you knock out the TCR [indiscernible] side. The second is you have to grapple with host-versus-graft disease or your immune system trying to kill the therapy. And here, we think we have a very unique system in our ability to hide cells from the immune system that is quite different from what's available from anybody. I think one of the things that the field has shown is that cellular persistence or how long cells live in your body is really important to making these cells work. So what we'll do is take a donor T cell, gene engineer it and create a hypogenic, hypo-immunologenic (sic) [ immunogenic ] CAR T cell. We've shown that we can hide these cells from the immune system in nonhuman primates. We have data that show that we can make CAR T cells, which are hidden from the immune system. So the next question for us is as we make all these gene changes, does it impact how well our cells function? And if -- this is a really important slide for us. And what it shows is that on the left-hand side, what it shows that the answer is no, that our CAR T cells function really well. The left-hand side is your typical mouse model of leukemia. And you can see by day 23, they're all red, meaning that they're taken over by the cancer. What you see is a regular CAR T cell in the middle, which again seems to take care of the cancer at least through these 28 days. And what you see is with our HIP modified cells comparable efficacy. And we see the same thing in a whole host of assays. So we think that with our HIP platform, we can really expand access we can make a higher quality consistent cell, meaning fewer side effects, and we can evade the immune system, which means that we will allow ourselves to last longer. And we think that, in it of itself, can create a tremendously valuable medicine. Beyond that, though, a number of patients lose CD19 or the target of these cells. And over the course of the last year, some -- a group at Stanford has shown that if you utilize a CD22 CAR T cell, this is an autologous CAR T cell. And you can really go after the CD19 failures. This is 24 patients, 23 were Yescarta failures. One had a primary cancer that was CD19 -- it was missing in CD19. And what you see is 56% of these patients have a durable complete response, and it's quite tolerable. This morning, we announced that we actually licensed the CD22 construct to put into our allogeneic CAR-T cell. And so that kind of puts together our strategy for you. We do expect -- we hope to file our first IND as early as this year targeting CD19. I think we can build the best-in-class CD19. We think CD19 and CD22 together give us a tremendous opportunity to increase both the number of complete responses and the durability of those complete responses in a way that we believe can have a tremendously positive impact for patients. The CD22 can also be utilized for CD19 failures. And we also announced overnight, we brought in a validated BCMA CAR that we want to utilize going forward to treat myeloma, which we get into. And finally, you'll see us go for another area. So we think this is a really exciting program or a platform, I should say, with a number of exciting programs, and you'll see a good bit going forward. But it's not where we stop. And if we are able to hide cells from the immune system, we think that an area we can have just a tremendous impact is type 1 diabetes. So human insulin has had a huge impact for patients. But despite all of the monitoring, the best insulins, a type 1 diabetic still has about a 15-year short expected life than a nondiabetic. And during that time, they have a host of challenges related to low blood sugar as well as complications of high blood sugar like blindness and kidney failure and stroke and -- this is a large market. It's around 4 million people just in the U.S. and Europe alone. And if we happen to get type 1 diabetes, right, we think we can go after type 2. That being said, there is a lot of evidence that suggests that you can take a beta islet cell whether it's from a cadaver or now we've even seen evidence from a stem cell. And you can make it into a beta cell, you can transplant it into a patient, heavily immunosuppressed that patient and their glucose will normalize. However, there aren't that many people for whom lifelong immunosuppression is safer than lifelong insulin. So our goal is to produce a hypo immune beta cell that can evade the immune system without immunosuppression in normalized blood glucose. So to do that, you have to make -- do 4 things. You have to make really good beta cells from iPS cells. You have to hide them from allogeneic rejection. You have to hide them from autoimmune rejection. Type 1 diabetes is a disease where the immune system kills the beta cell. And then you have to really turn this into a medicine. So where are we? This is an experiment where we show that we can make great beta cells. And on the left-hand side, what you see is that we have a glucose dependence secretion of insulin that is very similar to what you have from a normal beta cell. And as far as we're aware from the published literature, this is as good as it gets. And what you see on the right-hand side is that you can put this into an animal model of diabetes, and it will cure that animal of its diabetes. So that's an important first step. The second is you need to be able to hide it from allogeneic rejection. And you saw our nonhuman primate data earlier. So we're excited about that. And we wanted to show that we could also do that in the context of making beta cells. And so what we did here is we made hypoimmune or normal beta cells and transplant them again into an animal model of diabetes. On the left-hand side, what you see is if you put an allogeneic unmodified islet cells, they're killed by the immune system and the animals diabetes goes crazy. And what you see on the right-hand side is that our gene-modified cells both live and they function really normalizing glucose in these animals. So that's the allogeneic rejection. Now I have to do with autoimmune rejection. There is no mouse model or monkey model where you actually create an immune response to beta cells. We have to go about this a bit differently. So we took the blood from many patients with type 1 diabetes and really tested to see what would happen when they're exposed to our cells. On the left-hand side, what you see are T cells, and if you put T cells from a normal volunteer and expose them to either an unmodified or genetically modified HIP cells or islet cells, you see they don't kill, that's probably you and me. At least I don't have type 1 diabetes. When you see if you take it from a type 1 -- these T cells from a type 1 diabetic is they rapidly kill unmodified islet cells. And in the bottom, what you see is that they don't, in any way, affect our islet cells. So that's a real step forward. So next question is these animals -- these people also have antibodies and do the antibodies recognize and kill these islet cells? And what you see on the left-hand side are unmodified islet cells, they will bind to and recognize them. And then with our HIP cells, these antibodies, again, don't see them. So we've now shown that we can make great cells, we can evade allogeneic rejection. We can evade autoimmune rejections. So now we need to make a GMP supply chain. It's more complicated and takes more time than you might think. So we have to first make a GMP, genomically stable cell bank. And we've announced license with FUJI or FCDI as well as made some of our own cell lines. The second is you needed to modify the genome and we announced a license with Beam in the fourth quarter to go about our gene editing. And now where we are is we are making a GMP gene edit and master cell bank. All of this is complicated, it takes time. We remain right on track of what we talked about. And our goal is to file our IND here next year. It won't be this year, but next year. So I'm going to just switch tax for a minute and go to our in vivo platform. And as I mentioned earlier, the real challenge in the field to date has been delivery. Lipid nanoparticles, or LNPs, are good for getting things into the liver, but you also can't deliver genes. You can only deliver RNA or proteins. AAV allows you to do small DNA sequences, but pretty limited to where it goes and it's complicated to manufacture, and the payload size is quite small. And so what we wanted to do is create a better targeted delivery system. And again, when you're faced with a complex biologic problem, one of the things we like to do is see if mother nature's already solved it and if she has to exploit that system. And the left-hand picture is actually just coronavirus. We see it all the time. And there's little red spikes or the spike protein that we've learned so much about. That's actually something called the viral fusogen. And what it allows this virus to do is only target cells that express the H2 receptor. There are a host of different viral fusogens that target specific cell types. And we even use this as mammals for cellular communication and for fundamental processes like the delivery of genetic payload from sperm into an egg. The sperm does not stop anywhere along the way to deliver its payload, but instead goes specifically to the ovary, again, utilizing a fusogen. So in the middle is kind of a schematic of a viral fusogen. And it's -- think of it as nature's logic gate. There's a G protein or guide and there's an F protein infusion. The guide gives what gives cell specificity and the fusion, the F is what drives the fusion. What we do is we modify that G protein so that we can target basically any cell type we want. And what you see in the right-hand picture is we put those fusogens onto a virus like particle or VLP onto a virus. And that's something called a -- what we call a fusosome. The fusosome allows us to -- for cell specificity, so it binds to that target cell. It then -- what you see here is it will fuse the cell membranes in step 2, and we then have delivery of the genetic payload in step 3 where we can go into the nucleus and modify the genome. That's basically in a schematic how this system works. And we've shown that we can get this to work across a host of different cell types and a host of different genetic materials. So where are we going to apply it? We're starting with T cells, which I'll show you in a second, we're going to go to liver cells and HSCs as well in the not-too-distant future. So what's the idea of going after T cells. Our goal is in vivo creation of CAR T cells or inside the body. So the left-hand side is currently what happens with the manufacturing CAR T cells outside the body. And it's a complicated, expensive and time-consuming process. On the right-hand side, you see our goal, which is in a single doctor visit to be able to infuse our medicine without any conditioning chemotherapy and create the CAR T cells from that 1 visit. And the way this works is -- the orange is our [fusosomar] medicine, and it carries our payload and it will recognize the T cell and bind to it, we see that here. It then delivers a payload. We call it that minus sign here, a genetic payload, which creates the CAR, which is or chimeric antigen receptor. That's the yellow Y. And that CAR will recognize cancer cells. And when it sees cancer cells, it will kill it, the cancer cell, as well as amplify creating more CAR T cells. So does this system work? First of all, we've shown we can do this with targeting CD8 cells, We've shown we can do targeting CD4 cells. We've shown we can do this targeting CD3 more broadly. We can do this in a specific way. And this is a mouse model showing that it works. So this is that same mouse model leukemia that we showed up above. And on the left-hand side, what you see is what happens, if you make typical ex vivo CAR T cells. Again, no different than before. The sailing, the leukemia takes over the animal. With an ex vivo CAR T cell, it works very well over the course of a month to control the tumor. What you see on the right-hand panel is that with a single injection into the tail vein of the mouse of our medicine, we will create CAR T cells and we will eliminate the tumor. We've also shown that we can do this safely and effectively in nonhuman primates. So our next step is we're in the process of our manufacturing and GLP tox, and our goal is to file our first IND this year to move forward with these in vivo generation of CAR T cells. So where does that leave us? So we started the company over here on the left-hand side, trying to address the obstacles that we thought were most important and most tractable and really create engineered cells as medicines. We're moving forward with our first set of human studies. And we think that they will do 2 things: one, they can create really important medicines; and two, they will validate platforms. If we're able to hide our hypoimmune allo CAR T cells from allogeneic rejection. We think we'll have a technology that we can apply confidently across a host of much more prevalent and larger diseases. Similarly, if we're able to show with our fusosome platform in CAR T cells that we can deliver a genetic payload safely, efficiently and effectively to a specific cell in the body. We will be able to apply that, not only to make great medicines for cancer patients, but across a whole host of other payloads. So we're excited about where we are and really looking forward to the future. And with that, Cory, I'll turn it over to you for questions.

Terrific. Thanks, Steve. [Operator Instructions] We have about 10 minutes or so for Q&A. So I guess, Steve, just to start, I mean with the first IND you're expecting for CD19 later this year, what are the key gating factors to getting that done at this point? And how much risk is associated with it versus more execution, blocking and tackling?

I would say with the hypoimmune platform, right, it is probably more execution, blocking and tackling. We are -- we need to do the manufacturing scale up. And it is actually [indiscernible] scale upside. We do the manufacturing transfer and create the drug and get it into human testing. I think with the fusosome, this isn't the first time people are trying -- anybody who we're aware of is trying to deliver this genetic payload in vivo. And I think there's always some risk on 2 fronts. One, we could run into a safety challenge as we go through the GLP tox. We have dosed animals and for -- in the past, and we don't have any reason to think we will other than science is unpredictable. And two, we have to ensure that we get through the manufacturing scale-up and the manufacturing runs. The long-winded way of saying the hypoimmune platform is really pretty straightforward. And I'd say both preclinically and early testing is probably lower risk. And the fusosome having never been done before probably entails a bit more risk.

Okay. Makes sense. And then when you think about potentially developing better cell therapies than either what's out there today or what's in development, how does this ultimately get born out in the clinic? Like where do you expect to be better? Is it with the durability? Is it the overall activity? How do you think about this once you are in the clinic with your initial products?

Well, so I'll start with allogeneic T cells and move beyond that. So allogenic T-cells, really straightforward, I think data have been really clear to date that cells have to live -- have to persist for some period of time in order to have durable complete responses. So what we would expect is if we're able to hide ourselves from the immune system, we will have longer cell persistence, and that will lead to higher and more durable complete response rates. I think if you add on to that CD22, and you're now targeting CD19 and CD22, we think we can take an entire another step forward. If you then go into -- and by the way, the other big clinical differentiator is today, if you want it -- one of the ways that people deal with your immune system trying to kill my cells, right, is they just suppress the heck out of your immune system, and that comes with its own safety risks, right? So what we want to do is get away from that immunosuppression. And when you go beyond oncology, really that immunosuppression becomes intolerable, right? So as I said, there aren't that many patients with type 1 diabetes for whom lifelong high-dose immunosuppression is safer and preferable to human insulin. But there are a whole -- the entire field, if you didn't have to immunosuppress them, would like to get rid of the complexity of human insulin as a therapy. And so there, you just dramatically change the arm of the possible by not needing to use immunosuppression.

Okay. So I guess piggybacking off that last comment. I'm curious what you make of -- it's 1 patient, but the initial data that came out of Vertex in terms of starting to prove the concept of what you're doing, albeit with immunosuppression.

First of all, it's great data. I've got to be thrilled for that patient. And it's a real step forward for the field, right? To be able to take a stem cell and make it into a human islet that has glucose-dependent insulin secretion at a rate that looks to be a normalizing function for that patient, it's both a great outcome for Vertex and for that patient. And I have to say, it's not trivial to make a great cell at scale. And that's something that they've shown they can do. That being said, what it does is it validates that this is possible. And if you can add that to our hypoimmune platform and do this in a way that does not require immunosuppression. And you go from a disease that can treat a small number of patients, and I don't know what that number is. I'll let others decide that to one that is potentially viable for every patient. With type 1 diabetes and potentially viable for many with type 2 diabetes. You are now getting glucose sensitive in cell secretion, right? That's what we all want.

Right, right, right. You spent some time talking about the work you're doing in terms of the GMP supply chain. I just wanted to ask kind of a manufacturing overall of the progress that Sana is making here as well as unique challenges that you may or may not face in this area that might be different from other cell therapies? Or is it really pretty similar to others in the industry?

Yes. So I mean, within -- so I'll start by saying that we started this company, and we thought we would overinvest in manufacturing sciences and manufacturing capabilities. And it felt that way for a few years. As soon as you start getting a whole host of clinical data -- or I should say, preclinical data and trying to make medicines, you wish you had more right? It's almost impossible to overinvest right now in cell and gene therapy and process an analytical development. Second is our strategy around where we would manufacture was really clear. We would use -- we've built our own pilot plan. We're building our own plant for late-stage clinical and commercial, and we would use CMOs for Phase I. And I think one of the challenges of that is that these supply chains are tight and CMOs make their money by having more or less 100% capacity utilization. So getting slots, particularly at the time that you want them is not always easy. So far -- so that's a risk that we still have going forward. The third thing that I would say is that if you go -- if you look at our different programs, they have different levels of novelty to your last question, how different they are from others. So making a gene edited allogenic CAR T cell, there are a number of companies doing that, right? We're doing it in a different way. But I don't know, if there's anything from a supply chain perspective that is unique and hugely complicated. Making gene-edited iPS cells that we then turn into type 1 -- I should say, pancreatic islet cells or heart cells or clear progenitor cells, that's complicated. So the number of GMP, iPS cell banks in the world that meet our immunologic criteria are complicated or not many, I should say. The second thing you have to do is show that they're genomically stable. And that is an endeavor that I think the field is still trying to understand how to even do. And I think we've made progress. But to make a gene edited genomically stable iPS cell bank, one there's no-- no one's done it yet. So there isn't even a manufacturing facility, it's easy to do it at. There aren't people that are trained to do it. And then you have to figure out how you're going to measure this genomic stability after gene editing. So all of those things take time and they require a lot of resources internally. But if we get it right, we think we have a meaningful competitive [indiscernible]. That's where some of the novelty is. In terms of manufacturing fusosomes, I mean, it's a technology that others don't have. It's complicated to scale. It's -- so it's something where we're still working through getting this to a commercial scale. We think we've got it to a scale where we can move it into human testing, but we have work to do to get it to the scale that we want for the long run.

Okay. A question from the portal here is where is the genome locus do you have your knockout in place your CAR-T construct?

We haven't disclosed that.

Okay, I didn't think so. And then I'm just curious, and we're running low on time here, but when you think about your initial data, so assuming the CD19 program gets the clinic first, how much does that initial program derisk the overall platform from a hypoimmune standpoint? And kind of what you're trying to do? How much more -- I mean you've shown data in mice, you've shown data in nonhuman primates, when you get that initial data in humans, is that going to be like sort of your aha moment? Or is it going to have to be somewhere outside of CD19 as well?

Yes. So I would say, I believe it will be. I think if you see these cells and they hide from the human immune system, and you've seen this now in mice and monkeys, you can really get a lot more confidence as technology works. Some of our scientists would say, "Hey, it's winning this." Every situation is going to be different, and we're going to need to ensure that we look at beta cells and different cells as we go. But I think if you see these cells live for a long period of time, you have an aha moment. And I also think if you see with the fusosome platform that we can deliver the genetic payload specifically to T cells in vivo, right, that is an aha moment for the field of gene delivery. And regardless of what the cancer efficacy is -- I mean that's -- our goal is to make a great drug. But if it turns out that you have to -- provide cytokine support as an example, and will still be a really valuable step forward for the field in terms of delivery. So I do think that, that first set of data, while our goal is to make really important medicines with both of these drugs. At the same time, they provide really significant insights around both of our platforms.

Okay. And we get probably 30 seconds left. So a quick answer here from the portal. How long do modified islet cells last in mice. The chart showed 30 days, but have you seen them go longer?

They've lasted as long as we measure them.

Okay. Okay. That's good to know.

They were, by the way, the modified cells in nonhuman primates. So those mice and nonhuman primates, they last as long as we can watch them.

Okay. Terrific. Well, listen, this is fascinating stuff on which we can sit here and talk about it longer, but we are out of time. So Steve, thanks so much as always for your time today, and best of luck with all the progress.

Thank you, Cory. Great to see you. Enjoy the rest of your week, and thank you, everybody, for joining us.