CRISPR Therapeutics AG (CRSP) Earnings Call Transcript & Summary

Good morning. I'm Salveen Richter, biotechnology analyst at Goldman Sachs, and we're pleased to have CRISPR Therapeutics here with us. And we have Sam Kulkarni, CEO of the company. Sam, thanks for joining us and -- virtually.

And to start here at the EHA meeting this Friday, we expect additional data from 2 ongoing Phase I/II studies of CTX011 in transfusion-dependent thalassemia and sickle cell disease. Could you remind us of what we expect to see, including the number of patients, and just kind of put this in context of the data that we've seen to date?

Yes. Happy to, and thanks for having us here at this conference. We're about 5 years in -- after the start of the company. And as we sit here today in 2020, we're quite pleased with where we stand. We have more than 5 programs in the clinic. For our lead program, CTX001, we disclosed data last year in November, which was the first reported data for a CRISPR-based therapy in humans. And the data we showed was that for the first thalassemia and the first sickle cell patient that were treated with CTX001, they were -- one of them being a severe thalassemia patient, another being a severe sickle patient, they both had very encouraging data where they were -- one of the thalassemia patients was transfusion-independent. It's more than 6 months after being treated with CTX001, and the sickle cell patient had very high fetal hemoglobin levels more than 4 months after being treated. What we wanted to do was provide additional information on those 2 patients as we've had a longer follow-up and also include another patient, another thalassemia patient that's been treated with sufficient follow-up in our presentation at EHA. And I think while the data we disclosed in November were very exciting in terms of providing a potential cure, what we wanted to make sure was, is there enough durability? Do you see a sort of a steady state being reached? Do you see consistency of effect? And those are all questions we're trying to attempt to answer at the EHA presentation this Friday. So we're quite excited. And it is -- it will be the -- since the November data presentation was done as a company presentation, this will be the first scientific presentation of our data from our CTX001 trials in thalassemia and sickle cell. Obviously, we'll have more data coming this year, towards the end of this year with a larger cohort of patients. But at this time, we'll -- we hope to provide several updates as we move this program further along.

Great. And then we're -- you also have an allogeneic CAR-T program for CD19 positive B-cell malignancies with data expected by year-end '20. So that would be first data that we're seeing from this program. Is the time line still intact here? And can you just expand on what we might see and where the status of enrollment is with this program?

Yes. Yes. I think your question is with regards to our allogeneic CAR-T franchise. Essentially, as you know, we think about the business, the CRISPR Therapeutics business in 3 parts. There is the hemoglobinopathies franchise, which I just talked about, where we have the presentation coming up at EHA. We have the second part of the franchise, which is immuno-oncology, and we have a number of CAR-Ts in the clinic. We have 3 different CAR-Ts, one targeted towards CD19, one towards BCMA and one towards CD70 that are all in clinical trials right now. And then obviously, we have a third part of the business, which is on the remaining part of the pipeline in regen med and in vivo, which we'll talk about. But for our immuno-oncology franchise, we're quite bullish about the possibilities here. We have the allogeneic CAR-T platform, which can be very sophisticated and best-in-class because we're starting with the CRISPR platform where you can make several different edits to these CAR-Ts. And over time, these CAR-Ts are only going to get better. We showed at a recent poster, the ASGCT, that we can make 10 edits in these CAR-Ts without -- while ensuring viability of the cells and making sure they're robust. So this is just the beginning. We're scratching the surface of CAR-Ts and how potent they can be from an allogeneic standpoint. So we are excited for our lead program in the allogeneic CAR-T space, CTX110, targeted towards CD19 in terms of the progress on the trial and recruitment. We do hope to have data by the end of this year for the initial cohort of patients that are dosed. To remind you, what we're doing is a dose escalation study. It's a Phase I/II study where we're looking at safety of these CAR-Ts to make sure we can get them and give them to patients at high doses without causing any GvHD or other safety issues. The second part of it is we're doing dose escalation to see what the -- how -- what the different CAR-T doses do in terms of response rates and efficacy to identify what the dose expansion cohort is going to be like and what those we would pick for that expansion phase. And then we're also trying to optimize certain things. There's different degrees of freedom that you have with allogeneic CAR-Ts. As you know, you have lymphodepletion, and you can titrate that. You have the CAR-T doses itself. Eventually, we're going to see multidosing come into play and titration of what is the right appropriate multidosing schedule, but that's for a later part of the trial. And all those things will be parameters that we hope to study in this initial Phase I/II, and we'll have a data set for the initial cohort treated at the end of this year.

And what are your initial thoughts here on the allogeneic approach post the early data that you've seen from the field, including Allogene and Precision here? And how does that impact your perspective or even your trial design?

Yes. I think there's been -- as I talk to various pharma companies and others, there has been this hanging question in the last 2 years of whether -- what's better, autologous or allogeneic, right? And to me, the early data from other players in allogeneic CAR-Ts is a big signal that the time, at least in heme malignancies, allo is going to dominate autologous therapies. I think if you get in the same territory of efficacy of autologous therapies with an allogeneic construct, I don't know what the benefit of doing autologous therapies will be. And so -- and by the way, this is just a start. I think over time, you can optimize allogeneic so much more than autologous therapies because all the manufacturing is easy, it's off the shelf. And it works in a different way that the limit of improvement is a lot higher for allogeneic relative to the limit of improvement that you have for autologous. And so at least in CD19, I would say that the days of allo are just starting, and you're going to see allogeneic CAR-Ts be the main mode by which cell therapies are utilized in heme malignancies over the next 3, 4, 5 years. In solid tumors, it's still an open question, but I think, again, we're quite excited about CD70 and getting into solid tumors. And even in solid tumors, I think autologous, while there's excitement about some of the autologous therapies out there right now, ultimately, things are going to move to allogeneic once you show that these can be safely dosed. Beyond that, I think the other thing I would say -- learn, there's a tendency out there to use the same yardstick that was used for autologous for allogeneic cell therapies. And I think what we're seeing -- starting to see is that allogeneic cell therapies work differently than autologous therapies. There was a pervasive notion that these CAR-Ts need to stick around more than 6 months for you to get a durable CR, for example. And maybe that's true for autologous, but for allogeneic, it may be very different. If you come in with healthy donor CAR-Ts that are robust and you go and hit the tumor hard in the first place, you may not need the CAR-Ts to stick around that long. I mean you need a minimum level of persistence, whether it's 1 month or 2, but you don't -- you may not need 6 to 9 months. So there's a different lens by which you need to evaluate allogeneic CAR-Ts, and that realization is also starting to happen in the field. So I'm quite encouraged by data that's already put out there and quite excited about the data we're going to put out there by the end of this year. And I think we're just starting to see the beginning of the phase of belief in allogeneic. I think there was -- like a typical curve, there was initial excitement and then some disenchantment, and now we're coming back to the valley into the true belief cycle of allogeneic CAR-Ts.

And then alongside the ex vivo gene editing platform that you have, an allogeneic platform for oncology so far, you also have in vivo gene editing and a regenerative medicine platform. When you look across these 4 verticals, what do you think will have greatest impact on your future here? And what other directions are you thinking about, if any?

Yes. So that's the third part of our franchise, the rest of the pipeline. We spend a lot of time and effort and resources there. The rest of the pipeline includes a very promising candidate in regen med where we're actually taking pluripotent stem cells and making them into pancreatic islet cells. And we put that into device after gene editing those cells, and we implant them as artificial pancreas. That can be a potential cure for type 1 diabetes and potentially even type 2 diabetes. So that's something we're working together with our partner, ViaCyte, and are quite excited about. On our in vivo approach is we -- it's more like gene therapy versus cell therapy. We package the CRISPR/Cas9 in -- either in AAVs or lipid nanoparticles and target some rare diseases. We have a partnership with Vertex on a couple of programs, including DMD and DM1, where we're seeing good progress and good data towards the clinic and other programs as well, which utilize lipid nanoparticle delivery approach. But if I -- to answer your question, well, how I see this all coming, I think what we want to do, if you -- if I take a 10-year view of the company, we want to create significant growth catalysts every 18 to 24 months so that we can continue on a steep growth trajectory as we look to become the next $100 billion company. And what you have right now, you have this sort of baseline with our hemoglobinopathies franchise that provides us with a lot of baseline threshold value with significant value you can -- that we can build up from an enterprise value perspective just by executing on hemoglobinopathies. Beyond that, immuno-oncology, at this point, the data we're going to have end of this year and next year with our other pipeline programs are going to give us that sort of growth potential in liquid tumors. And then you have solid tumor data and solid tumors recalling 80% of the cancer market is solid tumors. And so any sort of dent in that space gives a significant potential. And then beyond that, we have data in diabetes, which can be significant in terms of impact on a large patient population. So we have sort of these growth drivers with reasonable bets over the next 4 to 5 years. And we're continuing to innovate beyond that, our pipeline, as we expand the applicability of the CRISPR platform not just in cell therapies but on other regenerative medicine fields and also other in vivo approaches.

And maybe just expanding on that last part. So we've seen just in the press and through news coming out of other academic centers, just various other new gene editing tool sets. We've seen prime editing and so forth. How do you think about incorporating technologies as you see the evolution of this field?

Yes. Now for us, I think we welcome all these improvements, and we would apply them if they are necessary for any indication we're going after. So far, we haven't seen anything where we said we have to use base editing or prime editing to create a superior approach because of classical CRISPR/Cas9 from strep pyogenes works pretty well, especially in ex vivo settings. It's derisked as a platform. So we're sticking with that for the programs we named so far. But in the future, we would apply whatever improvements there are. And the analogy I have is with antibodies, if you think about cloning and molecular recombinant protein production in the late '70s, early '80s, then you had the early antibody production, which were not humanized, but then you had technologies coming from Cambridge antibodies, technologies in Medarex or humanization of the antibodies that happened in the late '80s. And then in the '90s, you figure out how to make fully human antibodies. And those all were benefits that accrued to Genentech, even though Genentech was the pioneer and the original for recombinant biology. They, both from an IT perspective and a technology access perspective, benefited in creating great drugs out of all those improvements. So you don't have to be the one actually making that improvement, you can actually source it and be a smart buyer of those technologies at the right time and apply it. And that's what we would look to do. There's no sort of syndrome of not invented here, or sort of sticking with what we have, we would just use the best technology for whatever applications you would go after where gene editing makes sense.

So maybe focusing in on the hemoglobinopathy program here. So when we look at the data or per the EHA abstract here, you showed us follow-up data in 2 patients in transfusion-dependent thalassemia. The patient remains transfusion-independent at 12 months. And in sickle cell, the patient's [ POVs ] are, crux of crisis, at 6 months here. Can you describe what this means clinically and from a quality of life perspective for these patients? And then secondly, given the limited number of patients here, how do you get confidence about reproducing these results in a larger subset? And how are you managing for variability?

Yes. I think with these patients, it's a dramatic impact. I can talk about at least one of the patients pretty openly because that actually was a subject of an NPR article is a woman named Victoria Gray. She was a mother of 3 and was 33 years old and had significant number of hospitalizations every year before coming into our study. And after treatment with CTX001, had to stay in a hospital for about a month while the cells are engrafted. But after that, has been free of any VOCs or hospitalizations to the point at which we reported the data. And that's a dramatic difference in how she leads her life. And it's not just the hospitalizations where you -- hospitalizations are deadly because you walk in every time thinking, gosh, there's a chance I may die this time around. And that goes away. But you also -- there is notion of daily pain you have to deal with. All the sickle cell patients deal with pain on a chronic basis. There's organ damage that's happening. And all that leads to a very poor quality of life, and all that essentially is gone for this patient till the time of observation. And that just portends what the potential for this therapy is. It's a onetime therapy that can be a cure. Similarly, for the thalassemia patient, it was a 19-year-old woman who had a terrible lifestyle because would need more than one transfusion a month, didn't have access to the best quality of care, was -- knew that there is organ damage happening, had anemia, and all that disappeared after treatment with CTX001. So the impact is dramatic. Now the question is -- rightly, so the question that you have to ask is, is it durable and is it consistent? And the durability of something we would follow all these patients for a long time, and you'll see some data at EHA next week and more data later this year. But you want to follow them for a long time to make sure there's no drop-off, and this is a permanent effect. And the consistency aspect is something that we're quite confident in mainly because we're using CRISPR, and CRISPR editing always leads to a consistent level of edits in the cells we make, and that should hopefully translate to consistency of outcomes for the patients. But we're only able to know that once we've dosed a number of patients and look at the data in totality, and that's something we'll have later this year. But I think as long as we can get these cells manufactured in the right consistent fashion and also ensure that the cells engraft properly, I think we have something that could be a complete change in paradigm in how we treat patients with these diseases. It would be a onetime therapy, more like a surgery that results in a cure. I think you're on mute, Salveen.

Sorry. How do you think about the percentage of cells that are getting edited here, those that are biallelic versus monoallelic in terms of editing and the consistency here and what the boundaries are needed with regard to hemoglobin to prevent transfusions or sickling crisis?

Yes. I think the editing rate, so there is an editing rate that you have in the drug product when you manufacture the drug product. And then you put these cells into the body, and then you want these cells to go into the bone marrow and engraft in the bone marrow. So the key editing rate for us that we're watching and monitoring is the level of what the percent edit rate is in the bone marrow at certain periods of time. And that's really important because that's what tells you what's truly happening in terms of the patients thematic analytic system. So that's something we're tracking in the study and we'll disclose at the appropriate time. And then I think in terms of what needs to happen, what's the bar, the bar is different for thalassemia and sickle cell, right? In thalassemia, you want -- it's not as much about percentage of fetal hemoglobin. It's what's the total fetal hemoglobin produced. And then together with the endogenous beta-globin that's produced, are you getting to a level that renders you normal from a total hemoglobin level in oxygen transport, right? And that normal level is somewhere between 12 and 15 grams per deciliter in a normal adult. It differs a little bit by men and women. But the bar is, can you get enough fetal hemoglobin produced? And if you look at the first patient that we have, there was over 9 grams of fetal hemoglobin being produced, close to 10, and rising at the time we report that data. And that means that you're getting to those levels with our therapy. In sickle cell, we had 45% HbF at the time we reported data in November. Our primary endpoint is 20% fetal hemoglobin because that's what we believe based on all the natural history data. It's sufficient to actually provide a curative effect in sickle. And so we're well above that bar. Now we have to make sure it's consistent in all the patients we treat. But that's how we think about what's necessary from a fetal hemoglobin standpoint.

And then moving over to your cell therapy platform here. Can you touch on your approach to the lymphodepletion regimen and the puts and takes compared to the regimens that are used by some of the other players here with their CAR-Ts?

Yes. So in many ways, I'm glad that there are 3 different companies going after allo CAR-T and maybe more in the future with different approaches because, collectively, we're advancing the field. And I think at the end of the day, if you have an allo platform, we can all learn from each other and adopt what the best practices are and how these therapies work based on each other's experiences. So that said, I think the strategy we have is to edit beta2M to ensure persistence of the CAR-Ts. And because we have the beta2M edit, we don't -- we didn't start with a very high lymphodepletion the way you may have seen with CD52 antibodies or increasing the flu side combination and some data reported by Chinese companies. So we're starting with a relatively moderate level of Fludarabine and Cytoxan, which we believe quiets down the immune system to a reasonable extent and lymphodepletes it while -- and after that, we administer the CAR-Ts. Now we have -- we can always increase that dose of lymphodepletion as we go along, if we think that makes it better from a response standpoint. But the trade-off always is you can argue that better lymphodepletion is going to get you better response rates, et cetera. But ultimately, you want to make sure that this is commercially viable and that you actually have an advantage in -- when you get to community settings for oncology care. Because if you have very high lymphodepletion, you sacrifice 2 things. One is there is some safety risk that the community oncologists don't like dealing with. It's like wild reactivation and things like that. The second thing is when you try to think about redosing. Going through a heavy lymphodepletion every 3 to 4 months is not something that the patients and the community physicians or community oncologists desire. So for us, we think we're hitting a sweet spot in the sense that we have the beta2M edit to ensure a certain level of persistence. We have a reasonable amount of lymphodepletion to give us the best chance of success. And if that succeeds, then we'll have a much easier time in the -- in terms of commercialization and scalability because we can easily redose, and we can do all this in a community setting.

And when you look at your other levers or factors that you can optimize here, be it cell type manufacturing processes, redosing, what's your strategy for optimization? And how are you thinking about maybe bringing these in over time?

Yes. I think as with any oncology plan to trial, as a standard playbook, we want to look at all the variables here. The degrees of -- the axis along which you can optimize are the lymphodepletion regimen, the multidosing schedule that's on the clinical front. And then on the actual CAR-T front, you can actually change the construct, right? We can make more edits. We can actually change the level of editing and all those as well over time. So I think if you think about this, let's say, we have reasonably good data, but we may be able to titrate the level of responses versus the ability to multidose by change of lymphodepletion dose, right? We would do that, and we would test that out in a trial before we go to expansion phase. We would certainly interrogate multidosing because if you're getting durability of CRs for more than 3 to 4 months, you can just keep redosing every so often, and that should lead to a very long durability. And ultimately, I think we're -- we have a number of things that we're playing around in our labs where we have edits for NK cell attack on the CAR-Ts, for MHC class II-based attacks on the CAR-Ts. So we have -- we play out with different edits that make the CAR-Ts more robust, fitter. And so we can all bring those in play over time as well. So I suspect that over the next 4 or 5 years, you're going to see significant advances in how we think about allogeneic CAR-Ts. And I'm glad that there are many players all pursuing the same goal because we'll all learn from each other.

And can you remind us where clinical development stands for CTX120 and CTX130?

Yes. So we've started trials for both of those. We announced in our last quarterly update that we dosed patients for CTX120. So that's going through a standard dose escalation. The early doses for both 110 and 120 were sort of -- we knew they were subtherapeutic, but we need -- we wanted to make sure we established safety. We didn't want to have a GvHD event kind of derail the trial, so we're slowly moving through dose levels for both -- for 120. And then for 130, similarly, we'll move through those cells because we don't fully understand solid tumors yet and what the right dose levels are over there. But we're being deliberate and cautious because we don't want to create unnecessary risk to patients as we escalate these CAR-T doses. Because one of the things we're dealing with is these are CAR-Ts created from healthy donors, and they have robust T cells. They're much more cytotoxic at the bench side relative to autologous CAR-Ts. So when we compare head-to-head between autologous CAR-T from patient cell versus these allogeneic CAR-Ts, they're 5x more cancer-killing. And so we want to make sure we're not overdoing the dose early on.

And what are these kind of initial challenges that you're going to deal with as you look to solid tumors with your programs? And how do you assess it over time with regard to the trial?

Yes. So with solid tumors, we have a couple of additional challenges that you don't see with heme malignancies. One is trafficking to the tumor because these are 3-dimensional structures you need to get to the actual tumor cells. And the second is in solid tumors. The cancer cells have many ways of exhausting the CAR-Ts. So there is 2 different axis that we're trying to optimize. We picked renal cell carcinoma for CD70 because that's a tumor type where we know that the T cells traffic to as evidenced by all the data from the PD-1 and PD-L1 trials. You see plenty of T cells when you do histology sections in the tumor. So we don't have to solve everything at once. So we know that in renal cell carcinoma, these CAR-Ts get there or they will get there into the tumor environment. What we want to prevent is exhaustion. So we have an additional edit with our CD70 CAR-T that we believe actually gives it more robustness and reduces exhaustion that's something we'll have to see. There are a number of actually targets that we've looked at. We have a platform where we look at different targets to edit in these cells. And we found targets that are, frankly, much better than PD-1 in terms of editing out that would provide -- that will prevent exhaustion of these CAR-Ts. So we compare them with, say, here's a PD-1 edit on a CAR-T and then we compare them with other edits. And we found a number of other things that are -- and frankly, they could be antibody targets in the future, but they're much better modulators of exhaustion in these CAR-Ts.

And then you mentioned this earlier, but at ASGCT, you did show the ability to edit or to generate CAR-T cells with up to 10 edits here. What are you thinking about with your future programs with how many edits you may look to incorporate?

Well, I think over time, you're going to see more and more edits. I think this is going to be like iPhone versions. We're going to have version 6, version 7, version 8. I think we don't want to go with all the edits at once in our next version because what you're doing is confounding the experiment, right? This is -- we want to step through and see what is the appropriate commentarial set of edits that give us better efficacy and what's the individual contribution of each edit. This is analogous to the fight against cancer in the '50s and '60s when Sidney Farber was leading the use of all these different chemo agents, and we didn't know what the relative benefit of each was. And some people will try to dump all of them in one and put like 6 different agents. And it turned out they were actually not all better, right? It wasn't better to put all of them together. If you've done a systematic experiment of each of them and understood individual contribution, you would have helped a lot more patients along the way. And so we want to do a systematic external redo. We go from 2, 3 edits to 4, 5 edits to 7 or 8 edits. And at this point, our regen med program, which is our type 1 diabetes program, has more than 5 edits in those cells. So we've identified what those edits are. In oncology, we just want to be systematic in how we introduce the edits and not do it all at once.

So moving on to your in vivo pipeline, what are the limiting factors at this point in terms of getting the in vivo programs into the clinic?

Yes. I think at this point, if you ask me 2 years ago what the biggest challenge was in, in vivo programs, it was delivery. And it's not just the base level of delivery that you need to get these CRISPR/Cas9 molecules into the cell and to the nucleus, it's also controlling variability. I think what you have with gene therapy or something where you're putting these components into a virus, for example, or a lipid nanoparticle, is you get a lot of variability that you're seeing with AAVs and with AAV-based programs against the liver or other things, right? So how do you control the variability? How do you get sufficient amount of CRISPR/Cas9 into the cells you're targeting? Now we made tremendous progress in the last 2 years on delivery. I think our collaboration with StrideBio has given us a number of different new AAV constructs that are much better than the original AAV8s, AAV9s or AAV2s that people are working with that are much more tissue-tropic and get a higher cargo load into the cells. So we have our own AAVs now through StrideBio that we apply for programs, for example, like DM1 to get to the muscle. We also have lipid nanoparticles that we've made ourselves and using Cas9 as an mRNA that we can get into the liver with reasonably high efficiency. And we've also found ways to control that variability. So all those are moving forward. They just move on a different pace because you need to do many more species before you get to humans. You have to go from mice to rat to potentially mini pigs to nonhuman primates before you get to the clinic, and that just takes a longer time frame versus ex vivo cell therapies where you go from mouse experiments to humans essentially because you're dealing with human cells. So it's just on a different time scale, but that's moving along nicely. I think -- so that's why the company, we have many parts of the business. We've encapsulated within our business, we have something that competes with Bluebird on hemoglobinopathies, and it's that scale of franchise not too far behind. We have Allogene and Precision-embedded immuno-oncology franchise and competing with them on the solid tumors. We're competing with the likes of potentially Iovance in the future with our solid tumor candidates, and then type 1 diabetes can be a whole different franchise. And finally, in vivo fits more into an Alnylam-like mode, and it takes a little longer. But once you have a platform, it's much more scalable. And so that's why we continue our efforts in, in vivo, even though it's further down the road. But we have set up the pipeline in a way where we have catalysts of data every 18 months or so.

And with regard to the status of your Vertex-partnered programs here on DMD and DM1, how close are those to progressing towards INDs?

Yes. With the DMD program, we partnered with Vertex because they bring a lot to the table in terms of manufacturing capabilities that they're building with their cell and gene therapy franchise and the technology they acquired from Exonics. I think in terms of the updates in terms of IND time line, they are best positioned to provide those updates for DMD. I think on DM1, what you saw was a recent milestone payment from Vertex to us on the research end. And we're very pleased with how quickly we were able to achieve that milestone, which essentially signals that we're getting through the research phase very quickly and into IND enabling. But again, for both those programs, DMD and DM1, Vertex have the -- will take the lead in communicating to The Street around the progress of those programs.

Great. And then finally, here on the regenerative medicine platform, just remind us where the platform or the program stands now in development and when we might see an update here. And then secondly, beyond diabetes, are there any applications of the ViaCyte technology that could utilize CRISPR that we could see enter the clinic?

Yes. The key technology that ViaCyte developed over the last 18 years with over $400 million of capital spent was to develop a device that can encapsulate cells, and you can put the device into the body and not have a foreign body reaction or something that projects that device in the body. And that was done with a lot of experimentation around these polymers that go into making the device with the aid of GORE technologies, the company that makes all these ski jackets that are hydrophobic in nature. And so what we did is that we bring in the technology around differentiation of cells, iPS cells into these pancreatic islet cells or islet cell progenitors plus the device capabilities. We bring the editing. One of the things that they saw with their early data and they just put some updates out there with the recent financing is that they're seeing with their first-gen product, which is not edited, and if you put that into the body, you actually detect C-peptide. And C-peptide is a byproduct as proinsulin is converted to insulin. And so you're seeing these cells that are in humans and they've dosed a number of patients with it that are producing insulin from these islet cell progenitors that ultimately become islet cells, right? It's artificial pancreas, in other words. Now we're working on second-gen edited version because we know that the edits we're going to make them are going to make them that much more robust. You're going to prevent any sort of immune rejection. So the first-gen version that they have, they have to immunosuppress the patients. So it's not -- it's applicable to all the type 1, serious type 1 patients, but it's not widely scalable. But with the edits we're making, they can prevent immune rejection. So they could be potentially applicable to all type 1 patients and all type 2 patients as well. And so we're very excited about the fact that they're seeing early signs of efficacy in their clinical trials. And our hypothesis is the second-gen version is only going to be that much better than the first-gen version. So we're excited to get going. We're doing -- we're conducting IND-enabling study, but we'll provide an update on that program and when we expect to get to an IND at some point later this year or early next year.

Perfect. Well, with that, Sam, thank you so much for joining us today.

Okay. Thank you very much.