CRISPR Therapeutics AG (CRSP) Earnings Call Transcript & Summary

[Audio Gap]

Thank you, Salveen, for having us here. I think the lessons learned and will be applied to our strategy in how we build the company actually comes quite a bit from the antibody space and looking, in fact, at the last 40 or 50 years versus the last 6 years. In fact, as you might be looking at the last 100 years, there have been various discontinuities you've seen in terms of drug development. The -- it went from serendipitous drug discovery with wartime molecules or chemistry to more rational drug design. And then the '80s really heralded a new era with antibodies. And in the antibody space, you had about 25 companies start in the early days. And -- but while the technology was all similar, you had a big range in terms of value creation. You had one Genentech that created $100 billion of value, others that were really good science shops that never created much value, right? There were companies bought for $2.5 billion, $1 billion, some that went under. And the difference, I think, really was not as much the technology as it was the choice of indication, the portfolio, how you partner and how you build a company. And those are the lessons that we've learned or we're applying to how we build CRISPR. For example, we said, let's derisk the platform in a stepwise fashion. Let's start with ex vivo therapies, which is lower risk than in vivo approaches. In terms of portfolio, we said, let's diversify our portfolio, not just in 1 or 2 diseases, but with different pillars like the rare disease pillar we have with hemoglobinopathies, the immuno-oncology pillar and now the regenerative medicine pillar, so that diversification; how we did our partnerships, not just to bring validation in dollars, but how you bring capabilities to the mix with Vertex and Bayer; and lastly, how we build the company. Ultimately, you get -- you see various ranges of productivity from the companies that are similar size, and that's because of the people you bring in, the operating system you put in place and what output you get out of it. So those are all lessons that we've learned and I think have so far played out nicely for us. But we need to continue to apply those lessons from the last 20, 25, 30 years as we try to build the next Genentech with the CRISPR platform.

So as you look over the next 10 years, what can we expect from the company? And given the rapid ascent here of CRISPR/Cas9 gene editing, how do you see the technology evolving and the [indiscernible]? And I guess, where is there a key need for optimization at this...

Yes. If you look at the macro, the next 10 years. I think one of the things that we're going to see is cell and gene therapy is going to become a major part of the market, right? I think by 2030, we're probably going to have $2.5 trillion, $3 trillion of revenue in pharma. And I expect that at least 1/3 of it actually could be cell and gene therapies. People may say, gosh, that's impossible, but they said that 20 years ago at antibodies. Antibodies are now 50% of the whole market. So if it's -- if you're looking at a $700 million, $800 billion opportunity with cell and gene therapies, there is a macro wave that we're riding. And the large part of the cell and gene therapies will ultimately be based on some sort of manipulation technology, whether it's gene editing or some of the viruses. And CRISPR is one of the best tools we have to manipulate cells, to engineer AAV construct, whatever else you do. So I think CRISPR is going to be an engine -- the technology is going to be an engine for many of these cell and gene therapies in what could be a over $0.5 trillion market. And then beyond that, what does it mean for CRISPR Therapeutics? Besides have -- being the engine and providing that engine to potentially other players in the form of licenses or capabilities, we want to develop our own therapies, which will be the -- which actually provides more alpha, more value creation as we build the company. So if you look at Genentech, $100 billion of value created. Only about $6 billion of that came from their IP if you look at their IP licenses and decompose the value. Some $10 billion or so came from partnering and giving -- providing that capability elsewhere. But really the -- majority of the value creation came from the drugs they developed, whether it was Avastin or Herceptin. And so that's what we're trying to do is have drugs that transform diseases. And we'll take the example of sickle cell and beta thalassemia. We could have a curative approach that could transform that disease. So there's a lot of value you could create there for the entire ecosystem, not to mention these patients were -- and physicians who don't really have anything to grasp on to today, these diseases. And then in oncology, which is a very different dynamic from a competitive standpoint, we could have transformative therapies. I know there's a lot of debate about bispecifics versus cell therapies and everything else. But I think ultimately, these advanced therapies versus the old form of putting chemicals or cytotoxic chemicals into the body are going to win the race. And then regenerative medicine opens up an entirely new form of medicine that could regenerate all sorts of organ systems and really play a big part in how we treat various diseases. Ultimately, I think with these platforms -- there will be things that fail as well. I mean it's not everything is going to succeed off the bat. But if we take calculated risks in a macro market that's growing rapidly, I think we're set up to speed.

So when we look at CRISPR/Cas9, it's a relatively young technology, but it's obviously penetrated, I mean, clinical science relatively quickly. There are other alternate editing technologies that are coming out such as prime editing. How do you -- I guess how do you compare the capabilities -- how do these compare with the capabilities of CRISPR in terms of addressable diseases and other items? And would you look to bring in some of these future technologies as they progress and as they get derisked?

Yes. I look at some of the things like base editing and prime editing as continuations of CRISPR. The fundamental shift is, as you look at gene editing before with TALENs or zinc fingers, you had a protein-DNA interaction. You had a tool where you were trying to get the protein-DNA interaction to find and cut a place -- cut the DNA in places. Cutting DNA is not new. In the 1970s, people came with the restriction enzymes, EcoRI, HindIII and all these things came up in the '70s. And you could cut DNA, but they were indiscriminate, you're cutting it in all places, right? The whole innovation here is -- in gene editing is precise cutting. And the reality with the older platforms, while they can do the job, it takes a lot of time and effort to make them precise, right, but -- whether it's zinc fingers, TALENs, et cetera, they can do the editing. CRISPR is that much easier. You can actually very easily find guides because an RNA-DNA interaction, which is much more natural, and you can make an edit in a very precise fashion. Now if that's the innovation, you are going to have a lot of improvements, just like the antibody space. In 1978, it was the first time you had molecular cloning. Then you had protein-based therapeutics and then antibodies. But it took advances when you start humanizing these antibodies. First, to a certain extent, they don't have reactions. And then fully humanized antibodies, which is HUMIRA, that took some time. So I think this is analogous to that, which is you have CRISPR and gene editing and you know how you can precisely cut based on an RNA-DNA interaction. And now you lend that fundamental principle to do base editing or prime editing. And each of these will take some optimization and improvement, but value creation will come only when you apply these technologies to develop a therapeutic. So what's the use case and how you apply it is really important. So for us, we're a gene editing company. And if we find a technology that works better -- in fact, we do a lot of base editing and prime editing in our labs. If you find something better, we'll absorb that into our portfolio and apply it. So far, for the diseases we're tackling, we don't see a dramatic advantage of any of those technologies. In fact, there are some disadvantages, because they're still early in terms of target or characterization. And so we're sticking to the so-called classical CRISPR technology for our applications right now. But over time, we don't want to be -- we don't have an incumbent's curse. If there's something new that's better, we'll be good buyer -- smart buyers.

Great. So we've also seen you partner with companies such as Vertex and Bayer and KSQ recently and ViaCyte here in the regenerative medicine front. What's the common thread here between these business development decisions? And how are you thinking strategically about external partnerships moving forward?

Yes. So there are 2 types of deals. The Vertex and Bayer deal fall sort of in a sort of sell-side, whereas KSQ and ViaCyte are buy-side deals for us. The notion of doing Vertex and Bayer early on was to get validation for the technology, to be honest, to get some cash in the door. Because you always have the interplay between raising money through the equity markets and then raising money through partnerships. And if you have a good [ penchant ]there, you allow for the best outcome for the company in terms of cost of capital. So those were sort of validation sources of capital and then some capabilities. In the case of Bayer, there were some disease capabilities we were looking for. In the case of Vertex, there were some capabilities on translating the medicines that we were looking for at that time. And so those were targeted partnerships, right? ViaCyte and KSQ are very different. This is for us to be buyers in a market to acquire technologies, which we think are -- could have a great synergy with CRISPR. For instance, for the ViaCyte, they spent over 18 years trying to optimize the way you take iPS cells or embryonic stem cells and differentiate them into islet cell progenitors, right? So essentially, they can take embryonic stem cells and make pancreas out of it. And that work in regen med takes a long time. It was mainly funded by J&J to their BetaLogics subsidiary in the past. But we saw something that was an undervalued asset with great technology. And if you put iPS cells together with CRISPR, all of a sudden you create an allogeneic organ transplantation opportunity, right? So you can create islets, you can create pancreas, you can create liver and create any organs, which are allogeneic. The whole reason regen med never took off, even though it won a Nobel Prize in a relatively short time after discovery, for Dr. Yamanaka, it did not take off because to recreate a pancreas or liver for every patient took $2 million -- cost $2 million, but it also took 6 months to do it for each patient. It was never going to be at commercial scale. But if you do a CRISPR, you can make it allogeneic, where you can do one source of production for every patient or every recipient. And so you've all of a sudden taken away the scarcity of organs across a number of different important organs. And so that was the thesis behind regenerative medicine. But the deal thesis was capabilities that were going to take a long time for us to develop and had a synergy with CRISPR. Similarly KSQ. KSQ came up with massively parallel CRISPR screens in mice to look at what factors -- this is the era where we're saying there is PD-1, but what else is there, TIM-3, LAG-3, et cetera. And while some of the adjacent targets haven't worked out so far, there are lots of other signaling domains that are important in either exhaustion of immune cells or the trafficking of immune cells. And so KSQ had done quite a bit of work to identify the targets like that are the next-generation PD-1s, and we've done some similar work. And so as we look at the synergy there and said we have our Cas c with our immune CAR-Ts. So if you take these targets from KSQ and apply it to our therapies, that's going to actually make them that much more potent from exhaustion or trafficking standpoint. So it was a technology buy up, if you will, that allows us to enhance the pillars that we already have committed to.

Just moving forward here, how are you thinking about future collaborations? And do you still intend to wholly own your immuno-oncology franchise?

Yes. I think it's a very good question. I think if you look at the value creation trajectory market, largely defined by a lot of you sitting here in the audience, I think what you're seeing is earlier and earlier recognition of value creation from patients' data with -- a few patients worth of data or very small trials, right? So if you're in immuno-oncology, you have an asset that is not -- before an IND, you rarely get any value. Once you have data for 5 or 10 patients, you recognize 30%, 40% of the value of that asset. And then you got a 100-patient trial and you get -- you recognize 80% of the value of that asset. And then doing all the hard work beyond that to do global trials doesn't get you much more. And in fact, it's a disappointment after launch. So if you look at that value curve, partnering something before you hit that 100-patient trial or that size is almost giving up value. So you kind of want to -- if you believe you're playing the right cards, one, actually prosecute to that extent. And then there's a question, do you want to -- U.S., at least we definitely want to keep it. I think the old notion of saying small biotechs can't commercialize is going away, because the old model of commercialization is going away. It's no longer a sales force detailing kind of game. It's a medical affairs, kind of commercialization model that we're moving to. And in that setting, we can aspire to be the commercial entity for these assets. Ex U.S. is a real question. When you have a platform like CRISPR and you think about return on invested capital, there's so many things we could be doing with that platform, whether it makes sense to get the incremental value from launching in Lithuania, I'm not sure. So we'll make the right decisions when we get to that point. But first things first mean to execute to get to that 100-patient-or-so trial to see what -- and see what the data looks like.

Just to dig a little deeper into that, so when you expanded your partnership with Vertex, for instance, I understand DMD [indiscernible]. But something like DM1 at this point is [indiscernible] get proof-of-concept so early on a few numbers of patients. Why partner indications such as that, that aren't as competitive or indications in the future without waiting for proof-of-concept at this point?

Yes. So I think that nuance is important. On DMD, it's licensed to Vertex. On DM1, it's a 50-50 partnership, because we still haven't recognized much value there of what we could be doing with editing. The thesis there was these musculoskeletal diseases, production or manufacturing of AAV is at a very high dose. You actually need quite a bit of scale-up of manufacturing to produce that high doses where you're seeing response rates or you're seeing responses in patients. That requires quite a bit of capital investment to build that AAV [indiscernible], which was -- which didn't make sense for us, given all the other things we're doing with our immuno-oncology assets and our sickle and hemoglobinopathies assets. Vertex is highly committed to cell and gene therapy, and they're actually putting a lot of investments into building their own manufacturing and capabilities in this space. It just made sense in a very competitive environment, rather than divert a lot of our resources towards building these manufacturing capabilities, and in an indication like DMD, manufacturing is a single biggest driver of success or failure, it wasn't something that made sense for us, both from a bandwidth standpoint or a resource allocation standpoint.

Regard to manufacturing, what is your strategy in terms of in-house processes? Because it seems like you're in a bit of a dual process here?

Yes. I think it's -- I think you need to have the right balance. And try to do everything in-house is not efficient. It also means that you're not getting the best of capabilities from everything that's happening in the [indiscernible]. It's sort of how you think about the interplay between partnerships and equity markets for funding. I think you want to have a bit of a tension there, where you have some things in-house and some things outsourced. For example, guide manufacturing, which is somewhat commoditized and is available with multiple CMOs, not a real source of value accretion or IP protection, is something that we can easily do outside. Now for early pilot studies, we make the guides ourself, because we don't want to have the time delays to -- in terms of outsourcing it, nor do we want to -- we want to protect the IP there. So we shared that way. With cell therapies in immuno-oncology, we're going to build a lot of it in-house. Now we do use a CMO or CDMO, I should say, more appropriately, because they bring a lot to the table in terms of development of the assets. But ultimately, we want to have that flexibility to have an in-house option, but also have a CDMO producing these, because we have now 1, 2 -- we've 3 assets that we're moving forward at this point in immuno-oncology. And having sort of that capacity to not just get through clinical trials, but to get to early commercial is important in the early going, because you want to build that buffer capacity. If you see tremendous success early on, you want to be able to scale up very quickly. Or if you need to make more engineering steps and alter what you're doing, you'd rather do that in-house and then transfer to the CDMO versus trying to do it all in an external CDMO. So it's a multipronged approach as we look at a multi-component system that makes up CRISPR Therapeutics or CRISPR therapies. You have guides. You have protein. You have cells, mRNA, AAV donors. So we look across each of those individually and do a mix of in-sourcing and outsourcing.

So as we look at 2020, what are the key tests and milestones?

Yes. The philosophy from a data generation perspective and value creation perspective for our portfolio is we want to be in a continuous growth mode where there's something coming every so often. Because what you've seen in the space is people are betting on growth more so than -- mostly on growth, right, and [indiscernible]. So if you don't have some data coming every 6, 12 months, you have a stalling effect on your share price, you have a stalling effect on momentum as a company. And so we've constructed our portfolio such that we want to have data events for our programs every 6 or 12 months. So at this point, we had an early data, particularly on thalassemia. And so we'll have more data coming that hopefully is confirmatory with more patients in both thalassemia and sickle. Then we have early data that we'll have from immuno-oncology with our 110 program. And then beyond that, we have data for 120. And then as we look into following year, we'll have 130. So they're all stacked up in a way. And then beyond that, we have regen med. So we have more and more coming every so often that we unveil, and we can't guarantee success for everything. Obviously, there's going to be things that work and don't work. But at least there is that notion of our progress with our portfolio as we seek to answer the hypotheses that we've put forth with each of these products.

Jumping into the portfolio then, you recently reported impressive first-time data for CTX001 in 2 patients, 1 transfusion to beta thalassemia and 1 sickle cell. Could you just put this in context for us in terms of what is clinically meaningful and normal? And then given the limited number of patients that you treated, how do you get confidence about reporting future data sets in a larger subset and seeing these data sets reproduce? And would we look to patient variability or should we expect patient variability?

Yes. So for those who -- so many of you aren't aware, but we put data out for the first thalassemia patient, the first sickle patient. First thalassemia patient was a young woman who is -- didn't have access to the best quality care, was suffering from relatively serious beta thalassemia. It's an IVS-I-110 mutation, which is almost like a beta zero mutation, required transfusions very often. All these transfusions were leading to a lot of high iron levels, ultimately lead to organ damage in the patients. And so there was a patient with severe disease. And this patient came in, got treated with CTX001. And since then -- until the time we reported the data, had not required transfusion, right? So you've basically said, here's a patient who required transfusion all the time and you've functionally cured that patient in a sense by not -- by taking away the transfusion requirement. That for the patient is transformative and life-changing. So same with the physician, because the physician doesn't have to see that patient all the time. Now we can get into what all the measures are. There's the F levels in terms of fetal hemoglobin that we produced, how many cells or F cells versus not. In all those metrics, if you look at natural history data, there was a bar that we had in mind in terms of what would be curative or what could be functionally curative. And I think the data will well pass that bar, right, both in thalassemia and sickle cell. In sickle cell, we were thinking that 20% fetal hemoglobin could potentially be curative for these patients. And this patient that we had 4 months out was a sickle patient who -- this -- the woman name is in the public domain, so I can mention it, Victoria Gray, mother of 4, is leading a terrible life, because she was getting hospitalized several times a year. And each time you get hospitalized, you go to the ER, there is a reasonable chance you die. And that patient has been POC-free since being treated with CTX001. So it's a dramatic change for that patient. And again, we thought 20% fetal hemoglobin would be curative, and she's already at 46% fetal hemoglobin at the time we reported the data. So these are transformative data, but it's early. I want to caution that it's one patient, so it's early. And we do need to see that it holds true for all the other patients. But even if there's variability and there will be variability. There hasn't been a cell and gene therapy trial where there hasn't been variability. We're so above the bar, my hope is if you're clearing the bar by that much, that variability will make -- ensure that every patient does get benefit versus having a scenario where some patients get benefit and some don't. You go through myeloablation and a serious procedure like this, I think patients will need the confidence that a very high proportion of them, 80%, 90%, hopefully 100% of them are getting cured to go through this procedure. So we'll see. I think we expect to dose many more patients this year, and we'll have that data set at some point at the end of this year, where we hopefully confirm what we're seeing with the first couple of patients.

With the woman who [indiscernible] beta thalassemia, her hospitalization time, how does that link to [indiscernible] engraftment? Or what are the factors that are going to roll in that?

Yes. I think ultimately when we commercialized the -- hospitalization time is highly correlated to engraftment time, right? So once someone is engrafted, know that they have an immune system, then you let them go into some observation. In this case, the first patient, I think people were taking a lot of extra caution to make sure that [indiscernible] right place to go back. So we've been extra careful in terms of the hospitalization time. Ultimately, I think we want to -- serious patients will go through a procedure like this. But if you want to get to the entire market, we want to have lower intensity conditioning regimens, which also mean that there will be a lot less hospitalization time, can be done in centers that are beyond the top centers. So that's where we ultimately want to go with these therapies. We're in the early going. We do want to ensure success for these patients. So we're going with the full intensity conditioning and being very cautious with the hospitalization time after they engraft to make sure there's the best outcomes for patients.

[indiscernible] any additional?

We'll provide updates as we go along in terms of the patients dosed and have data at some point [indiscernible]. But I think we'll continue to enroll patients in both trials. Tremendous excitement after the ASH for us. I think while the data -- the -- we have people from countries -- every country in the world, people in Vietnam -- physicians from Vietnam, Thailand. Others saying, we really want a therapy like this, we want to be part of the trials. And ultimately, we want to have global trials. And so we now have trial sites -- multiple trial sites that we've opened. In fact, for immuno-oncology, where it's truly global in terms of how we're conducting the trials. And so we're quite pleased with the interest from both physicians and patients.

With regard to data releases, I know we'll get data over the year. But given that you have follow-up in the 2 additional patients, adding patients [indiscernible] updates like every 6 months at the EHA and ASH, is that how we should think about it?

So we haven't guided to when we'll have data this year. I think it makes sense to be at these important medical conferences. Dynamic where there is one, but more than one therapy that could be potentially curative. I think we want to be at these important meetings, meeting with the investigators. Now I think based on how the enrollment is going in the dosing, we'll take the appropriate times to disclose data. But we expect that we'll have more and more data accruing as we dose more patients. So that's something we'll guide to as we go on in the year.

Yes. Longer term and look towards commercialization products, how do you look to position this versus the existing products that may be on the market, the gene therapy products? And maybe you can just comment on how you would enter the market, and then at the end of the day, [indiscernible]?

Yes. But I think while there's all these therapies being approved in sickle cell and thalassemia, I think you really have to partition curative versus noncurative, right? I think for very serious patients, what they need is a curative therapy. If you're a sickle patient who has 7 hospitalizations a year, getting down to 5, not the change you're looking for. You're looking to be cured. So then you have the lentiviral virus-based product [indiscernible]. So at this point, as the sort of 2 lead approaches for a functionally curative possibility. I think the market is large and has room for more than one player. In fact, I doubt in a -- these indications where it's going to be supply-constrained launches that it's going to be a 90-10 type of split. I think you're going to have a medical device type launch where you have different players, and there's going to be market share split based on the different centers and the loyalty of the centers, but also based on how much you can actually manufacture. My suspicion is, demand will outstrip supply whenever these drugs will launch. Because these are autologous therapies, they're not easy to scale up from a manufacturing standpoint. And at this point, the bigger lever in success is market development, not market share. So I'm the biggest fan of Bluebird and their launch at this point, because we want that to be successful, because you want to train centers to be able to do these transplants. You want to have the payment systems all worked out. You want to have the payers used to covering these therapies. And you want to have patients comfortable with these therapies. We'll fight the market share battle later. At this point, it's all about market development, sort of like the Lucentis Eylea dynamic that we saw early in DME. So my -- the difference you're going to have is this is not anywhere close to HCV-type launch where you have a launch code that looks like this. You're not tapping the bolus. If there's 25,000 serious sickle patients in the U.S., we like if we dose a couple of thousand patients in the first 3 years and even that's tough, right? So you're going to have a slow penetration of the serious -- this market thing when there's the serious patients. But the good news is that's a very long-term growth type of trajectory from a revenue launch perspective. It's literally a linear launch curve for 7, 8 years. So from that perspective, if you look at Bluebird's price point of close to $2 million a patient and the number of patients we're talking about in thalassemia and sickle, there is significant value to be captured here. But more importantly, we're -- we can make a profound difference in the lives of these patients who are suffering from these diseases. And not only that, there were some of the questions asked earlier to the other speakers here, you can actually reduce system cost in sickle cell disease. Over time, you're actually reducing the cost of the system from all the hospitalizations and ER visits and everything else and could actually benefit the system in many ways, and obviously, benefit the ecosystem entirely.

Over to CAR-T therapies discussed earlier as well. Still on track for first data from your CAR-T program in [indiscernible] mid-2020?

Yes. I think with the CAR-Ts, it's a very different dynamic. What we're seeing is a very large addressable market, but many players, right? It's not like sickle and thalassemia where there's 1 or 2 players or maybe 3. In CAR-T, there's many different approaches in the immuno-oncology space. There's bispecifics, there's CAR-Ts, there's existing therapies and antibodies. And so you have this huge confluence in indications like lymphoma and multiple myeloma. In that setting, I think there's -- our belief is cell therapies, again, will be 1/3 of that [indiscernible] market, whether it's autologous or allogeneic. And if allogeneic can show efficacy and a safety profile that's within the range of autologous, you're going to see a dramatic shift to allogeneic. Not only that, we have a huge advantage in how fast we can crew the trials, because allogeneic is off the shelf. So once you identify a patient, dose them a couple of days with that or a week of that, and that allows you to dose patients at a much greater speed and expediency than we can do with autologous trials. Now in terms of data, so what you have is a new paradigm with CAR-Ts, right? You're going through dose levels as you go to allogeneic. You have to go from lower dose to high dose. And so a lot of factors come into play as you dose patients. Start with a relatively conservative dose, because what we did want to do is set back the whole franchise by going to a dose that we think could be efficacious will have a safety event. So we started with a very conservative dose. We're moving through that. And then we get to more reasonable doses as we get -- as we want to see what efficacy is. But it's not just solving or whether you're getting PRs or CRs in these patients and what response you're getting. We're trying to understand the basic phenomenon, how CAR-Ts work. And so you need to have lot of qualified assays, you need to have a lot of metrics around cell expansion, the cell dose, the cell profile as they do their work in these cancer patients and try to kill the cancer. So we want to be deliberate in how we put that data out there. And you did see -- if someone tries to go out with premature data, you have an effect that's undesirable on your -- not just your evaluation, but your company as a whole. So -- but I think we are on track. We'll have data mid-2020. But I think we want to be very deliberate with how we put the data out there.

Looks like -- I mean there are 3 companies that are kind of in [indiscernible] allogeneic CAR-T [indiscernible] going after the same population. Can you maybe just give us your thoughts as to the different regimens, why you've opted 3 parts?

Yes. I think there is -- if you [indiscernible] the big difference between us, Allogene and Precision Bio, right, there are 2 conditions to make something an allogeneic CAR-T or allogeneic therapy. One is prevent graft versus host, right? You don't want to have these CAR-Ts killing anything but the cancer cells. Second thing is you want these CAR-Ts to survive and persist in the body as long as they can to kill every cancer cell, right? In a sense, it's like adjuvant therapies or surgery. You get the margins out in colorectal cancer or something that -- and how cleanly you get the margins hugely impact your 5-year survival. So for preventing GvHD, which is graft versus host, all of us have the same construct. We're knocking out the TCR locus in different ways, right? The difference is in the preconditioning and the construct for ensuring persistence. But the preconditioning and the persistence constructs are all related to having the CAR-T survive longer. What Allogene is doing is using alemtuzumab or a variant of alemtuzumab to ensure that you completely mute the endogenous immune system of the patient so that the CAR-Ts can do their work, right? To us, it wasn't a great strategy. And I'm glad they are chasing down that hypothesis, because it was important hypothesis as we all try to strive to make this -- understand these CAR-Ts. It wasn't scalable. And the whole point of doing allogeneic versus autologous is to bring this into the community settings, bring it to outpatient settings. And if you use a therapy like conditioning regimen like alemtuzumab, it's a very serious conditioning of the human -- of the endogenous immune system, which puts the patient at risk. And so I doubt it's going to be very scalable, especially outside the specialized centers. So we want the different approach. What we said is, what is the main cause or the main pathway by which these CAR-Ts are going to be eliminated? It's from the endogenous T-cells. And they do it through a recognition domain through the beta-2M MHC class I recognition domain. So let's cut that out with our CRISPR system, right? So that's our approach. And the Precision has said, we don't need to worry about that. We'll just hit the cancer hard. And if you need to hit it again, we'll hit it again with the multidosing, which all the companies can. So there's 3 different approaches. And I'm glad there are 3 different approaches, because collectively, we're going to push the allogeneic field forward. And I think we'll see how the data come out, right? You're -- it's encouraging to have a Precision there is something happening in these patients. Cellectis originally showed that there is no GvHD from cells engineered this way. So you're seeing some cytotoxic killing of these cancer cells. This has now fundamentally changed and become an engineering problem. Most of drug discovery is usually a biology problem, and it still remains that way with some of the rare diseases and how we think about antibodies and small molecules. But in CAR-Ts, if you have these fundamental conditions solved, which is around GvHD and persistence, then you need to keep engineering these things to be more persistent to have -- be less exhausted and continue to increase their cancer killing, right? So I think that's what we're going to see over the next 10 years. We're already working on Gen 2 and Gen 3 of these CAR-Ts, because we want -- we don't want to wait for our data. We just want to start working on these now, just the way Apple or Samsung are working on iPhone 14 or 15 and Samsung respective models. So I think you'll have newer and newer generations, more and more manipulations, up to 10, 20 edits possibly, and they're going to be sophisticated cancer-killing machines.

[indiscernible] but recognizing [indiscernible] what were your thoughts on the [indiscernible] at ASH [indiscernible] for allogeneic CAR-T [indiscernible]?

Yes, it's generally not our practice to comment on competitors' data. But that said, I think the fact that there's some activity is very encouraging, right? I think the whole allogeneic hypothesis, kudos to Cellectis for showing that there's no GvHD, that you can actually put these engineered cells in and not kill anything but the cancer cells, right? And now we see that they actually do kill some of the cancer cells. As I said, if it's an engineering problem and we're here, we're going to play in this for years and continue to innovate with -- if it's reduced to an engineering problem, that's easier to get out of from a risk-reward perspective and your probability of success. So we're -- we think it's positive in terms of the data that was put out there. It's just -- there are a lot of questions that were raised, because it's very early and it's not clear what the mechanisms are. But so -- but that's why we want to put our data out there. We want to be very deliberate. We want to say, well, here's what you're seeing for each of the patients at different dose levels, but here's what the mechanisms may be, here's what we're seeing from a pharmacokinetics and pharmacodynamics perspective, here's what we're seeing from a cell profile perspective in the patients. And so we do quite a battery of tests on the -- in our trials, and we'll have a lot more data to not understand just the outcome, but the mechanistics of how therapies may work.

What's the [indiscernible] time frame here for optimization for in vivo platform? Walk us through as well what the limiting factors are to getting this technology [indiscernible]?

Yes. Our prediction on in vivo coming true, which is it takes a lot longer and costs a whole lot more to get things to work than you originally anticipate. Same movie played out with siRNA, right? In 2000, people were saying an siRNA therapy approved in 2006, and it took 8 more years after that to get to that point. And that's because when you go with in vivo approaches, one, you're inherently limited in terms of the pace, because you have to move from species to species, right? You're doing something -- you optimize in rats -- mouse, then rat, then potentially a dog or a mini pig and then nonhuman primates. And each step takes a lot of optimization. So inherently, it's slower. Then you have the variability issue. You're simply shooting in these drugs in vivo and then hoping that you get the same data, but you see tremendous variability, because everyone's liver is different or everyone's muscles are different. And it's hard to know what's causing that variability. And then you have a real tox issue, right, because while your intended organ may be something else like the muscle, everything goes to the liver. And then if you shot the liver, you're going to get sepsis or other various toxins. So fundamentally, I think with these cutting-edge therapies, we have these issues or understanding dosing and understanding PK. So that said, we've actually now made quite a bit of progress. Editing is not the issue. We can actually get edits and get high-level edits for what we're trying to do. It's, I think, engineering the system and the delivery of it in a way that's safe, consistent and reliable and only targeted towards the organs of interest, that's the challenge. And you're seeing that in terms of the R&D time lines and everything is stretched out on just for us, but for the competitors. So we expect -- we're making good progress in, in vivo. It does take some time before we get it to the clinic.

First program might enter the clinic?

We haven't put a time line on it. I think our goal is to have an IND every 6 months. I think if you look at our IO portfolio and the regen med portfolio, we'll have an IND and something go into the clinic every 6 to 8 months. So what we have right now is the way it's stacking up is we have our 3 immuno-oncology assets and then we have a regen med asset. And then that -- beyond that will be the in vivo assets coming in. And the other thing with in vivo is getting it to differentiate the viral versus nonviral approaches. With the viral approaches, you can stand on the shoulders of a lot of the gene therapy of a company that's doing the work to understand that variability in the dose levels. But again, you're seeing, even with the same targets and the same indications, 3 different companies have very different data when you look at delivery to the muscle. So there's inherent biology that we're trying to understand with the AAVs and the delivery. With lipid nanoparticles, there's bit of a formulation challenge in earnest and to make sure that you've got very consistent delivery, and it's -- these are 3-dimensional organs. So you need to get it not just in certain cell types in the liver, for example, you need to get it across the whole liver to get the desired effect. So quite a bit we're playing with in vivo. I have no doubt that in vivo approaches with the gene editing are going to work and will be a mainstay. It just takes longer to get to that point.

With that, I'll open it up for any questions. Maybe one more from me. Throat issue. On your partnership with ViaCyte, do you see regenerative medicine as a major or -- I mean you do see regenerative medicine as a major growth area for CRISPR. Can you just speak to the strategy behind the decision to move into this space? And then how you're looking to kind of bolster your capabilities into other organs? And how does this platform differ from Vertex's platform with the recent acquisition of Semma?

Yes. I think -- regenerative medicine. I think it's just something the -- we want to be ahead of what the market is appreciating. And we saw -- even though venture capitalists, for example, you're now seeing a number of venture capitalists jump into regen med. There's -- all these companies informed by various leading venture capitalists, because they're understanding now that regen med can be dramatic -- have dramatic influence across many different diseases, right? In fact, every organ can be regenerated. But the image I have in mind is one of 2 waves crashing. If you think about an iPad today, the first sort of device that looked like an iPad came out in the '90s. It was called PenPoint or something like that. It was a great device, but it never took off, because there was no Internet, there were no apps, there was nothing you could do with it, even though it was a great device. Regen med similarly had 2 companies. In 2000, regen med was huge. I think there were companies that were going IPO within few weeks of being formed. And then in 2006, after the -- after 2008, there was a little bit of crash. And then there was a Bush era cuts on stem cell research. So people said, "Gosh, I've set up all these labs, but I can't work on embryonic stem cells. What do I do? Let's find a way to make embryonic stem cells and take a hair cell or a skin cell and make an embryonic stem cell." And that led to the discovery of iPS cells in 2006 that ultimately won a Nobel Prize in 2012. But that also opened up this whole area of saying, "Can we make artificial retinas? Can we make artificial livers?" Now the problem was, you couldn't do it for every patient. In Japan, for example, there have been patients dosed with artificial corneas that were using regenerative medicine, and they seem to work. Problem is it takes a long time to do it. Just like Internet to the iPad or apps to iPad, also now you've got CRISPR and you can now make all these manipulations. In fact, for our regen med product, we have many more edits than we have for our immuno-oncology product. And if you cannot quite go to synthetic biology, but if you can get to 10, 20 edits in these cells, all of a sudden you can create any sort of -- type of cell you want. So artificial livers are not out of the question. Artificial pancreas is obviously the easiest one, because it's very simple. They'll add a feedback loop on a per cell basis. Heart is a little more complicated. In fact, there had been a lab that had made sheets of heart, cardiac tissue that beat together artificially regenerated. And so if you can get to all those things, that just opens up a tremendous new field of medicine. In fact, my prediction is medicine is going to resemble more like surgery. Patient is going to come in, go through more serious intervention, get cells in or something like that, conditioning regimens. But it's a onetime procedure and then it's done. And that's how we're treating diseases. There's no longer the paradigm of popping pills. And so in this new era, I think regen med is going to have a tremendous impact in medicine. But we're still at the early stages, but we want to get there early and be the preeminent player in that space.

Perfect. Thank you very much, Sam.

Thank you.