CRISPR Therapeutics AG (CRSP) Earnings Call Transcript & Summary

We can get started. I'm Yigal Nochomovitz, I'm one of the biotech analysts at Citi. Next session is with the CEO of CRISPR, Sam Kulkarni. I cover CRISPR, I think, since the very beginning, since the IPO. So Sam, welcome. Thanks for doing this. Appreciate it.

Thanks for having us.

Year went by pretty fast. Here we are again. So I think a lot of people are familiar with the story, but I think it might be helpful if you just kind of frame the business case, frame the investment case. What's the near-term mission of the company? What's the grand vision for CRISPR?

Yes. It's great to have this chat with you nearly 8 years after IPO. And in fact, you know this since you covered us from the beginning, we're about 10 years as a company in October this year. It's hard to believe that time flies by this quickly. Everyone said, "Oh, our gene-editing, is a new thing." And it's a new form of drug discovery first and then it's a new modality for drugs, and here we are on the cusp of what could be the first approved CRISPR-based medicine in the world with Axi-Cel later this year. Here we are having dosed over 200 patients across many different programs and platforms. And I'm also proud of the fact that over time, we've built our own manufacturing facility, we have our own development engine. So we have a fully -- apart from the commercial piece of fully integrated biotech company and where we hope to take on not just few diseases but tens of diseases over time. We've done all that while maintaining a very strong balance sheet and a capital base that allows us to take more risk. And I think we're very pleased with where we stand. And so the second part of your question, what's the business going forward here is we're through one stage of the company where we've derisked the platform brought, hopefully, we'll bring something to patients in a commercial setting in the near future. And the next part of the journey for the company is how do you become a $25 billion company in the next 3 to 4 years? And that has to happen with the parlay from Axi-Cel into other platforms. And as you know, we have our immuno-oncology platform, which we'll talk about more. We have our diabetes platform. And now we have a very scalable in vivo platform, which we don't talk about much. But essentially, if you can deliver CRISPR-Cas9 into the liver in a very efficient way and knock out genes or added genes, there are tens of diseases you can go after very much like what the RNAi companies did. And so that will allow us to build a very scalable but foundational platform that can be the MO for growth for us for a long time to come. And as you see these patterns over time, the antibodies, the first 10 years, there was a certain pattern in terms of market acceptance and market excitement, the same pattern you saw with RNAi but if you talk to big pharma companies now, RNAi is just an established modalities. It's just like any other drug development. It's not different than small molecules or antibodies in the way to think about it. And same thing is going to happen in gene-editing. And I'm very confident that every pharma is going to adopt gene-editing as an important arm for drug discovery within their own pipeline.

Okay. Well, let's start with Axi-Cel and hemoglobinopathies. You have -- the PDUFAs are announced, one in December, one in March. So can we talk a little bit about the Ad Comm? And how do you think the FDA is going to approach that? What is the tenure going to be? What types of questions might they ask? I think we've also talked in the past as to whether there'd be 1 or 2 because there's 2 indications, obviously, so anything you can elaborate on there? Some of it you may not be able to answer in great detail.

Well, it's hard to predict. I mean it's up to the FDA for sure. Definitely, I think the Ad Comm date by the way, was -- came out of the FDA registered today, October 31, so it'll be towards the end of October. And it's hard to say what kind of questions, but typically, these Ad Comm's will be one for a platform is my guess. Because you saw in the case of Bluebird, there was an Ad Comm for thalassemia, but not for sickle cell because a lot of the questions were covered around safety in the thalassemia Ad Comm. Similarly, I mean, it's not surprising they want to do an Ad Comm because this is the first CRISPR-based medicine that's going to go forward for approval. And I think that it would be good to get opinions from key experts, opinion leaders and we appreciate the agency leaning in a way, in support of a new platform like CRISPR. In terms of questions, I don't know, it's hard to be prepared for all the questions in terms of what may come up and what we need to answer. It's hard to say what questions are going to get picked. We'll only know when the briefing will comes out. But I wouldn't imagine that's too controversial in a way. I think they'll go through the standard questions around safety, efficacy and the data set that we have, but we'll be prepared for all eventualities.

Are there certain specific questions that you are really focused on making sure you answer very, very clearly if they come up that are more controversial specifically related to safety.

I can't imagine or think of any. I mean our data set is pretty strong data set. One that comes around every so years in pharma. It's not often that you see a data set like that. The efficacy is very clear. As opposed to the lentivirus type discussion, there was questions on durability. Our durability seems to be very clear. The bone marrow shows what the cells are once they're edited, they remain at that 80% editing level. The patient benefit from clinical markers are very clear. And from a safety standpoint, we've done a very robust preclinical package for going into clinic from an off-target standpoint and everything else. And we've continued to do more work to show that this is a very safe therapy, and it can be a one-and-done type therapy for patients. So it's a very strong base set. I can't -- there will be questions -- particular questions on one or another, but we'll just wait and see.

Okay. What about Europe? You're filing there or you have filed there. So that -- what do the time lines look like for that?

Well, unlike the U.S., you don't have a specific date where you're going to get approved or not approved or what the decision date is. And so we continue to work with the agencies in the U.K. or the rest of the EMA and we hope that they can continue reviewing the file in expected manner.

So obviously, you're partner with Vertex. So assuming you get the approval on the -- is it the 8th of December, how quickly can you be in the field and marketing the product? And what else has to happen in terms of ramp-up activities or pre-commercial activities before Vertex can launch?

Yes. Just so everyone knows, Vertex leading the commercialization of Axi-Cel. And they are an incredible organization from a commercial standpoint. They have a lot of experience with cystic fibrosis, commercializing a drug in the rare disease setting, they have a global footprint in a both -- in a -- right now, they have a very strong footprint in Europe in addition to the U.S., all in a capital-efficient manner. This is not giant commercial organization like some big pharmas do to get very effective. And there's a lot of work to do. This is not one of those where drugs where it's -- it's a new modality. So you can't just hang up the shingles and expect that patients will show up. I think what you have here is work that you are used to do on the payer front to make sure that they get all -- get comfortable with how the drug is priced, what the efficacy is, what the benefit is to patients. You do have a somewhat fragmented payer landscape because you have the commercial payers but then you have all the state Medicaids that come into play here in addition to CMS overall, so doing all the work around just making sure they understand the therapy, the evidence behind it, but then all the mechanisms are in place would allow us to make sure that there's prioritization and the therapy gets paid for. The second part of this is making sure the hospitals are all well equipped in terms of doing the transplant procedures. Now while there are not a lot of transplant procedures that are being done at volume in these centers for benign settings, let's say, Allo transplant for sickle, that's because you don't get matches. The hospitals can handle a lot more. In fact, in the cancer settings, they do a lot more transplants for AML and other indications. So the infrastructure is there, but we need to make sure that the qualified treatment centers know what they need to do. Even simple things like how do you freeze-thaw the product before you administer, all that obviously, we did that in the clinical trials, but we just need to make sure that's all there. And third, we need to make sure the patient experience is seamless when they come into the process because you have to collect their cells, manufacture it, let them know how, what's the status is of the manufacturing, when they're going to expect it back and then they can set up a date for when they can actually infuse the patient. So all that requires supply chain infrastructure requires account teams or require sort of a personal touch, patient service touch. And it's great to see Vertex's pushing forward very efficiently but very effectively on all fronts to make sure that everything is ready by the time the PDUFA date rolls around.

So what do you think the uptake curve is going to look like? I mean given all the operational details that you just outlined, and just streamlining. As an institution going to basically try with 1 patient and turn the crank and make sure everything is working before they start to dose 10 or more patients, for example, the same institution? I mean is that how this is going to go? Or...

Well, I think it will depend on the investigator's comfort level. A lot of the investigators, if you recall, half the qualified treatment centers were part of the trials. So they already know how this works. So it's not anything new for them. A lot of these centers have a bolus of patients that are already waiting, that couldn't be part of the trials, but they want to get the therapy. So I don't know that there's a lot of learning there. Obviously, if there's a new qualified treatment center that comes on board that were not part of the trial, they'll have to go through more of the learning process. But I would expect that the key metric that we look at here, which is patient starts, which patients have already been prior authorized and where the cells are collected, I think that -- we expect that there will be a strong signal once the drug is launched. It's very clear there is a strong unmet need for these patients. They're suffering from the terrible burden of the disease. They want to do something. There's no other options for them really and they're willing to undergo a transplant process for something that could be as transformative as Axi-Cel for them. So my expectation is that the -- it's going to be -- it made surprising upside in terms of the number of patients interested and the number of patients who are going to be part of this journey.

Have you commented on sort of the share you would expect in sickle cell disease? Among, I think, around 40,000 in the United States. I mean, we have it in our model I forgot exactly, but it's somewhere in the teens, I think, but I'm just curious where...

Well, there are two aspects. When you say share, you're talking about -- are you talking a competitive share or just...

No, your share.

Yes. So this -- if you build out the model for [indiscernible] play out and Vertex will do the job of guiding and saying what we may expect in '24 and '25, et cetera, but -- and we'll leave specifics to them. But if you think about this model, a lot of this is market development. What percent of patients are raising their hands to get a transplant-based therapy? It will be less about market share between us and bluebird, for instance. That said, our work independently CRISPR work as we are in the market talking to patients and physicians. There's an overwhelming sort of lean towards doing CRISPR-based therapy from patients as opposed to any of their modality. And how that's going to play out with physicians in the mix of the decision and the patient and everything else, we'll see, the market will tell us. But I think given the fact that it's notionally a much safer approach to CRISPR-based edit versus anything else and the fact that we have such strong data to date, I think will give us a tremendous tailwind as we launch this together with other companies. But we wish bluebird well, and we wish -- we want the whole segment to grow very rapidly.

Do you have any learnings from Zynteglo that are relevant here, that you want to adopt for your launch or not really?

Some, but I don't think it's a very strong parallel. I think -- obviously, what we're trying to do is take learnings from every launch in cell and gene therapy. There's learnings from Roctavian's, there's learnings from Yescarta's launch, there's learnings from the SMA launch at Novartis have. They all provide lessons. For instance, Zynteglo in Europe is, I don't think it's very relevant learning for us because we're going to do launch in our own way in Europe, for instance, with the payers, and we have our own processes and how we bring them on board. But each of these cell therapy launches, I think if you think about the Street, not everybody is able to predict them well, there's always been -- if you look at the range of expectations around some of the AV launches, there's a huge variance in what the expectations were. And I suspect the same thing is going to happen in Axi-Cel. But that said, I do -- if you look at the whole setting, which is sickle cell disease, no other option, transformative data to date and a very effective commercialization partner. I think that all points towards something that could be a very strong and durable launch. It's not just a strong bolus they're looking at, something that could keep growing for 7, 8 years. Once you get a strong start.

Yes. Okay. What about the move away from the current conditioning regimen? You have this development program, the Kit ADC, which I believe will potentially provide a softer, gentler conditioning regimen. And you've commented in the past that, that could potentially expand the market considerably. So can you just walk through your thinking there? And how close is that to being tested in the clinic?

Well, if you think about it, today, we're looking at the U.S. market, there's 100,000 sickle patients or so, right? Nearly 1/4 of them, 25,000 severe sickle cell patients will be candidates for Axi-Cel. And that's something that could Axi-Cel make a meaningful difference in their lives, whatever their baseline VOCs are and everything else, right? In our clinical trials, it was 2 and above from a VOC standpoint that defines severe sickle cell. As you think about a gentler conditioning agent, that number could be 50,000 or more patients that potentially could benefit from a therapy like Axi-Cel with the gentler conditioning agent. In fact, it could be even larger than that. And that's one effect, which is the TAM increases. The second is, there'll be more patients willing to raise their hands. So the penetration rate in the market that we have will also increase if you have a gentler conditioning agent. That said, I think there's realities of how you actually bring a general conditioning agent into the mix here. One, we have to show independently that this conditioning agent is safe and in a setting where we can measure chimerism, show that you can get a meaningful level of chimerism with the gentle conditioning agent. And once we demonstrate that, then we have to apply it in the sickle cell or thalassemia context, but we're -- this is a very high priority for us at CRISPR, and I think I'm pretty confident that the next 3, 4 years, we're going to see meaningful advances in that direction.

Okay. And then in terms of your response, do you want to just review quickly for everyone's benefit, just the Vertex, the economics of the Vertex arrangement for everyone that's less familiar?

Yes. So we're partnered with Vertex. It used to be a 50-50 collaboration, now its 60-40. Vertex has 60% of global profits, and we have 40% of global profits. And that's a similar ratio for cost shares we're developing right now. We do have milestones that we expect from Vertex if it get approved. There's a $200 million milestone that would come upon approval in a major market. That does -- for us, at this point, if you look at Axi-Cel alone as a franchise, we don't spend any more money if things go well in terms of that franchise. And everything else is profitable and eventually, it could a lot of our expenditures beyond that into other franchises as we look into 2026, 2027 and '28.

I should have asked this before, but just so we're clear, assuming you get approval on December 8, would you -- would Vertex launch in sickle right there? Or is there a scenario where you would wait to get both approvals before you do the launch?

No, no, I think we want to go for it. I think -- patients are waiting, I think the time is of the essence in this market for patients, for physicians and everyone else, we're expecting this therapy.

Okay, okay. All right. Let's talk about IO then unless you have anything else you want to add on Axi-Cel. All right. So CTX110 is the -- I guess, the original version of your CD19 CAR-T, but you have some additional edits, which you've made improvements with the Regnase and the TGF-beta, if I'm not mistaken. Can you just comment on how those advances are going to improve the product as far as making the original CTX110 more competitive and obviously, a very crowded space.

Yes. It's a crowded space, but we're very excited about these next-gen programs. I think what we've done, our original cost for CTX110, we inserted the CAR, we knocked on beta2M to give a little more persistence for these cells. And sort of that concept, we improved upon that by making an edit called Regnase-1 which is a lesser-known edit and TGFBR2, which is a better known edit in the space in cell therapies. And what we did is a very massive scale screening to say, we have no buys. And everybody was saying, oh, we should do a PD-1 edit, that's what we should do to make the cells more potent. We have no buyers here. Let's just do a massive screen to say what is the best edit that would make the cells more potent? And then we looked at the right quadrant and said, okay, these are the edits that were lesser -- these were not very well known genes like COX-1 or Regnase-1. And the ones that were really -- the expected ones that PD-1 didn't rise up to the top. So then we did another campaign to say, what is the best payer-wise edit because we can do more than one edit. What is the best payer-wise edit among the good edits to get transformative improvement in the potency of the cells. And we came upon Regnase-1 and TGFBR2. Now there were 2 other combos as well that we're getting there. And Carl June, for instance, has a different combination with Regnase and also somewhat equivalent in the cells. And the cell potency is 20x more. In manufacturing, the yields are logged greater. Cells seem to retain a very strong central memory phenotype even several days into a mouse model, for instance. And the sales seem to keep going for longer in a -- with a cytotoxic potential is a lot longer for these cells. So you combine all those sectors and what we could have is something that could be 10 to 20x more potent than CTX110, assuming we can manage the safety. So we've started dosing patients with CTX112 now. We want to see where it compares to CTX110. But one thing I'll say is already, I think what we're seeing is the expansion rates are much better for -- even at the dose level 1 or 112 versus 110 for instance, which is what we expect. And I think -- if that all plays out, you're going to have a greater AUC for CTX112 versus 110, and that should lead to greater cell killing. And I think -- tumor cell killing. So we'll see where the data plays out, but I think we're feeling very bullish. I think the key question for us is how does CTX112 positioned relative to 110 when we get to the end of the year? And do you want to...

Is there still a bit like a bake-off between the two or not?

It's not a bake-off per se. But right now, what we said is 110 is so advanced. If you get to the market, the math works for us. We launch with 110. It's a very -- it will have a place in the market. It's not going to be a multibillion-dollar product that could easily, I can see line of sight to $1 billion product there. But in 112 is a lot better, let's say it's much, much, much better than 110, then we do have a question that we have to face. Whether we want to do both or not. But at this point, we're all -- we're moving forward on 110. We're moving forward on 112. And once we have data for 112, we haven't said where we're going to show the data. But overall, I do feel like that could make a huge dent in the CD19 space, even if it's very crowded. In fact, ever since the Carl June papers come out, we've had inbounds from every pharma company that's in the space saying, we're very interested. Tell us what you see from your data. And even if it's very early data, we want to see what the PK/PD data look like. And ultimately, I think even had one pharma company say, we think ultimately, the best model for lymphoma, even in front lines may be a allo CAR-T to debulk the cancer, followed by 3 or 4 rounds of either antibody or bispecific, and that's the cure in the space. So the space is still very dynamic. I think, sure Yescarta has done really well with in the academic settings. But in the market will transform over the next 3 or 4 years as well.

So what -- is there some interplay between the Regnase-1 and the TGFBR2 that's causing this enhanced persistence as well as? Are they -- do these targets, do they interact in some way? Or is there some connection biology? Or is it just an observation from the screen you did?

I think the potency, there is an interaction. I think what you're having is the way Regnase-1 works is there -- it's pro-inflammatory by changing the expression of cytokines. Certain cytokines are overexpressed that allow the cells to be more pro-inflammatory. But at the same time, it's a transcription binding protein that reduces the differentiation of these cells into effector type. So you have more central memory phenotype. The TGFBR2 works by preventing TGF-beta from reducing the efficacy of the CAR-Ts, right? So it's a complementary mechanism but somehow that combination seem to work the best. We tried other combinations. We tried to Regnase-1 with PD-1, for instance. We you try to Regnase-1 with COX-1 or some other combinations. But this combination actually seems like the best from enhancing ultimately what the AUC is. We saw a very strong correlation with our 110 trials in terms of the amount of effector cells with the number of tumor cells and that ratio mattered. But also, we saw that our cells, the effector cells with 110 around the 12 or 13, there were -- cells were around, but their potency became less and less, right? So it's not the persistence in terms of elimination by the endogenous immune system was less the constraining factor. It was more the cell exhaustion. And if you can have these cells persist for a month, but still be cranking at the full effect, the entire time, you're going to see a multiple improvement in AUC and consequently see a multiple improvement in cell killing of the tumors.

Remind everyone what did you show for the 6-month CAR rate for the CTX110? And what do you want -- how much better would it need to be with 112 to pivot for that?

Well, I think where we are is, our single-dose CTX110 6-month CAR rate was about 20%. So 1 in 5 patients have a long-term durable response. I think that's where a lot of the [ a la carte's manned up ] in that zone. We did start an [ arm ] where we have 2 doses of CTX110. So we do 1 dose and then a month later do a second dose. And our expectation was that it's going to improve the 6-month CAR rate. We don't know about how much it will improve it. So that's the data we haven't disclosed yet.

That's coming up later this year?

That's coming up. And then we have 112 where if that ends up in sort of 30% range of 6-month CAR rate, that's equivalent to [indiscernible] or better and close to Yescarta. And that will change, in fact, better than the bispecifics. So that definitely changes the equation. So we're going to be best-in-class in that space beyond the bispecifics. So again, nothing we need to think about right now. We're just right now operationally focused on both assets and moving as fast as possible. And then as we get to the end of the year and early next year, we'll make that decision on how quickly you want to move both and what pace we want to move both of them forward in.

Okay. And I want to make sure we get to the other topics, but just on the renal cell program, the same 2 edits but that was not accidental that they seem to be the best choice. What's the plan there for 131 to...

We're dose escalating in solid tumors. In fact, TGFBR2 edit plays an even greater role in the solid tumors. I think a lot of what happens in the tumor microenvironment is TGF-beta-related suppression of the CAR-Ts and which is why auto CAR-Ts have not worked in solid tumor environments. There have been trials that have been done. But I think if you can prevent that TGF-beta-related suppression of these CAR-Ts, I think you could see a very strong signal. In fact, we saw responses with CTX130, which is the first-gen program in solid tumors. This is very encouraging. And if you can actually ramp that up and provide a greater exposure of the cytotoxic CAR-Ts to the tumor cells without suppression, I think you could see a longer activity.

And that data is when we're going to get that?

In some point next year. And I think solid tumors, the staggers are a bit longer. It takes a little longer to dose these patients.

112, next year?

112, I mean -- I think 112, I mean we're seeing the data, we just haven't made a determination when we want to show the data. I mean, in fact, I think are we going to provide some color commentary on PK/PD data? I don't know at this point. We are accruing data at this point.

All right. Do you want to make a decision? Is there a point where you want to make the decision between 110 and 112? Like is that...

Not at this point, we don't need to make any decisions. I think we're pushing both forward. We're built the organization in that fashion in an efficient way. Now I think we don't want to do it inefficiently. But at this point, playing in that franchise gives us that advantage of having both drugs there.

Okay. So let's talk about the diabetes regenerative medicine franchise. First of all, just give us the history there, like how did this even happen. How did it start? And it was ViaCyte and then it was Vertex and now you're at, right? So just...

So 20-year history in a away, which is nearly 23 years ago, the Edmonton protocol was discovered, which was the fact that you could take cadaveric islet cells, assuming the refresh cadaveric islet cells, and you can put a severe type 1 diabetes patient or immunosuppression and inject these cadaveric islet cells into the patient and typically to go and find their way into the liver or some other place that you could have effectively, because these patients can be -- there diabetes could be well controlled enough. And some of these patients actually still 20 years later, are still well controlled, but the downside is they to be an immunosuppression.

Is that -- was that all -- like all 3 variants of the islet cells like the alphas, deltas, the betas? Or was it a specific?

It was mainly betas.

Mainly betas. Okay.

And I remember in 2002, it led a lot of companies to pursue this field. I mean, J&J said they're going to make a big bet in this in 2002. In fact, all the way to build all in time. And what happened then is people realize that working with -- cadaveric cells are very hard to find. So people try to say, let's create stem cell-based products that do the same thing. It all turned out that embryonic stem cells are a little harder to work with. I think at the time, so it took some time. There were obviously some changes in the political environment and some of the big pharma companies dropped out are working with embryonic stem cells and it's also not scalable to do. And then we hit in 2006 to 2008 time frame, people start working on IPS cells and we had major breakthroughs. And all of a sudden, IPS cells became a reality. Right before the big recession, there's so much excitement about IPS cells because if you walk around Japan, there was all sorts of trials ongoing, in artificial corneas, in artificial readiness, because you could take IPS cells and make any organ that you wanted, you can differentiate them. But it turned out that it was also, again, not scalable because it's very expensive to do, because you have to put people on immunosuppression and you have to do it on an individualized basis. Otherwise, you can do it person by person, which is very expensive. And that's where CRISPR comes into play. If you can use CRISPR to edit these IPS cells to make them immune -- stealth to the immune system, then you have a very scalable platform. And if you ask them some of the leaders in the IPS space, they'll say the best thing to happen to IPS space is for editing to come into play. Similar to what happened with iPads and PenPads. In the early '90s, there was a product that was called a PenPad that did everything that an iPad can do now, but the WiFi was not very prevalent, apps were not there, you couldn't do anything with it. So you had this great technology like that was not very usable. IPS is the same way. IPS technology is amazing. It was there, but it's not been very scalable. So that's the fundamental thematic that we're making. So in 2017, we started looking at different options. In 2018, we did the deal with ViaCyte that had been working in the space for 18 years with type 1 diabetes to make -- use the stem cells to make pancreatic endodermal cells or precursors of basically the beta cells. And you -- they had already done trials to put them in a device and implant them into patients with immunosuppression. And our immediate response was let's edit them to make them stealth, so you don't have to have immunosuppression. That's a trial we're doing right now with 211, VCTX211. We've done a deal 50-50 deal with ViaCyte, but ViaCyte got acquired by Vertex, so now we're a 50-50 partner with Vertex in this space. Now it turned out that Vertex also had a parallel bed with the same thesis around IPS cells in editing and it acquired a company called Semma, they had a different cell line they were also progressing. So what we thought was actually may benefit everyone for us to take our -- license our CRISPR technology to the Semma cell line as well with Vertex and that's what gave us the deals we announced this year, which was in early part of the year, we announced the $100 million or so deal with Vertex. We had that $70 million milestone coming recently. That was 0 milestones, and then we have royalties associated with edited cells from the Semma cell lines. So we have some portion of participation in the Vertex efforts independent of us. And then we have our 50-50 program which we're pursuing for Vertex. So at this point, from a Vertex perspective, they're pursuing -- they're pushing everything forward. For us, we'll see where the 211 data come out. And that could be a transformative product if it works. But ultimately, I do think, given the data that Vertex has shown to date, the notion of edited cells in diabetes is going to be very powerful. In spite of everything you're seeing with GLP-1s and the transformative data with Ozempic and [ Mounjaro ] and everything else, it's still going to be a very important product.

What's the difference in terms of the edit with the Semma cell line and the one coming from ViaCyte, how are they different?

Edits with the Semma cells have not been disclosed by Vertex. So Vertex owns those edits and they will -- at the right point...

So you just license that technology. But it's not a 50-50.

We licensed technology. It's not -- all around it. That one, it's mainly a Vertex program. We have royalties and milestones associated with it. The program that we've developed, VCTX211 with the ViaCyte, is a 50-50 program.

And what have you disclosed what that edit is...

With 211, yes, we disclose all the edits. The immune cloaking is a combination of a beta2M at it, which we saw a benefit in the IO context. We do insert a PD-L1, which also clocks the cells and an HLA-E that also further prevents NK cell attack. So those are a set of edits for immune evasion. We actually have a couple of edits for -- to reduce cell stress. A lot of this -- there's a lot of endoplasmic reticulum stress that happens in these cells when they're cranking out of insulin. And we made a couple of edits -- these are lesser-known genes, 820 and MANF that prevent that sort of ER stress in the cells and increases their liability.

Okay. So what fold -- what factor are you increasing the beta islet cell population in the pancreas relative to the patient with -- in a type 1 diabetic, there's really none, right? There's hardly any?

Hardly any. There's -- when you put these -- we haven't disclosed the exact cell dose, but you're looking at close to 1 billion cells, potentially that you may need to replace to get reasonable efficacy from a C-peptide levels. So the way you measure the insulin levels that's been produced is that these endogenous cells produced proinsulin that's cleaved to make insulin and the byproduct to C-peptide and you can measure the C-peptide because all these -- a lot of these patients actually are taking insulin injections. So you don't want to measure that insulin that they're taking. But if you look at that requirement, I think you could be putting up to 1 billion cells in these patients. But again, 1 billion cells, because while the pancreas are sort of banana-shaped organ with a lot of cells, the beta cells itself is a very small fraction of those pancreas.

Yes, yes. Okay. And then as far as measuring the C-peptide, you use the glucose tolerance test to do it in a controlled way? Because sometime -- I mean because I cover this other company called Biomea which you may have heard of -- and they are the -- when you measure -- some of the diabetes experts have said that when you measure C-peptide, you need to really do it in a controlled way. Otherwise, it can be all over the map.

Yes. But I think in these severe diabetes patients that we're looking at they have no C-peptide, right? So it's really low. So the baseline -- you don't get that -- the signal to noise is a greater and you've seen that in a Vertex disclosed data, C-peptide data for their first couple of patients with their Semma cell line. And it's very clear that they go from baseline of nothing to do something. So it's different from the Biomea concept situation where there's some baseline in the C-peptide production.

And now this whole -- the cells are implanted in a in a basically half a credit card size wafer, right? Just explain why that's -- I mean, if they're immune privileged and cloaked, why do -- why can't they just be injected without this housing?

Yes. Yes. So I mean, that's a very good question. Ultimately, I think these cells will be just injected directly. I think our initial notions, we weren't sure what the regulatory environment is going to be. And if something goes right with the cells, do we need the ability to pull the cells out? If they're in a wafer-like device, remember, we were making these decisions in 2017, 2018 time frame, not moving how the regulatory -- buyers are going to espouse an approach like this. And we thought maybe having them in a device would give us enough information on immune evasion, but also if something goes wrong, we're able to pull it out. So okay, it provides safety. That said, I think where the field is coming out now is to say you just want to have fully differentiated cells that can be directly injected into the patient. Now most of the injections are done in the portal vein, so the cell is going to the liver and lodge themselves over there. But there may be other ways of injecting these cells as well. I think that's probably where the field is going to move to eventually.

All right. Let's make sure we touch on the in vivo programs. So you want to just review what are the time lines there to getting into the clinic. I think it's soon and the 2 programs, the ANGPTL3 and the Lp(a), how are they similar or different in terms of what you're targeting?

From a technology standpoint, they're very similar. I think eventually, this is a very scalable platform. What's been figured out over the last 3, 4 years is delivery of CRISPR/Cas9 to the liver is something that I think has been figured out across multiple. Obviously, Intellia showed the early data, but other companies have also learn how to make the LNPs. And I think we have best-in-class LNP platform with together with mRNA because that's how you deliver the Cas9. And I think we've shown very good data in primates, and we hope to get into the clinic this year. So we said that we'll be in the clinic with the ANGPTL3, one of the programs this year and then follow that soon with the second program next year. The two indications, while it's the same technology platform, and all changes the guide, for instance. That's what makes it very scalable. We can do 10 programs for gene knockout in the liver if things are working and very quickly follow the RNAi route essentially and do a lot of those targets that they're targeting, but better because the onetime treatment, you won't have any of the effects or safety issues with repeat administration. That's with the first 2 indications. The way you look commercially at ANGPTL3 versus LPA is very different. ANGPTL3 is more of a rare disease type of play. In HoFH a lot of these patients are refractory even to PCSK9 therapies. And the only option that would have is something that's like ANGPTL3. So that's the population that you initially did some of the trials. There are patients who have very high triglycerides. In fact, triglycerides over 1,000 sometimes where there's no other options and ANGPTL3 is known to reduce triglycerides. So that's another patient population.

So there's all these niche patient populations which you develop it in without having to think -- without having to do outcomes trials and get approval off of that.

So that's why we picked ANGPTL3 first. That's sort of more of a rare disease type setting where we would -- we can do all this on our own. LPA is a much bigger market. I mean there's like 11 million patients in the U.S. alone that have elevated LPA levels. And it's -- we're waiting for the data for Novartis next year. But if it's a clear outcomes benefit is demonstrated or a clear correlation between high LPA and advance as demonstrated independent of LDL, for instance, then this becomes the hottest target in cardiometabolic medicine in the whole space. And we'll be in a pole position there. Odds are, we would require the outcome trials -- outcome trial and everything else ultimately to make us a broad-based therapy. But again, we can define populations that are very, very high LPA for instance. There's the number of stories where the investigators have told us where there's people in their late 40s or early 50s, marathon runners are very fit, all of a sudden have a severe cardiac event because of high LPA where everything else is controlled. More and more of this emerging. As we...

This just explained to everyone the basics is that you're basically aggregating the LPA gene or what exactly?

LPA is a carrier of -- people simplify this to good cholesterol and bad cholesterol, right? ApoBs is really the intrinsic bad cholesterol that you're measuring, that's carried in different vehicles. It's either carried in a LDL vehicle, it could be carried in a triglyceride like particle vehicle or it could be carried in an LPA vehicle. The LPA vehicle is basically Apolipoprotein A that interacts with Apolipoprotein B into this domain. And that can carry a lot of ApoB in the bloodstream. And so it's independent of -- you could have very low LDL but still have high LPA that carries all the ApoB. And so it's an independent risk factor is what's determined. It's emerged over the last probably 6 or 7 years is a very important target. In the U.S. practice, it's still not a typical lab measure. You go to the doctor, they don't measure your LPA levels. In Europe, they're starting to do it. If you go to Germany, for instance, they start measuring LPA and it's going to be demonstrated as a very important predictor of cardiovascular outcomes.

What if this, the Horizon trial, I think, is if it doesn't work, is that going to change your thinking on this target or something cross that bridge and you get there.

Of course. Well, I mean, we will change our thinking, it would mean that the cutoff that they've used, I think, is about 150 for the LPA level is high. Is it that? Or is it 400? What is the population where LPA levels are so high that it overcomes any of the effect of LDL? And it's by itself the single biggest contributor of risk, that's something that those trials, both the Amgen and the Novartis trials were informed. They have slightly different cutoffs. But subpopulation analysis from that trial would be really important.

Okay. All right. Thanks, Sam. This is a lot of fun. Thank you very much.

Yes. And if I may do one last plug for CRISPR, actually, I know there's a lot of talk about next-gen gene-editing. We have -- people are saying CRISPR 2.0, 3.0, but we're doing all forms of next-gen gene-editing. In fact, you won't believe the type of edit we can do now to essentially gene right, and that's something that's a big focus for us. It's set up in California in the Mission Bay. We wanted to have a separate pool, a company within a company working on it. And I would say at this point that we're probably doing some of the most advanced gene writing technologies out there. We just don't talk about it until...

You're saying, you can do all scales from single base all the way to entire genes? And everything in between.

Absolutely. Yes. And ultimately, we want to target -- inserting entire genes that are 3,000, 4,000 base pairs. And so we're getting improving efficiency along all those fronts. And it's a huge part of my focus, and I think this is all going to change this -- the number of diseases that we can address with gene-editing from tens to hundreds, if that all works.

Well. Okay. Awesome.

Wonderful. Thank you.

All right. Thanks.