Okay, thanks everyone for joining us for the next half hour. My name is Dae Gon Ha, one of the biotech analysts here at Stifel. Joining for me for the next half hour is Prime Medicine, and specifically, Keith Gottesdiener, President and CEO. So Keith, thanks very much for for joining us. Keith, maybe I'll just turn it to you for just a brief intro of Prime Medicine, how you guys are distinct from the CRISPRs of the world and the base editors of the world, and once we have that, we'll jump into Q&A.
Okay, and I do apologize. After today, my voice is a little hoarse. It's just running out, so hopefully everybody can hear me. You know, the way we think about things is in generations. You know, CRISPR came along with an amazing technology. As time went on, people started to understand what CRISPR could do, but also what it couldn't. CRISPR seems, in general, to be very good at inactivating genes. Base editing, a second generation in our sort of conception of things, was a place where people first started being able to correct genes or inborn errors of metabolism, and it also was probably a little bit safer approach 'cause it doesn't cause double-strand breaks in the chromosome, and hence, you know, when it actually does the gene editing, there's a little less muck left behind.
That much may not always matter, but in some cases, maybe it might. But it was still limited in terms of the corrections that it could do, and in fact, sometimes base editing is even more effective at almost doing CRISPR-like things, but using a change in one of the base pairs, the limited base pairs it can make, to sort of inactivate something. For example, in some of the sickle cell programs that are ongoing at a company like Beam, or even the work that Verve is doing as well. Prime Editing was really different. The purpose was to very broadly be able to edit just about any kind of genetic edit that one wanted to make. At the beginning, we looked at single base pair mismatches in rare diseases.
Remember, if you have four base pairs, there are 12 ways they can mismatch if you do the math. Base editing can correct four of them. We can correct all of them. We correct small insertional and deletional mutations. A deletional mutation might be Del 508 in cystic fibrosis, the commonest mutation there. Tay-Sachs is an insertional mutation. But as time went on, it became very, very clear, fairly quickly, that Prime Editing can do much more. We can literally lay down 100, 150 base pairs of absolutely new DNA in a spot.
So for what I call peanut butter genes, places where there are mutations scattered literally throughout the genome, we can find a spot, and we can say, for 150 base pairs, it doesn't matter whether you have, you have mutation number one, I have mutation number seven. Everyone gets a perfectly clean, absolutely normal piece of DNA, and so we can very efficiently deal with many different mutations with one Prime Editor. We next were able to find that we could delete out of the genome very large pieces, very precisely. So the place we've shown that are in repeat expansion diseases like Huntington's disease. We literally can go in and just cut out the pathological repeats and leave the normal number of repeats behind, essentially doing a genetic cure. And we've talked since the beginning of this year about what we call our PASSIGE technology.
It's something we've been working on actually under the radar for literally since the company began, but it's a way to combine Prime Editing with recombinase to actually put in very large pieces of DNA, multi, multi kilobase pieces of DNA, but very, very precisely at a single location in the genome. So in the end, Prime Editing can really do a correction of just about any type of mutation or error that one would like to do.
Mm-hmm. So that's a good segue. Let's talk about your pipeline. You have quite a bit of possibilities, as you pointed out, different silos of verticals that you could go into. Sort of given the breadth that you have, but also bearing in mind sort of the realities of what you can do and cannot do, given the resource constraints, how are you thinking about your overall pipeline? Which verticals are you more focused on versus somewhat less prioritized for the time being? Or how are you thinking about your overall strategy?
It's a great question, and it's one we're still grappling with, and I'm certainly not gonna have a perfect answer today to talk to people. You know, we started out with a pipeline of 18 programs. We picked a very broad pipeline in many ways. The idea was really to test where Prime Editing might work most efficiently and where it might not. So it has a variety of different type of mutations to correct. It works in different organs. It matches with different delivery systems. Some of them are things where there's a very clear path to the clinic and past regulators. Some are a little more complex, but maybe higher value overall.
We expected that over the course of the last couple of years, many of those programs would declare themselves, i.e., these are places where Prime Editing probably isn't so easy to do, and others where it was very, very easy to do. We've... As a result, we have a really serious problem, but it's a good problem. The problem is pretty much every single one of those situations that we've looked at, all 18 of those programs, are still essentially moving towards the clinic, and if we had, as you pointed out, the bandwidth and the resources, there isn't a single one we couldn't be moving very rapidly to the clinic. It's a great problem to have. In 2021, it was an even better problem to have when people looked at companies a little bit differently than today.
Today, obviously, in this context, it's a much more complicated problem, and frankly, it was a problem in 2021 as well. People just had a little more rose-colored glasses on.... When we think about the things we do well, it's a pretty good list. We know that we can do some extraordinary ex vivo hematopoietic stem cell approaches. We know we can do some pretty special liver delivery with LNP approaches. We presented just recently a lot of in vivo data in our eye, and we have it in our ear. Some of the most interesting things are in the brain, repeat expansion diseases and other things like that, or potentially targets in Alzheimer's disease. We have really interesting things happening in the lung and very interesting things happening in muscle, and I'm probably forgetting a vertical or two.
I think what we're probably gonna do is try to figure out, because we started with this broad pipeline, which of those programs we want to take towards the clinic over the next year or two. My guess is we'll pick three, four, or five of them to really move forward over 2024, 2025, and 2026. But the important point is, some of the best things that we're doing are just a little bit lagging behind. Today, we have ways to get to the brain, but I'm not totally satisfied with our delivery to the brain, and we have some wonderful programs there. What also complicates it in a good way is there's a tremendous amount of strategic interest in our pipeline. In two ways, some people have come and looked at our pipeline, said, "You have 18 programs.
We'd like to take three off your hands." Okay, that's exciting. We're having conversations like that. Other people have come and have made my problem a little bit worse. They've come, and they've said, "We have problems we're trying to solve. We think Prime Editing is the solution. It's not on your pipeline. Would you like to work with us on that?" So as you can imagine, as an analyst, telling people our pipeline has now moved from 18 to 21 because we're doing BD is not a move in the right direction, but it is still very positive for the Prime Editing, and I think even for the company, just a little difficult to deal with.
And so because those things are all sort of hitting and that interest we're sorting out, some cases literally as we speak, I think it's gonna be a little bit more loosey-goosey than probably people would like to see for the next six months till we really figure out where the places are we're gonna partner, where are the places we're gonna out-license, where are the places we're certain we can move these things forward expeditiously, and leaving a little room for some of the more exciting things that, you know, are not, pun intended, ready for prime time today.
Mm-hmm.
It's gonna be a little bit of a complicated dance in the next six months.
So we'll obviously stay tuned to the next six months for any updates on that, but you have already announced the Simcere deal.
Correct.
I guess maybe using that as an example, what excited you about Simcere, the terms of that, and to what extent is that a blueprint of what the subsequent transactions or BD activities may look like?
Let me start with the last part. It is definitely not a blueprint-
Okay.
for what we're thinking about in the future. Pretty straightforward. We may do another deal like that. That was technology, a matching technology deal. I'll explain it in a minute, but the idea is we are looking for places where something matches with Prime Editing and makes it synergistic more than the whole. For example, a great delivery method today that could get into the brain, every part of the brain today, that would match with Prime Editing, would be something I'd love to have. We're working on it. We're talking to other companies. If that came forward as a BD deal, okay, it wouldn't be the typical deal, but it may, in fact, add great value to Prime as a company, 'cause some of our biggest indications and the most exceptional ones are in the brain.
Simcere was a good example of that, and then I'll talk in a minute about what a more typical strategic discussion we're having today might look like. Simcere is a wonderful example 'cause they came to us for a reason. They were developing something, something called a shielding approach for hematopoietic diseases, and the idea really was is these are diseases that are treated by bone marrow, but in order to make a bone marrow work with these new cells that are coming in, you really have to clear out the old bone marrow, and to do that is very, very toxic to patients. So they're trying to find new ways to do that in a less toxic approach. And what they came up with something called shielding.
They have the antibodies to the stem cells, these critical stem cells that make all of your blood cells. And the idea really was is if we had an antibody that could kill all those stem cells, wouldn't that be great for AML? 'Cause then we could put new cells in. How can we make this all work together? So what they did is they, with their antibody that they used to kill all these cells, they found mutations that protect cells from that antibody. So they came to us, and they said, "What we'd like to do is we'd like to use this antibody to clear out all the old cells. We'd like to license Prime Editing to make the correction in these cells that'll make them resistant to the antibody, and then we can put that together." Great.
At one point, we clear out the old ones, we do the new ones at the same time. Everybody knows who's supposed to survive. Old die, new live. We pointed out to them that they're really thinking small with that. Now, they are still developing that for AML, and we've licensed that to them, and we get wonderful royalties and milestones. But what they missed is we can multiplex edit. If we're gonna make that edit to shield, we can also do a correction. So any kind of hematopoietic stem cell disease, we can do a bunch of edits, and at the same way, we can do something really exciting, like for hemophilia or thalassemia, and do the edit, and then people won't have to go through a toxic bone marrow transplant because of that. We probably need-...
6-12 months to really put these technologies together to make sure everything's working the way we want, even though the data today looks very, very exciting. If we get to that point, there are all kinds of diseases where people don't want to get a bone marrow transplant to solve, which we can now correct. A good example of that might be sickle cell disease, where I know many people wonder how many people are gonna get a toxic bone marrow transplant to get the CRISPR-Vertex correction. I don't know what that number is. I can't even hazard to guess, but think how much greater that number would be if you didn't have to make people go through a toxic bone marrow. They came in for a day, they got a shot of antibody, got the new cells at exactly the same time.
They were never neutropenic. They were never immunosuppressed. That's a perfect response overall, and with the Simcere, I think that's possible.
Mm-hmm.
A more typical thing would be to pick one or two of our products in our pipeline and to do either a co-co type of deal or for someone to outlicense them and just say, "We're gonna develop them and give you milestones and royalties.
Right. Just as a clarification, you do have that partnership with Beam on the sickle program, but just wanted to... I think this is a natural segue into your 359 program, the CGD program, the hematopoietic stem cell-targeting edit.
Correct.
Walk us through sort of the rationale for choosing this, because this isn't a typical indication, and, you know, the pushback we've gotten is that this is too small of a marketable indication. So how did the economics come into your overall decision-making here?
So when we picked our original programs, economics was not our number one factor overall. It was important, but the idea really was we need to make sure that Prime Editing actually works, and so we picked places where we thought they were big enough to be economically attractive, but clearly not indications that anybody should value the company on. CGD probably, you know, is hundreds of patients at best in the U.S., maybe an equal number in Europe. Hard to judge. When we look at the economics, we take a very hard-nosed opinion. In this sense, we look at it as things that will fund the development of the company, not necessarily return things to investors.
So if we end up treating 100 or 200 patients in the three or four years after we come out, folks like you have estimated that the price for a product like ours could be $2 million-$3 million. The margins still remain fairly high. That's the kind of information or kind of finances that could keep the company running and really fund all of the work we need to do till some of the bigger things come forward. We didn't pick CGD as the lead. We picked a bunch of programs, a whole wave of them, four or five. This is the one that really rose to the top because everything clicked in this disease. It just went as smooth as silk overall. Now, I'm kind of glad we picked it. Not picked it, we ended up with it.
A couple reasons: It's a huge unmet medical need. It's small. It's wonderful to have a regulatory discussion when the FDA recognizes this huge unmet medical need, and they realize that when you do this, you're not gonna treat thousands, tens of thousands, hundreds of thousands of patients with that. That isn't so good for investors, but it's exactly the kind of thing that warms regulators' hearts when a new technology is going into the clinic. It also helped that these patients get an allogeneic bone marrow transplant. It's the only cure, and we're offering something that's much less toxic to them and potentially curative. And so many of the patients already are primed, this case, no pun intended. They're primed, basically, for, you know, for a bone marrow transplant. And when we off...
Put in front of them the idea that they can get a simpler, easier, less immunosuppressive one that may cure them. It's the kind of thing that investigators think, who care for these patients, that people are gonna line up rapidly to get that. We can't do too many of these programs that are so small, but I think we could do one or two of them to move things along fairly quickly, but we're gonna have to back them up with programs that matter more, Wilson's, the repeat expansion diseases.
So on the development side, I think you look at smaller indications as benefiting from a trial regimen that's probably not as extreme as some of the, like, cardiovascular disease, for example.
I would say a lot less extreme.
A lot less. Correct. But I guess people always want to look for sort of the precedent of how to think about the overall risk profile of such programs. So we do have an ex vivo gene therapy program from Orchard-
Mm-hmm.
That was in the X-linked version of CGD. How are you thinking about the trial design and the regulatory path vis-à-vis what Orchard was able to do?
Well, Orchard took that to a certain point and stopped.
Correct, yeah.
I'm not sure anyone's 100% exactly certain what the full set of reasons were that they stopped it. I think part of it were business reasons as much as anything else. But in the end, what they were able to do, the investigators who have worked on that think that it generally showed some efficacy as well, and they think that they were on the path to helping the FDA and how to think about those things. I personally think that there's a little bit of precedence to bring forward, but not a huge amount in many ways. In some ways, it's easier to explain to the FDA what happens if you have a genetic cure. Essentially, we're bringing people back to normal and what you might expect with that.
But on the other hand, I think Orchard set the paradigm that this is a place where a surrogate marker might be very, very impactful overall. And certainly, we believe that the clinical trials in this disease will be very small, possibly long in terms of duration, two years, for example. Hard to know, but they'll be very, very small. There'll be the ability to go from a phase 1, 2 type trial into whatever is considered pivotal. Certainly, we believe the number of patients that we would have to treat, you could count on your fingers and toes. I'm not saying the FDA has given us any feedback. We haven't talked about any of our FDA interactions or even if we've had any on that, but we hope that that's the case.
Keep in mind, I came from a company, Rhythm, where that's exactly how we did three indications. Clearly, there are precedents in the small molecule or protein space where that could actually occur, and we hope the same thing will, you know, actually happen here. That's why the ability to use these programs to fund the company become important. If we, in fact, get into the clinic in 2024, we haven't promised that yet, it's the kind of situation where three years later, you could potentially have an approval if everything were to go well.
Mm-hmm.
I think that's really optimistic, but, you know, when you start to think about that, that's a point where, you know, having $200 million a year from a small program could be very, very impactful in our financial situation.
I see. So looking ahead, what are some remaining steps or the gating factors, as investors call it, for this lead program to enter the clinic and for us to envision that three-year-
Yeah
-time horizon?
There's a number of things you have to do to get your IND passed, and pretty much we've said all of them, we believe, are on track. You have to have a good preclinical models. We've shown that data numerous times at conferences. You need to have a toxicology package that passes the FDA. We've pointed out that the toxicology package is essentially an identical study to what we need for the preclinical proof of concept study with a few bells and whistles. We've also shown that data on numerous occasions, so fairly de-risked, but we have to do the work. You need to have a very robust off-target profile that's clear from the recent, you know, CRISPR-Vertex approach. We have it. You've seen some of the data, but our off-target profile today looks outstanding.
But we haven't shown all of it, but it's something we have a few pieces to fill in, but it's moving along really quite well. You need to be able to manufacture GLP, good laboratory practice, and good GMP, good, you know, good manufacturing practice product, in order to go into a clinical trial. We haven't talked about what the status of that is, but we have one of the world's best CTOs, truly, you know, Ann Lee, who's really is the doyen of really CMC and has worked at a number of companies, and most of the folks who work at big companies are people who worked under her in previous years. She's just extraordinary.
Last thing you need is a clinical trial that the FDA feels is appropriate overall, but to a great degree, that's not really a gating factor because you get a chance to have some interactions with the FDA about that trial, and if you don't listen to them, then you're really not so smart, but in the end, you almost always come up with a trial that makes sense. We're really on track with every single one of those things.
Great. Maybe on that off-target, just wanted to highlight that because your technology is still a derivative of CRISPR-Cas9, I think Exa-cel provides a lot of learning or lessons. Can you just remind us what off-target editing are, are you actually implementing? And since you're adding that extra layer of Prime Editor versus just a CRISPR-Cas9, how are you envisioning providing more data or, you know, preemptively covering your bases, if you will, in case FDA asks for it?
Well, first of all, we have decided that one of the strategic imperatives of the company is to really demonstrate an incredibly robust off-target activity. Part of the reason is for a variety of practical reasons, data-driven reasons, and mechanistically driven reasons. We believe we have the potential to be the best, truly best-in-class off-target. We have work to do, but it's important to realize that at this particular point, using assays that we believe are ready for IND type assays, you know, the ones that we're gonna use and we believe will meet the FDA requirements for, you know, for submission for the IND. We can't find an off-target edit in any of our programs that are currently moving towards the clinic. No, also no rearrangements, no translocations.
That's very the package we put together is very different than what you saw at the Exa-cel advisory committee meeting. That package, no reflections on Vertex, it's an older program or CRISPR. That package is five years out of date, okay? It uses two orthogonal assays to make predictions where an off-target might occur. We're currently using four or five. It only did effectively one other assay to look for rearrangements and translocations, which was a karyotype. Frankly, you could rearrange whole parts of chromosomes, and you'd never pick it up on a karyotype. It's completely insensitive overall. If I had been the FDA, I would have had a very different discussion on that. I'm not saying that any of those gaps would have changed the outcome, but it's a very out-of-date program.
We're working with something like 10-15 assays that we work. Many of them are variants of what other people are doing, but we also have assays that look at Prime Editors a little bit different, so, for example, we look as if we were a double-strand breaking company, just to show that we don't find any of the things that would happen if those were to occur. We also look, because we nick or to basically break one strand, we repeat variants of those assays that are used for nicking. That's a lot of extra work, considering nicking is so physiological. To actually prove that nicking doesn't matter, you have to go much deeper into the genome to really show that. We have a whole variety of assays that look at translocations and rearrangements.
There is one assay that is specific to Prime Editing besides the nicking, which by the way, we would share potentially with Beam, though I don't think Beam has outlined exactly what their assays are yet, so I'm not 100% sure. And that's-- we use a reverse transcriptase, and we have a special assay that looks to see if that has any untoward effects. We haven't seen anything, but then again, we wouldn't expect it. Your cells are filled with reverse transcriptases. You know, we have retrotransposons and retroviruses that are working every day on, on piece... you know, in the cell and in the genome. So though there's a huge amount of RT activity already there, we just show that the pattern with and without Prime Editing looks essentially identical, and that's just, again, a reassurance.
There are other assays in development as well, but our goal is just to really come forward with the most robust off-target activity. Frankly, I think we're greatly exceeding what the FDA would need. I can't promise that exactly, but I believe it. But our purpose is really aiming at the wider world. We think if you really want people to believe in gene editing as a therapeutic approach, you got to be able to say to them, "We're going into your genome. We're making an edit, where it's the therapeutic edit. We're not doing anything else. We're not going to muck with your genome.
When we get out, you either get the therapeutic edit or nothing happens." Now, I don't think we're ever going to be able to say absolutely 100% nothing happens, but to the limit of the detection of these assays, because remember, many of them use methods like next-generation sequencing, which have their own error rate, so there's a limit to how deep you can go. We, we just don't see anything at this particular point.
Got it. Is there any question from the audience? Yep.
I was just curious, when you feel like you'll have sort of clinical proof of concept for one of your programs? Do you have visibility into that?
So the question was, when are we going to get this, you know, proof of concept clinically for one of our programs? Yes, I do, and no, I'm not going to tell you. So in practice, sorry to be a little glib about that, we haven't predicted anything about our clinical programs besides saying, we expect that as early as 2024, we're going to file our IND. I should point out, by the way, we're not just filing our IND, but we're also filing in other countries as well. That's just our shorthand for that. We will probably update people early next year on when that IND is going to come, when we expect to dose the first patient, a little better idea about what our flow-through should be.
I think that was the purpose of Dae Gon's question, you know, trying to understand by his, what's gating the delivery, trying to get a sneaky estimate of when that IND is going to go in. At the moment, we're trying hard not to comment on that, but we understand there's a tremendous amount of interest, and I don't think it'll be particularly long. We just want to kind of get our ducks in a row and try to figure out what that flow looks like. I don't think it's going to be very long, though.
No, I understand. May I ask one unrelated follow-up? I guess,
Certainly okay with me.
Thank you. All right. Sorry, I've never got on one of these before. It should be-
Yeah, it's not on, I don't think, anyway.
Okay. Well, it's-
I'll repeat the question then.
Yeah. I don't have a soft voice anyway, so what I was interested in is if you think that there's applications for your technology within sort of tough-to-treat cancers like multiple myeloma in addition to AML?
So the question was, could it work in cancers, tough cancers, like not only AML, but multiple myeloma? Keep in mind that in the AML case for Simcere, our edit has no therapeutic effect. It just protects the cells that are supposed to deliver the therapeutic effect. So that's not a great case of whether we can work in cancer or not, as much as we can be a great tool to improve the ability for other approaches to work much better than they could. Cancer is a place we haven't focused that much yet. There are probably two reasons: one, the company, more by coincidence, doesn't have a tremendous amount of cancer expertise. I ran late development in Merck before Keytruda for almost a decade. Merck didn't ever have a cancer compound. I just am really not a cancer person.
None of our folks really are, so we'd have to build that up. I think the second reason is cancers often require a tremendously high percentage of editing. For many of the rare diseases we're dealing with, if you hit 50% editing or 60%, you're going to have a major therapeutic effect. If in a cancer, you fix 50% of the cancer cells in a tumor, the other 50% may grow out very quickly. So I think really, Prime Editing may be more an indirect advantage. For example, we're working in very innovative forms of CAR T-cells that may have a very great impact for cancer, because we can just do such sophisticated gene editing. But keep in mind that that isn't really necessarily the same thing as thinking we should be a cancer company.
Now, having said that, there are a couple of strategics, pharma companies, who have come to us to explore their ideas about how we could be used in cancer, and we're very, very happy to work with them. Some of them have some very creative ideas. Whether they'll actually turn out to be business development activities or not, I'm not certain.
... maybe a twofold last question, given we're out of time. Cash, which is what everyone is focused on in this market, but also when you think about insertion, deletion, and passage, which were all other arsenals that's in your toolbox, what's after that we should be really keeping an eye for?
What's the next one?
Yep.
Okay. Cash, we're about $160-$165 million. I don't have the exact number. I should. We just put out our 10-Q, but in the end, that number changed a couple of times, so it's somewhere approximately that. That'll take us, you know, till the end of 2024. So we have decent cash, but not outstanding at this particular point. In terms of what'll come next, very, very hard to predict. When you start to think we can correct any small mutation, we can lay down big pieces, we can take out pretty much any size piece we want, we can put in any about size piece we want, we can put it in very, very precisely, we can do it without off-target editing, and we can use a whole variety of modalities of delivery.
Myself, David Liu, our founder, many of the companies, we've really tried hard to think about what are we missing? There's a lot of little technical details around the outside, where if we could solve a teeny little problem, it would make this part or that part just a little bit easier. Frankly, we're struggling to come up with things that you would want to do, if you said to yourself, "What is Prime Editing missing?" And actually, David Liu often gives that in his lectures. He points out to people, you know, from gene therapy to CRISPR was about 40 years, and then we hit a bunch of very quick things, you know, with CRISPR, Base editing, Prime Editing. And I would point out there are another number of technologies out there people call gene writing.
I have to tell you, from the prime side, they look awful like prime editing to me. I suspect some of those companies are gonna run into our very, very, very robust patent approaches someday. But when you think beyond that, the extraordinary thing is there have been so many generations of improvements just in a short 10- or 15-year period of time. There's no reason to suspect that there's gonna be new breakthroughs. Could happen. There certainly will be slightly different ways to do what we can do, and frankly, I'm sure someone's gonna find one of those ways that works outside our patent if they work hard enough. But having said that, I don't think they're gonna have new modalities that we can do. It's more a matter of slightly different ways to get to the same point. Now, I could easily be wrong.
We may both read a new science paper next week, but, you know, when Base Editing was there, everybody could look at it and say, "This is what I want to improve, good as this is." And I think with Prime, it's been very, very difficult. A lot of words, a lot of hand-waving, but no one's really come up with anything that I'm aware of that really would be a major step forward.
All right. Well, I guess we'll stay tuned for your January update-
Yeah
... for what to look for in 2024. Thanks so much.
Thanks so much. Bye-bye. All right.