Good afternoon, and thank you for hanging around. Day three of Guggenheim's Inaugural Healthcare Innovations Conference. I am Debjit, one of the therapeutic analysts, and my privilege to welcome Keith Gottesdiener.
That's definitely close enough. Thank you.
Keith Gottesdiener, CEO and President of Prime, and Allan Reine, the CFO. Thank you both. Thanks for your time.
Oh, it's our pleasure. Thank you very much for having us.
Since Allan has come on board, he's taken 18 of your children and cut it down to three.
It's cut it down to six, but it still hurts, I would agree.
Let's talk about the first three, and then we can go to the next three. Why did you keep those three first?
I'm not sure exactly which three you're talking about. Our pipeline today has six elements, and you can tell me which ones you'd like to hear me talk about. We have two programs in chronic granulomatous disease, one called p47 chronic granulomatous, and one that is X-linked granulomatous. The first one is clearly ahead in the clinic. We have two liver programs. One of them is Wilson's disease, and there's another undisclosed program. Cystic fibrosis is another focus of ours, and then the collaboration with BMS that we recently announced. So I assume you'd really like to hear about CGD and Wilson's disease and CF among the three.
Because the others are one not disclosed, one.
Totally agree. I'm sorry. It's been a long day. I apologize just if I start with a little bit of a lighter tone. We came down to those things, not specifically to those things, but we went through a whole process. Actually, in a minute, I'm gonna ask Allan to talk a little bit about that process, 'cause one of the things he did when he came in. It's great to bring a CFO as a very definite viewpoint and really can add value to the company, and Allan certainly has. He looked around and he said to us basically, "I really think you need to be more careful about where you're allocating your cash and deciding what we're gonna do." So he helped us to do a process.
I think we had done quite a bit of the thinking already, but he really helped us do the process. I'll let him describe it in one second, and in the end, we looked at all the things in our pipeline. Previously, we had 18 things in our pipeline, and we asked the question, "What's gonna really add the most value to the company," so you wanna take a minute to walk through a little bit, Allan?
Yeah. So, you know, I think he's right. When I joined, you know, I think the company had been thinking about this, but it, you know, makes sense as you think about, you know, Prime at that point was only, you know, a few years old, and really there were so many places that you could take this technology. So the idea is let's really figure out everywhere this technology can work, and I think, you know, Keith likes to say this, and I agree, from seeing all the data that I've seen generated from this company, there really wasn't a lot of attrition. Everything they really put their minds to really did work in all these different fields.
But that did create the, the problem was there was, you know, really 18 programs within the pipeline, which, you know, for a company where Prime is today, you know, you really do at this stage really need to figure out where to focus. So the idea is to really put together what I would say a, what we called our value framework exercise that we went through to really look at all the different parameters that we can use to assess what the right investments or capital allocation strategy is for today.
So different things that we looked at without going through the entire process, really understanding, you know, obviously top of the list, what's the commercial opportunity for a program, what's the clinical tractability for a program that we wanna take forward, meaning can we get to true value inflection points in the clinic within a reasonable period of time. So critical to this is not just what we think can be successful, but it can, can it be successful in a period of time that can really impact Prime's cost of capital. We looked at obviously technical feasibility. So I think there's certain areas like the liver, and ex vivo applications where we know delivery has been, you know, quote unquote solved today, or we feel like there's high probability of achieving delivery.
And other areas like the brain, muscle, and others where, you know, we do think delivery will likely get solved one day, but we think, you know, there is, you know, significant risk today, which affects sort of what that probability is of achieving what I would say value inflections in the next few years. You know, there's some other parameters we looked at as well, but I think, you know, important is where do we really think we can build value over the next few years to ultimately build the company we're trying to build to then continue to invest in even more programs over time. So we're not forgoing all, I'll call it sort of future innovation.
We're still investing in innovation, investing in delivery, investing in other things, but from a program perspective, really investing in programs where we think we can really generate value for the company to be able to expand things out further in the future.
I think Wilson's disease is a perfect example of that. We've had a variety of liver programs in our pipeline. We started with a small one, glycogen storage disease 1b. It became very clear that all the work we did for that was immediately transferable to our other liver programs, what we call a modular level, because we basically use all the same reagents, the same delivery techniques, et cetera. Here you had a small program, and we, with that transfer, we were able to bring up Wilson's really right behind it. In the end, we paused that program, even though we probably would've filed an IND in 2025 to bring Wilson's forward. We were filing an IND in the first half of 2026. You know, that's an indication that had tens of thousands of people potentially in the U.S.
So these are the kinds of things we thought about when we put that list back together again.
Got it. So, on the CGD program, now this is a smallish kind of a, an asset.
Mm-hmm.
You've got three cohorts planned for the, initially.
Correct. For the trial we're doing in phase I/II , there are three cohorts: two adults, two adolescents, two children. These patients are typically immunosuppressed. They are, I wouldn't exactly call them immunosuppressed. They, they're missing a key enzyme to fight off bacteria, and so maybe in a sense they are unable to fight off things, but the net result of them, it's usually I wouldn't use word immunosuppressed 'cause people think of, you know, more cancer and other things that cause that immunosuppression, but because they in fact can't fight off infection and they're missing the enzymatic activity, it's a perfect place for Prime Editing to go in, recreate that enzymatic activity, and actually restore them to normal health.
So do you expect any difference between the younger patients versus older patients?
No. We really don't.
Did the FDA ask for a stagger between cohorts or between patients? How should we think about it?
For the first study, 'cause it's a brand new technology, FDA asked for a stagger between the first and the second patient and the adults. Very quick stagger, but they wanted to see probably the most important thing is the safety. We can Prime Edit these cells ex vivo, put them back in, and they engraft well. In all of our animal studies, we were able to demonstrate that. It's worth noting that our animal studies used human CD34 cells. We did all the editing with the same guides that we're using in humans. Up to that point, everything is exactly identical to what we're gonna do with humans. The difference in the animal studies is you put them back into an immunocompromised mouse so that it can populate the marrow, as opposed to putting it into a human, you know, to populate the marrow.
But we will need to show that in at least one adult patient before we go to the second, and we need to step down to the children as well. Do keep in mind though, this isn't just a sequential approach in the sense of moving down in age, because when we finish with the adults, it's our intention to go to the FDA and say, "It's time to start the pivotal trial in adults. We've shown you the data that you wanted to see. Let's get the pivotal trial done." So we hope we'll move very rapidly into pivotal trial. And then as we get the adolescents data, move them into the pivotal trial, and then the same for children.
Given how quickly you could get a read on the engraftment.
Mm-hmm.
Frame up, frame for us at least what 2025 data should look like.
Sure. So actually there's three pieces of data we'd like to deliver, almost certainly in adults. One of them is the engraftment, which typically happens within a month, but certainly happens within one or two months. The second thing we would like to do is to be able to show editing data. How good is that editing? In the animals, we've been able to hit well above 90%. In practice, to really have a clinically relevant, you know, impact, it's probably closer to 15%-20% that you need. In humans, we don't wanna make a prediction, but really anything north of 20% is gonna be basically curative based on some very good data from human bone marrow patients who've been treated with this disease. And the third thing is we'd like to see reconstitution of the enzymatic activity that's missing.
That uses a test called DHR, which is how people are diagnosed with this disease and also are followed with this disease. We believe, though we haven't talked to the FDA yet about this, that this would be a great candidate for a, you know, potentially surrogate marker to support accelerated approval.
So given how quickly this could happen, what do you think the scope of the pivotal study could look like if it sort of meets your expectations?
I think in our planning, we certainly expect the whole population for this particular indication will be south of 20. Frankly, I'm hoping a pivotal trial will be closer to 10 patients. When you keep in mind with Casgevy, the approval was based on 40 patients in a much, much larger indication in sickle cell disease. Obviously that would require, you know, safety, but it would also require very robust data. And that's a place where our animal model at least gives us hope that we could develop that kind of dataset for the FDA.
So you could potentially think of a 2026 launch in this indication?
Oh, definitely not. Keep in mind, first of all, we have to transition to a pivotal trial with data first coming in in 2025. And that means, you know, we certainly wouldn't start a pivotal trial until we get closer to 2026. I'm just, you know, talking hypothetically. But more importantly, the FDA certainly wants to see persistence of this effect. So we're almost certain, though we haven't discussed it with them, that we would have to show a year's worth of data that the editing, the cells continue to produce what's necessary. We're guessing a year, and then you could go ahead and you could file. But you are correct in your general thinking, without the specific dates attached.
We are hoping that there are paths that could be a fairly rapid path from when we dose the first patient until we, you know, when we are able to get approval.
Got it. So, moving on to the Wilson's disease, IND in 2026, what do you need to show here versus the effort from Ultragenyx with gene therapy?
I think first of all.
Not that we have seen any data or remarkable data from this.
Yeah. I think that was a little bit more complicated. Probably a better way, easier way for me to answer the question is talk about the differentiation. Why would someone pay attention to a gene editing approach versus a gene therapy approach? And this is something where Allan usually answers the question. So I'll turn it back over to him to lead you through the way he, you know, that we've been answering the question.
Yeah. So as we think about one approach versus the other, when you take, let's focus first on the gene therapy approach. So there you're using a truncated portion of the gene. So you can't fit the entire gene into the AAV vector. So they're using a smaller amount of the gene, a smaller form of the gene. So the gene still does have activity, but there's, you know, question of if you're gonna have the same type of activity with the truncated gene. The second thing is this is, you know, hepatocytes are low turnover. I think the entire liver turns over in something like over two years or something, right? So the expectation is, you know, daughter cells there, there'll be some dilution of effect.
So it's, you know, sort of the risk over a five-10 year period, you know, what type of effect will you lose from a gene therapy over time? When we think about gene editing, we're actually going in and making that correction directly in the genome. So as a result, every daughter cell is gonna maintain that correction. So there should be no waning of any type of improvement over time. You know, secondly, we're making the correction right at that specific place in the genome. So it's gonna maintain that correction under endogenous control, which means that that protein or that enzyme should get expressed at the level that it should get expressed versus a gene therapy where you lose that endogenous control and you also risk, you know, overexpressing a gene.
And, you know, you can overexpress or underexpress, but if you overexpress, you know, even some theoretical risks of having copper that's too low in the body. So, I think for numerous reasons, I think gene editing, the gene editing approach is to me one that I'd expect patients to choose. Again, we're not gonna be able to treat every patient with Wilson's disease today. We can only treat the patients with the mutations that we're targeting, which we think could be, you know, 50% + of the patients out there. So, there'll still be a market, I think, for gene therapy if they choose to go there, for some of those patients that we can't treat today.
You know, ultimately in the future, if, you know, we decide to take our Prime Editing technology to the liver, maybe there is that opportunity, but that's not something that we're working on today.
I would also just point out we've tested this with quite a bit of thought leaders, just really asking them what do they think they would really like for their patients overall, and a little bit to our surprise, but it was a pleasant surprise. They've told us really the chelators, which are sort of the standard of care, really aren't effective. We knew that already, but it was great to hear it. Second of all, they really seemed uninterested in the gene therapy approaches, and they were incredibly enthusiastic about the idea of doing this for their patients, gene editing, and basically, except for end stage cirrhosis patients, each and every one of them we talked to at a recent ad board said this would be the first choice for their patients, and they would sign up just about every patient they could.
Obviously, we have a lot to prove before they're gonna sign things up, but we found it just very, very encouraging, you know, feedback.
From a delivery perspective, how is this medicine being delivered?
So it's being delivered by lipid nanoparticle or LNP intravenously. And we're, we have developed a proprietary LNP, which we've used in each of our liver programs. It was developed originally for GSD. It's a very, we call it a universal LNP. It's our jargon. The reason we do it is we developed it to be used in program after program. So we didn't fine tune and optimize it for each program. Maybe that would be useful or maybe not. What we said to ourselves was, let's make an LNP that's really, really good that we can use over and over again. And we have discussed informally with the FDA and formally as well, you know, how would that work?
The FDA has made it very, very clear that if we continue to use a universal LNP over and over again, there are all sorts of synergies in terms of the regulatory path, the probability success, the cost of getting there. 'Cause in the simplest form, you could just imagine that if you take the components of an LNP, in our LNPs, there are eight components, two guides and six other, you could make one batch of each of the other six, and you could be pulling out of that for every program for the next couple of years.
We're not putting that much money into making those big batches today, but with the manufacturing and the assays and the other things, the idea really is, is anytime we need to do one, we'll make a batch, and maybe it would've been for the GSD 1b program, and now it's for the Wilson's program, and maybe for our third liver program that we haven't disclosed. It's a really efficient way to move from gene to gene as you move forward. The editing, sorry, the delivery with that has been pretty outstanding. Just recently we showed some data at ESGCT about delivery of our LNP to humanized mice. We use humanized mice because then we can use the human guides specifically and the LNP we're gonna use in humans, and we actually show that it actually does high level editing.
We got 80%+ in that particular data, and we showed that in fact, liver hepatic copper stores were markedly decreased within two to four weeks of the time that we did the therapy. So that's great, but you know, if you really wanna test delivery, most people feel more comfortable with non-human primates. We can't use the human guides in non-human primates because our guides are so specific to the actual sequence, but we use a surrogate guide aimed at exactly the same spot. We don't have to optimize that guide as much as we would the human guides because there's no therapeutic purpose in using them, but they're a great way to test delivery. So we optimized the guide, a surrogate guide for non-human primates, to about 50% editing in vitro.
We put it together with the same LNP, delivered it to the hepatocytes, and we got 50% editing in the hepatocytes as well. No unanticipated edits, no off-target edits, okay? And most importantly, incredibly good tolerability. We saw very mild, very transient blips in liver function tests and otherwise nothing.
So when you think about a Wilson's disease patient's liver, are the hepatocytes compromised in any way? So you could think about sequential dosing, not go with a, you know, therapeutic dose of, let's say, penicillamine or trientine or something like that?
It's interesting. That question comes up often. We would like to deliver a once and done therapy, and we've designed it that way. It turns out all of our non-human primate experiments are not done. They're not terminal experiments. What we do is we use the animals over and over and over again, where they get an LNP and delivery at one locus, and then, you know, x number of months later, we use them again in a third place and a fourth place. We've dosed some of these animals three or four times. There's no drop in efficiency. There's no change in the editing from the first time we use it to later. We have years' worth of durability under the circumstances. We've looked very, very closely for immunogenicity. We don't see any.
So while it isn't our intention to dose that way, there's absolutely no reason we couldn't either of two ways. We could wait and see what happened and come back and dose a patient if there was a need. Or potentially, though it isn't high on our list of things to do, we could use a multiple dosing regimen right from the beginning if it was necessary.
So, Allan, when you talked about the gene therapy approaches, one thing you probably didn't mention was, the high variability we see in response rates with gene therapy.
Mm-hmm.
With the LNPs, how confident are you? You absolutely won't see any, it'll mimic what we have seen with every other genome editing.
I think those are mostly gene therapy comments you're talking about, which is very different delivery issue. What we often do is we take a look and we eventually stain the liver and look to see are there, is there evidence in the hepatocyte, how widespread they are as we're looking for those things. So you could look at edits, but you could see that all the functioning, you know, ATP7B, I can't ever say the gene name, but the Wilson's gene, we can actually look for the staining of the intact gene in the cells. And we see very, very high levels of the cells actually have the intact gene. So it looks very, very consistent within any individual liver. So far in the mice we've been doing, there's some variability, but it's reasonably tight overall from what we're going to see.
Now, all of these are markedly in excess of what we think is gonna be a clinically relevant approach. So in some ways, you know, the variability is a little bit less important because if you get up to, you know, 20% editing of cells or 20% of cells editing, already you're hitting therapeutic levels. So if you get 50%, 60%, and 70%, which is not necessarily what we're seeing, you should be just fine in each and every one of those patients, even though there's some interpatient variability as well.
Got it. So let's touch upon the BMS collaboration that you entered very recently. The first gut reaction to that was, well, interesting, but from a financial perspective, not necessarily transformational. Kind of kicks the ball down the road. So given your balance sheet considerations, what's the next, non-dilutive avenue, available to you?
Can I just say something about that? I must admit I don't share that view, personally. I think, $110 million upfront, $185 million are preclinical milestones, you know, that are really short-term milestones, $1.2 billion potentially of clinical milestones before we get to the bio bucks. To me, that's a transformational deal. Does it solve every one of our problems? Absolutely not. I don't wanna give you the wrong impression, but I would just point in gene editing, there hasn't been a deal like that for, you know, a couple of years, okay? And even then, if you go back to the, the heyday, you know, they were bigger, but they weren't extraordinarily bigger than that overall. Don't get me wrong, there are a couple of deals. I mean, back from 2020 and 2021 in a different environment, but I think this is fairly transformative.
Having said that, we do think BD's very important for us, and there are many sources of BD. Number one is that we are basically, we stop programs to pause them. And there are places like eye and ear where we have great data. It wasn't because of lack of any efficacy. And there's been quite a bit of interest looking at those kinds of things, and we hope we'll get partners with them. Second of all, people have looked at other elements of our pipeline and come to us and asked us questions about it.
But third is there's been a number of companies, which has pretty much happened, what happened with BMS, who came to us very coincidentally as we were getting ready to go out and talk to people about the amazing things we could do at editing CAR-T and said, we know what we wanna do in CAR-T, we just don't know how to do it. You have the technology, we wanna work with you, which is essentially how the two of us came together. They're not the only company or pharma company that's come to us and really put us in that situation. So we hope there'll be some things that maybe are a little surprising 'cause they weren't on anybody's radar screen a year ago, and frankly, you know, they weren't on ours a year ago.
But these are places where a company has decided our technique is a way to solve a really critical problem. We're very complimented by that.
Awesome. So one last thing. There's obviously one copycat. How aggressively are you willing to prosecute your patents?
So if you ask me at a high level, the answer is very. If you ask me, does that translate into the fact that we're gonna start a dozen suits, you know, tomorrow? You know, you have to really look closely at how patent prosecutions work. Keep in mind that until people have a product and they're out from safe harbor, there is really no product. There is no point in doing patent prosecutions because you can't prevent anybody from doing any work under safe harbor. You can attack people's patents under those circumstances, but you have to be very careful in how you do it. In many cases, you only get one chance to do it, and it may depend really on bolstering your patent approach. We are totally convinced we own Prime Editing.
We are absolutely convinced that there are people out there who almost certainly are doing Prime Editing and that is prevented or falls under the scope of our patents. I personally intend to cash an awful lot of royalty checks sometime in the future, okay? But if you're asking me exactly what my strategy is between now and the royalty.