Prime Medicine, Inc. ($PRME)

Hi, everyone. My name is Maury Raycroft and I'm one of the biotech analysts at Jefferies. I'd like to welcome Allan Reine, CEO of Prime Medicine. Thanks so much for joining us today, Allan.

Thanks for having us here.

And we're doing a fireside chat format. So maybe for those who are new to the story, if you can give a brief intro to the company.

Yes. So Prime Medicine is a company that's focused on progressing our technology, which is Prime Editing. This is a revolutionary gene editing technology, sort of the 1-minute pitch without going into all the details around the technology is think about this as the way I like to say it is it's the safest way to edit the genome versus any other type of gene editing technology that's come before this, but it's also the most versatile way to edit the genome. So think of Prime Editing as we can do any type of edit that any other technology can do, but we can also do so much more, right? So we can do large insertions. We can do -- fix all these different types of edits that other technologies can't do. For example, in Wilson disease, going after what are called transversion edits, et cetera. So again, the most versatile and the safest gene editing technology, we think that exists out there.

Got it. Yes, it's a great intro to the company and the technology. Maybe starting off with your Wilson's disease program. You've got the Prime Editor PM577, which is entering the clinic first half of this year. You're guiding to file an IND and/or CTA first half of the year. What can you say about where you are with the submission package at this point? And what are the remaining gating factors?

Yes. So I think all we've said all along is what our guidance is, and our guidance remains the same as the regulatory filing for Wilson disease in the first half of this year. What you typically expect from us is to announce once an application has been accepted. So look to that as the next announcement there.

Got it. Okay. And is there more you can say on the strategy here? Are you only focused on getting the study going and getting to initial safety and efficacy? Or would you want to pursue a global Phase I/II and quickly try to accelerate development?

Yes. So this is -- we're planning on even this initial Phase I/II study to be a global study. So we've got a number of sites that we've already selected and are already working very closely with a number of KOLs out there. And I can tell you, there is a lot of interest. I've attended some of these KOL sessions at some of the conferences, and there's a lot of interest amongst the KOLs for the study. So this is going to be a global study. But there's also a lot of interest from the patient standpoint as well. So a lot of patients have also reached out here. So this is an area of very high unmet need. We're hearing that from KOLs. We're hearing that from patients, and we're very excited to get the study going. And data dependent, we'll expedite this as much as we can.

Got it. And you've already -- you've enrolled an estimated approximately 30 Wilson's disease patients across 7 U.S. academic centers in a prescreening observational study. Just stopped enrolling earlier this month. How will this impact your enrollment time lines for the dose escalation and eventual expansion?

So this is U.S.-based. So it's not going to impact, obviously, our ex-U.S. sites. But for the U.S., what this allowed us to do is to really identify a number of patients -- and I think that number is a little bit over 30 now, a number of patients that we'll have genotyped to confirm that H1069Q mutation, which is the initial mutation we'll be addressing in -- for our first Wilson disease study. And so that should help us sort of fast track getting those patients ultimately enrolled into our treatment study and hopefully help us move those cohorts along faster. But again, that's U.S. focused. So it won't impact what we're doing ex U.S., but should help us enroll faster in the U.S.

And it's observational. So some of those patients may not qualify? Or how should we think about that?

Think about that as we're trying to enroll patients into that study that we think could be good candidates for the clinical study is my understanding.

Okay. Got it. And is it fair to assume that by the end of this year, you could have sites open and potentially a few patients dosed?

That's a fair assumption.

Yes. Okay. Any more granularity on that...

We're not going to give updates on patient-by-patient enrollment into our study. But based on the time lines we've put out there, I think that's a reasonable assumption.

Got it. Okay. And at the ASGCT meeting recently, you showed preclinical data at 0.4 mg per kg and 0.8 mg per kg doses, suggesting approximately 45% hepatocyte editing is the threshold for near normal liver copper in mice. How many dose cohorts? Should we expect an escalation in your clinical study? And based on your human-equivalent dose modeling, are you starting at a dose where a functional cure is achievable?

So that depends on, obviously, the question you're asking what your starting dose is, right? You're going to go in with a starting dose that you'd like to go at and then the regulatory jurisdiction where you're going into has to agree on that starting dose. So you can't necessarily answer that question until you know exactly what your starting dose is. That being said, we're very encouraged by what we see even at 0.4 mg per kilogram. Even at that dose, where if you think about bulk liver editing, that's somewhere in the 20s, right? So hepatocyte editing in the, call it, 40% to 45% range, we're seeing almost complete normalization of copper metabolism preclinically. So that suggests that at that fairly low dose level in patients, we could potentially see normalization. Normalization might not even be where you need to get to because remember, a heterozygote, so a person that has one mutation but doesn't have that second actually doesn't have disease, right? So they might have increased copper in their liver and some increased copper, but they actually don't get sort of "clinical" Wilson disease. So we're very encouraged by that data, and it suggests a dose of 0.4 or potentially even lower, if the preclinical data translates well, could be very effective in these patients. If you look at where other companies have started in the clinic, that suggests we could be at a dose that's not too far below that. So you may not be at your optimal dose when you start out, but the goal is to be at a biologically-active dose when you start out, and we do expect to be there. And the better the data translates, hopefully, the closer we are to that optimal biologic dose.

Got it. And so based on what you said, can you put finer points on just the amount of editing efficiency you need in order to get efficacy in the clinic?

Yes. Again, this all comes down to preclinical to clinical translation. So I think preclinically at fairly what I would call low editing rates, like below maybe at 40% hepatocyte editing or maybe even as low as 20% bulk liver editing, maybe even lower, you might get to levels that are equivalent to essentially where a heterozygote is in terms of copper load in the liver. We'll see ultimately what that looks like clinically, but at least suggests you can be at lower levels of editing to derive benefit. That being said, we're at very high levels of editing, right? So if you look at where we're getting to at 0.8 and even slightly above that, we're getting to levels of 70-plus percent editing, and we've even shown data getting to 80%, 90% editing. And we think we've got an LNP that at least looks to us could have a good wide therapeutic index here. So the ability to dose higher is there. We're not saying it's necessary. But at least what we've seen preclinically, we're very encouraged about that dataset.

Got it. Makes sense. And you're leveraging traditional Wilson's biomarkers like ceruloplasmin...

I can never say that word either. Ceruloplasmin.

And serum copper, urinary copper and the fecal copper, what threshold or magnitude of change versus baseline would you consider sufficient to demonstrate functional cure restoration?

Yes. I don't know that those numbers are sort of understood, right? I don't think there's sort of an exact number to say what a functional cure is when you're looking at serum copper, urinary copper, fecal copper. So the way I like to think about it is the way that copper is normally excreted, it's down to an enzyme in the liver, it's excreted through the bile, so ultimately through the feces. That's normal copper metabolism and excretion. In a Wilson disease patient, they lack that enzyme that's able to normally metabolize that copper. And so you don't get that fecal excretion. That copper builds up in the liver. It ultimately spills out into the bloodstream. It collects in different tissues in the body. So you obviously get that liver disease, but you can get pretty severe neuro and psychiatric disease because of the buildup of copper in the brain. And ultimately, that copper gets excreted predominantly through the urine -- through the kidneys and through the urine. So that's sort of abnormal copper excretion. So when you treat, what we see preclinically is we can fully normalize fecal copper, we can fully normalize urinary copper. We're not going to have necessarily like a baseline measure on these patients, but you'll see what sort of that normal range is. And the expectation is that we can get these patients, if we get to a high enough level of impact, we can get them into that normal range. But I can't sort of off the top of my head, quantify exactly what those numbers are.

Got it. So it sounds like the urinary copper that's going to be something that you're highly focused on.

I think -- as I think about the endpoints that we're looking at, if the copper PET data translates well preclinically to the clinical setting, that should be the most compelling data because we'll be doing a baseline measure and then we'll be doing a measure 6 to 8 weeks after that patient is dosed. And that should show you have you -- do you have a significant improvement in the way that copper is being metabolized because you actually take these patients off of standard of care for those few days before they get the radiolabeled copper. So you're seeing the sort of true copper metabolism when they have full disease and after they've been edited. I think things like urinary copper, fecal copper, there are some other measures of ceruloplasmin and ceruloplasmin-like measures something called NCC, that you can look at, those will also all be very helpful measures to sort of gauge where you are in terms of disease improvement.

Got it. And for the radiolabeled copper, is that going to be included in the initial study? And how important is that going to be for regulators?

So that will be included in the initial study. So yes, it's actually -- it's not a large effort, but it is actually a separate IND you have to do for that -- for the radiolabel copper assessment. So that's something that's in progress as well, but that will be included in the initial study. From a regulatory standpoint, there are other companies that have included this in their study. It's relatively new, I would say, but there are a couple of companies in Wilson disease that have started to include this as part of their studies. So it's possible they have some experience with that now. I think the better the data looks, the more likely the FDA will allow us to use this as a good measure of, one, hopefully, removing standard of care and showing that these patients remain sort of -- have normalized copper once you've removed standard of care. But there's also the possibility of could this be something that's used as an endpoint in a registrational trial. I'm not sitting here today saying that's the case. Like I think the base case is always going to be can you remove standard of care and show that these patients are maintained. But as we always say, the stronger your dataset, I think the more flexibility the FDA could have.

Got it. Makes sense. And you're enrolling patients with moderate liver disease. How are you thinking about the safety in that population, particularly in the context of prior liver safety signals seen with other gene editing approaches? And are you incorporating steroid prophylaxis or more frequent liver monitoring in the study?

Yes. So we are -- we do have some prophylactic or some drugs that are given to sort of right -- kind of just before you treat. So that does exist here, but you don't have to do significant steroid loads or anything like that. From a safety standpoint, we can just look at what we've done preclinically, and we have benchmarked our lipid against some of the other lipids that have gone into the clinic. And we do see improved measures when we compare across liver function tests, we compare across some of the inflammatory markers and we compare across some of the coagulation markers that you see. So if that plays out as we go from sort of the preclinical setting to the clinical setting, it's possible that you do have a wider therapeutic index. And as you're going into patients that are going to have some element of liver disease, it's moderate liver disease here, we hope that therapeutic index is something that hopefully plays in here, and this is very safe in that population.

Yes. That makes sense. And as you look ahead to initial data expected in 2027, what do you plan to report? Will there be an opportunity to show proof of concept in an early cohort? Or would you want to wait for more mature and complete dataset?

Yes. I mean I'll never say never to anything. I mean I think everything is kind of always on the table with this stuff. I think we -- I mean, obviously, with CGD, we came out with 1 patient and then a second patient. We ultimately only dosed 2 patients. And then in the end, we're actually going to be filing for a BLA just on 2 patients' worth of data. But I think in general, it's better to put out a more fulsome dataset where investors can really and physicians and patients can really understand what that data is. I don't love the idea of you throw out one patient, you've got proof of concept and then that second patient doesn't validate what you did in the first patient. So I always find more data is better. But when we feel like we have a dataset that we think is validating, and it's the right time to share that dataset publicly, we'll do that. And I think if it's important to share for patients and for others, when we have a dataset we feel is ready, we'll report it.

Got it. Okay. Let's shift gears and talk about alpha-1. For this program, you're going to file an IND or CTA in the middle of this year and then show initial data in 2027. Maybe just starting off, what's the latest you're seeing on the status of the filing and what are remaining gating factors there?

Yes. So same thing at Wilson. I'll sort of not answer the question in the same way and just say same -- no change to guidance at this time.

Okay. I guess for both programs, is there anything critical that you're waiting on?

Again, just refer you to our guidance.

Okay. And for the Phase I, do you plan for your Phase I to follow a similar staggered Part A and Part B structure as Beam, where you separate lung and liver patients to establish LNP safety? Or is there an opportunity for cohorts to enroll concurrently to accelerate development time lines?

Yes, it's a good question. Wilson is a little bit ahead of alpha-1. Whether that's going to sort of play into -- in terms of when we dose, if we can leverage one to the other, don't know, right? If one was very far ahead, then you could potentially leverage safety as you think about dosing in your alpha-1 study. But I think they're close enough at this standpoint. I don't know if that's something that's going to be sort of meaningful here. I think for now, I think about it as probably going into different cohorts, not so much from a safety standpoint. I actually think it more so from a -- just efficacy standpoint. I think seeing efficacy for lung disease, you're really just kind of looking at levels, right? I think seeing sort of an improvement in liver disease is a little bit different in terms of the biomarkers or what you want to look at to get some comfort there. You're just going to get a much faster read likely as you think about lung disease versus liver disease, although there's a few things you can look at in liver that can get you there. So my expectation is we'll start off similarly with lung disease, but the goal will be to pivot to liver disease patients as well.

Got it. And how are you thinking about biomarker endpoints relative to Beam? And is there anything you're planning to do differently that could allow you to better understand or demonstrate impact on functional measures?

Well, I think early on, you're really looking at biomarkers, and it's really hard to get a read on functional measures. I think functional measures is something that's going to take a much longer period of time. Some companies, and this is more sort of luck of the draw, if you get a patient that's undergoing some type of inflammatory response to infection and you see elevated CRP, et cetera, yes, you can show -- or am I getting the biomarker increase, that response there. But in terms of a functional benefit, those are longer-term studies. And I think what we've seen today, most companies are going to be going forward, looking at biomarkers, predominantly AAT. There might be another couple of things that you get to include in there as well as a basis for an accelerated approval, but there'll be confirmatory studies where there'll be some different measures that we'll have to look at as well.

Got it. And kind of a similar question to the Wilson's one earlier. Just based on your preclinical data for alpha-1 in the mouse model, you've got 2 dose levels that restore AAT above the normal human range. Can you confirm if you're starting the dose escalation at an equivalent dose or below those doses [ validated ] in mice? And with your data and others' data, what are the key things to know about your dose range strategy?

Yes. I mean I'll caveat this in the same way, right? Until you have sort of what you agreed upon dose level is, it's -- you can say you want to start at a certain dose, but you don't have that go-ahead yet. In terms of thinking about those dose levels, if I think about the lower dose level there, if we're successful starting where we want to start in that specific study, that should be fairly close to that lower dose level.

Got it. Okay. And should we expect a more comprehensive preclinical update for your AATD program at a medical conference in 2026?

I don't think so. I mean I think we've put out a lot of our data already. I think we've got a really strong profile as we're moving forward. We do have some additional data internally. But frankly, I think we've got a really good profile from what we've seen. So not to say that we're not going to share any additional data, but I don't know that we have any plans at this time to share any additional data.

Got it. Okay. And you're currently in arbitration with Beam related to your AATD program and previously guided a resolution by first half of this year. What is the status and latest you're seeing on time lines for potential resolution?

Yes. So we've refined that somewhat. So now we've said so the first half of this year, we expect a decision by the end of July. It could come sooner, but we have a date that we know there's a commitment to get us a decision by. So by that date or before. So we're within, call it, 7, 8 weeks of having a decision or, again, it could come earlier.

Got it. Okay. And can you talk more about how PM647 clearly falls within Prime's retained field under the collaboration agreement? For example, does PM647's Prime Editing mechanism introduce edits beyond direct correction of the E342K mutation that impacts hepatic secretion and functional restoration?

Yes. So what Prime's editor does is we have a number of measures when we think about improving editing efficiency that impacts something called mismatch repair. So when you think about a Prime editor and not to go too far into the cellular biology, but you're always fighting against mismatch repair. We're always trying to do things that can kind of tamp down that sort of mismatch repair. And so we have a number of tools and one of those tools is doing something that we call synonymous editing, where we make some additional edits that don't -- they impact -- they impact the sequence, the coding sequence, but they don't impact the amino acid sequence because the redundancy built -- that's built into that translation. And so when we do those additional synonymous edits, that's sort of one of the tools that can impact mismatch repair. And so that's something that we do here. So there's additional edits there that really get us to the product that we have and to the editing efficiency that we're able to achieve.

Got it. And so that basically should be very independent from the collaboration agreement.

Yes. I mean if you read the language of the collaboration agreement and you read across what it says in terms of the Beam field and the Prime field, at least we believe it's pretty clear that anything that is a -- does a transition-only edit that falls within the Beam field. Anything that does a transition-only edit and -- does a transition edit and a non-transition edit, clearly falls within the Prime field. This is not an agreement that was separated based on disease. The one disease that is called out is sickle cell disease and Beam has rights to sickle cell disease for Prime Editing, and we don't contend that.

Got it. Okay. Grifols is going to report pivotal AATD data second half of this year using PD15 to measure emphysema progression at 2 different doses. What are you looking for in the data? And what type of read-through will it have to Prime?

Yes. So this is another -- this is a replacement therapy?

Yes.

Yes. We haven't seen a lot of great data as we think about sort of functional data with replacement therapy. So I think it's important to see something that can really validate kind of what they're doing. I don't see that as having a real impact in terms of will patients, will KOLs opt for this one-and-done therapy if they have that option. I -- let's see what the data shows. But even a very positive data event there, I still think a lot of this market would be willing and will prefer to go to a sort of once-and-done gene editing treatment. But let's see what that data shows. I -- for patient's sake, honestly, I hope it's good data. I hope patients have options that actually work because they're getting these therapies today. I still think we all question on are they really having a strong benefit.

Yes. Okay. Makes sense. And for cystic fibrosis, moving on with your pipeline to cystic fibrosis, what are the specific hurdles in optimizing LNP versus AAV delivery to basal cells? And what exactly do you need to show in your anticipated 2026 in-vivo proof-of-concept data to advance into IND-enabling studies?

Yes. So there's a couple of different approaches here. So I think we're focusing more on LNP versus AAV for probably just in the 5 minutes, I won't go into it. Probably for obvious reasons to a lot of people, I think LNP would be a preferable approach here. And among -- and thinking about that approach, there's also different target product profiles that we're thinking about. So one is how do you get to those basal cells? Because if you can get to those basal cells, you can truly have a once and done or treat a few times and you're done kind of therapy in cystic fibrosis. And that would be a huge win for patients. But another huge win for patients could be, what if I could get to the secretory cells. And they make it some of those -- that basal cell population. So over time, I can dose less frequently. But if I can get to those secretory cells that are turning over every month, 2 months, 3 months or even longer potentially, depending on that could be different for a cystic fibrosis patient. So maybe there's actually a drug I can dose with a Prime Editor that is dosed, I don't know, twice a year or 3 times a year, even 4 times a year. We know some of these drugs with LNPs have been dosed daily. And there's been some tox daily, but we know others that have been dosed even 3 times a week where we haven't seen high tox burden. So this is something that can be dosed. And if you can dose it a lot less frequently and have this be very efficacious, maybe you don't have to get to those basal cells. So there's different sort of profiles I think we're looking at here, and I think there's a number of different ways to win. This is a very important population of patients where, call it, 10% to 15% of patients that, for whatever reason, either because of their mutation or because of tolerability just can't take current standard of care. These are drugs that have revolutionized the treatment for CF, right? This has dramatically changed the lives of these patients. But the Cystic Fibrosis Foundation is looking at these patients, as are we, and saying, there's a number of patients that have been left behind. How do we treat those patients effectively? And that's what we're working really hard internally and with the Cystic Fibrosis Foundation to really try and figure out. And we're hopeful that we're going to get to some real proof-of-concept data this year preclinically that we can really get towards a DC and ultimately get to IND-enabling studies here because we see this as a really, really important population to treat. We're also not going to stop there. So even one of the mutations we're going after is very close to del 508, the same editor could address both that 507 and 508 mutation. And I think this could be a very desirable solution for patients that are also on TRIKAFTA that are treated for -- that have the del 508 and it could be in addition to their correctors, this could be something that's very beneficial to those patients as well. So we're not just going to stop at that one population. We know that's a very important population to target first, but we also are thinking broadly about the CF population.

Got it. And for the 507, 508, just out of curiosity, is that a pan-genotype editor there or...

So what we do -- so if you think about the editing window, like usually, we're covering -- I don't know, we could do longer base pair edits, but you can very easily do 10, 15, 20 base pairs. And so you can cover -- if 2 mutations are sort of right beside each other and think about it as one editor can really cover both of those mutations. We call this hotspot editing when one editor could do multiple mutations. But I'm not talking about hundreds of base pairs where you're editing amongst that. It's sort of like if you have 1 or 2 or 3 mutations that are very close together, then one editor can address all of those.

Got it. And you've got technology that could address pan-genotype PASSIGE technology...

Oh, PASSIGE, yes. Sorry, I didn't know that's where you're going. Yes. So PASSIGE would be if we're -- whole genes maybe are -- maybe it's not a whole gene, but you can do pretty significant almost whole introns. So there, you can do, yes, pan, large insertions, you can cover the majority of mutations essentially with one large insertion. That's something we are also working on with the CF Foundation. That is just when you're doing a whole or a very large multi-kilobase insertion, that's going to be a more complex problem to solve with lung delivery. I hope we get there, and we're working on that as well. But I think we've got a great approach just looking at the current Prime Editors. And if we can get there with PASSIGE, that's going to be a great pan solution as well.

Got it. And you've highlighted potential to expand into larger liver-based indications and know the development plans are still being finalized. When should we expect more clarity on that?

We would hope to provide more clarity on that as we kind of get through the year. There's a lot of really interesting indications where I think Prime Editing is uniquely suited for. Indications that are, frankly, a lot larger even than Wilson disease and alpha-1 antitrypsin deficiency. And so hopefully, we have more to say on that as we move along there.

Got it. Any update on the collaboration with Bristol?

No, continues to progress well. A lot of big preclinical milestones there. So continue to make good progress there. There's only -- they don't let us say anything, so not more I can say there, but good progress.

Good. Okay. So we're pretty much out of time. Maybe just to close up, if you can recap key catalysts ahead over the next 6 to 12 months that investors should be focused on.

Yes. So obviously, you mentioned the arbitration, we'll have that coming very soon, getting these 2 programs ultimately into the clinic, getting to data and what I think is going to be proof-of-concept data from our first 2 in-vivo clinical programs in 2027. And I think importantly, we didn't talk much about CGD, but filing a BLA for CGD, hopefully sometime later this year, sometime in the first half of next year and getting that important medicine to patients as well.

Got it. Allan, thanks so much for joining us today.

All right. Thank you.