Prime Medicine, Inc. (PRME) Earnings Call Transcript & Summary

Okay. Well, thanks, everybody, for joining us. I'm Terence Flynn, Morgan Stanley's U.S. biopharma analyst. Very pleased to be hosting Prime Medicine this afternoon. From the company, we have Allan Reine, the company's CEO. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. But Allan, thanks so much for joining us today. Really appreciate the time this afternoon.

Yes, Terence, thank you. Thank you for hosting us.

Absolutely. Maybe -- I know we were talking about this before, but you've recently become CEO of the company. And so maybe you could just kind of outline your strategic priorities, what you've been focused on, how you've been focusing your time and effort in this role now at the company?

Yes. So as we think about -- as I think about strategic priorities and sort of capital allocation plan and how we really just build the best company that we can over the long term, I think that even started before I became CEO in May, really when I joined the company even back in January of last year. I think at that time, there were probably about 18 programs in the pipeline. Obviously, this is a technology that can be used pretty broadly. And with that, there are a lot of difficult decisions to make because we're having a lot of success across a lot of different programs. So one of the first things I did once I joined the company is really trying to figure out, we call it sort of a value framework exercise, but really understanding from a strategic standpoint, how we can really build value. And so really focused on what programs do we think can derisk early? What programs do we think have high probabilities of success? Where is there a true unmet need where we could really make a difference in patients' lives? What's the competition look like? And ultimately, what's the commercial opportunity look like? And we put that all together, what really came to the top of the list is really focusing on our 2 liver programs, where we think there's one high probability of success because we know that we can deliver very effectively or at least it's been shown that other gene editing companies have been able to safely and effectively get that cargo to the liver. And so that's our Wilson's program, where there's really no other gene editing technology that can go after those types of mutations. And then our alpha-1 program where we think bringing a patient back to wild-type protein really is a potential best-in-class therapy. The third area that we're also focusing on is cystic fibrosis. So there, I wouldn't say the delivery problem is sort of it's still a bit of a hurdle, right? We're not as far along, and we can talk more about that as we are, say for liver disease. But we're making a lot of progress there, and that program is essentially funded by the Cystic Fibrosis Foundation. So we think about capital allocation, we essentially have that capital provided for today. And then we're pushing forward our ex vivo CAR-T therapies, and that's in collaboration with Bristol Myers, and we can talk a little bit more about that collaboration as well. No, I think beyond that, that's sort of how I'm thinking about the short term, right? What are the short-term to medium-term opportunities for the company. But there's still a ton of long-term value outside of those core areas. And the expectation is we'll come back to them in the future, even as -- either as proprietary programs or potential programs that we move forward with additional partners. So areas where I think prime editing is differentiated from any other type of gene editing technology, I think there's a lot of neurological diseases that we can go after that are either repeat expansion excisions, like that's one area where we really have a unique technology to target those diseases. There's different sort of transversion mutation diseases to go after in the brain, so whether it's Rett's disease, ALS, Friedreich's ataxia. There's dozens and dozens of potential high-value programs to go after. It becomes a bit of a delivery question. We think there's been a lot of advancements over the last year, and we've seen some data, whether you're looking at direct injection or otherwise. So over the long term, that's one area that I could see creating a lot of value for the company. Cell therapy. We think this is a best-in-class cell therapy approach, whether it's what we're doing in T cells, what we can do in iPSCs and other areas as well. And then as we think about other indications in the liver where we can really leverage a lot of the data that we already have. So the vision is, yes, let's see how we can create value over the next few years. But ultimately, how we can ensure we're creating value over the next 5 to 10 years as well.

Yes. No, that's great. We'll unpack some of this in more detail here. But maybe the first thing is just as you think about differentiation of the Prime approach versus some of the other genetic medicine approaches, just remind us kind of what those key features are. I know delivery has been a challenge for everybody, but what are the key things from a prime editing perspective that you really think makes it stand out here and amendable to such a broad opportunity you talked about?

Yes. So think about -- prime editing can really do everything that CRISPR, Cas9 editing can do, which I really think of you make a double-stranded break, and it's really good at knocking down the gene. I can do that very effectively. Because we're making that double-stranded break, there are off targets. There's obviously a lot of indels that are formed, genetic chromosomal translocations, rearrangements, et cetera. So it's -- we could definitely do what CRISPR is doing in a safer way because we make a single-stranded break. We don't have those off-target sequelae that they have. And obviously, the way that we edit, we won't have those indels. So it can do those in a safer way. It's not the area we've chosen to go because we don't want to be the third or fourth company kind of going after the same target and doing a knockdown. So as we think about differentiation, it's really -- what we're doing is -- so if you think about the guide that CRISPR is using, the RNA guide, we're adding an enzyme into the mix, called the reverse transcriptase. A reverse transcriptase is what's converting the RNA into DNA. So we're actually adding a template onto that guide that now gets written into your genome. So instead of just knocking something out by cutting your DNA at a precise spot or with base editing, they use an enzyme that can change just one letter to another letter, so call it an A to C or a T to a G. This is something that can actually go in and write in the number of base pairs in a row. So it opens up a tremendous amount of opportunity in the types of diseases and edits that we can do that really were not attainable beforehand.

What -- and then maybe the other question we get a lot now is just there's been a lot of changes at the FDA. And so as you think about kind of your interactions with the agency and their stance on genetic medicines, are there any areas where you feel like there are changes that maybe either benefit you guys or maybe will make it a little bit more challenging from an operations perspective as you move into having a broader pipeline here?

Yes. So we don't comment directly on our own FDA interactions, but I think I can comment on sort of what the FDA has been openly talking about both publicly and in some of the closed meetings that they've done like the CEO listening tour. So they seem to be, at least from everything that I've heard, fairly positive in terms of their approach to cell and gene therapy. They've talked -- they've made certain comments such as let's see if these drugs appear safe in these really difficult indications, should there be a pathway that potentially you can test these things in the commercial setting versus the clinical setting. Obviously, there was a lot of fear initially, if everyone has to do a randomized controlled study. I think those fears have been allayed somewhat. And if you look at the guidance they put out last week, they wanted to make it, I think, pretty crystal clear that even the patient size is under 1,000, that there'd be another designation, so you can ensure in these very difficult-to-treat indications that there's a pathway. But when we think about prime editing, again, this is -- when we're taking a patient back to wild-type protein. Again, we talked about -- when you asked the question before about differentiation, that's very different than what other technologies are doing. And so when you're just taking a patient -- when I say wild type, like it's going back to the normal protein, that you and I would have, assuming we don't have a mutation. We're doing that without those other off-target effects. And in the case of base -- being in base editing, at times, depending on what they're doing, you could have bystander edit. So you're not always getting back to wild-type protein. So we think from a regulatory standpoint, the fact that we don't have all these sort of off-target effects and that it's really the native protein that it should make a lot of those conversations hopefully a little bit easier just based on the biology and what we're doing.

Yes. Great. Maybe we'll transition over to 359, which, again, you guys had some first proof-of-concept data here for the platform. But maybe you could just recap for everyone kind of the key findings from that Phase I/II data and then we'll dig in a little bit further.

Yes. So this is a program -- it was our first clinical program in a disease called chronic granulomatous disease. So I didn't mention it before as one of our focus areas because it's a program that we're not investing in more clinically. And I think we'll get to kind of how we're thinking about this program going forward. So we treated 2 patients. It's a -- this is a disease where you have defective -- your neutrophils are defective. So it's a type of white cell and because of this defect, you get both infections that are very difficult to treat and oftentimes can become recalcitrant and you can get inflammatory like illnesses, like an inflammatory bowel disease is a prominent one and CGD. These patients tend to have a reduced mortality, so they tend to live into their 40s or 50s. And the only available treatment to them that can really cure them is an allogeneic transplant. And some patients don't have a match and some patients, this is a therapy that also becomes less effective as these patients get older. So there are a number of patients out there that have not been able to get an allogeneic transplant. And that's effectively, I think, the patients that we've initially treated. So there's a biomarker that we use in this indication and something called DHR and through allogeneic transplant experience, we know if you get above a certain threshold which is about 20% positivity that these patients should be sort of effectively cured and should be at a normal rate for infections and their inflammatory like illnesses should go back to normal. In the 2 patients we treated, we see DHR levels in one patient, I think we've gotten above 70% in both patients, one patient well above 60%. The other one, I think, lasted well above 70%. And there's a marker of inflammatory bowel disease that we're measuring too that was sort of 15x normal, that's now come back to normal within, I think, a month of treating that patient. So it's showing that we -- you got to obviously show a little bit longer duration, but it should be these patients engrafted well in about 15 days, which is really impressive versus engraftment you've seen with CRISPR and other CRISPR-based therapies, and we think that's the gentleness of the single-stranded break. So it's pretty -- what we think of as groundbreaking data here. Given the commercial outlook, it's just a very small patient population that have not been transplanted, we think there's probably only a few dozen patients out there to ultimately treat. So it forced us to make a sort of very difficult decision to not invest more clinically. But given -- and this comes back to how you were talking about the FDA earlier, given the sort of FDA's apparent sort of willingness to think about these therapies for these difficult-to-treat indications, we think it warrants at least going and having a conversation with the FDA to see if there is a path forward just based off of the current data set, and we'll see what happens there.

Okay. When and what's the timing on when you'd anticipate to have that kind of discussion?

Yes. I mean, we probably -- we're not going to comment on time lines necessarily, but I'd say something we'll do over the next 6-ish months.

Okay. Got it. Okay. And then meanwhile, there's no -- are there any other patients you're going to treat or that's pretty much it...

No. That's it. So from a capital allocation standpoint, there's no further investment into that program.

Yes. Okay. And maybe just talk about the read across to the rest of the platform. Obviously, this is an ex vivo approach. Again, some of the newer approaches are in vivo, but how should we think about translatability, de-risking the platform, those kind of questions as you look at the data through that lens?

Yes. I mean I think about gene editing as sort of 2 steps, right? So you've got can you edit with high efficiency, right? There's a technology that started in David Liu's labs like 6 years ago, where you could do this incredible thing in the cell, but at a lower efficiency, right? So now we've demonstrated in multiple tissue types as is David Liu in his lab as well, that we can edit at very high efficiency in a number of different tissue types. So then the next part of the equation becomes delivery, right? And where can you effectively deliver this cargo. So we've shown that we could do it very effectively ex vivo and we could do that and these cells have high fidelity and they can, as I said, engraft very quickly. So I think that's really important to show that in human cells, we can get high efficiency editing, which is maintained when it goes into the human body. So now it just becomes a question of delivery, not necessarily can we edit effectively. For the liver, we use a lipid nanoparticle to deliver that has a moiety on it that really helps hone these cells to the liver. And we've shown preclinically, we can get very high levels of editing efficiency. So similar to CGD, if we can translate our preclinical data to what we -- to clinical data and really see that just even similar levels of editing efficiency, then we should make a tremendous difference for patients with both Wilson's disease and alpha-1 antitrypsin deficiency.

Great. I know we'll come back to delivery in a little bit here. But you mentioned your 2 programs for Wilson's, AATD, I think you've guided to an IND, first half of '26, for Wilson's and mid-'26 for AATD. So maybe just what's gating to each of those programs? And then is there any opportunity to maybe pull that forward somewhat in terms of time lines?

Yes. I mean -- so we're working as hard as we can to get those INDs in as early as we can. It's not lost on us how important of a value driver those 2 programs are for us. I can promise you us getting to those time lines is -- there are certain things that are going to be rate limiting. Wilson's, we're trying to obviously get in as early as we can. And for both, I'd say, obviously, you've got your CMC, right? So your CMC time lines are your CMC time lines. There's a lot of things that are stepwise there. This is, frankly, a complex -- these are complex drug products to make. There's a lot of different components. And so I think you've got a specific time line that we're going to, which we've got just an incredible manufacturing team at the company. And Ann Lee, I think, probably one of the top people doing this in the business. And I think we're doing an incredible job in getting those components made, but that's probably sort of a step that's going to take us to getting the IND. So remember, you've got to make your GLP product, you've got to make your GMP product, you got to take your GLP product into your GLP study and obviously complete those studies, get your INDs written, submit your INDs and get your INDs approved. So we haven't said exactly where we are, but we are absolutely on track to meet those time lines.

Okay. And we'll come to some of the kind of preclinical proof-of-concept data. But maybe first, as you think about the market opportunity here, it sounds like you guys have done a little bit more work in terms of trying to frame this out, both on Wilson's and AATD. So maybe you could kind of walk us through how you're thinking about those 2 opportunities.

Yes. So for -- I'll start with Wilson's disease. So in some ways, it could end up being an even bigger opportunity than alpha-1, which sort of surprised me as we really started to go through the numbers. So it's a little bit of a different disease in the sense of the mutational backdrop, right? So if you think about alpha-1, it's like 98%, 99% of patients have what's called the PiZZ mutation, right? So it's not a heterogeneous population and most of the patients you're going after have the same thing. So from a genetic disease standpoint, that's great, right, because you only need one gene editor essentially to correct that entire population. As a result of that, for obvious reasons, there's a ton of competition and other people in the field. For Wilson's disease, it's probably about a similar patient size. So we think it's about 10,000 to 11,000 patients with Wilson's disease in the U.S., probably a similar number, maybe 10,000 to 15,000 diagnosed patients with alpha-1 even though the number that have the mutation are a lot higher in the U.S. It's something called incomplete penetration of that disease. And so of those patients then obviously, you can ultimately target, call it, 99% of the alpha-1 patients. For Wilson's disease in the U.S., we're initially going after one mutation called the 1069Q mutation that might make up 40% -- 30% to 50% of those patients. With a few different editors, we ultimately think we can get to about 60-ish percent of the population. So if there's 10,000 patients, call that 6,000 patients in the U.S. versus, let's say, 10,000 patients with alpha-1. In Europe, it's probably similar numbers, maybe slightly higher for Wilson's disease. So another 10,000 patients for alpha-1 and maybe another 6,000-plus patients maybe a little higher with Wilson's disease. But then as you go into Asia, the market is a little different. So alpha-1 actually doesn't exist in those populations. So it's actually nonexistent. So the patient population is 0. And if you think about Wilson's disease, in some Asian countries, the prevalence of the mutation is actually higher. So one example is Japan, and the mutational backdrop is different. So the prevalence rate is higher -- the incidence rate and the prevalence rate is higher, and the mutational backdrop, we think we might be able to get to 70-plus percent of those patients. So as we think about the patient numbers there, it could be even higher than the number of patients that we could ultimately treat in the U.S. depending on sort of where those assumptions come out. So if you look at that sort of in total, the Wilson's total population could be the same or even bigger than alpha-1. And as I said before, there's no other gene editing technology that could really address this. So in many ways, this is a lot less competition and a similar to potentially bigger opportunity.

And is there an opportunity you mentioned Wilson's. So you've got -- you're going to have one kind of your lead editor, but then you're going to have these follow-on ones. Is there a way to leverage that first clinical data and CMC package and all the stuff such that the second, third one that come behind to get you to the 60%, 70% number can be quicker and maybe it's only a couple of patients or something you have to treat? Or is there -- do you have any sense of how that would play?

Yes. So we feel pretty confident we're going to be able to leverage the majority of our tox work for sure, the CMC work, wherein -- and we've kind of run the whole business model for Wilson's disease. And yes, there'll be the upfront costs, are what it takes to get the 1069 mutation through. But I would say it's a very small fraction of that, that we anticipate to get those additional mutations through. So we don't think we'll need to repeat any of the tox work. Obviously, the manufacturing, it's the same LNP. The only difference is going to be a slight change to the guide. So we think there's minimal extra work, and then we can debate sort of what's going to be necessary on the clinical side. Maybe for the first mutation, it might be a little bit more in terms of patients, but then ultimately, we think, hopefully, we'll need a lot less clinical data to ultimately get to even approval. And we do believe we can do all these mutations under one IND. But it's not just leveraging again towards other Wilson's mutations, but it's also leveraging towards alpha-1. So for alpha-1 program, we'll be able to, we believe -- well, one, we're using, again, the same LNP. So we will leverage that manufacturing towards alpha-1. But given that it's same LNP and most of the tox that we see here, you expect to come from your LNP, the hope is that we can leverage a lot of our tox work in Wilson's for alpha-1 as well.

Okay. Great. Maybe -- and again, I said we'd come back to it, but just the preclinical data you have right now for both these programs as you think about having confidence and success on the efficacy side. Maybe just talk to us about the preclinical data you've generated such -- so far that gives you that confidence to go forward into the clinic here.

Yes. So for Wilson's disease, we've got very high levels of editing efficiency at what I call low doses. So what you want is an editor that has high efficiency and is very potent. We've shown both, obviously, the editing efficiency, but in addition to that, really strong phenotypic data. So we see a pretty marked reduction in liver copper even over the first 28 days. We show an increase as you'd expect in fecal copper or a reduction in urinary copper. So across the board, we're seeing really what I think is impressive data in terms of what you'd expect to see, hopefully, when you get to a patient if that data is translatable and plays out. In terms of alpha-1 and other companies have shown similar data sets, right? The difference here is we're going back to wild-type protein, but we're seeing high levels of editing efficiency, again at low doses. So we think it's, again, a potent editor. And then we're also seeing normalized levels of M protein and essentially almost all of your protein in the blood converted from [ D ] to M, which is what you want to see.

Yes. What -- and then any early insight in terms of what the Phase I trials might look like? Is this going to be pretty standard for genetic medicine type trial design? Or is there anything novel that you guys are thinking about in either case for the initial design of these studies?

Yes. I think for alpha-1, it's pretty straightforward. I don't think there's much novel there. I mean, you've got a pretty established biomarker and AAT levels. I also think you're now looking at AAT levels for wild-type protein, which is even different for us versus potentially some of the others that you're looking at. So to me, that's pretty straightforward. When it comes to Wilson's disease, there are a number -- these patients initially at least, are going to be on standard of care. So for those patients, they already have, what should be normal serum copper, they'll have other things, right? They'll have high urinary copper because that's their primary way of excreting it. They'll have low fecal copper. And so there's different things that you can measure. There's enzyme ceruloplasmin you can measure that really shouldn't be affected by standard of care. That's one enzyme that can give us a keynote what's going on. There's something that we're looking at called a copper PET study, to use radio labeled copper that can help you visualize where the copper is going in the body. You actually take patients off standard of care, and you do it before and after. And you can show that these patients on -- now that they've been treated, are they still able to mobilize copper in the right way? Are they getting copper going into their bile instead of into the blood? Are you excreting it through your feces in the right way? So that could be a pretty important study to really help us to hone in on dose and really to validate efficacy.

Great. And you mentioned this, but the AATD a little more competition than Wilson's. Maybe just what do you see as like the key differentiated features of your approach versus someone else like Beam or something like that as you think about going into AATD?

Yes. I mean, just to reiterate the fact that this is a -- this is really the only gene editing approach that I've seen where you're a wild-type protein, meaning you don't have protein that is bystander edits or anything else. So to me, that's what's really differentiating here that you know what you're getting is normal protein.

And last one before we go to the rest of the pipeline, the proof-of-concept data, given what we just talked through in terms of trial design, et cetera, is it reasonable to think about 2027 for both programs? Or is...

Yes. We've guided to 2027 for data for both.

Okay. Got it. Okay. Great. We were talking about this a little bit, but obviously, delivery has been one of the big still hurdles for the field in terms of broadening the potential of the genetic medicines and you guys have focused on LNP for some of your first efforts. So maybe just give us an update on kind of where that stands? And then is this a totally internal effort? Or are you also looking externally to supplement that as you think about broadening to other tissue types because I know every company has a little bit of a different approach here as they think about how much of this to do proprietary in-house versus looking outside?

Yes. I mean -- so I think the -- as you think about delivery, it's almost like you've got a number of companies out there that are just delivery companies, right? We're not looking to compete with the just delivery companies. That's what they're doing all day every day. We've obviously got an effective LNP for the liver, right? And we have some in-house experience both in know-how from the LNP side, AAV side and other. But I would say, as we move forward, I'll take CF as a good example. For CF, we have both AAV and LNP approaches. We're sort of parallel tracking both. Ultimately, we hope we're successful with both, but we can kind of choose which one we prefer to take forward based on safety and other things. But we're very open to figuring out what the best technology is and whether that's internal or external. And if we decide there's something better external, then we'll figure out how to partner with that company to move something forward. Another area is for AAV, right? As we talked -- I talked a little bit before about neurologic disease, that's ultimately today, maybe that will change in the future, but those will be AAV approaches at least for right now. We've got some internal AAVs that we work with, but we're always evaluating sort of external technologies as well to really figure out what the best approach is. Because my hope is that other companies really help to solve some of these delivery issues. Let us focus sort of on editing and let these other companies sort of focus on delivery. And then pick and choose the best delivery technologies to help us move forward if it's not going to come internally.

Yes. On the CF side, I mean, I think one challenge with that disease, at least from my perspective because I know a lot of companies are working on gene therapy a decade ago, is getting the drug through the mucus in the lungs. And so as you think about that challenge, is there something novel or differentiated that you guys can do there? Because I still think that to me is still like the kind of first principles like big picture issue for CF in particular, just getting through that mucus in the lungs to get it to the cell types?

Yes. So I mean, there's a couple of things there. First off, when you're trying to evaluate sort of the gene therapy approaches, the complication there is there's very little actual CFTR expressed. And so to really dial in the right amount of expression and then trying to elucidate clinical data when you're not sure what level should be expressed and how and why and where becomes very difficult in my opinion. With a gene editing approach, the difference is, again, you're under endogenous control. So think of how differentiated is that. If you can fix the mutation in the gene, you're now under the -- your body's sort of normal expression levels, how it wants to express that gene. To the point of sort of getting through the right barriers, I mean, there's a number of -- even in vitro assays that you can do today, one is you use something called air-liquid interface or ALI to try and sort of mimic what a CF lung is like. And so it's not just can we get to the right cells, can we show high editing in vitro, but we actually look to show high editing in vitro under these conditions that are somewhat mimicking what's going on in a CF lung. And then ultimately, there's different animal models that you can use that can kind of help validate that as well.

Okay. Great. Maybe the last program to touch on, again, not sure how much you can share here at this point is just the Bristol collaboration. You talked about this earlier. Maybe just remind us of kind of the structure, the terms and then time lines of what you can say in terms of when we might get more insight here in terms of how that collaboration is going?

Yes. So there's not a lot we can say on collaborations, unfortunately. The great thing about pharma collaborations is they're pushing forward your technology, obviously, the capital support but the negative is they don't want you to say anything about your programs anymore. But in terms of our CAR-T, so is the $110 million upfront from BMS. There's $185 million in preclinical milestones as well. We haven't said specifically kind of what and when. But we do think these are milestones that are achievable somewhat early, obviously, in development if they're preclinical milestones. I would say -- the most I can say is the collaboration is progressing well. We think we're sort of delivering on track for them, but I can't talk about targets or obviously, anything else or timing, but I think they're happy with the collaboration today.

Okay. Great. And maybe just last question, remind us kind of expenses in the second half of the year, cash runway. I know you guys recently raised capital. So just remind us kind of where you stand and then could other partnerships play a role in bringing in non-dilutive capital you kind of alluded to?

Yes. So we did a number of things. So we did raise capital that took our cash reach into 2027. But that happened through a couple of things. So one, we did announce in May. So on -- it was May 19, the say day that I took over as CEO, we did -- we announced a reduction in force that day. And essentially, that we're really going to bring down our expense -- kind of expenses over the next few years and even reduce -- I think the number we put out there was reduce our total cash needs by like half over that period of time. So expect sort of as we go into next year, our burn is going to -- our burn this year is much lower than it was last year. Our burn next year is actually going to continue to come down. And we kind of expect that 2027 is not going to be a huge increase or anything from there, something that could kind of be maintained as we look out going forward, obviously, depending on a rationale to increase burn. From a collaboration standpoint, we just think there's such breadth to this technology and there's so many areas to explore, and we expect a lot of this can and will be done ultimately with partners as well. So we think the ability to extend runway with a collaborator, I hope if we're obviously on the stage with you next year, I would hope that we're doing additional deals and have more than just the BMS deal under our belt.

Great. Well, thanks so much, Allan. Really appreciate the time today.

Great. Thanks, Terence. Thanks for having us.