Jeremy Reine Crofton, one of the biotech analysts at Jefferies. It's with great pleasure that I'd like to welcome Allan Reine, the CEO of Prime Medicine. Thanks so much for joining us today, Allan. Thank you for having us here, Maury. We're going to do fireside chat format. Maybe for those who are new to this story, give us an introduction to Prime. Yeah, so it's a really exciting time for me to speak about Prime Medicine. I recently took over the role as CEO and am tremendously excited about the promise of our technology and the promise of the organization. The company was founded with a singular goal: to deliver on the transformative and widespread potential of prime editing technology. Prime editing technology, think about it as a next-generation gene editing platform. I think about it as the most versatile and the safest way that we can edit a person's genome. Anyone that's doing any type of gene editing, I believe, should be done with prime editing today. This could really be a best-in-class platform as we think about curing some of the most devastating genetic diseases. There are a lot of applications as well for prime editing outside of just genetic therapy as we think about areas in oncology, immunology, cell therapy, and others as well. When I first joined the company, I really saw the promise of what this technology can do. I could tell you, in stepping into this role today and how I feel currently, my conviction level cannot be higher on what we're doing. I'm really impressed both by the talent that we have at the company, the dedication, the passion of our team, which I think is important when we're pushing forward such an important technology. I think this is highlighted by the recent news that we actually showed data for our first patient in humans. We have now shown that this application can work well in humans. I think what we've demonstrated is that this is potentially a cure for a disease called chronic granulomatous disease, and I'm sure we'll talk about it a little later. As I embark on this role, I think the other thing I'd say about the company is we are really laser-focused and really acting with urgency to advance our current pipeline, which is focused on three, I would say, large commercial opportunities with unmet need. One is Wilson's disease. The other is alpha-1 antitrypsin disease and cystic fibrosis. We are also in a partnership with BMS for our ex-vivo CAR T cell therapies, in addition to other assets that I hope to move forward in the future, also with partners at some point. I think there's tremendous promise here and a very exciting time for the company. Got it. Yeah, it's a great intro. Yeah, congrats on taking over as CEO of the company. One of the common questions that comes up is just how do you compare and contrast the different editing approaches? Maybe talk a little bit more about what you're doing with prime editing versus some of the other approaches. Yeah, so I kind of think about it as a continuum. As you think about CRISPR-Cas9 editing, I mean, this is now a Nobel Award-winning technology that was created. There are certain things that it can do very, very well. For the first time, you can now, with very high efficiency, get a location in the genome that you want to edit. That's what CRISPR does very, very well. Using an RNA guide, you can locate or hone to the exact place in the genome that you want to edit. Now, how does it do this? You marry that to what's called a Cas9 enzyme that essentially makes a double-strand break in your DNA. It's very good at knocking things out. That's how I think about the use of that technology. It's very good at knocking something down, but it can't really make corrections well and do other things well. There are also some downsides to making that double-strand break. You have a high degree of off-target edits that occur when you're editing using CRISPR technology. Obviously, the way that the cellular mechanisms sort of repair after that, everything that comes in is going to be sort of indulged that form, right? It is not the perfect way to edit, but an incredible discovery. Out of David Liu's lab, who's also our scientific founder, came something called base editing. This is a very novel way to edit the genome, where he added into this, so you still use that search piece, right? You can still go hone into the exact point within your DNA that you want to edit. Now he's added an enzyme, a deaminase. Now you can transform one letter to another letter. It cannot really rewrite multiple base pairs of your DNA, but it could change an A to a G, a C to a T, or vice versa. It can kind of change a pyrimidine into a pyrimidine or purine into a purine. The one issue with that technology is it does cause bystander edits. When you edit, you are going to bind a little bit, and you will, let's say, bind here, and here is your editing window. Every A within that window is now going to be converted to a G, or some percentage in some of the edits you make will be an A to a G. Those are called bystander edits. The protein that you are creating, not every protein is going to be exactly back to wild type, depending on the disease. Sometimes they'll have high rates of bystander edits. Sometimes they'll have low rates of bystander edits. Out of David Liu's lab also came another technology, and this is prime editing. Instead of using a deaminase, he's now added a reverse transcriptase. Reverse transcriptase converts RNA into DNA. With that reverse transcriptase, now you have that guide. That guide is going to find the exact place within the DNA, exact place within your genome that you want to edit. Onto that guide, onto that same molecule, you can now add base pairs. You use that reverse transcriptase to transcribe those base pairs into the genome. It is this now, for the first time, you can actually really just rewrite the genome in any way you want. You could use that to fix frameshift mutations, missense mutations. We can do hotspot editing, where we're editing a longer piece of DNA, where different people might have different mutations, but we can fix it all with one editor. With our passage technology, we can do very large inserts as well. It is a very versatile and, again, very safe way to do gene editing. The other important point that both base editing and prime editing use is we use a sort of modified Cas9 enzyme. Instead of the double-stranded break, you now have a single-stranded break. With that single-stranded break, you really get minimal to no off-target editing as well. You do not really see translocations and chromosomal rearrangements. Again, we think if you're using any type of gene editing technology out there today, it should be prime editing. Got it. Yeah, I think that's a great summary and overview of the different gene editing technologies and what you're doing with prime. One of the other key points with the gene editing approach is that you have to deliver to the right place. You can use AAV or LNPs for that. Maybe talk about the latest developments with your universal LNP platform and what differentiates this platform from others in the space. Yeah, so it's a really important question because when you think about gene editing companies, there's the editing piece, right? And then there's the delivery piece. You have to get both right to have a successful medicine when you're a successful medicine. We've got a very robust LNP platform. Internally, we've developed an LNP to the liver that works extremely well. We've shared a lot of data preclinically, both in mouse models and in non-human primates, to show that we can get very, very high levels of editing in the liver. We've got some data now where we're looking at cystic fibrosis, where we can get or working on evaluating different LNP constructs to get good editing in the lung as well. I think there are many companies out there that I'd say have LNPs that look to be effective and many that look to be safe. There's a number of clinical studies that have now gone on. I think the way at least I see it is the field has somewhat de-risked liver delivery. I think what's yet to be de-risked is LNP delivery to other tissue types. I think there's a lot of very interesting technologies out there. Like our cystic fibrosis program, that's an area where we're evaluating a number of different technologies to figure out how to solve that problem, either with LNP or potentially with AAV. I think from a differentiation standpoint, all I can say is we've done a lot of modeling and modeling against some of the benchmarks that we know have been used in clinical studies today. It appears we have a good safety profile as we compare and contrast to those LNP constructs. Got it. Makes sense. You mentioned the CGD program and the data that you've shown there. From our understanding, part of that is to de-risk prime editing. I think you show a good example of that. Maybe talk a little bit more about the data you show there and how that de-risks the prime editing platform. Yeah, so chronic granulomatous disease is a disease where we're essentially thinking about it as having defective neutrophils. As a result of that, these patients get recurrent infections. They also get some inflammatory-like illnesses and typically, unfortunately, have the morbidity mortality where they can maybe live into the fourth decade of life. It's a pretty severe disease. The only curable treatment that exists today is an allogeneic transplant. There's a lot of drawbacks to an allogeneic transplant. With prime editing, we can do an autologous transplant, which is going to have a slightly more benign conditioning regimen. You don't have the risk of graft versus host disease. What we saw in the first patient of data that we presented is, one, we're getting very rapid time to engraftment. I think if you look at sort of there's obviously the sickle cell experience with some of the CRISPR companies where engraftment usually takes over 30 days. We're seeing evidence of engraftment even at the 15-day time point or earlier. DHR is sort of what you want to measure here to determine if the neutrophils are now functional. It's a great biomarker, and it's actually the biomarker they also use when you look at allogeneic transplant. There's a lot of data to suggest if you're even in the 10%-20% range, these patients can benefit. Probably about 20% of these patients can essentially be cured. Even at 15 days after dosing, we saw a DHR level, I believe, was 58%. At 30 days it was 65% or 66%. It really shows that initial patient, it's the first time that we've really demonstrated that this approach, this technology can work very effectively in humans. In a severe genetic disease, really provide a potential cure. Got it. Why is the engraftment faster with prime? There's some theories around that. I think the theory that I hear talked about most is the idea of making the double-strand break versus a single-strand break is just a little bit more gentler on the cells. It's the functionality of the cells after you do the edit and then how quickly they're essentially able to transplant. Again, this is our engraft. This is our belief. I don't have data to support that, but that we think is likely what's going on. Got it. Strategically, you're not going to focus on this setting. The indications you're going to pursue, which you've mentioned, Wilson's and AATD and CF, much bigger market opportunities. Wondering, though, for CGD, if you dose that second patient. Yeah, so we haven't commented on sort of additional patients and dosing. I think what we have said is we're going to finish that first cohort. As a reminder, the cohorts we've said previously are two to three patients. So assume that first cohort will get completed. Got it. Okay. Maybe talk more about just decision to move into Wilson's disease. You have a prime editor there that could address about 20,000 patients in the United States and the European Union. You are going to be filing an IND or CTA in the first half of 2026 and could have clinical data in 2027. Maybe talk about the preclinical data you have already shown to date and what gives you confidence in this approach. Yeah, so just on the patient numbers, just because I want to be clear, it is 20,000 patients that we think have Wilson's disease. Of that, we can go after different mutations. If we look at the mutational status for H1069Q, which is the first editor that we're taking to the clinic, that covers about 30%-50% of the Wilson's disease population. There are other editors that we're working on as well. One example is 778 that can cover about, call it somewhere in the 5% range in the Caucasian population, but maybe is a similar 30%-40% range in the Asian population, which could be an interesting place to consider a partnership with an Asian company in the future. The numbers are a little bit below that, but still a very significant commercial prospect for the company. In terms of data for Wilson's disease, we've shown data now, I think just at a recent conference at ASGCT, that we can get editing levels in the 1069 mutation up into, I think it's around the high 80s or even 90% range at relatively low doses. We are at a very great place there in terms of our drug candidate that's pushing forward to the clinic. We also, at that same conference, showed data for R778, where we showed also, I think, editing rates that were in the high 80s-90% range. From an editing efficiency standpoint, we are at very, very high levels. With Wilson's disease, you can also look at phenotypic data in our mouse models. We see really extensive copper reductions after four weeks of about 75%. We also see sort of an increase in fecal copper, the reduction in urinary copper, which is what you'd want to see in this disease, that you're excreting the copper in the right way. Got it. That's helpful. For getting to that 80%-90% editing, it's pretty impressive to be able to do that. For other gene editing approaches with the liver, it seems to be translatable when you go from the preclinical data to humans. Any perspective on how you're thinking about that? Translatability? Yeah, I mean, so that's the, I was going to say million-dollar question, maybe the billion-dollar question. Look, I think everyone in the field is trying to understand what that translatability is. I think mouse models are different, and they're different for every disease. Trying to look at one mouse model to another and one disease to another, you can't say with certainty that the translation is going to be there. With those caveats, I do think what we've seen so far has actually demonstrated reasonable translatability. I don't want to sit and promise it's going to be the exact same for our Wilson's disease program. I do think we can gain some comfort seeing that we've seen nice translation for what other companies have shown. I think there's two important aspects to that: how potent's your molecule, how high is the editing, and how safe is your LNP, right? Every LNP at a high enough dose is going to cause liver toxicity. Some LNPs cause additional coagulation and other issues as well, right? It's always about what's your therapeutic index going to be. We think we have a good therapeutic index with our LNP. We think we've got a potent enough drug for our first Wilson's disease program in 1069Q. When we put that together, we expect we'll be able to translate that. Again, this has to get tested in the clinic before we can say anything with certainty. Got it. You have already got the experience and success with the IND for CGD. Talk about how you can leverage that for Wilson's disease and what are key next steps for your Wilson's filing? Yeah, I think one big positive is obviously there were a lot of health authority discussions that went on before we filed our CGD IND. There's also been a lot of discussions that have gone on for some of the other indications that we're working on as well. We have a good idea of sort of what's necessary to get one of these INDs across the finish line. We showed that with CGD where that IND was generally accepted, and we were able to go forward and even under the 30-day time clock, which for gene editing therapy is a pretty incredible feat. There are a number of assays that we've developed for the CGD program that we can also use for every other program going forward, some of the off-target assays that we look at and other things. There are a lot of learnings from that program that will, I think, transfer over to the Wilson's program and ultimately the A1AT program. Where we get even more leverage and more synergy is when we go from our Wilson's program to our A1AT program. There we have synergies both from a manufacturing standpoint. If you think about the LNP, the LNP is going to be the same. We're only changing a couple of the components that are going to be packaged within the LNP. We believe as a result, it's highly likely that we'll be able to leverage some of the TOC studies and other things. Again, those will all be pending sort of discussions with the FDA, but we're hopeful that we can leverage a lot of those learnings and a lot of those studies from Wilson's to A1AT. We even hope that we can leverage some of the clinical data as we kind of learn from a dosing perspective from one disease to the other. There are significant learnings as you go from CGD to Wilson's and then even more significant when you go from Wilson's to A1AT. Got it. For Wilson's, how are you thinking about starting dose for the study and how many doses you'd explore in that first clinical study? Yeah, I mean, look, we've talked about an IND in the first half of last year. That is really going to depend on getting through all of our IND enabling studies. That will give you the suggestion of what the right dose to go into is. We are not going to discuss sort of specific doses. I think if you look across companies that have gone into the clinic, and I do not have all the data, so do not quote me on this, but I think companies have generally started in the probably 0.1-0.3 mg/kg range initially and dosed up from there. Based upon what we see preclinically from a safety standpoint, it will determine what dose we ultimately end up starting at. It will depend how many doses you can go up. It's going to depend on, again, back to the translation from the models that we've used to human and how high you have to go to get that level of efficiency. Hopefully, some of the safety data translates as well where we're able to dose higher. Anything else you could say on just what the first study would look like as far as the design endpoints and patient selection? Yeah, so Wilson's disease, there's a lot of interesting biomarkers that you can look at to determine effect. I think one thing that you'll ultimately want to look at is because many of these patients are going to be on chelation therapy, which is not a perfect answer for Wilson's disease and the reason there's still a very high unmet medical need here, is you can remove chelators and see if patients remain stable, their copper levels remain stable. That's an endpoint that might take a little bit longer to play out. You could look at other markers, other biomarkers that we're still, some that we're still evaluating that I think can be very useful early on in a clinical study to really get to proof of concept. If we're getting to the efficiency rates that we believe is possible in these patients, and we believe this is something that should work in all or most of the patients that you're treating, this should be data that you're going to be able to see across the full patient population and understand even at low doses the effect you're having. Obviously, you'll want to see a dose response there. Got it. Once the IND or CTA is filed in the first half of next year, what could timelines look like for activating sites and dosing initial? Yeah, so Mohammed Amsal, our Chief Medical Officer, he's got a phenomenal team under him as well. They're working with urgency to get sites up and running as quickly as possible. We've got certain ideas of how we can even make that done, how we can get that done even faster with certain things you can do that sort of before you start the official clinical study. I would say we're working as fast as we can. I would say Wilson's, A1AT, as we think about the value-creating events for this company, it's how can we get to clinical data, clinical de-risking data for large commercial opportunities. This offers two in 2027. You can be sure we're going to be working as quickly as we can to get to that data. Got it. For initial data, I guess, what do you want to show in that update? For Wilson's disease? Yeah. Yeah, look, I think you're going to want to see, back to what I was saying before, evidence, good evidence that you were very effectively binding the copper, mobilizing that copper to be excreted through the bile, not go into the blood and be excreted through the urine. I think there's a number of different biomarker measures that can really demonstrate that. Ultimately, you'll want to see that they can come off their chelation therapy. We're very hopeful that we'll be able to show that proof of concept in 2027. There's a couple, I would say, more investigational biomarkers that could also potentially really demonstrate that. We're still working through some of those as well. Got it. Okay. You mentioned the 778 mutation where there's a large proportion of patients in Asia that have that. How are you thinking about that as potential BD? Yeah, no, I think it would, I think that's something that makes sense, I think, for one of those Asian companies to look at. Again, there's a large population, large in quotes. I mean, these are orphan diseases, but for an orphan disease, there's a nice commercial opportunity and I think an unmet need in those patients. I think if you're a sort of Asia-packed company, this should be the type of asset that would make sense for them. It is a strategy that I think would make sense for us to pursue. Got it. You disclosed your AATD program recently, soon after Beam had reported their initial data for AATD. Maybe just talk more about the strategy there and how you can differentiate versus Beam's program. Yeah, so I always look at, again, it comes back to what I said at the beginning, like what's the best way and the safest way to edit the genome. And I truly believe that's with prime editing. When it comes to A1AT, there are a number of different approaches out there, including ADAR approaches. Obviously, there's replacement therapy, that's standard of care today. And then there's obviously what Beam showed. I thought the Beam data looked really good. I think it's very promising to show that they can get what likely translates to a pretty high level of editing efficiency in the liver. But they are not taking the patient back to wild-type protein. The proteins that they generate, the majority of them do have these bystander edits. That protein is still functional, but it's less functional, call it 70%-90% as functional as the wild-type protein. We do think there's room for improvement there. I think for any of these diseases, if you can bring a patient back to exactly wild-type protein, that is the kind of true treatment these patients should get. A1AT or AAT protein, I think there's a lot of discussion, I'll say, around the analyst community and the investor community on sort of what are the right levels. Everyone's very focused on, can you get above 11, which is the trough level that historically has been used for approval. People talk about 20 as maybe being the trough level that may be needed for approval going forward. I think when you're talking about replacement therapy and you're talking about a gene editing therapy or an ADAR therapy, these are different things. Understanding what these levels mean is different. You're sort of comparing apples to oranges to some extent. AAT protein, the normal range called 20-40, 20-50 in a normal human. Those levels, this is an acute phase protein. In reaction to, in response to an infection or the response to inflammation, AAT protein is going to go up two- to fourfold. The question is, do you have the right level of activity when you need it, right? It's not necessarily like you want 11 or you want 20. It's, do I have the ability to get to 40 or 80 or whatever that level is? What we're doing is we're actually just changing your genome. What does that mean? It means you're under endogenous control. It means you can have that acute phase response when you need it. To me, that's what's critical here to understand when you're thinking about a gene editing approach to any other approach. For Beam, they're also going to be under endogenous control. For good reason, that should make them better than replacement therapy or I would say even other therapies that might not have that acute phase response, albeit, again, I think the differentiation is with the number of the bystander edits and it not being wild-type protein. Got it. Makes sense. Beam's probably the key competitor that you're kind of going after there. Maybe talk about where you're at with just development for your prime editor here and if your study could look different from Beam's. Yeah, so where we are in development, we're very close to sort of DC there. We've shown really strong data already where I think we've got something that sort of could be a great drug today. For Wilson's disease, for example, we showed really high levels of editing when we kind of came out with data in the fall. We weren't quite at our DC yet. There's a number of different tweaks you can do around the guide, the LNP, mRNA, etc. That resulted in even higher levels of editing in what we recently presented where we're, again, at the 90% range versus 75% range, as an example. We're doing the same with A1AT and taking a lot of those learnings there. We're getting, I would say, very close to that nomination. Ultimately, that's going to drive us to an IND that we've already disclosed, which we're planning for the middle of 2026. Got it. Okay. Also, as it relates to Beam, I want you to just highlight the collaboration agreement there and just elaborate on the Prime Editors' ability to incorporate both transition and transversion edits as well. Sorry, the first part of your question was? Just the collaboration agreement with Beam. Oh, gotcha. Okay. I think you're probably talking about the arbitration then when you say collaboration agreement. A couple of things. One, incorporating other edits. The point there was just we're making, when we edit, there were a couple of mutations. They're very, they only occur in like 1% or 2%. One's called the Z allele mutation. We can repair those. That's not going to be our target here, right? It's just you're really repairing that Z mutation. I think the point we had made that we can repair both, that's sort of the, think about it as the hotspot where you can repair more than one mutation with one guide. In terms of the collaboration agreement, there is an arbitration that is ongoing with Beam. We believe very strongly that what we're doing is within our field and we have rights to and Beam does not have rights to. To be clear, Beam does have the right to prime editing, but they can only do prime editing using transition-only edits. I encourage everyone to go read the contract. There's a lot of it that's unredacted, so it's pretty clear when you read it there. Anything aside from a transition edit, if it's intentional, if it's a material change, falls outside of the Beam field and is within our field. With our A1AT program, we're making intentional non-transition changes that are material to the product. If you kind of follow that logic, it's clearly outside of the Beam field. Got it. Okay. That's helpful. We're pretty much out of time, but I want to ask a last question, just maybe a multi-part question. If you could just talk about cash position, how you think about just optimizing expenses going forward, and then just key events investors should be focused on. Yeah, so cash position is $158 million was our last reported end of March. That takes us, call it first half of next year. We did announce a restructuring a couple of weeks ago. That does extend our cash runway by months, call it. We have put ourselves in a position where the cost to fund the company beyond that is significantly reduced by like 50% of what it would have cost us otherwise to get through our important inflection points. Key catalysts coming up, I would say it's really just head-down execution and getting our programs into the clinic and getting those INDs done as soon as we can for Wilson's disease and A1AT so that investors understand we're getting to those important value-creating inflection points that are, again, not too far away that can, in my opinion, completely revalue this company. We are heads-down executing and we are going to get there. We are trying to be as efficient as we possibly can. I can promise you there is a lot of discipline in how we allocate our capital. We really believe in the allocate to value going forward, which is how we are going to operate this company. Could anything come up on BD this year or how are you thinking? Yeah, no, that's a good thing to close on. There continues to be a lot of interest in what prime editing is doing. Again, I'll say it again, if you're going to be doing any type of gene editing, you should be doing it using the Prime Editing platform. There are a lot of conversations ongoing. We do not ever promise anything, but I'm hopeful we'll do deals in the future. Got it. Thanks for joining us today, Allan.