Thank you all for being here. My name is Yanan Zhu. I'm one of the biotech analysts here at Wells Fargo. We are fortunate to have Metagenomi presenting for us in this fireside chat. With us are Brian Thomas, CEO of the company, Sarah Noonberg, Chief Medical Officer, and Alan Brooks, SVP Preclinical. Thank you for being here. Thank you. Great. So I wonder, Brian, could you give us a brief overview of Metagenomi's gene editing platform and how it is differentiated from other gene editing players? Sure, happy to. So at Metagenomi, we are using a fundamental science called metagenomics to go out into the natural world and to identify novel enzymatic capabilities that can be useful in gene editing applications. We've done this for the last several years and now have a toolbox that spans a huge variety of programmable nucleases that show some very unique characteristics. They're highly efficient, they're very specific, so they don't show off-target activity, and they're very small, so we can really leverage them for a variety of delivery capabilities. Importantly, though, we've taken this differentiated toolbox and been able to build on it so that we have other capabilities beyond programmable nucleases. So we've used these enzymes as a chassis so that we can build out other capabilities, such as adding deaminases to create base editors or adding reverse transcriptases to create what we call RNA-mediated integration systems, also known as prime editing, when you're using them for very small edits. But then also, we've found variations on reverse transcriptases that allow us to do very large integrations, so the reverse transcriptase can read a much larger message, much longer message, and it doesn't make errors in the process, so we've been able to scale the amount that we can use RNA-mediated integration to achieve much larger integrations at specific locations, and then finally, we have a system that's called CAST, CRISPR-associated transposases, and that system is a DNA-mediated integration system, but also targetable throughout the genome. We're very excited about what we have built in our technology capabilities, and we're leveraging this toolbox to really do a specific translation of these tools into the clinic, where we can really show meaningful gene editing progress, and so I think we're gonna talk about some of the programs and the applications later in the talk. Right. Yeah, thanks for that overview. By the way, congrats on this morning's announcement. Thank you ... of the twelve-month data for NHP for the lead program. We'll definitely touch upon the details of that later in the discussion. So I was just wondering, maybe let's look at what you have in your platform and dive into a little more details about differentiation. The first-generation CRISPR-Cas has, you know, come a long way. Now we have a product approved and more in later stage clinical development. How can next-generation CRISPR-Cas nucleases improve upon the existing modalities? And are you just talking programmable nucleases? Yeah, the- Yeah. Yeah. So for programmable nucleases, what we've just been, you know, astonished by is the real diversity of enzymes that are out there. I mean, this is something that really gets to the core of the science that I really love, is being able to tap into the world around us to find novelty. And we've seen this in so many different instances, where we can find a nuclease that works extremely well in specific locations. It shows really high levels of efficiency, and we find these systems that have these other characteristics that really make them unique. And it hasn't been just one, it hasn't just been two. We literally have thousands of these systems that we have identified and characterized. We probably have about 10 of those, which we've moved into sort of production mode, which are driving most of our preclinical translational efforts there. And so these systems, I think also one of the strategies that Metagenomi has always had is to move forward with technology, gene editing technology, that's ready now, but we can also couple that with delivery capabilities that are ready now. And so that really has driven our first wave of translation efforts. It includes our Lead program in Heme, it includes all of our partnered programs with Ionis, and those are really leveraging these unique characteristics of the enzymes that are out there relative to Cas9, that first generation. Got it. I would add, you know, Cas9 being a great first-generation system, our metagenomics approach has allowed us to identify enzymes that are much smaller, so can fit in the capsid carrying constraints of an AAV, that have higher inherent specificity. Mm-hmm. Particularly moving toward Type V as opposed to Type II nucleases. So we really take a look at that first generation and think about how can we enhance specificity, enhance efficiency, and then ensure that we get to areas just outside of the liver that you might be able to put into an AAV. So I think those are some of the areas of differentiation. Great. Yeah, I think targetability is also, you know, one of the limitations of Cas9 or single system. You're limited by the PAM and where that PAM exists in the genome, and so having multiple systems to address that, we can target, you know, a large proportion of the base pairs in the human genome. Yeah. ... Got it. Very helpful. Can we then touch upon your base and prime editing capabilities? I think, Brian, you touched upon the longer stretch that can be transcribed reverse transcribed without error or undesirable effects. Could you elaborate on those fronts for- Sure The base and prime editing? Sure. Let me start with the base editing. We actually had another press release last week where we were able to demonstrate some of our smallest nucleases that we call SMARTs. And the SMART nucleases are really unique. They're incredibly rare. They're an ancient representative of an enzyme that that's very hard to study these organisms, and you only actually find these organisms when you're using an unbiased method like metagenomics. And these SMART systems are amazing. First of all, they're very small. That's what attracted us to them, because as Sarah just mentioned, as we think about getting outside the liver, this small size gives us flexibility of packaging. So that was the first thing. But then, when we started to realize their size was small enough, we could start to think about coupling a deaminase activity with them and creating a base editor. And so one of the problems that we hit with that system, though, is it naturally doesn't want to work. It's was very challenging to make it actually show any activity. And so we were actually able to leverage some some AI efforts to really go back and reconstruct the ancestral state of this enzyme so that we can now actually do targeted gene engineering, protein engineering to really enhance the activity. And now we've been able to see with our base editor, very small base editor, that could entirely be packaged inside an AAV, upwards of 80% activity of editing, which is really astonishing. So that's the first component is really how we're kind of using AI to accelerate some of the activity in our technology development. As far as in the base editing, as far as in the RNA-mediated integration, it's a wonderful system because you've got a template and your guide RNA that's all one molecule. And so it in the case of prime editing, it's a very small insertion or deletion that you are carrying out. So th e enzyme as a chassis targets you to a location in the genome, and then the reverse transcription actually happens at that location, and you do an insertion or a deletion there. And so where did we find our reverse transcriptase? We actually went into our metagenomic database, where we have, you know, genomes of unknown organisms that we've recreated from our discovery efforts, and we found novel variations on the reverse transcriptase that have new characteristics. And so one of them was high fidelity and the lack of the high processivity, so it gets through a much longer message without making an error, which is really important because these reverse transcriptases usually come from a viral origin, where making those changes is part of the life cycle of that virus. It needs to mutate its genome on a regular basis. So being able to find these variants that don't show that activity really allowed us to connect them to our chassis and achieve... We've now demonstrated in human cells that we can get upwards of a thousand base pairs based on a template integrated at a specific location in the genome. So we're very excited about the progress that we're making with all of our gene editing capabilities. Great. Thank you. That indeed is very differentiating- Yeah -in terms of how long can you do the prime editing? Yeah in the introduction of the new sequences. So maybe another part of the platform is large gene integration. Mm-hmm. Here it impinges on the lead program in Heme. I was just wondering, what's the benefit of large gene integration through gene editing as opposed to conventional gene therapy, and if you can talk about that, and then how you are going about to do large gene integration. Great. So I'll answer the first part, and then I'll hand it off to Sarah or Alan, to answer the second part about the application and why it's differentiated from gene therapy. But for the first part, I just would say Metagenomi doesn't have just one platform for doing large gene integration. What we announced today with Heme was our very first platform using a programmable nuclease in combination with delivering a donor DNA to do the integration. So that's the first platform, and really beautiful data on that. We're very excited about where it goes. But the other two mechanisms we've talked about are what we call our RNA-mediated integration for large genes, which I just described, and then finally, the CAST CRISPR-associated transposases. So these systems are really unique in that they're programmable, just like a Cas, you know, a Cas enzyme like Cas9 is. But instead of cutting the genome, it actually relies on the transposase components to do an active integration of the donor DNA at that site. So while the prime editing or the Big RIGs, what we call RNA-mediated integration, while that and Cas are newer technology that we're still working on translating and still working on delivery, primarily, we still have this platform that really demonstrates beautifully that we can integrate in a site-specific way and take advantage of existing promoters. But as far as the next question, I'll let Sarah- Yeah, I'll. You know, thinking about how you differentiate from gene therapy, i t's really around ensuring persistence of the DNA and ensuring persistence of the expressed protein. With gene therapy, particularly with an AAV, you deliver the given gene, and it's off of an exogenous promoter as opposed to a natural promoter, and it's also in an extrachromosomal state, so it's episomal as opposed to integrated. So by virtue of having an integrated gene that's driven off a native promoter, you avoid any kind of transcriptional silencing that can be invoked by cellular mechanisms sensing extrachromosomal protein transcription, translation that can quiet, and we've seen evidence of that with Roctavian. And so by using a gene-editing approach, particularly for large gene integration, we believe that we can overcome some of the durability challenges that have been seen with Roctavian and with other gene therapy programs, at least in hemophilia A. Additionally, by integrating at a specific site within the human genome and having hepatocytes divide and keep that gene at a given proportion, we become a really viable and attractive option for pediatric patients, not only adult patients. Whereas for AAV gene therapy, it's really only suitable for adults because the vector copy number will get progressively diluted as hepatocytes divide over time. Got it. Very helpful, and I think that's a nice way to dovetail into the Hem A program. You know, you recently designated MGX-001 as your development candidate, so congrats on that. Thank you. Can you describe the editing strategy before we, you know, dive into some of the data you presented today? Sure. Alan, you wanna do that? Yeah. Yeah, I can take that. So the editing strategy, it's a two-component system, so we edit with an MG nuclease that's delivered in an LNP, so it's a lipid nanoparticle, it's a non-viral delivery system. So we package an mRNA encoding the nuclease and a guide that directs that nuclease to the specific site in the genome, and that site is albumin intron one. So an intronic sequence is not a coding sequence. So we can make edits there without affecting expression of that endogenous gene. Okay? And at the same time, we have our donor template encoding Factor VIII. This is DNA that's delivered with an AAV, and so when that edit happens, that donor DNA gets integrated at that cut site, and this integration is mediated by natural repair processes in the cell. That integration frequency is relatively low, in the range of 1%-3%, in the monkeys that we presented on today. But that's sufficient to drive therapeutic levels of Factor VIII because that promoter is very strong and because you don't need a lot of Factor VIII to cure the disease. So that's really the molecular mechanism by which this editing happens. We have a lot of data in rodent studies showing that this happens. We've characterized that very carefully, and we now have, as we presented today, non-human primate data showing durability from that approach going out, a whole year. Got it, so durability is an important attribute of gene therapy because one and done, right? That comes with the name. Roctavian obviously under-delivered on that front, but I was wondering, how much is the contribution of that versus contribution of, for example, availability of Hemlibra? Right, so i.e., if you address the durability problem, what could we expect in terms of market adoption? So sorry to jump ahead into that. Sure. Sure. Sure. but I thought it's helpful to get that. Sure ... answer. Sure. Yeah, and you know, this morning on our call, we had one of the really world-renowned key opinion leaders to comment on that very point. What we've heard repeatedly both from Glenn, who's a member of our scientific advisory board, Glenn Pierce, as well as other international leaders, is that the field continues to push for a cure. Ever since the gene for factor VIII was identified, cloned, there was a goal of a hemophilia-free mindset. So no longer having to think about availability, breakthrough bleeding, differences in manufacturing, changes in insurance, the ability to live normally without having to think about bleeding or availability of medications. You know, Roctavian was poised to deliver a curative approach for adults with hemophilia and hemophilia A, and that lack of durability has really hindered its commercial uptake. We believe through a gene-editing approach, we can solve that loss of durability, and you know, from our discussions with advocacy leaders and key opinion leaders as well as patients, we believe that there's a continued drive for a cure, and one of the reasons that people aren't satisfied with even a transient hemophilia-free mindset with Roctavian is they do believe a curative approach is coming, and they wanna wait for it. Very nice. Thank you. Thank you for that. Mm-hmm. That's also exactly why we chose to really get ahead of this and do that durability study as soon as possible really to get around that durability question, so. Right. So we see the data that we released this morning as a major de-risking event for this program. Got it. Got it. Can you describe that data and what do they mean for us? Yeah, I'm happy to describe it. I mean, it's the culmination of, you know, several years of work at Metagenomi to get to that point. And we'd preface it by saying this is not a development candidate, but it's a pre-DC, in that it-- we've improved a number of the criteria or the, you know, characteristics of the molecule beyond that. So we had three animals in this study, and we picked the doses basically based upon our rodent studies, so these are not optimized doses at this point. But despite that, we're able to achieve therapeutically relevant levels of factor VIII in all three animals. I think, as I recall, 8-9% in the lowest animal and 75% of normal in the highest animal. Actually, 85%, I think, yeah. Eighty-two percent- Eighty-two percent. Twelve months. Yeah. But, you know, those ranges are well within the range that one would like to target for a cure. So, we were really happy to see that data, and that study is now 12 months out, and the levels of factor VIII have not dropped at all in that study. They retained that level throughout the duration of that 12-month period. And, you know, it's a study where we can gonna continue to follow those animals out for probably another 6 months to a year. Got it. And, does that, you know, how did that translate into human? I'm not sure, have that done before with this model or- Very good question. How do you predict what happens in human patients? Yeah, I mean, from my knowledge, and I think yours as well, Sarah, no one has done this sort of durability study previously with other treatments like the AAV gene therapy. And the reason for that is that those previous studies in preclinical models, like the primate, were done with human factor VIII, which of course makes sense because that's your development candidate, right? So why would you not use it? The reason you would not use it for durability study is that all those animals develop antibodies against human factor VIII. Human factor VIII is a highly immunogenic protein, so you give human factor VIII to a monkey, they develop antibodies, and then you lose the ability to detect or see expression of factor VIII. So we actually did a trick, which is, we believe, fairly unique to us, which is to use the cynomolgus factor VIII, the monkey sequence, instead of the human sequence. So okay, great, so they don't develop antibodies. So then, how do you actually detect that protein as distinct from the endogenous monkey factor VIII? Well, we have a trick there. So we introduced a single amino acid change in the protein, and we can detect that protein specifically because of that change. And so we believe this is probably the first publicly available durability data for this kind of a study ever being shown. So, yeah, we're very proud of that. Yeah. Okay. Got it. Got it. And the expectation for the ultimate durability I guess scientifically, probably there's no reason to decline due to gene editing, but do you have any, you know, comment on the ultimate durability? As I said, we see it as a major de-risking event. This wasn't done for Roctavian. Obviously, true durability in the clinic will come when we get to the clinic, and that's why we are continuing to execute and move forward to our IND-enabling plans. Another aspect of Roctavian that I think limits commercial interest is the dose of AAV used. 6e13 vector genomes per kilogram, pretty high. In the phase III study, it required a fairly prolonged course of corticosteroids, which is not without its own, toxicity aspects. As we move forward with our development candidate, we really look to ensure that we were going to be able to get therapeutic levels in the clinic with a much lower dose of, AAV. And so, that's another differentiating aspect that we hope to prove in the clinic. Y eah. I mean, I would also say that the drop-off in expression, the factor VIII from Roctavian in the clinical trials happened starting basically a year to two years post-dose, right? So we want to extend to two years so that we can see or demonstrate that we don't see that change. Got it. Got it. And, one observation from, the data you described was that there is a variability in the level of, factor VIII expression, ranging from 9% to 82%, right? So I was wondering, is that a feature of this, pre-DC in a non-human candidate, or, you know, is that an inherent variability that cannot be controlled? Wanna talk about, Sarah? Okay. You know, I think there's going to be some inherent variability that we see in the clinic. We've seen it with gene therapy. Unlike gene therapy, we did not see intra-subject variability. Often with gene therapy, you see very high expression and low and high and a lot of variability just within the same animal or with clinical data within the same patient. Because we're expressing off the albumin promoter, we see within a given animal it's fairly consistent, so that's an important difference. We do believe there's going to be variability, but we also have some levers that we can pull as we move forward into clinical development to try and reduce that. First, we are enhancing our manufacturing. This is research-grade, a very early-stage AAV. We continue to improve on our process as we move into GLP and GMP studies, so I think that's an important aspect that we think can minimize variability. We are also, with our development candidate, using a bioengineered variant to enhance factor VIII expression. With that enhanced factor VIII expression, we can further reduce doses of LNP and AAV, and we believe that enhances the degree of consistency. Again, this is gonna have to be proven out in the clinic, but it's one of the reasons why we moved forward with that variant. And then we've optimized other aspects, again, to try and minimize variability. Despite the variability that we believe will occur in the clinic, though, this is a disease area with pretty wide goalposts. I mean, if you are above 10%, if you're below 150%, that's fairly wide variability. That is a true functional cure, and it's also one of the reasons that we went with hemophilia A to begin with, because this isn't a disease with just a narrow range that r epresents a cure. You can't go too high, and you can't go too low. Obviously, we wanna be above 10%, we wanna be below 150%, but that's a pretty wide area for a true functional benefit. Got it. Great, great to hear about all the additional improvements now in MGX-001. Right? So, looking forward to that. You touched on AAV. It's good to hear that you're targeting a lower dose. You know, given your modality, that, that's, that, you know, can only be a good, good thing. What about the LNP component? You know, how confident that this wouldn't, you know, will be safe and effective? Mm-hmm. We've also made enhancements to our mRNA, to the coding sequence of the nuclease MG29-1, as well as its mRNA, to enhance nuclear localization, to enhance expression. That we believe will further reduce the LNP, in addition to the bioengineered variant that we think will allow us to not only reduce AAV, but also LNP dose. I mean, a core criteria for commercial uptake is going to be safety of this approach, given the standard of care, and so we're doing multiple things in our development candidate to ensure that we use the lowest possible AAV and LNP dose. There's a component of our durability study that we read out with safety t hat we might wanna talk about. Sure. Excuse me. So even with the doses that we use, we believe are gonna be higher than what we ultimately translate into the clinic, we saw only moderate transient elevations of transaminases within the first week or so after AAV and LNP administration that, over the long term, didn't recur. Albumin levels, as expected, remained within the normal range. We also looked at other safety parameters, coagulation studies, hematologic parameters, chemistry parameters. Saw no changes besides some acute inflammation as expected, particularly after the LNP, and so that's very reassuring to see. Got it. Got it. We're also gonna learn a lot more about dose response when we do our DRF studies later this year. Right. Can you talk about... Yeah, sure. One question is, what's next- Yeah ... for the Hem A program, right? Where do we go, right? Can we describe that program please? Yeah. So we're now into the IND-enabling phase. We've locked our DC components, and we've planned a series of NHP studies to essentially complete the IND pharmacology package together with the tox package, and so we're targeting IND in 2026. We're on track for that. We've remained on timelines since about a year ago, and so we're very confident we're gonna hit that IND date. As I said, those studies are planned. We're, you know, producing material right now. We have our AAV manufacturing ongoing. We make actually mRNA in-house, so we have that under control, and we're very confident we can do that ourselves, similarly with LNP. And I would add, you know, the other aspects that we're really preparing for is on the regulatory side and the advocacy side. On the regulatory side, we've had our first FDA interaction. It was very positive, a high degree of engagement, and really outlined a clear roadmap that we'll need to follow to get to an IND. There were no surprises there, so we feel very good about that. We'll also be broadening that to include ex U.S. interactions in the future. And then on the advocacy side, not only do we have Dr. Glenn Pierce as part of our scientific advisory board, but we've really broadened our community of experts and advisors because, as you mentioned earlier, you wanna make sure that you are making a drug and a therapy that's going to be exciting and acceptable for physicians, patients, payers. And they're really helping to guide us and continue to be very encouraging of our efforts. Got it. Great! I think that that's a very thorough look into the Hem A program, right? Can we maybe talk about some earlier or other programs? I think most notable, if you have four collaboration programs with Ionis, it looks like they have progressed very rapidly. Can you talk about what's happening- Sure ... in those, gene editing programs? Sure. Maybe I'll start, and Sarah, you can talk about the specifics. I think I'd like to just comment on the rapid nature of how those programs are progressing. The partnership with Ionis is fantastic. There are four programs that are included with that partnership. There are an additional four that are coming after the first four, but for the first four, they've actually chosen all four of those targets. We've only announced two, but we plan to make announcements on those others, and I think it's really the learnings that we have had from the last couple of years of both that partnership, as well as our development of the technology, that's allowed us to very quickly have the ability to go from a novel nuclease directed at a specific target to generating that important data that gives us the confidence that we're on the right track, we have the right nuclease and location targeted. And I think that really speaks to something that I'm super proud of, which is we built a company to really be the premier gene editing company in the space that includes all of these capabilities, both having the novel technology but also having the system underlying it to really translate those technologies rapidly. So we love having a partnership with Ionis that we're very aligned, we're very interested in the cardiometabolic space and the application of gene editing technology to that space, and we anticipate that those programs will move rapidly, continue to move rapidly. Maybe if there's anything you want to add about the specifics. Yeah, it's, you know, working together with Ionis is really a great partnership. We bring the gene editing expertise, they have the disease area expertise, having worked in, preclinical, clinical, and now even preparing, in the commercial space, for some of these targets. So they have developed preclinical models that help us, accelerate our development. They have really broad experience with, different regions and different key opinion leaders with advocacy groups. So we really, have complementary technologies that are working together to move these programs forward quickly. Yeah, and to build on what Brian said earlier, I think underlying all of this is a lot of technology platform establishment that we've done at the company over the last two to three years, which is essential to take an editing system from, you know, early stages all the way to a development candidate, and there's a lot of optimization that goes on to get there. And we put in place a lot of high-throughput systems for screening and so on, that really accelerated our ability to bring forward a target towards a DC quicker. So, the first two programs with Ionis took us, you know, a while to get to where we are now, but then the second two, that started six months later, actually caught up because of the learnings we gained during that process. Right. That, that's great to hear. Yeah. You also have a collaboration with Affini-T. Any update on how that program is going? For cell- You want to do that, Sarah? Sure. You know, we believe it is going very well. We've filed drug master files to support their IND. We believe that they are planning an IND this year, so that will be the first time our gene editing cargo goes into the clinic. Very exciting. It's a very exciting partnership, and I think really shows that our gene editing technology can be used ex vivo. It can be used in vivo, and there's just a lot of continued opportunity. Wonderful. Maybe let's touch upon cash and, you know, resources. You have guided the cash runway into 2027. That's quite a significant runway. Can you talk about your cash burn rate that you have put into those assumptions, and what milestones could be achieved by the end of that runway? Sure. So I'll say, you know, these are very challenging times. In terms of fundraising, we made a decision earlier this year to become a public company, and I think that was an important one that allowed us to build our resources to the point where we think we can get our lead program through the IND filing. You know, I think I feel really confident that given the resources that we currently have and the strong partnerships that we have, that we should be able to continue on our mission of translating this technology. Even though it's very complex systems, we really want to be able to move this technology quickly, and we think that we have the runway in order to see some appreciable milestones in both our wholly owned program as well as our partnered program with Ionis. Right. Are you continuing to pursue partnerships, to We are- further? We are. We, we feel that... You know, again, I think Metagenomi is very unique in that it has a toolbox that has a variety of capabilities. I think cell therapy is a great example of an area we've decided to target that, we can do more partnerships. And but if you think about the larger toolbox that we have, we're really, you know. Of course, the most important thing for us is moving that technology forward quickly, and if a partnership makes sense there, it either unlocks a new disease space or, you know, capabilities that we don't have, we would absolutely continue to consider partnering. Got it. Great! It looks like we are out of time. Thank you. Thanks for the team- Thank you. Yeah, appreciate it. For a great discussion. Great. Thanks. Thank you.