Hi everyone, and welcome to the 43rd Annual JP Morgan Healthcare Conference. My name is Natalie Collins, and I'm an associate on the JP Morgan Healthcare Investment Banking team. Just a reminder, we'll follow 20 minutes of presentation time, followed by 20 minutes of Q&A. And with that, I'm pleased to introduce the CEO of Korro Bio, Ram Aiyar. Thank you, Natalie, and thank you for the JP Morgan team for the presentation. My disclaimers and forward-looking statements. I thought we'll keep this conversational rather than a presentation so that we can both interact rather than just me present. So hopefully I'll tell you a little bit about what we're doing. If you lead with anything today, it's three things. We are a genetic medicines company bringing a novel modality to go into prevalent diseases. Three years ago, we set some stakes in the ground in terms of what we will execute on, and I'll tell you what we have done so far. The second thing I want to leave with you is our lead program. As of today, we announced KRRO-110. We dosed our first two participants in Australia. It's a big milestone for the company to go from a novel modality all the way to the clinic in a very short period of time. So I'll tell you a little bit more about it, as well as the data that we anticipate to get in the second half of this year. And lastly, much like we did three years ago, we're going to set the stage for the next three years in terms of what we're going to do, which is three programs in the clinic across two tissue types with one platform. That will be the crux of our 3-2-1 strategy that I will talk to you a little bit about. Okay, so bear with me as we walk through the presentation. Korro is not my first company. I've started multiple companies. This is my seventh company. And the last company I was at, we had identified a point mutation, a single missense variant that created large outcomes for patients with chronic kidney disease. We're talking about a 50% reduction in mortality. We looked at epidemiology based on that genetics and identified that a single variant had a huge impact on a large patient population. When Korro was started, the idea was a lot of the founders were from the CRISPR-Cas9 world. They were some of the early pioneers that developed it. The idea was, can we get beyond where DNA editing was? Can we get beyond rare Mendelian diseases and start to think about areas where you can make specific edits? Can we actually impact and learn from genetics and go after highly prevalent chronic diseases? So that's our mission. Our mission is to get to prevalent indications with a drug modality that people know of, which is in this case oligonucleotide. We do that using RNA editing, so we don't touch genomic material. We make a single alphabet change on RNA so it can be transient, it can be specific, it can be reversible. We do that with a modality that's an oligonucleotide. Think inclisiran, think siRNA, think antisense oligonucleotide. It's a chemically modified synthetic compound. Regulators have seen it, patients have seen it, physicians have seen it. So that's really the platform that we're building. The secret sauce is, of course, the design, how we make them, how we deploy them. And finally, we wanted to, being a patient space or an indication space, our sandbox, if you want to call it, is where we don't compete with any of the others, which is all the others are focused on silencing and we're focused on activating. So we either repair a certain protein and activate it, or we make single edits to activate a certain biological pathway. And so I'll walk you through what all of that looks like. So when I say RNA editing, what does that really mean? Well, it turns out there is, much like Argonaute and much like the RNAi mechanism, there happens to be an enzyme that already does that for us in every cell in the body. It recognizes specific constructs in cellular RNA depending on how the structure forms, and it can go and edit or change an adenosine to an inosine. That inosine is usually read as a guanosine through the translation machinery. So it's an inherent system, much like how CRISPR-Cas9 systems exist in bacteria. This is a human editing system that already exists in each one of us. All we did is create a similar profile that this enzyme recognizes. We create an environment where the enzyme can find the adenosine of choice and convert that adenosine to an inosine, and so we can affect an A-to-I change. And we do that with a chemically modified synthetic oligonucleotide. So all the compounds that we leverage, all the benefits of the Ionis and Alnylam and the Arrowhead of the world that have spent time developing chemistry and delivery, we sort of leverage that to identify the right designs to do this. You may ask me, okay, so where can you apply this? You talk about editing. People immediately go to rare Mendelian diseases, and so I want to paint a picture of that's not our playbook. That's not our playbook. Our focus is really to get beyond that, so when you think about central dogma, when you go from DNA on the left all the way to protein on the right, there are many places in RNA you can play a role. The first one is when DNA gets transcribed or pre-mRNA, where it can be impacted by regulatory regions. We can make an edit, learning from genetics to convert an A to an I, and either increase gene expression or decrease gene expression. All we have to do is look at genetics, look at epidemiology, and we know where to make those changes, so that's one place when an edit makes a difference. The other place where an edit makes a difference is you're thinking about the protein. You can change the structure and therefore the function of a protein by making a single edit. Now, that has a couple of different flavors. If you have a pathogenic Mendelian disease, you can actually revert that back to its original protein and therefore fully functional. And I'll walk you through a couple of examples. The other way we can actually play a role is just making a single edit. You can change 12 amino acid sequences. Now, you may say, 12 out of 20, who cares? But there's a lot of biology you can play with that. And so just by making a single edit, 25% of your transcriptome that is made of adenosines, you can really impact the space. And so when you think about the vastness of products, the vastness of biology that you can play with, you can do this with this platform. Easier said than done. You won't believe me if I tell you you can do that. We intend to show you with products that we can take it step by step and show that we can actually get there. It sounds easy on the surface. It's chemically modified oligos. Ionis and Alnylam have done it. Anybody can do it. It turns out it's a little bit more complicated, like most things in life. The key to making a good drug or a good product candidate in this context is to understand this enzyme in and of itself. We have expertise in terms of purifying this that nobody else has. The second component is chemistry and how it applies to a certain target and a certain construct. We've spent a lot of time both empirically understanding what that looks like, as well as learning from computational models. We leverage known delivery. There's only so many things we can work on. Delivery is not something that we want to work and spend a whole lot of time on at the current moment. And so we leverage known delivery. The beauty of that is that so many other companies have now brought to bear delivery to liver, delivery to skeletal muscle, delivery to the eye, delivery to the lung. And so we have this ability to leverage known delivery modalities and tack on our editing oligo. And lastly, all of this is capable through a computational platform that we put together where we can design these compounds in a very, very rapid fashion. So we have an industry-leading less than four-week turnaround time from a design test iterate perspective for over 200 oligos at any given point in time. That takes time to build, so that's what we've been doing over the last 3-4 years as we take our first candidate forward. When we started out the company, we wanted to take out risk through layers. And the way that we wanted to do that was, first, we wanted to ensure that we go after an indication where you can make a single edit, understand biologically the impact, and then know that looking at it in the periphery or in blood, you have an impact. You have a very good idea very soon if the drug is working or not. So alpha-1 antitrypsin deficiency, which is our lead program, where KRRO-110 is our lead product candidate, which is in the clinic right now, fits that bill. Within 100 days, within 100 patients, we will know if the drug works or not. The challenge there was to get super high levels of editing because the protein is present at a very high level, which is why we picked a delivery modality that will cater to getting that level of editing efficiency, which in this case was a lipid nanoparticle that we licensed. That is our large indication. It's a rare population, but it's a pretty prevalent rare population. It's a known market, and we'll know if the modality works or not. Everything beyond that is what I talked to you about in terms of making a single edit to impact protein biology. And so our second indication, which we're going to disclose later in the year with the development candidate and a data package that can highlight the benefit of an RNA editing outside of rare genetic disease, is something we're going to do pretty rapidly. There, we're stabilizing the protein by removing a ubiquitination site, increasing the half-life and therefore the function. And we picked a subQ delivery for that because we didn't need a high level of editing. We needed a long duration and durability for that patient population, which is why we went to subQ. So we've been very thoughtful in terms of how we built out these indications and built out the delivery associated with them. Finally, today, I'm not going to talk much about it, but personally, both my Chief Medical Officer, Kemi, and I are super excited about it. We've made some inroads in terms of ALS with a target called TDP-43 that's moved into preclinical development. We're not going to present data today, so I don't want to tease you too much, but hopefully over the next six months to a year, we'll start sharing the data. That will highlight what the modality is capable of that others cannot do. As I look back at 2024, we accomplished a lot as a company. We changed and converted to a clinical stage organization from being entirely a discovery stage. We advanced multiple discovery targets with novel biology into preclinical modalities. We announced a partnership with Novo Nordisk for just two targets in cardiometabolic indications, again, with the vision of getting into prevalent indications. We closed a PIPE of $70 million in Q2 of last year to extend our runway. Then as you think about our team, not just the management team, in addition to the board and the clinical advisory board, we've brought all of that together in just a matter of 12 months. Every year, I stand up in front of the company and say, it's a big year for us. 2025 is going to be a big year for us. So it's 2026, so it's 2027. But in 2025, we intend to accomplish quite a lot, which is, at least for the lead program in the second half of this year, demonstrate what we have demonstrated with our preclinical studies that we have the potential for a best-in-class compound in alpha-1 antitrypsin deficiency at 100% efficiency with KRRO-110. This study is called REWRITE. It's a phase one two, and we'll present SAD data in both healthy and PiZZ individuals. I'll walk you through the clinical study design. We'll nominate a second program, as I said, sometime in 2025 that can expand the visibility of what RNA editing would look like. And then we're going to progress both our wholly owned as well as partner programs in a very short period of time to get to some specific milestones. The partnership with Novo Nordisk has been fabulous. As we get into cardiometabolic indications, we've just started the collaboration and the work in October, but we've made significant progress towards that end. All of this is supported by a cash runway into the second half of 26 that will enable us to finish the study for alpha-1, as well as move some of our other programs into the clinic. Much like what we did for the last three years, where we put a stake in the ground in terms of where we're going to go, we're going to today put a stake in the ground as to where we are for the next three years. As I said, this is our three, two, one strategy, three clinical programs, two tissue types with a single platform. We want to do that thoughtfully. We started to do that with alpha-1. We said we'll nominate our second candidate, a third one in a different tissue type, and then really expand and iterate on our platform as we move over the next three years to improve potency, improve delivery, and show the biology that can come beyond this. I'm going to spend, for the next couple of sections, a little bit of time on the platform, and then for the majority of it, I'll spend time on KRRO-110 and the data that we've generated. Our platform is called OPERA, and the designs that we have as our first generation are called CHORDs, no pun intended. It's a single-stranded oligonucleotide. The designs that we have done are based on known chemistries as well as novel chemistries. We've shown and demonstrated that it's highly specific, very efficient in terms of delivery. I mentioned a little bit about the computational tools that we've applied, and we believe that this will get us to our first couple of compounds. In 2021, we had intellectual property filed to identify specific nucleobases and chemistries that can improve potency. We did that through computational chemistry. We've identified them to have in vitro efficacy, again, learning from ADAR and the binding tools that existed, and then showing in vivo potency, and some of those compounds that have already been in manner now are being applied towards our lead candidate. We wanted to expand on delivery so over the last year, we've shown that, one, we can deliver using lipid nanoparticles, which is part of our lead program, which is on the left-hand side. You can see the profile on this graph. It starts very high at super high editing levels, as soon as, even as early as a day from dosing, and then it starts tapering depending on the stability of the oligonucleotide. We've then dosed in the middle panel with a subQ delivery, showing that the profile is slightly different. It starts low and then increases, and then it's very durable, even in mice, even at 10 mg per kg. And then what is new today is data that we share from intrathecal delivery in CNS, where we wanted to get to certain tissue types in mice. And we've done that all with a single target, demonstrating that this mechanism not just exists in the liver, but also outside of it. Over time, we'll demonstrate more into the different tissue types that we can get to, but this is a start in terms of the areas that we want to go after. So for the next few minutes, spend time on our lead program, KRRO-110, on AATD, which is alpha-1 antitrypsin deficiency. I'll spend a little bit of time talking about the indication, a little bit of time on the preclinical data that we have, as well as finally on the clinical design, which I'm sure you're looking for. So the indication itself is a very interesting one. When you think about the biology, the disease is caused by a single missense variant. So one alphabet change of a guanosine to an adenosine in the SERPINA1 gene causes this protein to misfold. So when you have the homozygous variant of this disease, which is the ZZ phenotype, you get a pretty severe manifestation of the disease. So the protein misfolds, accumulates like plaque, and polymerizes in the hepatocytes and the liver cells, causes liver damage through an inflammatory process, and then cirrhosis. Unfortunately, the function of this protein is across the body, and so you need pretty high levels pretty quickly because it's an acute phase response. It is in response to a pathogen or an injury, and so when you have all of this protein stuck in the liver, you don't have enough of it, and where it manifests itself the most is in the lung, so you have two organ pathologies with a single point mutation in a single source, which is the liver, so that was our solution. Our solution was, let's target the liver, in this case with lipid nanoparticle that we've licensed from Genevant. That lipid nanoparticle has already been in man before. We've run experiments to test with different vendors to look at which would be the right lipid nanoparticle for us. We wanted to make sure the amount of oligo that we deliver was pretty high because the amount of protein in this space is a lot. Encapsulated in that is an antisense oligonucleotide with chemistries that we have, and we deliver it IV. The hope is that we edit a certain percentage of the transcripts, ideally 100%. We don't think we can get there in the first generation of compounds, but we'll get pretty close. But the idea is some proportion of those transcripts into the normal M variant versus the pathogenic Z variant. So that's how the mechanism works. What are we shooting for? And here we want to learn from genetics. At the end of the day, nature has done some of the experiments for us. It's a great place to understand what to shoot for. If you look at the graph on the left-hand side, on the top, it shows, based on the different genotypes, being homozygous all the way through ZZ being homozygous with a Z allele, with the MZ being a heterozygous individual, you can see median levels of the protein in circulation with the diamonds. It's about 35% in an MM. Some of us, or most of us, hopefully, fall in that category. When you look at the ZZ phenotype, the circulating protein levels are close to four or five. Severe decrease. There's actually a straight line between the three. You can connect the heterozygous to a homozygous and an MZ, which is great. You have an idea that if you have a linear dose response, you can actually fall on that line. Depending on what the protein level is in circulation, you have a good idea of what editing efficiencies are inside the liver. If you then look underneath, you find the odds ratio for each of those phenotypes. So if you have a MM, you're normal, or if it's non-smoker, non-alcoholic odds ratio of one. And then if you move to the ZZ phenotype, it's nine. It's almost nine. You have a lifetime risk of having either COPD emphysema or cirrhosis or both. And so what we're trying to do is move them left towards the. The higher the amount of editing you get, the bigger the benefit, or the quicker the benefit, or the more consistent the benefit. So that's our goal. We're trying to shoot for getting patients as close to normal as possible, the highest amount of editing as we can. And we hope to demonstrate that in human studies. What have we done so far? Over the next few slides, I'm going to show you two things. One is in transgenic mouse models, rodent models, which have the ZZ phenotype. How does our drug fare? We've done the same thing in multiple mouse models, and the editing efficiency is very similar. Then we wanted to see, because we didn't know about this modality, how it would work in humans. We ran non-human primate studies to look at allometric scaling, to look at how these compounds translate across species. So I'll show that data. It is not KRRO-110. It's a different compound. It's a surrogate. And then finally, I'll walk you through the clinical study. So in this first study, this is the NSG-PiZ mice. It's an immunocompromised mice. It has somewhere between 10-20 copies of this gene inserted. So we've dosed every other week at 2 mg/kg, and we've taken out seven days post the first dose, seven days post the fifth dose, and then seven days post the last dose. On the right-hand side, it shows the editing efficiency. You can see that the editing efficiency is north of 50%, even as early as seven days post the first dose. The reason I say that, if you remember earlier on the slide, that has shown you an LNP profile, as early as day one, we saw about 70-80% editing, and then tapering on, very much similar to an IV profile. We didn't sample early on, but that would be our expectation. It should manifest itself in the protein that I will show you in the next slide. On the left-hand side is protein levels. The dark shaded region are ZZ aggregates that you see come out. The light blue is the M protein that you see, and you can see we go from somewhere around 15 micromolar all the way to about 60 micromolar by the end of the study. That is a big change, and so when you think about not only do you see a decrease or an increase in the M, you also see the Z protein sort of stabilize over time. What you're not seeing here is the other window that's there in the liver, because all of these polymers exist in the liver. So when you're editing it, that M protein has to come from somewhere. So where it's not there is in the liver. So you see that liver benefit also manifest. Our hope and goal is that we can produce this therapy that benefits both clinical benefit both in the liver as well as in the lung. We have the preclinical data to demonstrate that. On the right-hand side, even though we can show that using mass spec that this is the real protein, we wanted to make sure that it's active. It's an indicator that at week 13, we showed that relative to the Z protein, do we have an active protein or not, which it is. I mentioned to you about the surrogate. The surrogate is about seven base pairs away from the E342K site that is pathogenic. It is in a region that's homologous with monkeys and humans. It's a different construct, same length, same LNP. What you see on the right is how that manifests itself both in the PiZ mouse model as well as in monkeys, and then what that modeling looks like that would translate into humans. Note that the efficiency of editing of this compound is probably a fourth of what our drug candidate is in a similar model. You can see that we can get durability and allometric scaling as we go up in species. You can see that the stability of the compounds increase as we go up in species. Our hope and what we have conviction to demonstrate in the clinical study is that we can get Q monthly dosing with high levels of protein in the clinical study. Just to recap, we think that we have a potential best-in-class compound. We've shown the preclinical efficacy. We've demonstrated in our hands preclinical safety in the context of we haven't seen any off-target effects. We don't take ADAR away from its day job. We've shown that it's well tolerated in monkeys as well as in rodents at levels probably tenfold higher than patisiran, which is an approved product from Alnylam, and then we've also shown the translation into higher species, so for a new modality to have all of that together from a package standpoint is a great place to be as we sort of get into the clinic. What does the clinical study look like? This is our phase I/II study called REWRITE. It's comprised of two parts. Part one, which is the SAD portion, part two, which is the MAD portion. In the SAD portion, we're going to study both healthy volunteers, PiMM individuals. We'll dose escalate up to six cohorts. At doses that we think are therapeutically relevant, which we believe will be around the third cohort or so, we'll start staggering and in parallel dosing PiZZ individuals. The part one is going to be placebo control, two drug, two active, one placebo. And we have up to 64 subjects that we look at in the part one. So the idea would be at the end of this SAD study to show the data in its entirety, both safety as well as target engagement and efficacy with a single dose. We would then at the same time start the MAD portion as the healthy volunteer portion still continues to dose escalate and hope to show that we can hit steady state in a pretty rapid interval. As I mentioned, data is anticipated in the second half of this year, and we intend to finish the study in 2026. It's been a tremendous four years for me, and definitely over the last year, it's been a whirlwind of activities across a lot of areas. But 2025 is going to be a big year. This is the first time that we'll show with our platform within the next, call it nine months or so, that the platform is working, that we have the likelihood to show a best-in-class compound through this interim data from our SAD study. We're going to nominate a second program that will expand the visibility of what RNA editing can do. So the question about which is better, DNA versus RNA, shouldn't really exist. And then as we push forward on our other pipeline programs, hopefully we can share more of the data as to how differentiated these programs are. And not just in the context of where our competitive advantage is, which is understanding the biology product fit. So you'll see that come through more and more over the next few years. I want to end with that and happy to take questions. Thank you for your time. Thank you so much.