...Here's our customary forward-looking statements. So if there's anything you walk away today, there are three things I really wanna highlight in terms of what Korro is doing, right? The first one is, we're building a modality with the bases of oligonucleotides, as the foundation to edit a single base on RNA to convert an adenosine to an inosine. Okay, so that's the first thing. That specific platform is called OPERA, and I'll tell you a little bit more about it. The second thing I wanna talk about is that using this modality, we've nominated our first development candidate for an indication called alpha-1 antitrypsin deficiency , and I wanna present how we have generated a best-in-class profile in animal models and how we've shown translation to larger species, increasing our likelihood of success in the clinic.
The last thing I would say is that for that asset, we're going to have a regulatory filing at the end of this year to be in the clinic. We'll generate data in the second half of next year, and we have sufficient capital to get to late 2026. So if there's anything you leave today, those are the three things you wanna, you wanna grasp onto as we go through the presentation. So what's our mission at Korro? Our, our mission at Korro is to create transformative genetic medicines, okay? Specifically in diseases of high prevalence. And that goal is important because you want to generate commercially feasible medicines that are going to have an impact on patients' lives. So how are we doing that? As I mentioned, we're doing that by creating a transient and reversible way to make a single A to I edit on RNA.
The way we do that is much like siRNAs and how antisense oligonucleotides do by recruiting and co-opting an endogenous system. In this case, that system is called ADAR, or that editor is called ADAR. One of the key focuses or pillars for us as we, as we think about what we do in the company is really focus on areas where others are not. So how do we provide a modality to increase the expression of a pathway or activate a certain pathway? There are very few modalities that can do that, and we believe that we have one that can combine a combination of both genetics and pharmacology to really expand the genetics medicine's profile. The third component for us is that, you know, we've this platform has the ability to generate multiple candidates.
Going after any adenosine in the transcriptome across multiple genes is very, very unique and approach, and especially in the activation pathway. Having process discoveries, intellectual property, and know-how, both from a manufacturing as well as from a disease indication standpoint, is very critical for us. We start like every other novel technology in areas where there's lowest risk. In this case, starting with liver indications, where we know we can deliver oligonucleotides with multiple modalities, and then expanding as the delivery modalities come to bear. Again, remember, the focus for us is how do we get into larger patient populations or subsets of larger patient populations where we can provide an activation modality? This is where genetics has really helped us.
Over the last two decades, data has shown that there are single nucleotide polymorphisms or one alphabet changes, either on DNA or on RNA, that can have profound impact on biology. Folks typically pay attention to rare Mendelian diseases, where you know a causal pathway, you know a single variant creating to a devastating disease like ALS or, or many other indications. But now more data has come out where this correlation has not only existed for Mendelian diseases, but also for large chronic indications, such as Parkinson's, Alzheimer's, MASH, cardiovascular disease, et cetera, et cetera. So how do we, how do we approach this in a way that you can't touch the genome but have the same approach to go after it from a transient perspective? So our way of doing this is, as I mentioned, is to edit RNA.
Specifically, we transiently create a highly efficient edit of an adenosine to an inosine on RNA using a synthetic oligonucleotide. We do that by recruiting and co-opting this enzyme called ADAR. ADAR naturally converts an adenosine to an inosine through a catalytic process. We create an environment by delivering this oligonucleotide such that ADAR is recruited and go after a specific adenosine, and we can change that adenosine to an inosine. In most cases, that inosine, through the translational machinery, is read as a guanosine, so it is an inherent A-to-I gene mutation correction that we can do or change that we can do. How do we do that? The pillars for our platform are built on four aspects. They are very critical to us.
We are the only company that has been started based on RNA editing, so we have a deep knowledge and understanding of ADAR, the enzyme, the way it functions, the way it behaves in different tissues. It's critical for how we think about building therapeutics. The second is really having an expertise in oligonucleotide chemistry. We have been fortunate, or the entire field has been fortunate, where we have two giants that have developed a toolbox of chemistries to both modify RNA as well as deliver RNA using oligonucleotides, and we leverage on that by building novel amidites, novel chemistries, as well as using existing chemistries to design some of these compounds. The third aspect of it is our focus is on patients. So when we start focusing on patients, the most important aspect is: What is the need? What does the TPP look like?
And then we layer on delivery. And so as we think about each indication, we really think about it as fit for purpose. Where is the modality? What is the delivery tool that we need on board? And I can walk you through what we're doing for some of our programs. And lastly, although we don't talk much about it, for efficiency, given that these oligonucleotides are a string of alphabets with chemical modifications on top that can be coded into a language, we use computational methods to design or at least get to the initial designs of our compounds. That has made it very, very efficient for us to go from target, iterate on it, and get to a drug product.
Just as a case in point, it has taken us just 2 years to get our first development candidate from the time we see initial activity, and we hope to increase that pace as we move forward. We have a platform of over 32 patent portfolios covering various aspects of the platform. And so we believe that we have freedom to operate on a lot of the drugs that we're currently developing. So where do we use this tool? 'Cause without the patients, you know, a tool is just a tool.
So our focus has been, as you think about central dogma, going from the left, from DNA, all the way through the protein, we can make changes on pre-mRNA to alter gene expression, or we can make changes to the mRNA and thereby changing the amino acid sequence to impact the protein structure and function. So our focus over the last three years has been to focus on the right-hand side of central dogma, closer to the protein, where we make a single amino acid change, change the amino acid codon, and therefore change the protein and create de novo proteins to impact the disease. We also have an ability to go after and repair pathogenic G to A mutations rather than on DNA, but on RNA.
But we only focus on areas where we think that we have a differentiated way to go after that disease and patient population. So I'll walk you through some of these examples. So our current pipeline consists of our, as I mentioned, our lead program with our lead asset of KRRO-110, is focused on alpha-1 deficiency. As I mentioned, we'll have a regulatory filing in the second half of this year with data next year. And I'll walk you through how we believe we have the best-in-class profile. That is an area where we're repairing a pathogenic G to A variant, much like the second on this table for LRRK2 for Parkinson's, which is in the CNS. Again, leveraging the same technology of ADAR to go and correct a pathogenic variant of G2019S variant back to its original self.
Every other program that we will have in the pipeline, both in the liver and the CNS, is in the aspect of creating a de novo variant. We don't have much time to talk about it today, but in later presentations, we'll share with you why that is differentiated relative to any other modality for each of those indications. So I'll go very briefly over our approach in terms of what OPERA is. Our first-generation constructs are single antisense oligonucleotides, which we call CHORDs. They have high target efficiency, highly specific, based on the chemistry that we put. I mentioned briefly around the computational efficiency in terms of how we can iterate on these compounds very quickly.
For the first-generation compounds, we are leveraging existing chemistry and existing delivery so that we can show proof of concept in humans that this mechanism works, and the mechanism works at high efficiency. Just to give you a sense of why we strongly believe that the application of this technology is really wide, on this slide, it shows two sets of graphs. One on the left is focused primarily on liver cells or hepatocytes. On the right is focused on neuronal cells, and so specifically here, in this case, patient-derived cells, you can see that we can edit almost any gene in either tissue types. The reason you see varying levels is because depending on the chemistry, you can modulate the amount of editing efficiency that you really need.
One thing that differentiates us is, as I said, you know, we spend a lot of time understanding ADAR and what it can do in this context, especially from a drug development perspective. So one of the questions that we had as we started developing it is: Would we have enough of this ADAR substrate to ensure that we can get the level of editing that we need and continue to get it over a period of time? So what we did is we looked at, given that it's an endogenous process that happens in all of us at any given point in time, we looked at naturally occurring editing sites without an oligonucleotide, and then we came on top with oligonucleotides against multiple targets to see, are we altering that level of editing?
What you see is that despite high levels of potency across... We show three sites here, COG3, COPA, and AJUBA. These are three sites in hepatocytes that get endogenously edited. We show that despite us putting high levels of oligo, we are not changing that level. And we don't change that across more than these three sites, but these are just representative across the different levels. That's a big deal, right? Because as you think about the amount of times that we need to redose, the efficiency that we need, we know that we're not gonna exhaust the system, and you're gonna get high levels of editing over a period of time, given that this is chronically administered. Again, nobody's shown data like this before. We spoke a little bit about the computational efficiency.
I said we've figured out how to think about these oligos as a string of features and use computational methods to understand how do you design the best compound. So in this case, on this chart, I show where a certain chemistry is favored versus not relative to the editing site, and we can continue to iterate on this over time. We also think about this from computationally from a rational design perspective. We've looked at predicted structures of the editing site relative to the enzyme, and we've identified ways in which we can improve potency. On the right-hand side, I can show you that in vivo in mice, we've shown that by making a single change based on these models, we can actually increase potency almost threefold for one of the changes that we've made.
We continue to iterate and innovate along these lines. I talked a little bit about fit-for-purpose delivery, and so here I show that no matter what delivery we use, we are able to provide high levels of editing efficiency. On the left-hand side, we show in mice GalNAc conjugates that we can get above 50%, and on the right-hand side, I show the same similar construct that's delivered via an LNP to show that irrespective of the delivery modality, we can actually deliver these oligos at very high efficiency. Okay, all of that sounds really great. Most of that is done in the context of a Petri dish as well as mice.
Now, I'm gonna take you through how we think about alpha-1 antitrypsin deficiency and walk you through the data that we think is, you know, delivering a potential best-in-class compound that we can take to the clinic later this year or early next year. So I'll spend a brief minute on the disease itself. alpha-1 antitrypsin deficiency is caused by a single alphabet change in a gene called SERPINA1. SERPINA1 is expressed primarily in the liver. Very few cells, primarily immune cells, express it in the lung, but really, more than 98% of it is in the liver, in hepatocytes. Under normal scenario, you have a genotype called MM.
In this setting, the liver produces this protein at a steady state, and as it produces this protein at a steady state, the function of this protein is to be there when you have an insult, a wound, an exacerbation in the lung, something in the skin. And as soon as you have it, you see a spike in this protein to stop neutrophils or one of the white blood cells from attacking your own body, okay? At basal level, it's at about 35 micromolar. When you have an infection or a wound, it's about at 90 micromolar, okay? When you have this mutation, or what we specifically call the Z mutation or the Z allele, when you have two copies of that, this protein misfolds in the liver, starts to aggregate inside the liver, polymerizes almost like plaque, and gets stuck.
Because it gets stuck in the liver, it starts to destroy the liver cells. The second thing that happens is because it's stuck, it also is not present in circulation. So when you have a wound, your body is not able to generate enough of this protein to stop neutrophils from attacking your body. So you have two pathologies that are represented: one, that leads to cirrhosis in the liver, and two, that causes lung damage over a period of time. I told you a little. This graph shows you a little bit about where these patients are from an outcome standpoint, as well as protein levels. On the top half, you can see across the three genotypes of MM, heterozygous MZ, and homozygous ZZ.
You see broad ranges of protein levels, but from a median standpoint, an MM is about 35, an MZ is about 20, and then a ZZ is about 5, okay? Big, big, drastic changes between those three patient populations or three genotypes. If you look at the bottom, it talks about odds ratio of both the lung as well as the liver injury. You can see that if you're a ZZ individual, you have about eightfold risk of getting either COPD and/or cirrhosis. We've heard stories from patients that have double transplants. They're walking around with oxygen tanks up and down stairs. For some individuals, this is a very, very debilitating disease, and given that it's been passed, you can see when you have a sibling or a parent that has the disease, it can actually be pretty bad.
There is no functional cure for this patient population, and at least in the U.S., there's nothing approved for the liver disease, and from a standard of care right now is once a week IV infusion that I know some patients have been taking for the last 10 years, and she's only 35 years old. So what is our goal? KRRO-110 is a antisense oligonucleotide that is encapsulated in a lipid nanoparticle. This lipid nanoparticle we've in-licensed from Genevant, again, removing layers of risk because this has clinical precedent. We deliver it via IV to this patient. Our anticipated dosing paradigm is about Q3W or once in three weeks and above. And the goal is to convert some of those Z alleles to M alleles.
In this case, our intent is to get to above that 50% editing with this, the MZ phenotype, and get as close to normal as possible. That is a high bar for us, but we believe that that is what these patients need, and that is what these patients deserve from a clinical benefit perspective. One question that always comes up is, we get lumped in from an editing standpoint in terms of DNA versus RNA. Even though it's transient, we wanna showcase that the protein that we're finally making post-edit is entirely the normal variant. And so what we show in this graph here is that in vitro, in cells, we've looked at heterozygous MZ patient hepatocytes.
We've added our 110 compound, and we show that we don't get any edit anywhere close to the site we edit or across the gene, plus or minus 100 base pairs on either side, highlighting the specificity of our approach. As we iterated through these compounds, we did see off-target effects, and we removed all of them by layering on chemistry. You see the dotted line at the bottom that is blown up on the right-hand side? That dotted line is a lower level of detection, so we don't see anything from an off-target standpoint above that dotted line. The next goal was, okay, from a PK/PD standpoint, how does this look from an animal model perspective? There is an animal model called the NSG-PiZ model. This is a humanized mouse that is immunocompromised.
It has about 10-20 copies of the SERPINA1 gene, and we created a paradigm where we would dose this mouse every two weeks at 2 mg per kg, and then look at both editing efficiency one week post the first dose, and then one week post the last dose, which is at week 13. On the bottom, you see that even after the first dose, we're getting close to 50% editing. That increases over time, very, very close to 60%. I don't think anybody has shown this level of editing before, and not such high levels with a single dose. The second thing we wanted to look at is on this graph, it's a little bit busy, so bear with me. On the left-hand side is the protein, on the right-hand side is the function.
You can see on the left-hand side, the control animals have the shaded blue graph. That is the Z polymerized protein that gets out in circulation, and the light blue represents the normal M variant. I just showed you a couple of slides ago that we created no bystanders. It is the protein that is fully functional, that we've detected through mass spec. And so you see that with just a single dose, a week after that first dose, we get about close to 50%—50 micromoles of protein in circulation. Again, not something that people have been able to show before. Of that, of that 50 micromolar, we're getting 35 micromolar of M protein, so we go from 0 at control up to 35 micromolar in just 7 days.
Okay, and as we extend it further, you see because of the accumulation of the oligonucleotide and because of the accumulation of the editing, you see higher and higher levels. And again, if you look at the ratio of M to the Z protein, we're close to 75% or so, indicating that the amount of editing that we are seeing internally is commensurate with what we see externally. Mice are great. We've cured cancer in mice. We've created lots of therapies for mice. But one of the things with novel technologies is that you just don't know how it translates as you go into higher species. So we took time to understand what does that look like?
So we created a new construct, that is synonymous with human and a monkey, and wanted to edit an adenosine that is not at the site of the mutation, but just to be able to show that you can translate from what you see in the PiZ mice to what you see in monkeys. So this edit site is seven amino acids away from the E342K site, which is what the Z allele is. And we are able to show on the left-hand side in, with a single dose in mice, on the right-hand side, in the single dose in monkeys, that there is very good translation between the mouse and the monkey. This is the first kind of data that anybody has shown.
In our prior constructs, we've shown that the amount of protein that we generate in circulation is commensurate with the amount of editing that we see. So if you start connecting the dots, we see it work in mice, we see it work in monkeys, we see it work in human cells, so we are very confident that this will likely show up in the human studies where we generate the data next year. So what's next for us? We've looked at efficacy. We believe that we have one of the best profiles out there from a compound, and we think that we can demonstrate this in humans very, very soon. From a safety standpoint, we spoke about the off-target effects.
When we've tested this in mice and in monkeys, we've shown that it's very well tolerated, better than Onpattro or patisiran, which is an approved construct. And we've also shown that it can translate to higher species. So you add all that together, you know, our goal now is to submit this regulatory package, and then start our clinical study as soon as possible. I'll take one second to highlight the team. We are very fortunate to build a strong team that comes from a drug development background, as well as having a strong expertise in chemistry, delivery, as well as development experience at the table. So we have Steve Colletti, who spent 25 years at Merck. Kemi joined us as our Chief Medical Officer. She was at Ultragenyx most recently.
Vineet, our CFO, is at 14 years at J.P. Morgan. Todd has started 4 companies, with multiple IND filings, and the list goes on. This is just the management team, and the layer below brings a lot of experience and, should I say, gray hair to the table, from a drug development standpoint. With that, I'll stop, and happy to take any questions if there are any.
Great job, Ram. Congratulations on a very great translational study. Maybe you can comment on the timeline for the lead pipeline to, you know, the IND?
Yeah. So what we've said is that by the end of this year, we'll file from a regulatory standpoint. We haven't said which jurisdiction yet. We anticipate getting data in the second half of next year. We don't, we haven't said what kind of data, but it's anticipated that we will show something in the ZZ homozygous individuals with levels of protein.
Fantastic, thank you.
I know you had another question, but I think I'm out of time, unfortunately. Thank you.