Hi everyone, thanks for coming out today to listen to Korro Bio. My name is Ishita, and I'm from TD Cowen. I'm pleased to introduce Korro's CEO, Ram Aiyar. I'll pass it over to the company to go ahead and get started.
Thanks, Ishita. Good morning. Yeah, maybe let's take a moment out of the day to have a conversation rather than, you know, glean into the gloom and doom that's happening outside. Just a brief moment of levity. Bear with me. We'll keep this, you know, feel free to stop me, and I'm happy to take you through what we're doing at Korro. Okay, my name is Ram Aiyar. I'm the President and CEO of Korro. Here are the normal forward-looking statements. Some of you may know this, but just to reset expectations with everybody, at Korro, what we do is develop genetic medicines for both rare as well as prevalent indications. The goal for me as I joined into Korro was to build an organization that can start to take the combination of both genetics and pharmacology to a larger patient population.
My belief is that Korro is one of the ways to do that. What do we do at Korro? We modify a single alphabet on RNA, convert an adenosine to an inosine. We change RNA, we do not touch DNA, and we're able to make a very highly specific edit, which is transient and modular and titratable. We do that using a platform, using chemically modified RNA or oligonucleotides like an siRNA or an ASO, precedented product. That means that we need to build a modular platform, building on other companies over the last two decades' chemistry deliveries so that we don't need to reinvent the wheel on that drug modality in and of itself. We can deliver multiple products to multiple tissue types.
Lastly, you know, all of that sounds great, but you know, our sandbox, as we play, is in activating biological pathways. There are many ways to knock something down, to insert a gene, but what we are talking about is activating a pathway in which, you know, others cannot do. That is really, you know, an area where we leverage genetics. We can look at RNA, understand where some of those variants are, and start to think about modulating biology. Earlier this year, we laid out our 3-2-1 strategy. We need to make T-shirts. We have not made T-shirts yet, so those are coming. The idea is to get three products to the clinic over the next three years. Our lead program, which is KRRO-110 and Alpha-1 antitrypsin deficiency, is already in the clinic.
I will touch on that in a minute. We are going to nominate our second program that's going to be in the liver, delivered sub-Q, and then the third program in a tissue outside of the liver. The goal for two here is to access more than one tissue type, starting in the liver and then expanding beyond that just to show that this biology can work everywhere. Finally, you know, my belief is that you don't have a platform until you iterate and see it works across multiple tissue types, and you can generate compounds pretty rapidly. Our anticipation is that over these three years, we would have a solid platform that we can work on to access any tissue, any target, any delivery modality over the next three years.
I'm going to touch a little bit on Alpha-1, specifically our KRRO-110 program, because this is front and center for everybody. I'm going to lay out why we believe that this is a potentially best-in-class therapy for patients with this terrible disease. It is an RNA-editing oligonucleotide that's encapsulated in a lipid nanoparticle. It will be delivered IV. Our belief is that we have a modality that's highly specific. We had shared data in preclinical studies, both in rodents as well as non-human primates. We have the ability to make a very specific edit, in this case to make the normal protein, which is Alpha-1 in these patients where the protein is mutated. We don't have any bystander edits, off-target edits. We're not creating a mixture of proteins. It is the highly specific M protein that is a normal protein that we're making.
I believe that the high specificity is important as we think about activating or ensuring that we can block neutrophil elastase, which is important in this patient population. We also think from a regulatory standpoint that gives us a leg up from a favorability standpoint across different jurisdictions that we make the actual protein in and of itself. The second component is really around efficiency. One of the big theses for us was we needed to get protein levels high and editing levels high, and I'll get into the details in a second. The goal was to get and learn from genetics to get as close to normal as possible. Close to normal is about 50% editing. We've been able to demonstrate that in preclinical studies to show that on an AUC basis, we are at about 60% editing in these rodent models.
We hope to translate that into humans at reasonable doses. This is, again, important because it is relevant for this patient population that is very heterogeneous to get to levels that are as close to normal levels of Alpha-1. Only instance where I would say more is more, where, you know, the more editing you get, the more protein you get, the better the liver and the better the lung. That is why the efficiency that we provide with a lipid nanoparticle with our data is very favorable. The last component is, of course, safety. Most folks will be like, "Yeah, sure, you can get that with an LNP. Who really cares? You know, are you really going to be safe?" We are starting to demonstrate that we are highly differentiated.
Again, when you think about an LNP encapsulating an siRNA or an oligonucleotide, it has a very different profile than a lipid that encapsulates an mRNA. Those two profiles are very, very different. You can see that with our NOAEL, which is at about 5 mg per kg in non-human primates with a single dose. That is a pretty high dose from an LNP standpoint. As we think about, you know, scaling into humans, and we have shown that with our non-human primate data, as we scale allometrically in higher species, our efficacious dose goes down and our safety window continues to increase. We believe that we have a large, you know, favorable dose regimen and favorable safety margins as we get to the clinic. The big news, of course, is that we stated that we dosed patients in January.
I'm pleased to announce today that we've enrolled two cohorts already, and we're starting to dose escalate in our SAD and our Part 1 study, which we call REWRITE. A lot of progress in a very, very short period of time. Just taking a step back, so we're all on the same page. What is RNA editing? RNA editing is delivery of a single-stranded oligonucleotide in our case, binding the target. The design of the oligo goes and binds a specific target. When we bind that target, we are able to recruit an enzyme called ADAR. This is an endogenous enzyme that's present in every cell. It can come in, latch on to the double-stranded complex, and convert that adenosine or an A to an inosine very, very specifically with an editing window of just one, so very high specificity.
Therefore, the resultant protein that's made has a guanosine in it. When it's translated, that inosine gets translated as guanosine. It's an A to G edit for all intents and purposes from a protein standpoint. Where would we apply it? You know, as you think about central dogma going from DNA all the way to protein, there are many different places you can intervene. You can intervene pre-mRNA before the splicing has occurred. You can intervene post-mRNA translation. Both of those mechanisms are slightly different. Our current focus is on the mRNA portion where we know that the protein that's going to be translated. We are able to come on board, make a very specific edit, change an amino acid sequence, and therefore the structure and function of that protein. I could spend a lot of time on that.
I'm not going to do that today and happy to spend more time down the line in terms of why that is important and all the fun things that you can do to impact biology through that mechanism. Our platform, as I said, you know, we have the beginnings of a solid platform. You know, we are one of the only companies that have been started in the space from an ADAR standpoint. Some of our founders have deep experience in ADAR biology. That is a crux of how we design some of our compounds. We then leverage on existing chemistry as well as novel chemistries that we've built over the last four years to improve editing. I'll show some data to show how we've been able to do that. Finally, we're already doing all this work on target ID, all this work on designing these compounds.
We don't want to take the risk on delivery. We want to make sure we can leverage existing delivery. That is how, at least for our first three programs, we want to make sure that we can get there and deliver these compounds and then come later in terms of novel targets, novel indications, et cetera, et cetera. I won't touch on this too much today. We've spent and built a large component around computational biology, both to help on target ID as well as driving efficiency. That has been the engine for us to make this rapid progress. Our pipeline, as it stands today, is a combination of things that we call repair a certain protein sequence versus things that create a novel mutation in a protein sequence to provide structure function benefits. You can see that it's separated into two tissue types.
Our lead program and our second program is in the liver with two different delivery types. The first one with a lipid nanoparticle, the second one with a GalNac sub-Q. As we go into additional targets, you know, we'll leverage, as I said, existing delivery to access some of those tissue types in a very stepwise fashion. We have a collaboration for two targets with Novo Nordisk that we signed late last year. The idea there is to get into large patient populations as we think about, you know, accessing a known drug modality in large indications. That partnership's going really well. What do I anticipate for this year?
We're going to share interim data from our SAD study, and I'll get into the study design in a little bit, really focused on sharing safety tolerability as well as activity in the second half of this year. We'll nominate our second candidate this year and disclose the indications and the potential for a pipeline and a program with that second product later this year. You know, as we think about progressing our wholly owned pipeline, you know, we'll start to share data in other extra hepatic tissue types where this technology can actually be used. We have a cash runway into the second half of 2026, so definitely more than a year's worth of runway post-data that's going to be sometime in the second half of this year.
In the interest of time, I'm going to walk through this very quickly from a platform perspective because the proof is always in the pudding. There are two things I wanted to touch on. The first one is, you know, we mentioned we have a deep expertise in terms of ADAR. We've understood structure function relationships here and the binding of RNA to the ADAR protein. We've made some discoveries about two or three years ago in terms of how that happens. We've used computational models to come up with chemistries that aid in improving the editing efficiency. That has translated to improved efficiency in vivo. We've used a very systematic way of answering certain questions as to how do we improve potency and activity using this mechanism.
Just to put it in comparison, RNA-editing oligonucleotides, no matter which company it is, when you compare that with siRNA or ASO, the potencies are very, very different. We're still not there in the context of once-in-six-months dosing as an siRNA. It took them 20 years. We think we can get there faster, but we're not there yet. The second component that we've been able to leverage is really delivery. I wanted to highlight that editing does not necessarily work only in the liver. It has an ability to work outside of the liver. Even within the liver, the idea is to think about the profiles you need from a target product profile perspective, and you have an ability to mix and match in terms of what is best for that indication.
On the left side, I show delivery using a lipid nanoparticle of an oligo that edits beta-actin. The idea there is an MC3 LNP precedented used in Patisiran. You can see that it gets to 60%-80% editing early in timeframe, much like an IV profile, and then comes down versus a sub-Q therapy with GalNAc. It takes a little bit of time because it's delivered sub-Q. The distribution takes a little bit of while. The enzyme escape takes a little bit of while, and the profile is slightly different. You can achieve, with potent compounds, high levels of editing depending on the profiles. On the last graph on the right-hand side, I share a little bit about the spinal cord because for our indication in ALS, we're going to be focused on that.
I wanted to highlight that we can edit in the spinal cord and edit at different locations in the spinal cord using just naked delivery. The most important asset for us at the current moment and, I imagine, of interest for you guys as well. I'm not going to spend a whole lot of time on the indication. It is a Mendelian disease with a single-point mutation that's a G2A variant. What we're really doing is the protein that's made, we're converting it back to wild type. The unique thing about this is that you have a gain of function in the disease where you have aggregation of the polymers in the liver that leads to cirrhosis.
You also have a loss of function where this protein is not in circulation and causes, you know, degradation of the liver or the skin where it's really needed to stop neutrophils from its activation. What do we do? KRRO-110 is an RNA-editing oligonucleotide. We've encapsulated it in an LNP that we've licensed from Genevant Sciences. This specific LNP has human exposure and has human precedence at doses that we think will be efficacious. The idea is once we deliver this into the liver, we can use the oligo with the endogenous ADAR, fix the edit, and then you have an amount of M protein that's produced, which is the wild-type protein, the normal protein in circulation. The nuance here is that you have, when you make this edit, you will have some amount of Z protein in the liver and some amount of Z protein in circulation.
It is contingent on the amount of editing, i.e., if you have 100% editing, you have nothing in the liver. All M protein is out. If you have 50% editing, 20%-40% of the Z protein gets out. The remaining 10% gets stuck in the liver. Why is that relevant? It's relevant because when you look at the odds ratio of these patients over a lifetime, which is at the bottom of this graph on the left, you see that although you can get enough protein in circulation, the odds ratio is still 1.5 for liver disease, even if you are a heterozygous individual or have, for our case, 50% editing. There is a market decrease, right? You're going from 8 to 1 or 1.5. That's still a benefit. I think the goal for us as Korro is to get between that MZ and phenotype.
MZ being heterozygous, being a normal individual, that is a range that we want to be in, in the context of where editing is and what will be beneficial for patients. Also from a regulatory standpoint, what will be beneficial from approvability. What is shown above the graph is the levels of protein or the 95% interquartile, 95% confidence intervals. You can see the diamonds indicate the median. The median is about 28 for 18-20 for MZ, and about 4.5-5 for ZZ. That is the range. That direct line, depending on protein, you have a pretty good idea of how much editing you are getting. You can look at gene dose effects in terms of just based off of protein levels, you have a sense of, you know, how much editing you are getting.
The goal for us, as I said, is to be at that 50% or above at most points in time. What have we demonstrated to show that we can actually get there? We've shown in there is a transgenic mouse where you replace and insert the human gene. It leads to a phenotype of both the liver as well as the, it doesn't replicate the lung phenotype, but it does replicate the liver phenotype. The first thing we want to do is can we get the amount of protein in circulation? What I share on the left-hand side is ranges in terms of what our sawtooth of editing looks like. It's a predictive modeling value where we get 80% at peak all the way down to 45% at trough levels with medians around 60%. That translates to high levels of protein in circulation.
It's that level of editing that leads to a greater than 50% M protein in circulation. If you imagine some of it is still stuck in the liver, somewhere between 10-15%, that's why you're seeing the disconnect. You need to see more amount of M protein in circulation and at total numbers that are very high to be able to get to 50% editing and above. This has been dosed every two weeks in this mouse at 2 mg/kg. We can show that over this 13-week timeframe, we're able to both accumulate as well as increase the amount of total protein in circulation. In the liver, we shared this data almost 18 months ago. We show that concomitant with the increase in protein, you're seeing a decrease in the aggregation.
What you're seeing here is two images, one gross histology and then a magnification of that region. On the top is a reduction in the aggregates. At the bottom is a reduction in the Z protein between the vehicle and 110 treated. We then wanted to make sure that we actually see translation in humans. We started to do higher species. The Z mutation doesn't exist in monkeys. We created a de novo variant in normal monkeys to show that we can measure it in the monkeys. It is a human version, a human sequence, so that we can look at the mouse as well. You can see that on the right-hand side, when you look at the actual data, durability goes up, editing goes up, very similar to allometric scaling for small molecules or other ASOs.
Actually, as we go up higher in species, we anticipate getting that. This is a surrogate. This is not 110. If you notice, the surrogate edits only at about 20% at day seven, whereas 110 edits at 60% at day seven. Those are the reasons why we believe that we have a best-in-class profile. I think that this is as a generation one, as we think about providing the benefit for these patients, it's really, you know, we're pretty excited about the opportunities that this can bring. Finally, I wanted to touch on the clinical study that we call REWRITE. As I said, we've finished two cohorts so far. You know, the study started in Australia. We're starting to expand into other jurisdictions as we stagger the study and start to get into PiZ individuals in the SAD portion.
That SAD dataset at base case is what we would want to share at the second half of this year. It is placebo control. That way, you can look at both safety as well as tolerability as well as activity all at the same time in at least two dose cohorts showing dose responsiveness as well as activity in PiZ individuals. Are we going to see all 64 patients? It is the 64 patients also includes the MAD portion. You are likely going to see close to 48 individuals in the SAD. The single-assigned dose consists of healthy volunteers just to escalate and then add a dose cohort that we think is relevant, either three or four, based off of PK that we see in plasma that correlates with exposure. We will be able to transition to ZZ individuals at the same time.
[audio distortion]What efficacy are you going to be happy with at this interim?
That's a loaded question, Jay. What I would say is that at best, you know, at base case, we would probably want to beat any of our competitors with the data that they share. I think, as I said, our goal was to get as close to normal as possible. That would be a huge win if we can get that. If we can achieve that with monthly dosing, I think that's a home run for us. Again, we need to wait and see what dosing looks like. You know, we have a large therapeutic index. I think that, you know, as we think about taking this into the clinic and as we've started to dose up, we feel relatively confident in terms of the goals that we sort of set out to achieve.
I'll end with that. Happy to take more questions.
You know, best guess on the size of this market?
I think what we had estimated was peak sales in the U.S. about $3 billion. That includes, you know, when you think about the patient population, about 100,000 in the U.S., 30,000 of them are diagnosed, only 10,000 of them are actually treated. I think 30,000 base case and expanding into the undiagnosed COPD misdiagnosed patient population, I think that's a pretty reasonable estimate as we think about, you know, this market opportunity. That's just sales, right? We're not counting Europe in that number. That's roughly the human dose range. We might start to see some efficacy. I guess, how does that dose range compare to LNPs in the past? Is it higher than we're editing programs at those with LNPs?
I'd like to separate out LNPs in the context of what the cargo is. There are like two layers. The first layer is what is the lipid that's used. The second layer is what is the cargo. When you start to think about, for example, Patisiran is an MC3 lipid with an siRNA that is partially modified. They've gone in patients up to 1.3 mg per kg, right? They decided to go with 0.3 mg per kg because the safety risk benefit was fine, but they could dose pretty high. When you look at other companies that have taken an siRNA in an LNP, they've dosed at greater than 1 mg per kg for as long as 36 weeks and haven't seen any changes. This is in patient populations that have fibrosis. That was not really an issue either.
When you start to think about mRNA in an LNP, the profile changes a little bit. The TI changes a little bit. I think that people are still learning and evolving the LNP, but that's a very, very different therapeutic profile. As far as I'm concerned, an oligonucleotide in a precedented lipid from Genevant, we have a long way to go. That was indicated by a NOAEL of 5 mg/kg with a single dose. Do you think the safety margin benefit comes mostly from the cargo being safe, or do you think it's that the lipid is actually quite different? I think it's a bit of both. Because when you think about the ratio of lipid to cargo, most, so Patisiran, for example, the ratio is somewhere between 12:1 or 10:1. When you start to think about mRNA inside an LNP, it's 20:1.
There is a lot more lipid on a weight-to-weight basis with an mRNA cargo versus an oligonucleotide. The devil is unfortunately in the details. Do you mind just taking a second and comparing yourselves to the competitive data that we have seen today? I know Krystal was a little bit hard to interpret, but even they feel like they have a path forward. I do not want to comment on competitive data.
What I would say is that, you know, what wins here, right, for this patient population that has been underserved, underdiagnosed, misdiagnosed, I think that if you can get to a dosing frequency more than once a week or more or less frequent than once a week, and if you can get protein levels as close to normal as possible that can benefit both the liver and have enough protein for the lung and is the M protein and can increase during a time of exacerbation, I think that is as close to cure as you can get. It has to work in both the liver and the lung. I think that's the more elegant solution. I think there are technologies like us that can provide that.