All right. Good morning, and welcome back to the Cantor Global Healthcare Conference. My name is Rick Makowski. I'm one of the analysts here on the biotech team, and today we have the pleasure to be hosting Ram Aiyer, the CEO of Korro Bio.
It's a pleasure to be here, Rick.
Welcome. Yeah, so, you know, I'd like to begin, if we could just do a little bit of level setting for the audience. Can you give us introduction to the company and also to the RNA editing technology?
Happy to do that. So, I'm President and CEO of Korro Bio. We are based in Cambridge, Massachusetts. We've been founded since 2019. We have a platform on a technology called RNA editing, and that sounds like gibberish, but we're just changing a single base on RNA from an adenosine to inosine. And much like other oligonucleotide-based companies or mechanisms like siRNA or antisense oligonucleotides, we co-opt an endogenous enzyme called ADAR and redirect it to go make the single alphabet change. Funnily enough, this enzyme is active on a regular basis in each of our bodies, so all we're doing is, you know, redirecting it. Our lead program that we nominated in December of last year is in a rare Mendelian disease called alpha-1 antitrypsin deficiency.
We anticipate a regulatory filing at the end of this year, with clinical data at the end of the next year and the study completion in 2026. We have a large pipeline. Some of it is early discovery work. We haven't nominated our second program yet. We will over the next 12 to 15 months, and the hope is to showcase not just what we can do on the lead program, but also, you know, where the technology can really be. The one thing on the platform is, we use precedented drug product, like, unlike DNA editing or otherwise, it's not very complicated. It's a modified RNA, chemically modified, very precedented, can manufacture in multiple CDMOs, and we deliver it using precedented modalities for other oligonucleotides that have been in there.
So that makes it accessible to large patient populations to access biology in very, very meaningful ways. And the last thing I would say is that, as a company, you know, yesterday, we announced a partnership with Novo Nordisk in the context of developing two targets in the cardiometabolic space, so really springing forward our vision in terms of accessing large patient populations.
All right. Fantastic. That was a great overview, so I'd still like to start at a very high level here before we dive into some of the programs. So RNA editing technology, ADAR editing, you could change one base in the RNA, so could you talk a little bit about kind of the range of potential edits you can make to RNA here? I know for the lead program, you're correcting one of the mutations responsible for a genetically inherited disease, but could you maybe elaborate on the types of changes you can make to RNA and what the broad range of applications are?
Oh, sure. You know, when we say it's just a single alphabet change from an adenosine to an inosine, that's like almost 20-25% of your transcriptome is adenosine, right? So one of 5-6 alphabets that can be in your transcriptome. And so, the vastness of what we can do from a biology standpoint is pretty large. But that doesn't mean, you know, where you actually focus is really important to develop medicines, and that takes some thinking. And so when we started off, the goal was, one, showing that this modality can exist in humans, i.e., you can deliver an oligo, you can co-opt an ADAR and actually make the edit. And so in that context, alpha-1 makes a lot of sense.
Single point mutation, pathogenic G to A mutation, you can correct it in the liver, where there are multiple delivery modalities, and in the biomarker that you see in serum is actually what's going to be, potentially for approval. So it checks all the boxes in terms of, okay, from a biotech perspective, you're actually gonna develop a drug that you're gonna know proof of mechanism and proof of concept very, very quickly in the clinic. Beyond that, when we started, you know, we put a lot of thought into, okay, what's the next step, right? early in the process, it was, do we identify novel targets that can fit this bill? Then you have to explain to investors, you know, where this fits from a biology standpoint and start to think about, you know, how you actually go after indications.
Or do you go after precedented targets, where people have tried to do this many, many times and have failed, either because of efficacy or because of toxicity or both? And so, our focus was, we're not gonna compete with siRNA and antisense. You don't want to knock down a protein, and we focused on activation of certain pathways or modulating certain proteins in a very differentiated way. And so we identified targets and Biology where we can make a single edit, we can change gene expression, or we can change the structure of a protein and therefore its function. And so we haven't nominated a lot of those compounds yet, but you can see inklings of what we can do with some of our pipeline programs that we have, i.e., you know, we have a program in TDP-43 looking at ALS.
We can make a single edit and destroy the toxic gain of function that happens with accumulation of TDP-43, but also continue signaling downstream with stathmin-2, and you can't really do that with any technology. You can either knock down the protein and remove the fibrils but no signaling, or you can add more TDP-43 from an mRNA standpoint and continue signaling, but still the cell dies. And so this is really a very, very unique way of showcasing what we can do.
All right, you know, the RNA editing technology is pretty new to the scene. It's newer than DNA editing, which investors may have a little bit more background in. So I wanted to ask you just if we can compare and contrast DNA editing to RNA editing. What do you see as some of the major differences, maybe beyond the obvious, making permanent changes versus a more transient change? I guess more what I was going for was, what are the implications of that to the universe of diseases you could potentially treat or where an RNA editing therapeutic might be more applicable?
So this is a moment in time, right? So I believe in the science, I believe in innovation, and over time, I think everything will evolve, okay? But where we are today, from a DNA editing standpoint, is the focus to show the technology really works is to focus on a point mutation or a gene deletion or a gene insertion, and starting to fix those rare Mendelian diseases. The indication itself needs to be severe enough, where mortality is very high, or intervention is needed, where you can bear the risk-benefit profile of a permanent change. Hypothetical safety issue, right?
I think when you think about rare Mendelian diseases, where the patient populations are somewhere south of a thousand patients per indication, DNA editing makes a lot of sense because you sort of go in and fix the error. Where we think that this fits is, think about the space where you have an intracellular protein, you want to activate it. You don't want to activate it forever, because if it has an energy utilization, it could lead to overactivation, and you don't want that either. You don't want to permanently activate a certain pathway. Now you can come with a technology that can titrate the amount, that can make it very specific. Think of all small molecule activators that have been out there that have not succeeded. You know, we have...
that have off-target effects, either cardiotox or kidney tox or liver tox. You can now come in and say, "Okay, that intracellular protein, I am going to activate that pathway. I know it because I can target the mRNA," so super specific. "I know I can titrate it, so I can go in acute settings, such as pain, or I can go in chronic settings, such as cardiometabolic diseases, and say, 'Okay, I'm gonna come in and do this at... I don't need to edit it at 100%. I need a variant of the protein to be active for a little bit.'" Think of it as a boost, right? So 20% edit, 30% edit, half longer half-life of the protein, higher activity of the protein. Those are not things you want to touch with DNA editing, right? You don't want to permanently change that.
You want to be able to toggle this. The second component of it is, when you go into a large patient population, there is huge diversity. There's epigenetic factors, there's environmental factors. Like, you can have kidney disease, but then the etiology of that is very, very broad. And so how do you then look at risk-benefit in that patient population? And so you want something that's titratable and let the physicians do that. So that's where RNA editing sort of fits, where you want something pharmacologically active, you want something titratable, you want something less permanent, and you want to be able to do it in a way that's patient specific. And there are many, all chronic indications fall under that category.
Right. That makes a lot of sense. One of the things I also wanted to touch on was delivery. The problem of delivering genetic medicines to the correct tissue types has been a problem the field has been tackling with for only about 25 years or so. But first I wanted to ask, since you did mention ADAR as an endogenous protein that all of our cells express, all you need to do is deliver a small oligonucleotide to the cells to create a therapeutic effect. So first, I guess, could you comment on just some of the implications of using an endogenous protein and using a oligonucleotide to more or less hijack that mechanism?
Secondly, for your lead program, which we'll be diving into in a minute, you're choosing LNPs to deliver your therapeutic to the liver. Could you maybe go into why you chose LNPs and why you think that's the best method to deliver?
Yep. So on the first point, delivering a small, chemically modified, RNA or oligonucleotide in this case, we're not the first ones to do that, right? So there have been many companies that have developed a synthetic oligonucleotide and delivered it to multiple tissue types, showing that they could also co-opt an endogenous protein and go, redirect that to make an edit. And what I'm saying specifically is, in the siRNA space, to silence a gene or in the antisense oligonucleotides to silence a gene, or in the case of some companies, steric blockers to activate gene expression, so all of that has been done. Nusinersen is a perfect example of that, from an SMA perspective.
I think the idea of doing that exists, so that makes it a little bit easier for us to go to regulators, to go to manufacturers, to, like CDMOs, to make the drug product. And that infrastructure, we don't really have to build. And so it gives us a lot of confidence that what we think about, we can do in monkeys and start to plan for human studies, that's a tried and tested path. So that's. I feel more confident about using a oligonucleotide, rather than a complicated bacterial protein or a complicated base editor packaged in a viral vector or otherwise, and deliver it. The second component of is delivery, right? And so over the last twenty years, companies like Alnylam and Ionis have been figuring out how to deliver oligonucleotides to various tissue types.
We picked Alpha-1 because we knew it was in the liver, and we knew that there was many ways to deliver to the liver. And so in that context, you know, we didn't want to reinvent the wheel from a delivery perspective. But when you think about delivery, you know, there's only so many things a smaller company can do. And so our focus has been: how do you design these compounds to go and do RNA editing to edit RNA, and then leverage and work with other folks from a delivery standpoint that we know have shown this delivery in the clinic? So that piggybacks up into our Alpha-1 program. You know, it really goes towards a company build strategy, which is, you know, we want to lay a foundation and then build brick by brick.
If we shoot for the moons in the first shot and fail, that's not really how you build biotech companies. And so, what we said was, you know, look back at the clinical study and say, "All right, if the clinical study were not to work, for whatever reason, why would that be," right? There are only two reasons: one, the mechanism doesn't work, and two, you don't know if you have enough drug in the system, right? So we wanted to remove the second component off the risk factor and say, "All right, we know with a lipid nanoparticle, we can deliver it at very high concentrations in the cytoplasm and the nucleus, and we know we can get a response." And that's what we are seeing both in the mice and the monkeys, and hopefully we'll see in humans.
The second component is, we worked and licensed the lipid nanoparticle and the formulation from Genevant, because they've actually taken that formulation all the way to the clinic and have shown tested it now in humans, and it's shown to be safe with different modalities. And so for us, that's another risk off the table in terms of, you know, we're not gonna reinvent the wheel in terms of the safety profile for some of these compounds. And so that's why we picked the lipid nanoparticle for our lead program from an Alpha-1 perspective. Alpha-1 also has this unique element to it. It's the fifth most concentrated protein in plasma. It's very transcriptionally regulated at a very high level.
The half-life of the protein is only a week, and so you wanna make sure you have enough oligonucleotide in there to edit enough. So if you can show high levels of editing in Alpha-1, you know that the mechanism can work across the space. Right, and so that-
Yeah.
That was the intent.
All right. Definitely sounds like a first good indication to show the mechanism is working. Yeah, so you mentioned some of the preclinical data you've generated to date. I know the program is still in preclinical stages, but I was hoping you could maybe highlight what you've seen in animal studies and why that gives you confidence for this, you know, potentially working well in the clinical setting.
Yeah, with a novel technology, I think that's the most critical component, right? So you have to make sure from a translation standpoint, you have a high likelihood of success going into humans. And the second component is, you know, the dose that you would pick as a high likelihood of success, showing that efficacy in humans. And so when we started off, designing these compounds, most of the designs or all of the designs have been in systems where we had human ADAR and the human gene, specifically the Z allele. So we've done a lot of work in that space to understand what designs work, what don't. RNA-binding proteins are various, and so you wanna make sure that you're actually pulling ADAR and not something else. And so we spend a lot of time thinking about the off-target effects. Are we exhausting ADAR?
Are we not? So there's an immense amount of preclinical data around cell lines with the Z allele that we've generated over the last two and a half years. The second component of that is, okay, are these oligonucleotides stable from a translation perspective? Are they going to show the PK/PD that we need in humans? And so we used two mechanisms to do that. One is in rodents with the human transgenic model, we've been able to show that we can, one, correct and repair this protein. Two, we can see high levels of protein in circulation. When I say high levels, I think we have the best preclinical data out there in terms of the amount of protein that we've generated in just a single dose, where others have taken about 13-14 weeks to be able to do that.
And then the second component of it was, can we show benefit on both the liver as well as the lung? 'Cause this disease has accumulation of this protein in the liver, and therefore leads to cirrhosis, and it also needs sufficient protein in circulation to help the lung phenotype. And so having that balance of being able to show both in preclinical models was important, and we were able to demonstrate both of that. So that gives us a higher likelihood of success in terms of showing both of those phenotypes in humans as we take this forward. We don't know how long it would take, but at least we have a good sense of the high likelihood of doing that.
The last component where companies have struggled with novel technologies is, especially in genetic medicines, is: does the allometric scaling that happens for our small molecules and antibodies actually apply to novel technologies that come together? And so we decided to make a surrogate compound to edit the serpin gene in non-human primates that has homology with the mouse, and test it both in the mice and monkeys, like, do we see translation across? Because the ADAR in non-human primates has a 99% homology with humans, and so if you can show that it works in monkeys, the higher the likelihood of success.
And so we were able to demonstrate that, what we see in the mice, we saw about 25% editing of the human gene, and we saw about 45% editing in the monkey gene at the same dose. It also turns out that the stability of the oligonucleotide is higher in the monkeys, and so we were able to see a longer half-life in non-human primates than we did in the rodents. So again, and that's what people see with siRNAs, that's what people see with antisense oligonucleotides, so just walking that through step by step, I think was important. The last component was obviously safety. I think when people hear LNPs, there is a "Oh my God, you know, LNPs are toxic.
You know, this is gonna be problematic from a chronic dosing perspective." What we've been able to demonstrate, both in the non-GLP studies as well as in the GLP studies, is that we can actually dose at a very high level and not hit MTD. And so when we dosed monkeys at 2 mg per kg, we did not see any changes in LFTs, we did not see any changes in cytokine levels, and so even at 2 mg per kg, we were actually pretty safe. Needless to say, we've pushed that and seen that, you know, we have a large therapeutic index. To put that in context, onpattro, which is patisiran, that's approved by Alnylam, is the dose that they're going after, that is approved, is 0.3 mg per kg.
The NOAEL in mice is 0.3 mg per kg. The NOAEL in monkeys or the no effect dose in monkeys is 0.3 mg per kg, and so we are already at two and above for some of the lipids that we're looking at, so it gives us a lot of confidence that we can do the dosing frequency that we need, as well as achieve the dose levels that we need to, to see efficacy.
All right, I wanted to ask you about dosing frequency. So I know we're still in early stages of development, but based on the preclinical data, what's the range of potential dosing schedules do you think a drug like this would be? Would this potentially be once-a-week dosing, twice-a-month dosing, or what are the ranges we could see?
I wish I had a crystal ball that I can tell you. The data suggests that based on stability, and half-life, we should be able to hit Q3W dosing, depending on the dose that we reach. How much we can go beyond that is to be seen.
All right. I also want to touch on, you know, the pathophysiology of alpha-1 antitrypsin deficiency is pretty well understood, especially for lung disease. So thinking about how much expression of the alpha-1 protein you need in the serum to make a protective effect, I know there is, you know, one level that's thought of that is needed for a disease modification, where you'd see some level of protection, and then when you go to the other end of the scale, when you get to the heterozygous phenotype, where basically you don't see any disease at all. Where does the preclinical data fall in that range, and how does that compare to any of the competitors in the space?
Oh, that's a loaded question at the very end. What I would say is the following: I think if you look at the genetics, okay? MM being normal, MZ being heterozygous individuals, and ZZ being homozygous individuals. ZZ individuals have a protein level of a median of five micromolar. MZ individuals have a median level of somewhere around 20-22, and MM, which is normal, has about 30, okay? Or somewhere between 30 and 35. When you have an exacerbation in the lung, COVID, RSV, TB, some viral infection, the amount of protein in circulation is 90 micromolar, okay? So it jumps to three times as much under a baseline situation. So remember, ZZ homozygous individuals have five, right? Relative to what you really need under some of those scenarios. So for us, the...
And based on KOL interactions, the more protein we get and the more normal the protein, the better. And I think we've heard that very consistently across geographies, that higher the protein level of M, the better off we are, and that's why people are pretty excited about our preclinical data. 50% represents the MZ phenotype. The range there, when you look at individuals, is somewhere between 11 micromolar and 30 to 35 micromolar, okay? So the lifetime benefit of those individuals is an odds ratio of one, right? And so, but if you are an MZ phenotype, and you're a firefighter, guess what? You're gonna end up with COPD. If you are an MZ phenotype, and you smoke, you're gonna end up with lung disease.
So it's not off the table that if you have an MZ phenotype, you are actually, quote-unquote, protected. So the goal would be is to get to as close to normal as possible or at least within that range, and the lower bound of normal is about twenty micromolar. What will we see in humans? I think from a regulatory standpoint, the bar is eleven, at least as of today. It may change over time. I think when you look at the RAPID study, which was a study run by Grifols with augmentation therapy, with once-a-week dosing, keeping it at eleven micromolar, they saw that, above seventeen micromolar, they were protected from a lung standpoint, and everything about that is gravy. So what the regulators will want and what we'll have to see, my experience is that, you know, we can speculate.
What we can do is show up with data, with our modality and say, "Hey, this is what we want to do, and this is the compendium of data," and, you know, we'll find out.
All right, and if we can touch on the competition here. You know, this is an area that is being investigated by a number of companies with different mechanisms of action, different modes of delivery. Just thinking at a high level, what could potentially differentiate Korro from all those other areas of investigation?
... so you know, drug development has showed that there's never a one person wins everything sort of a market, right? You have, depending on the patient population, depending on the severity, different people do it. The statins have shown this, and LDL cholesterol lowering have shown this, right? If you are mild, you take statins, if you're not, you take PCSK9. If you want once in six months, you take Leqvio, and then if you want something more permanent for familial chylomicronemic syndrome, you can sort of get, you know, the potential therapies in development. So that's a broad range of various therapies across the board.
When we think about this modality for us, I actually think RNA editing is appropriate for this patient phenotype because there is a large diversity and heterogeneity in this patient population, so everybody can benefit. It can go after both ZZ homozygous individuals, as well as MZ and ZS heterozygous individuals who are being treated, because you don't know which allele you're gonna change when you go after DNA editing or base editing. Whereas from a RNA standpoint, we can actually go after both, and we've demonstrated that in MZ phenotype, we can actually do it. And so I think when you think about that entire market opportunity, I actually think RNA editing sort of fits the bill.
Different patients may need different doses, and over time, you know, we can find out post-approval, you know, how that plays out, but those are things that we can sort of play with, whereas from a DNA editing standpoint, that's not really what you can do, right? You have a shot at this only once, so to speak.
All right. Understood. Looks like we're coming in on the closing minutes here. Wanted to make sure we could touch on some of the pipeline beyond AATTD, sorry, alpha-1 antitrypsin deficiency. So you said TDP, TDP-43, and also pain are two of the areas of investigation. Why did you choose to go into, I guess, neuro or CNS indications as your second area?
So we haven't nominated any of those candidates yet. Delivery in the CNS is still a up-and-coming area. We know we can go after TDP-43 and ALS because it's spinal cord delivery with intrathecal, and at least for ALS patients, that's something that could be feasible. Now, is that gonna be our second program? It's not likely. Our second program is likely going to be something in the liver again, and it's likely going to be something that we deliver subQ that can last a longer period of dosing or more infrequent dosing. We haven't nominated that yet.
I think that, no matter what we come out with from a novel biology standpoint, I think we'd rather come out with a drug candidate, showing the data associated with it and, you know, where we're going after. So hopefully imminently, we should be able to show something in that space. The other thing that it will showcase is likely how RNA editing is different than DNA editing in terms of overcoming some mutations or overcoming a large patient populations in a very meaningful way, and so that's the other hope that we can do in terms of showcasing what the technology can do.
Got it, and just yesterday, the partnership with Novo Nordisk was announced, which you briefly touched on already, but I was hoping you could maybe elaborate a little bit more on the structure of that deal, what exactly the partnership will be working on, and I guess what it means for, from the perspective of Korro in general, and maybe for the balance sheet looking ahead.
So let me start with the scope of the agreement. The scope of the agreement is for us to work on two targets. We didn't decide to do 10 targets. We could have done that from an upfront standpoint if we needed to, but we wanted to focus on developing medicines, and we wanted their attention also focused on the two targets that we would go after together. It is in the cardiometabolic space. I understand cardiometabolic means a lot and different things to different people, but Novo's strategy is to really go after, you know, very specific areas, so we will be focused in that space. You know, what we haven't said publicly is the tissue type that we'll go after.
You know, it'd be presumable that we've demonstrated data on delivery of this technology to specific tissue types, and it'll be one of those things that we've demonstrated. The goal would be for us to work together to get to a development candidate and then hand it off to Novo Nordisk in terms of taking it forward from a development standpoint, and towards that end, you know, we get an upfront payment, some development milestones, as well as some commercial milestones, and finally, a royalty stream. Along the way, we'll also get, till we get to a development candidate, we'll get R&D cost reimbursed on a per-target basis.
All right. Got it. And just as a, I guess, a closing statement, could you maybe say again, you know, when we expect for the IND filing for the alpha-1 program, and when, when is the first time we may see clinical data there?
Yeah. So our guidance has not changed from what we did at Q2, which is that we expect to file before the end of the year with a regulatory agency. We didn't specify specifically an IND or otherwise. We anticipate data in the second half of next year, and by data, likely in ZZ individuals, showcasing levels of protein. And then we hope to finish the study in 2026.
All right. Well, thank you for the time. This has been fantastic, and I look forward to following the program.