Thanks very much, everybody. Happy to be here moderating a fireside chat with Jonathan Violin, President and CEO of Korsana Biosciences. Some folks listening may be new to the story, so I was kind of thinking maybe, Jonathan, you can give sort of a three-minute overview and background of the company and KRSA-028, and then we can do Q&A. How does that sound?
Perfect. Thank you, Paul.
Thank you.
Thank you, Paul and the Stifel team for having us here. Really a pleasure to be at the conference and to have you covering the story. Korsana's a relatively new company. We just came public in September via reverse merger, very well capitalized. We reported $456 million on the balance sheet, as of the end of September. And we are here to try to make a dent in Alzheimer's disease, obviously one of the biggest unmet needs in medicine today. And we think that this indication is at an inflection point. We have two disease-modifying therapies on the market, but while they slow cognitive decline, they are far from optimal.
Then we have trontinemab from Roche, a true breakthrough, a next-generation medication that both clears amyloid plaques much faster and is much safer in terms of amyloid-related imaging abnormalities, ARIA, microedema, microhemorrhage, which has been a series of adverse events that has plagued this category. Trontinemab is better in both of those. Faster plaque clearance, far lower rates of ARIA, but still is not the end of the road. It is given intravenously; it has a very high rate of infusion-related reactions, such that all patients have been pretreated with corticosteroids in phase III, and it has a moderate rate of anemia. So while we think Roche has done a fantastic job, we think it is really a true breakthrough product; there is more to do yet at this target for these patients.
So when we set out at Korsana to make a new molecule in this class, we really thought carefully about what does it take to maximize the benefit we can give patients in clearing amyloid? We think, at least based on our preclinical data, that is what we have achieved. We are approaching the clinic and expecting our first clinical data next year, healthy volunteer data mid-year next year, and then our interim proof-of-concept data in patients by the end of next year or the first quarter of 2028. So a lot of excitement ahead for us.
Yeah. Okay, great. Well, maybe we can sort of break down the thought process behind KRSA-028 and the different sort of points of potential differentiation. But starting with the mechanism of action, maybe talk about the thought process for targeting pyroglutamate A-beta and why you thought this was the right mechanism.
There is a really nice correlation if you look at all the phase III studies, so the biggest, best-run studies for amyloid-targeted antibodies. If you look at the first-generation unshuttled antibodies, so require very high doses to get enough drug in the brain, so excluding trontinemab, there is a striking correlation between plaque clearance as measured by PET tracers, which is extremely sensitive, and slowing of cognitive decline measured by CDR-Sum of Boxes. If you look at that set of data points, there are some molecules that failed, and there are explanations for why they failed. But the very best plaque clearance, and at least numerically, the best slowing of cognitive decline is donanemab, which targets pyroglutamate-modified forms of amyloid. That pyroglutamate post-translational modification makes amyloid more sticky, helps promote plaque formation.
You can think of donanemab as more plaque-selective as opposed to various other forms of amyloid. When we look at the donanemab data, in addition, remternetug, which is a next-generation molecule Lilly is advancing, also selective for pyroglutamate forms of amyloid, that target, that form of the target, clearly delivers the fastest and deepest plaque reduction. If you're starting from scratch thinking, how do I make a molecule that's hard to beat? drugs like lecanemab that are more broad-spectrum amyloid binders, clearly work. The data suggests that pyroglutamate is the place to be for a starting point.
On the safety side, I mean, I think the biggest revelation within the trontinemab data is the ability to decouple plaque lowering from ARIA. What do we understand about this biologically in terms of why this is happening, and how confident can we be that this is an extrapolatable finding to other shuttle programs?
Yeah, that was the, I think, biggest surprise in the trontinemab program. It used to be we thought that ARIA was inextricable from clearing plaque and that if the antibody bound plaque and promoted its clearance, that you would cause ARIA. The reason that happens, though, is because to get enough unshuttled antibody into the brain, you have to give very high doses, leading to very high peripheral exposures, and you end up with a crowding of antibodies with all the immune function that antibodies bring to bear in the large vessels of the brain, where it can accumulate and just damage the local blood vessels. That's why when you have these plaques driving peripheral concentrations very high, can result in these findings. With trontinemab, with this shuttling technology, we can get much higher brain levels with much lower doses, lower peripheral exposure.
You can see that in studies that Roche and others have done, that rather than requiring huge doses to crowd in through the large vessels of the brain, by binding the transferrin receptor and enabling shuttling across the blood-brain barrier, these lower doses, lower peripheral concentrations get more evenly throughout the brain. That's how we can selectively attack the plaques that are causing amyloid, that are causing Alzheimer's disease, with a much, much lower rate of ARIA.
Yep. Okay, great. I think the other question with transferrin binders in general is anemia risk. I probably wouldn't rank that as important as ARIA, but how comfortable are you that you'll be able to have lower rates of anemia than some of the other competing programs out there with KRSA-028?
Yeah. Anemia is certainly one of the factors that has been a problem for the entire field of transferrin receptor shuttling. Today, we're sitting on two or three decades worth of lessons in how can you engage transferrin receptor, which is part of iron homeostasis, right, a core biological process. How can you do that without causing harm? We've learned along the way you need to target the transferrin receptor with the right affinity, with the right epitope, so you still allow transferrin to bind and do its job. The remaining issue that we see with trontinemab is an acute reticulocyte depletion, which we reasoned could have been caused by CDC or ADCC, two of the effector functions that a normal wild-type antibody brings to bear. When we designed KRSA-028, we wanted to preserve the beneficial effects of phagocytosis. That's how plaques are cleared.
But we wanted to try to spare reticulocytes with knocking down CDC and ADCC, and our preclinical data suggests that's what we've done. That's just one piece of how we think we can make a best-in-class molecule. We're aiming to be subQ, better local tolerability. Hopefully, we can reduce anemia. Then the benefits of the transferrin receptor are dramatically lower, potentially even placebo-like rates of ARIA.
Maybe talk a little bit about the different work you've done on the preclinical side, and walk us through Korsana's thought process in taking the preclinical data and being able to make a confident prediction around human dose range, frequency of subQ, things like that. I guess the sort of bottom line, what is your level of confidence that all this preclinical work should reliably translate?
Our philosophy in discovery programs is to really minimize risk, take risk only where we need to. When we designed our shuttle, we carefully looked at all the shuttle data we could find, including trontinemab, and the shuttling capacity of trontinemab is great. The opportunities to improve were elsewhere on the molecule, so we designed a novel sequence, but it's the same target, same epitope, same affinity as trontinemab. Then we compared head-to-head in mice and non-human primates, which gives us a lot of confidence that our shuttle is working the way it should. That extends then to the IgG backbone. Again, we're targeting the pyroglutamate form of amyloid. We think that's the optimal starting point.
But when we put that on the shuttle, we can measure the peripheral PK, we can measure brain concentration, and leverage all the modeling that we've created based on the programs that came before us tell us that we should need somewhere around 150 mg once a month dose subcutaneously to achieve brain concentrations comparable to what trontinemab in their phase III dose has achieved. That's great for us because we've already pushed our concentration up to 150 mg per mL, so that means we're targeting somewhere around one mL once a month subcutaneously.
At the same brain concentration, might you have a more potent drug anyways because you're not leveraging gantenerumab?
That's correct. For that, we need to look at what we can learn from donanemab and remternetug, the pyroglutamate-targeted antibodies, to think, okay, even if we match trontinemab's brain exposure, might we be more potent? That's all gone into our clinical trial planning.
Yep. Okay. Well, maybe let's talk about the development plan. You'll first be going into healthy volunteer studies, read out some data next year there. Maybe talk about the design, and also what will we learn from the initial readout?
We designed our phase I/II study to be phase III enabling. To do that, we've strung together a three-part study. Part one, healthy volunteer single ascending dose. Part two, Alzheimer's disease, early Alzheimer's disease patients, multiple ascending dose. Then part three, expansion cohort, so we can pick one or more doses from the multiple ascending dose and increase the sample size so that we've adequately characterized safety to move to phase III. There's precedent for integrating SAD/MAD. There's precedent for MAD going into expansion cohort, so we feel strongly that this should be a phase III enabling first trial.
Okay. As it relates to the data you'll ultimately be generating patients in the initial readout, I think it's either late next year or early the year after, to what degree can you de-risk your TPP in a study like that? Will we get the firm answer on ARIA? Can we take early plaque lowering data and extrapolate it forward? Maybe talk about what you plan on disclosing and what the implications will be.
Sure. The clinical profile will be de-risked incrementally, as is normal, but what we really like about Alzheimer's disease, and it certainly didn't used to be the case, but we can learn a lot about our molecule very, very quickly. That actually starts with a single ascending dose in healthy volunteers. So, we're beginning clinical development with a subQ route, so we'll see how well the drug gets in the bloodstream. We'll see how well the half-life extension is working, and feeling very comfortable with both of those. We'll measure levels in CSF. That will tell us if the shuttle is working. We'll get a first read on safety and tolerability in the single ascending dose. That data is targeted for middle of next year.
What we are describing is interim proof of concept, so the first one, maybe two cohorts from the multiple ascending dose, where the major readout is plaque clearance measured by PET. For that, even five or 10 patients gives you a very, very precise read of the speed and depth of plaque clearance. What we are looking to see, and that data would be end of next year or first quarter of 2028, what we are looking to see is that we are on a dose response curve that can match the trontinemab rate of plaque clearance, which is the fastest anyone has ever seen plaque be cleared before. That would put us in a great position.
We will have early reads on ARIA, but since shuttling reduces ARIA rates to single digit percentage, we are looking for very low rates, and then it will be the expansion cohorts where we really learn over time what the true safety profile is in terms of ARIA.
Yep. Okay. As it relates to the phase III design, is it as simple as kind of looking at what Lilly and Biogen/Eisai did, or any other nuances that you might kind of consider there for yourself?
Well, too early to comment on the specifics, but certainly there is a lot of precedent to lean on, both in terms of what regulators want and in terms of what performs in terms of the clinical trial itself. Our first port of call here will be the early Alzheimer's disease population. That is the population where we have got the most evidence. It is the lowest risk. We know that this mechanism should deliver efficacy. Longer term, we are certainly interested in the preclinical population, people who have plaques but have not yet developed symptoms, and there is high interest in the potential that clearing plaques in those patients could delay or even forestall the development of disease symptoms.
Yeah. Okay. Makes sense. I think the other big event in the Alzheimer's space that people are looking to is the pre-symptomatic trials. I guess, one, what's your perspective on the likelihood that those studies will show a greater benefit? And two, how do you view that as a real accessible commercial market?
Yeah. We're getting lots of questions about preclinical disease, and maybe I should start by saying that we grew conviction in the value-creating potential of Korsana just based on early Alzheimer's disease. Huge population, a lot of unmet need, a lot of room to develop a new best-in-class drug. But certainly are interested in this preclinical or pre-symptomatic population. There's a fair bit of admittedly retrospective evidence suggesting that the earlier one clears plaques, the better patients will do. For example, Lilly looking at their donanemab phase III, if you look at where patients are in their natural history of disease, if you look at patients who are low in tau, which would suggest they're earlier in disease, you see bigger clinical benefit.
If you extrapolate that thinking backwards in the natural history, when patients have plaques, they're on the trajectory towards becoming symptomatic, can you, by clearing plaques, deliver an even greater clinical benefit? The studies that are ongoing, most proximally Lilly's TRAILBLAZER-ALZ 3 study of donanemab in preclinical, is not looking at the same primary endpoint. It's a time-to-event endpoint, so it's a little bit apples to oranges. But any efficacy at all would be a huge boon, right? Taking patients and preventing or delaying development of disease would be fantastic. Hard to say what the magnitude of benefit will be, but when we think about accessing that part of the market, safety, tolerability, and convenience are going to be even more important, right?
Because you're talking about treating patients, some of whom might never develop symptoms, and so you really want to make sure you've got optimal safety, tolerability, and convenience. We think, all credit to Lilly and Biogen/Eisai if they're able to access that market, but a shuttled drug like KRSA-028 we think would really be the eventual winner in that indication.
Yeah. Maybe on that point, do you want to talk a little bit about the competitive landscape on the transferrin side? I think that there are these different views out there in the investor community, and obviously there's so much white space left in neurology in general, but there is one school of thought that with the amount of companies that have popped up here now, it's sort of starting to become like GalNAc was in the liver, where people can kind of, I don't want to say it's commoditized, but it's doable for a number of parties, right? Arrowhead Pharmaceuticals just had some success too, which I thought was pretty interesting. I guess in that vein, what makes Korsana's platform here different and special versus some of the other peer companies out there?
Yeah. You're right that it seems like every day we hear about another transferrin receptor shuttle program.
A nother group of NHPs that thought that was super and well-tolerated, you know? No, I'm kidding, but yeah.
Yeah, I think, again, we're sitting here today 20 or 30 years into the research on how to target the transferrin receptor to shuttle molecules. So yeah, it's no longer too heavy a lift to get molecules into the brain. It's still very difficult to design a great drug. And so when we look at the competitive landscape, yeah, there's more and more entrants. Some, I'd say, are closer to trontinemab, full effector function, and trying to carve out pieces of differentiation. On the other end of the spectrum, there are some entrants that are completely effector function null, so no CDC or ADCC, no phagocytosis. So probably quite safe, but open question as to how well you can clear plaque absent phagocytosis, since that's how the approved drugs are thought to work.
W e've really targeted what we think is a sweet spot, where we've preserved phagocytosis, we've knocked down CDC and ADCC, we've improved the pharmacokinetics to allow us to get to that lower dose, that infrequent low-volume subQ. I think it shouldn't be any surprise, given that Alzheimer's disease is one of the bigger unmet needs in medicine today, that there's a lot of people trying to crack this problem. But I love our position in this. We're well-funded. This is our lead program. We're focused on it, and I think there's a lot of good we can do in the world with our efforts here.
Yeah. Where else do you see the greatest opportunity with your platform? Tell us your other programs.
Yeah, so we are actively working on a couple discovery programs. We are not disclosing them primarily for competitive reasons, and also because there's a lot of value to create by focusing on our lead asset, and there is a competitive advantage in focus. The way we describe it is, I think we've got a great shot at having a best-in-class molecule for clearing amyloid, but to the extent we have a crystal ball, pretty good bet that five, 10, 15 years from now, just like in most oncology indications, to optimally treat some or maybe even most Alzheimer's disease patients, we'll need more than one target. So we're actively thinking through what other targets might combine with KRSA-028, so clear amyloid, and then what other target might maximize benefit for patients. We think that amyloid clearance will remain the backbone of therapy. People should not have amyloid plaques.
We can remove them, we think, effectively, safely, conveniently. But there may be more we need to do. So we're actively thinking through not just how to win an amyloid clearance, how to win in Alzheimer's disease, but of course, we have our eyes on a few other indications as well. Just too early for us to talk about it.
Okay. All right. Look forward to that. Maybe to wrap things up, just talk about your balance sheet and how long your cash runway takes you.
Yep. I like to tell people that Korsana's off to a very good start in life. We're grateful to a core group of investors who have supported us quite robustly. We've raised over $500 million in our short two-year history. And as of the end of September, $456 million in the balance sheet. That funds the company into 2029, so well past the couple catalysts that we talked about.
Okay, great. Anything else you want to add before we wrap up?
Just that, to underscore what I opened with, we really think Alzheimer's disease is at an inflection point. The first two approved products are already annualizing over $1 billion, and that is just scratching the surface. The amount of unmet need, the number of patients who are untreated is staggering, really, especially as we think about aging demographics. We've got a chance here to develop a drug that can benefit a huge number of patients, and it's really gratifying to have the opportunity to do so.
Great. Awesome. Thank you. Best of luck, Jonathan.