All right, perfect. We can get started. I'm Tiago Fauth. I'm a Biotech Analyst here at Wells Fargo. Joined Keros Therapeutics. Have Jasbir here. Thank you so much for attending our conference. Appreciate it.
Thank you for giving us the opportunity to have this chat.
Of course,
Looking forward to it.
Again, quick intro remarks on Keros, where you guys are at in terms of key programs, stage of those programs. Then I have a list of fairly detailed questions we can work through.
Sure. Keros is a clinical-stage biopharmaceutical company. We have focused entirely on the TGF-beta pathway. We have three product candidates in the clinic. The most advanced of those is our elritercept, zero five zero, that is in MDS and MF. We have advanced through Phase II on the MDS, and now moving to Phase III, registration trial. We have had multiple data readouts with that over the years and will continue to provide more data. Our second program is KER-012, zero one two, that is in PAH Phase II trial, and just two days ago, we announced that we had closed screening. Tremendous enthusiasm for the product candidate, and that sets us on a timeline to have data readout for that in second quarter of 2025.
And our third product candidate builds skeletal muscle and decreases fat mass, and that's in a healthy volunteer study, where the data readout will be in Q1 of 2025 . So a lot happening and lots of,
Indeed.
Data readouts over the next 9-12 months.
Perfect. And again, we'll spend some time discussing every one of these programs. Since you just had news on zero twelve on the completion of enrollment, I guess we can probably start there. The key question that we get from investors is, again, potential differentiation versus sotatercept, right? Mostly because we're both targeting TGF-beta in different ways. So again, can you just briefly say, "Hey, what exactly are we doing that is differentiated, and how that can actually play out in clinic and in practice?
Yeah. So I think it's worth understanding that sotatercept was not designed for this indication. It started life off as being a bone drug, became an anemia drug, and when those opportunities were not fully capitalized, it became a PAH drug. And that whole journey from being a bone drug to an anemia drug is driven by the dominant pharmacology of the drug, which is increases in red blood cells. That becomes your dose-limiting pharmacology. So in 2017, when Acceleron decided to take it forward into PAH, they had to thread the needle whereby you could get enough target engagement and still have increases in hemoglobin that were small enough that regulators would allow you to do that. And that's really limiting it to one and a half grams per deciliter.
As a consequence, they explored two dose levels, 0.3 and 0.7, on a 3-week schedule, so that you can get accumulation of the drug, and with the 0.7, you'll probably get to a point after the sixth dose, where you get to about 60% target engagement. And what the Phase II trial showed, and with the two dose levels, was a dose response. So the question is, are you leaving 40% target engagement on the table, which is... can be translated into efficacy? So that's the first place. If you don't have a drug that increases red blood cell, you can go to maximum target engagement. That's zero one two. And then on top of that, increase in red blood cells is not without consequences.
Yes, you have a benefit in terms of increased oxygenation, but you're increasing the viscosity, which increases strain on the heart, but also in the vasculature that is already narrowed, you're having to push harder. And you see it in the label, that there are bleeding events, and the regulators looked at all of those bleeding events, and they aggregated them into the first warning and precaution, which is erythrocytosis and hyperviscosity syndrome. Okay? So in thinking about KER-012 , we were always thinking about capping the beneficial aspects of sotatercept's biology without the limiting biology of sotatercept, and that's exactly what KER-012 is.
Got it, and I think that's a great summary of at least the proposed mechanism of action and all those things. So, again, when we get questions from investors, there's a question on an incremental efficacy, right? And I think that's a reasonable therapeutic hypothesis to pursue, given what you saw with some of those response benefit and a few of the endpoints and things like that. But the key aspect that a lot of investors are focused on is on safety, right?
Mm-hmm.
So you know that based on the phase one data, you don't have the same impact on hemoglobin, right? So that should allow you to have a wider therapeutic index and all those things. But as it relates to potential risk of telangiectasia and bleeding, how much science or how much biology is behind to give you confidence that you can actually differentiate on those parts as well? Is just not having an impact on hemoglobin a good enough proxy to assume that you're not going to have other hematological associated AEs, or is it a little bit more complicated than that?
It's a little bit complicated, okay. Right. In PAH patients, because they're hypoxic, you've already perturbed the vascular remodeling. And it's not in just the lungs, it's throughout the body, okay? So you see a background of bleeding events in patients, and from sotatercept STELLAR trial, that background bleeding was about 10%. However, on treatment, it was 20% at 24 weeks, increasing to 30% at 32 weeks. So the drug is increasing that. Now, how much of it is due to the mechanism of vascular remodeling and how much of it is due to the increase in red blood cells? We can debate. Okay, right? But the two are additive, right? And an increase in red blood cells of one, one and a half gram per deciliter is increasing viscosity by up to about 12%.
That is not going to be without consequence. Do you see it in the first few weeks or a few months? The increase in hemoglobin with sotatercept, and it's true with our KER-050 as well, by the way, it happens on day two, so it's immediate.
Uh.
Right? It's not like with erythropoietin, where it takes three weeks, okay, for an incremental slow increase. Here, it's rapid, okay? You get that first half gram, okay, right, on day two through day five. Okay, right. So it's a rapid increase. So some of those bleeding events are likely to be due to the increased red blood cells. But with time, by inhibiting the bone morphogenetic proteins that sotatercept is able to do because it is the wild type activin receptor extracellular domain, where the activin receptors are used by bone morphogenetic protein for signaling as well, you are inhibiting the BMPs, and that is going to have consequences. The best-known biology there is in a disease, genetic disease, called Hereditary Hemorrhagic Telangiectasia, where inhibition of 50% is sufficient, whether it's the ligand or the receptors or the co-receptor.
Regardless of which it is, the phenotype is exactly the same, which is that in the first decade of life, patients are perfectly fine, but as they get to the end of that decade, they should begin to show these telangiectasia. And then in the second and third decade of life, these telangiectasia are leading to now much more severe bleeding events, gastrointestinal bleeds, cranial bleeds that can lead to death. Okay? So in this pathway, just one ligand inhibiting that takes time to see that. But sotatercept, because it binds multiple BMP ligands, you see this accelerated pathophysiology showing up at 24 weeks, where it was 10% telangiectasia versus 5%-
Yeah
in the placebo, increasing to 14% at 32 weeks. Okay, right. So time on the drug is increasing, okay, right, these potential AEs.
Got it. That, that's fair. And again, as we're just trying to piece that apart, the whole debate on differentiation. So on safety, you just gave us a compelling argument, biologically, why they could differentiate. You have some evidence from a Phase I that kind of shows that it's doing what you said it would do.
Yeah.
Right? So I feel like the safety profile should be differentiated. One pushback that I get, surprisingly, is sotatercept label is clean enough, and the monitoring requirements are not overly demanding. Is there really the need for a cleaner drug, even if that's the case? But again, you indicated-
Yeah
some additional long-term data, but how are we squaring that off on the safety side?
Yeah, I think, okay, right. The appreciation of the safety side of the equation was not there eighteen months ago when the data initially came. If you now talk to the physicians, they are much more aware of it and concerned, to the point where some of the physicians are only giving their most severe patients sotatercept and are recommending monitoring beyond the first six doses, okay? Because they are concerned about the hemoglobin changes. So I think, okay, right, what you're seeing is that as the community is understanding the drug better, some of the shine has gone away, and they are appreciative of the safety signal. So I think it's with time, you're going to get a better appreciation of it. Even to the point, the patients are seeing this and are talking about it.
So you go into the chat rooms, okay, right? Yes, of course, there's bias. Everybody, okay, that's in the chat room is, you know, feeding upon each other, right? But you're hearing about the bleeding event happening after the first or second dose with sotatercept. And there was the report earlier this year where a patient on sotatercept was getting gastrointestinal bleeds starting at day 44 through day 400 , and when they came off of the treatment, those bleeding events stopped.
Got it. No, and again, I think it all adds up to tell a nice story, right? And for a drug whose peak estimates are north of $7 billion, like, even marginal differentiation on safety to us should lead to fairly substantial opportunity, right? At least that's how we're thinking about that. But let's pivot, I guess, to the efficacy side of things. Again, two components there. The first one is just mechanistically. Can you talk about the PD markers, your FSH data, how that compares to an equivalent dose of sotatercept?
I guess one of the questions we always get is related to some early data that we saw at higher doses they ended up taking to PAH, and that shows more target engagement for sotatercept, but that's not what you would expect in PAH patients, precisely because they're dosing that at a lower level. What is the comparable basis from a PD perspective, FSH target engagement between the molecules?
The first study was the healthy volunteer study with sotatercept.
Yep.
And in that study, the drug was dosed by IV route of administration. Dose levels started at very, very low and went up to as high as three milligrams per kg. At three milligrams per kg, there was a 45% reduction in follicle-stimulating hormone. A lot of biology to indicate that in post-menopausal women, roughly half of the FSH is through the activin receptor signaling pathway. Therefore, if you're inhibiting totally all of the activin signal, you should see roughly 50%, and therefore, at three mgs per kg, they achieved-
Full.
Full target engagement.
Yeah.
Okay? We see the same thing with our zero one two, where at the 4.5 mg/kg, we saw 50% reduction in FSH. They're using 0.7, but 0.7 on a 3-week dosing schedule with a drug that has a half-life of 23 days. Therefore, you will get accumulation of the drug. From the first cycle, at Cmax, you should be seeing about 40% target engagement. However, with that accumulation, after the sixth dose, you're gonna see about 60% at Cmax, probably about 40% at the trough level, so you're cycling between 60% and 40% thereafter.
Okay. Totally.
Okay.
All right. Yeah.
On a milligram-per-milligram basis with KER-012 , at the one and a half mgs per kg, you get more target engagement at Cmax, and at four and a half, you're gonna get full target engagement. We believe that at three mgs, you're gonna probably be around 80%, and then at four and a half, you're at, you know, greater than 90%.
Okay, right.
There's gonna be overlap between those two dose levels. But we also get down to a trough level because of the shorter half-life of twelve days, and in this pathway, it's about taking a population of cells, stem cells, allowing that differentiation to occur and not to exhaust those stem cells. So you've got to allow renewal, and I think what we've learned with 25 years of working in this area is that having constantly high levels results in tissue remodeling, that is aberrant tissue remodeling. And the cycling is actually leading to a more normal tissue architecture.
Got it. No, that makes sense. And again, we talked a lot about biology, so from a molecule perspective, how we could differentiate what you should expect to see. Let's move to how that's gonna translate into clinical data so far. So, the triples, right, like, and can you just compare relative to sotatercept or relative to other contemporary PAH studies, what's the inclusion and exclusion criteria? Like, where could this trial fail to capture the benefit of the drug, I guess, is where I'm trying to get at?
Yeah. Look, the trial was designed to have very similar inclusion, exclusion criteria to the sotatercept and other trials that have been done in the last five years.
Okay.
It's not any different from that. What we know is that pulmonary vascular resistance is a good indicator of disease worsening, okay? You wanna capture patients with a certain PVR, and that was generally above 600, okay, right, in PULSAR, and STELLAR was around 800, 850. You're trying to make sure you get patients that are around there. The studies are designed similarly, right? Same inclusion, exclusion criteria. Primary endpoint will be PVR. Secondary endpoint is the six-minute walk test, changes in NT-proBNP, functional class, and all, all that. Not very different. What can we anticipate? Okay, great. I think that's the question.
That's always a key question.
Right. I think what I'm looking to see is that with sotatercept, with the point three and the point seven resulted in a mean change when you pull the two together of roughly 200 dynes, is something along that lines, okay? If you see a 200 dyne, okay, right, that says it's great. If you see 150, it's still good, okay, right, because patient populations are different. So then the next question is, are you seeing a dose response? Okay. And if, okay, right, we don't see a dose response, and we've capped out at the low dose or the mid dose, then going into a Phase III, it's a fixed dose on a monthly interval with no monitoring for hemoglobin. There's your first differentiation, okay? And then when you start looking at the safety side of the equation, you get additional.
I believe we have the opportunity to capture more biology. That can result in greater efficacy. But it- in the worst-case scenario, you're capturing the same biology and maybe, okay, you end up with the same efficacy. But-
Perfect.
But now, okay, right, you've got convenience going into the marketplace, and then you wait for the safety. But I think this difference between binding, okay, the bad guys, the activins, and the good guys, the BMPs, that window is small with sotatercept. It's a much bigger window. That could also translate into efficacy, but takes longer. You go back, okay, to PULSAR. In the open-label extension, you saw that the benefit that was achieved in the first 24 weeks was not improved upon.
Right about...
Right. And if and when you think about it, those studies had patients that were diagnosed for eight years or more, right? Are you telling me that that biology that has been cumulative can all be corrected in 24 weeks, right, and no further, okay, right, thereafter? Well, maybe it's a case of you don't have enough of a window. So again, we'll see the 24-week data, but the open-label extension may give you additional clues, okay, as to additional benefit, right?
Yeah. No, for sure. And again, I think we have a pretty good idea of the setup. In terms of news flow, the final readout, Q2 of next year, may be a little too early to talk about that. Do you intend to disclose baseline characteristics or things like that?
No.
The next time we're gonna talk about it, it's gonna be when you flip the card.
That's exactly right. And, and the reason is very simple. We have three programs. We're really busy. It's a small team. I think the last time we reported something like 140 employees, okay? A Phase III program that the teams focus on executing on-
Yeah.
- this Phase II program, as well as a Phase I program. I don't need to distract them with, you know, trying to generate data, which is not really that meaningful.
Fair enough. Fair enough. And great. Again, we spend way too much time on the O12. Let me move to a couple of other things, because we still get a lot of inbounds on the other programs. So O65, again, a Q1 readout in the healthy volunteers that were enriched. A lot of hype earlier in the year that kind of dialed down, maybe because of Lilly, maybe because of some regulatory uncertainty. What's the, what's the value proposition of this asset relative to the competitive landscape?
So I think the value proposition for a potent muscle-building agent is that you can ameliorate the loss of lean mass with GLP-1 receptor agonists. That's one value proposition. The other is that when patients are intolerant of GLP receptor agonists, they're on it for a short period of time. They will lose weight. But now, can you use a muscle agent thereafter to further build that skeletal muscle, lose fat mass, where you might only see a 5%-10% reduction in body weight, but your body composition is a better body composition, which is leading to better metabolic parameters, reduced A1C, better insulin sensitivity, reduced circulating lipids in circulation. Those will all have better outcomes in the long term in terms of cardiovascular risk, in terms of renal risk, right? But it takes time, okay, right, to show those.
We have been very, very transparent that the opportunity in obesity is there, but it is not one that we can fully exploit, and therefore, we will only take this product through a proof-of-concept phase II, and that's as far as we're gonna take it. And to your point, if partnership is there, we'll do it. But there's also lots of other indications that a potent muscle agent can be exploited. And again, when we unveiled O65 in August of last year, the indication was Duchenne, right? And it was really the excitement around the obesity, and frankly, okay, many investors came to us. Some even came with a corporate presentation, retooled, okay, right? And we saw that, yes, there is enough interest here for us to pivot to do and commit to doing a proof of concept in obesity.
No, that makes a lot of sense. And again, a few data points there or similar mechanisms of action that could be comparable here, but feels like the landscape sometimes feels crowded, sometimes feel like there's only two or three actually viable assets over there. So how should we think about your mechanism and differentiation versus peers?
Yeah. I think if you want a potent drug, you have to inhibit multiple ligands. You have to inhibit myostatin and activin A. And the three approaches in the clinical setting that are doing that is the Versanis-Lilly approach, where they're targeting the receptor that these two ligand signal through, Regeneron, through two different antibodies, right? Each targeting that, and ours. Then there's the approach of inhibiting myostatin, and two and a half decades of data shows that that is not a very potent approach, okay? You will get some increases in lean mass, but there are relatively small increases in lean mass.
In fact, in going back and reviewing all of the data from Phase I studies, I did not see a single piece of data that says that you get to half the increase in lean mass that you observe with the Versanis approach, with the our approach or the Regeneron approach, with longer term treatment with a myostatin antibody. The best number I saw was in one study at one time point of 3% increase. Otherwise, it's less than 2% increase in lean, in every study over six months, okay? So I think, could there be an opportunity there? It is, but it's gonna be not the best molecule. And many will talk about, well, if you're inhibiting other ligands, there's safety signals.
I'll tell you that Lilly's antibody, that they took into a number of indications, including the frail elderly population, that's a GDF8 myostatin and a GDF11 antibody, okay? It was safe, right? They abandoned it because it wasn't potent enough and to give them enough efficacy, and instead went out and bought Versanis. I think money talks here.
Sure. That's always the best indicator. No, that makes sense. And again, I think it's a little harder just to handicap expectations for the Q1 update. There's a different mechanism of action, so do you feel like investors are mostly focused on the absolute top line, weight loss number? Is it more about body recomposition? What are some of the key benchmarks and what you think would be a successful profile in the first quarter of 2025?
Yeah. So what I keep stressing with investors is, number one, this is a safety study. It's a first-in-human safety study, that's been enriched to give you PD markers, okay? And what are those PD markers? It is not weight loss, because if you got an agent that increases muscle, you're actually going to see an increase in body weight initially.
Yeah.
Right?
Yeah.
And then you're gonna get a decrease. I mean, anybody, okay, that goes on to an exercise program, right, you don't lose weight instantaneously. And actually, if you're vigorously active, you actually see the needle go in the wrong direction. It's very disheartening, but it does go in the wrong direction. But if you persevere, then it starts going in the right direction. So small study, small and not looking for weight loss, okay? Period. What we're looking for are changes in body composition by DEXA scan, and what I've said again to investors is, directionally, we're looking for an increase in lean mass and a decrease in fat mass. And in the MAD part of the study, multiple ascending dose, we have enriched it for a population that has a BMI of 27-33, so overweight to mildly obese, okay? So that's what we can anticipate.
Answer the question of safety, show the biology is translating and directionality on the change of lean and fat mass. By DEXA, as well as by looking at MRI, at the thigh muscle volume, where you can look at the size of the muscle and also, fatty infiltrates into that muscle.
Got it. Okay. That makes sense. Again, based off of those data, you're gonna figure out Phase II and-
Yeah.
Okay. Makes a lot of sense. And I know we still have another asset to discuss. So my last question, just on KER-065 . When I'm thinking about DMD, there are some questions, and that's mostly because of the myostatin agents that were studied in both SMA, DMD, and there's some skepticism that you can actually drive functional benefit. Perhaps less skepticism that you can increase muscle mass, but some skepticism where you can see functional benefit. So why do you think this mechanism could actually be appropriate and lead to that sort of outcome, which we haven't really seen with more limited MOAs?
Yeah. I think it's again, the potency of the molecule and the myostatin potency is just not sufficient. But if you go back and look at all of that data closely, what you see is signals of improved functional benefit. It just didn't reach statistical significance to be able to get a registration study completed, okay? So you look at the six-minute walk test, okay, right, there weren't big changes. However, if you look at the Acceleron ACE-031 study. Going back to 2010-2011 ? Okay, right timeframe. With a more potent molecule in a small study, what was noted was that in the placebo boys, they lost 39 meters of ambulation over five months. Three months of treatment, but over five months, 39 meters, which should be compared to the natural history study, showing 85 meters lost for in the course of a year.
So mathematically, spot on, right? In the treatment arm, the boys gained eleven meters, i.e., muscle stabilization function. So given all of the snippets of data in the myostatin antibodies with longer-term treatment showing signals and the Acceleron data, I'm confident, okay, right, that you can stabilize muscle function. How long for is not known. With glucocorticoids, what do you do? You stabilize that muscle, and you delay ambulation by two to three years. So maybe that's all it is. But then on top of that, we know by inhibiting activin A, you will have antifibrotic effects. These boys eventually die from loss of respiratory capacity. They also die from cardiac failure, and we know that that, based upon all of our work with the other molecules, KER-050 , KER-012 , you're gonna have benefit there.
So if you're trying to treat the dystrophin, we're not gonna do that. But if you're trying to treat the patient, there is actually lots of benefits from treating the patient and the various pathophysiologies that arise from the fundamental lack of dystrophin.
Got it. No, I think that's very fair, and I know we only have a couple of minutes left.
Just last point I want to make.
Yeah, quick.
All of the gene therapy, the exon skipping strategies have not shown benefit-
Yeah
To date. The gene therapy looks exciting. Okay, right. Everything always does.
Right.
Right? but what we have also seen in preclinical models is that you can increase the efficiency of exon skipping with KER-065 , so a combination could potentially have additive benefit, okay, right, in the marketplace.
Got it. No, that's helpful, and I just wanted to touch onKER-050, and again, we get fewer inbounds on that one, which is surprising. It's a Phase III-ready asset, right? So the questions are mostly on differentiation versus Reblozyl, and now that you have Geron in that space, like, what's the actual unmet need and how that product profile can differentiate? So if you can just kind of summarize what you've seen so far and how you think that may play out in a Phase III trial design, how it compares to imetelstat, perhaps.
Yeah.
Just a quick refresher.
So look, we're moving to Phase III because we saw a differentiated profile. ESAs have been the standard of care for a long time. They work well in a bone marrow that is somewhat compromised but essentially okay functioning. luspatercept takes over from that, and luspatercept data has shown that in a bone marrow that is more healthy, it works, but as the bone marrow becomes unhealthy, it doesn't. So in high transfusion burden, in the non-RS, the response rate is very poor, okay, right?
Yeah.
You got 20% in the high transfusion burden, RS-positive patients, whereas we're seeing 40%-50% response rate. So it is KER-050 mechanistically was hypothesized to be able to treat a broad patient population and a more severe, bone marrow, and we've seen that. That can translate into, providing benefit to those patients that are more severe, but also the durability of the response. With luspatercept, it's 30 weeks. We have not reached the median duration of treatment. Two-thirds of our patients in the trial are high transfusion burden, and at the last readout, one-third of the patients had exceeded a year of treatment. So we are likely to have a longer durability of response. That means the patient can go on it, and they stay on it longer.
Okay.
Where Rytelo comes in, Geron, is in the high transfusion burden, RS and non-RS, okay, right? We think the safety profile of that is challenging, and therefore, okay, right, you can be competitive with that. So we see ourselves as the best option with KER-050 , competing with Luspatercept, but then just continuing with those patients until they're no longer responding. Now, how long is that? We don't know. Actually, inhibition of activin A has other potential benefits that will not be realized in low trans, in, low-risk patients, okay? You're gonna have to go to high-risk patients.
Sure.
And activin A promotes increased motility of malignant cells, both hematologic and solid tumors, and therefore that you could actually slow down progression to AML, but you're never gonna see that in the low-risk patient. So we see broader patient population, greater durability. And one of the other liabilities of luspatercept is, how do patients feel? They feel fatigued. Even when they achieve transfusion independence-
Yeah.
They're feeling fatigued. So now you've got this battle between the physician who wants to get them, keep them on, the drug because they're becoming transfusion-independent, and the patient who feels lousy, okay, on it and wants to come off, and therefore, the patients don't stay on luspatercept for long. On the other hand, with KER-050 , we have shown that your patients are feeling better. They have an improved fatigue score, okay, right? Therefore, they want to stay on it. We've had patients, okay, that are responding but aren't achieving transfusion independence, and they want to stay on the-
They still want to stay on the drug.
Yeah.
Gotcha. We've run out of time, but again, I think it was fairly comprehensive. I appreciate you making the time for us.
Thank you for the time.
Appreciate it.