All right, great. Good afternoon, everyone. Thanks for joining us here at the Leerink Partners Global Healthcare Conference. My name is Tom Smith, one of the Senior Biotech Analysts here at Leerink. It's my pleasure to welcome our next company to the stage, Keros Therapeutics, and happy to be joined up here by CEO Jasbir Seehra. Jas, thanks for joining us.
Thank you, Tom.
Great. I know the company's been busy. We have a number of data readouts coming up. I'm sure we'll get into a lot more detail. Jas, why don't you kick us off with a brief overview of Keros, for those in the audience who may be a little less familiar with the story.
Thank you. Keros is a clinical stage company focused on the TGF-β pathway. Our first product candidate, the most advanced, targets the pathway to increase red blood cells and platelets. That we partnered in December with Takeda after having demonstrated efficacy in a Phase II study in patients with MDS and ameliorating the anemia there, as well as in myelofibrosis. That was a good partnership where they are aligned with our vision for the drug. Our second product candidate, cibotercept, was and is in a PH trial where we saw some safety signals. I am sure we will spend more time on that. Our third product candidate targets myostatin and Activin to increase skeletal muscle and reduce fat mass. By targeting Activin, it also is a bone anabolic agent.
We announced earlier this year that we're taking that into neuromuscular indication, the first of those being Duchenne's. Data readout from that this month. With the cibotercept, we'll have the top line data in the second quarter of this year when we will make the decision for what the path forward is with cibotercept.
Okay, great. That is right, I guess, where I want to start and kind of drill down into the updates around c ibotercept . Maybe if you could just provide a little bit of context here. I know you announced pausing of dosing of some arms of the study in December and then subsequently early terminated the TROPOS study. Maybe if you could just kind of walk us through some of the events that you observed there, I guess the level of visibility you have into the data, kind of like where it sits now. I know we are waiting for the sort of top line release in Q2. Any sort of biologic hypothesis around those observed events.
I have no visibility into the data. Okay. I'm still blinded to it. There is a small team within Keros that did unblind the data to look at assess the overall risk-benefit back in December. On the basis of the Drug Safety Monitoring Committee asking us to stop dosing at the highest dose, we did that evaluation. Together, the team came forward with the recommendation that there was value in continuing at the one and a half mg per kg dose. Just remember, Tom, that by December, because we had completed enrollment into the trial in September, by December, everybody would have got at least three months of treatment in the study. The DSM was continuing to monitor the one and a half mg per kg.
In January, they advised us to stop dosing because of an imbalance between the placebo and the one and a half. That is what I know today. I do not know anything more than that. When the data is unblinded, I will learn. I am trying to be patient in the meantime. In terms of hypothesis, I do not have a hypothesis as to what causes pericardial effusion. I do believe that it is specific to the patients with PH. The reason is that in our tox studies, dosing at 50 milligrams per kilogram every two weeks, we did not see any evidence that would point us to pericardial effusion. Nor did we see it in the healthy volunteers study. Other molecules that I worked on, you see that in, excuse me, you see it in the tox studies or in the Phase I studies. We did not see it.
What could it be? I think the vasculature in PH patients is sufficiently altered that we are augmenting the pericardial effusion that's the background in these patients. We've talked about the approved product, Winrevair, sotatercept, in the past. It has demonstrated bleeding events. We've talked about how the biology of this pathway could be altering that, right? However, one of the things that we observed looking at the Summary Basis of Approval is that bleeding events occur from the first dose, the second dose. That's not further changing the biology. It's there. We looked at the PH model preclinical and showed that increasing red blood cells, which cibotercept does with a different agent, erythropoietin, is sufficient to cause bleeding and shorten survival.
I think in PH, all vasculature is altered, not just the vasculature in the pulmonary where you got pruning and thickening of the blood vessels. It looks like other blood vessels have a different pathophysiology where they're actually getting weaker. I don't understand what's going on in terms of pericardial effusion today. There is no preclinical model to help me understand that.
Understood. Do we know in PH patients? I mean, I know there's registry data. There's a number of different publications that have tried to look at what is the background rate of pericardial effusion. Do we have a good sense for what that background rate is in these patients?
No, we don't. Okay. Right. When you look at those publications, you come up with different numbers. Okay. We don't know. It's got to be in the 10% or more. Some of the studies suggest higher. I think 10% as a baseline or above that, okay, right, could be the case.
Understood. We actually ran a MEDACorp survey that happened to be in the field kind of as these announcements came forth from TROPOS. That did pick up reports of Winrevair-treated patients who experienced pericardial effusion, who subsequently paused or discontinued due to that. It also picked up from the clinicians who reported this kind of mixed opinions as to whether they thought it could be drug-related or not. I guess, what are your thoughts on those data points? Is it possible that just with cibotercept through their clinical development program, perhaps they were not monitoring as closely? I guess, what are your thoughts on the association in the real world with Winrevair and pericardial effusion?
I think there's a background. I think this biology is altering it on top of the background. Okay. Is it real with Winrevair at the moment? I don't know. Okay. I think the point I would make there is that with Winrevair, you're dosing at very low doses. You may not be having sufficient target engagement, right, to see it. Okay. Because we don't have any increase in red blood cells with cibotercept, we were targeting sufficient engagement even at the lowest dose of 1.5 mg per kg. It could well be that because of the increase in hemoglobin, you're just keeping it at a very low dose and you're not seeing significant pericardial effusion above the background.
Okay. That makes sense. When we think forward to the Tropos top line data in Q2, it's a truncated data set. Do we have a sense of, I guess I'd be curious what your expectations are for the top line. Do we have a sense of number of patients, kind of like average treatment duration, anything like that to help sort of frame expectations for that top line?
Yeah. I think the first thing is because the study was terminated earlier and everybody got their Pulmonary Vascular Resistance measured at baseline, and then the next one was at 24 weeks, we're not going to have many patients that would have seen 24 weeks of PVR. We're going to have to rely upon other measures like Six-Minute Walk Test, functional class, and so on, right, in order to assess the efficacy. Hopefully, there are enough patients that did get six months of treatment that we can get some signal on PVR as well in terms of a dose response. What do I want to see? I want to see was our original hypothesis correct, right? In terms of efficacy, are we seeing a dose response? Are we seeing a plateaued dose response? What, right.
Our hypothesis was without the increase in red blood cells and a decrease in platelets, there should be a reduced bleeding. Right. Are we seeing that? With respect to pericardial effusion, what do the numbers look like between the different doses? What is the time to onset? Right. Can we glean anything from that? More importantly, is there a certain baseline characteristic or a co-treatment that is leading to pericardial effusion? Because if we can figure that out, then there is a path forward in pH. If we cannot see a good correlation between the baseline characteristics or co-treatment, the path forward in pH is very challenging.
Understood. Yeah. I guess the follow-up to that would just be what would you need to see to feel comfortable that there is a path forward? I know we have a whole library of other compounds. Are there other backup compounds that you've been working on that you think could be potentially advanced? I guess it's all kind of pending data.
Yeah. I think if we find a particular baseline characteristic that says that all the patients had this and therefore you can put that in your exclusion criteria, other than that, I don't think, okay, we can move forward. With respect to a backup molecule, we've got a library of ligand traps of probably more than 100 at this point in time. I think the question for me is how do I select one that doesn't have the same biology? I need a preclinical model. Unfortunately, no such thing exists. We have to develop it. Once we've developed it, then we can test in that model and find a molecule that doesn't have this liability to go forward in PH. Until then, it will be repeating the same experiment.
Understood. No, that makes sense. Let's shift gears and talk about KER-065. And maybe you could just highlight the mechanism and what you're doing differently here versus some of the other activin and myostatin approaches from companies like Lilly or Regeneron or Scholar Rock.
Sure. I can do that. Before I do that, just one comment on cibotercept. While we may not have a path forward in PH, the drug was very clean in the healthy volunteers. There are other opportunities. We demonstrated in that Phase I increases in bone mineral density in postmenopausal women. There are other opportunities for the drug. It will not be that there are no opportunities for the molecule. With respect to 065, 065 targets two molecules, negative regulators in the TGF-β pathway, myostatin and activin A. Both signal through the same receptors. You can take different approaches to increase skeletal muscle. People have gone after myostatin alone. That is the Scholar Rock approach. You have two ligands that are negative regulators.
As long as the other one's around, the magnitude of the increase is going to be small. You need to inhibit both of ligands. You can do that with two different antibodies. The Regeneron approach, where they've got an antibody to myostatin and an antibody to Activin A. You can use a ligand trap like we do where you're binding those ligands and additionally other ligands like Activin B, which also signals through the same receptors. You can take the approach of inhibiting the receptor, the Versanis now Lilly approach. Regardless of which of the approaches that inhibit signaling by Activin and myostatin, you will get to the same magnitude of increase. There's nothing to differentiate between a combination of two antibodies to activin and myostatin, targeting the receptor or using a ligand.
You're going to get to the same magnitude, which is much greater than a myostatin or an activin A antibody on its own.
Got it. That's helpful. You mentioned this in your opening remarks. Currently looking at 065 in a Phase I SAD/MAD study in overweight individuals, but otherwise healthy individuals. Maybe you could just help frame expectations for that data set later this month. What are you measuring? What would you like to see to consider that a successful sort of proof of mechanism signal-generating study?
Yeah. It's a safety study first and foremost. As always with this biology, you can glean the signals of that biology in healthy volunteers study. In the multiple ascending dose, we have subjects that have a BMI of 27-33. Treatment for three months, anticipate that we should be able to see increases in lean mass, so muscle, by DEXA scan. We should be able to see reductions in fat mass. This is the first time that any bone anabolic agent has ever been taken into men of 18 to 55. I don't know what to expect from bone mineral density changes in that population. If it was postmenopausal women, I'll tell you we're going to see an increase in bone mineral density. In addition, there's MRI in the study looking at imaging of the thigh muscle.
I think there you can look at changes in fat as well as the size of the muscle, so muscle volume, right. I think that's the main things. There's, of course, biomarkers of metabolism also included in the study.
Okay. Are you expecting to see any changes in patient weight? I guess, is there any instruments that you've implemented here to get a sense of changes in kind of like functional muscle mass or strength?
Yeah. We have not included any functional measures whatsoever in the study. They're very noisy. These are very small cohorts of six or eight. You never see changes in those small cohorts. With respect to changes in weight, muscle anabolic will always cause an increase in body weight first. You've got to build that muscle, which will then consume the fat to reduce fat. I think, again, small ends, it's really hard to see in small cohorts. On top of that, with three months of treatment, I think if you had a large enough end, you'd actually see a small increase in body weight.
Understood. Earlier this year, you announced kind of a strategic shift away from obesity, kind of the general obesity indication into more of the targeted neuromuscular disease. Do you share as a lead there? Maybe just talk about what went into that strategic shift and how you think about the market opportunity there.
Yes. The strategic shift was really driven by the fact that we had found a partner for our most advanced asset. We were partnering our most advanced asset. And then our third asset going into obesity meant it was also a part will be a partnered asset at some point. We've talked about it in the past. Neuromuscular indications is a real big opportunity because the biology of these molecules aligns perfectly with the biology of these diseases. Regardless of what the mutations in the structural components of muscle are, the pathophysiology that arises is the same. The time of onset and the progression varies from one mutation to another, but the pathophysiology is the same. That muscle is weak. As a consequence upon use, it breaks down, leads to inflammation, which then inhibits regeneration of that muscle, and then you lose function.
It is the same biology regardless of which of these indications it is. What does 065 do? It can build muscle. It increases muscle regeneration. It has anti-inflammatory properties. Boys with DMD are on glucocorticoids, which are catabolic. They are catabolic because they increase expression of myostatin and activin A, the two negative regulators. In bone, it is Activin A. Therefore, you get osteoporosis in these boys. Their weakened muscle is breaking down. Inflammation is then leading to fatty infiltrates, fibrotic tissue. 065 is really reversing and ameliorating all of the pathophysiology that you see in boys with DMD as a consequence of that single mutation in the dystrophin gene. What we have also seen in preclinical models is that because you are increasing regeneration, you actually increase expression of the homologue of dystrophin, utrophin, which is present in boys with DMD at very low levels.
In preclinical models using exon skippers, we see enhanced expression of dystrophin when you use a combination of the exon skipper and 065. I think it marries extremely well with the pathophysiology of the disease in patients with DMD.
That makes sense. Assuming you see a good signal out of this phase one study, maybe just walk us through how you're thinking about next steps in DMD and sort of like timeline to data in patients.
Yeah. I think once we see the Phase I data, we have to engage with the regulators. Okay. I think the opportunities that I just spelled out are opportunities in boys that are ambulatory, boys that are non-ambulatory as well. At the same time, patients that are on glucocorticoids, glucocorticoids and exon skippers, or gene therapy as well. I mean, if nature thought it was appropriate to have dystrophin instead of utrophin after birth, can you imagine, okay, right, how good a micro or mini dystrophin is going to be when you only have got three to five repeats in those and utrophin has 17, dystrophin having 20. Okay. I think it can be an all-comers sort of trial where we look for changes in biomarkers that show proof and then using that to have a study that is a registration study.
We can do this study as an open label study. Therefore, we're generating data as we go along. I think this is going to be in young boys. Therefore, we are going to have to have those discussions with regulators early. I think this puts us in a position where the start of a study is not this year, likely, okay, towards the end of this year into 2026. With an open label study, okay, as soon as you got half a dozen patients, right, you can start looking at what's going on.
Got it. That is helpful. Let's switch gears and talk about elritercept, which is your most advanced asset, really nice Phase II data in MDS and some good signal as well in myelofibrosis. This asset, as you alluded to in the intro remarks, you partnered with Takeda. Maybe just talk through why Takeda, how this kind of fits into the Takeda broader HEMONC plans. I know we have a Phase III MDS study that is underway. Maybe just talk through kind of like how you're thinking about market opportunity and why Takeda for this asset.
Yeah. I think as we have discussed in the past, the opportunities of ameliorating the anemia in a broad set of indications was always there with elritercept. We focused on MDS and MF. I think when we talked to strategics over the years, there were those that were interested in MDS. There were those that were interested in MF. In Takeda, we found a company that really fully appreciated both opportunities, but beyond. Of course, you're going to go about it in a stepwise manner. We as a small company could pursue MDS in a second line setting. Once we get approval, then go into front line setting where ESAs and now ESAs and luspatercept is being used. To compete there, you do need to get into the front line setting. We would have had to wait to do that.
Takeda has the means to be able to pursue the indication in the most optimal way. They have the capital to be able to deploy in multiple studies at the same time. We could not do that. I think in terms of the market opportunity, Takeda in their R&D day said in MDS alone, they see $2 billion-$3 billion as an opportunity for elritercept. That's pretty nice. That's their number.
Yeah. Yeah, maybe just remind us on some of the nuanced differences in biology, but then also the data that you've generated in Phase II, help us compare it to obviously Reblozyl is out there, luspatercept. There's also a drug, Rytelo, recently approved, launching now in MDS. You mentioned Takeda more aggressively moving into front line, hopefully. Where do you see this kind of fitting into the anemia treatment paradigm in MDS?
I think the data that we've generated shows that elritercept is able to treat patients early on in their disease as well as those that have advanced in their disease. I think that's where the differentiation starts with luspatercept, where luspatercept works very well when your bone marrow is functional but compromised. As the disease progresses, the efficacy with Reblozyl declines. We've shown that in the MDS trial that the durability of response with elritercept is over 130 weeks, whereas in a similar setting with Reblozyl, it was only 30 weeks. I think we've demonstrated a more durable response because it's able to treat the bone marrow as it progresses in its disease. Therefore, in terms of where it fits in, it can fit in right in the front line setting and then continue as the disease progresses.
Rytelo, on the other hand, works well, okay, right, in that late stage, but has a very challenging safety profile, which academic centers are able to manage well, but in the broader community, I don't know.
Right. Okay. Maybe last question, remind us of balance sheet, kind of like cash on hand. And then as we think through this library of proprietary compounds, like how you're thinking about advancing things from the platform towards the clinic, is there a certain goal in terms of like IND or cadence or opportunity or just how are you thinking about resource allocation?
Yeah. We had $560 million of cash at the end of December. That did not include the $200 million from Takeda. We have a very healthy cash balance. I think we will always continue to move preclinical assets forward so that we replace the molecules that have moved on. We have not guided to what the goal is. I think you've seen in the past, we tend to wait until we have the candidate ready to go into the clinic before we announce it. I think that we're not going to stray from that policy, right, at this time. I think we have lots of opportunities. The first of those would be to make the decision on cibotercept and then to move 065 into DMD and additional neuromuscular indications.
That makes sense. All right. Unfortunately, we're up against time, but thank you, Jas, for joining us. We'll stay tuned to the Keros story.
Thank you.