All right, good afternoon, everyone. Thanks for joining us here at the Leerink Partners Global Biopharma Conference. My name's Tom Smith, I'm one of the senior biotech analysts here at Leerink, and it's my pleasure to welcome our next company to the stage, Keros Therapeutics, and really happy to be joined by CEO Jasbir Seehra. Jaz, thanks for joining us.
Tom, thank you for the invite, and wonderful opportunity to have a chat.
Yeah, looking forward to it. Why don't, Jas, you kick us off here with an overview of the company, some of the key upcoming milestones, and walk us through the setup for the rest of 2024?
Well, Keros is a clinical-stage biopharmaceutical company focused on the TGF beta pathway. We have three activin receptor ligand traps in our pipeline in clinical evaluation. KER-050 is the furthest advanced, and MDS and myelofibrosis we've been sharing data from the open-label studies for a few years now, so the profile has become pretty clear over the last few years. And with that program in MDS, we're moving to engage with the regulators on the design of the registration trial, in the first half of this year. The myelofibrosis data that we had read out at ASH of last year was very exciting.
For the first time, we're seeing not only correction of the cytopenias, but also, as was hypothesized, if we are restoring hematopoiesis in the bone marrow, the need for hematopoiesis in the spleen would be reduced, and therefore you should see improvements in symptom score as well as spleen size reduction, and we saw that at the lowest doses. As the year progresses, we'll have two data readouts on both of those programs, one mid-year, one towards the end of the year. You can sort of guess at what those venues could be. And with the MDS, you'll see more maturing data, in terms of, durability of response. And in the myelofibrosis, now you'll get to see the higher-dose cohorts as to the efficacy there. Our second product candidate is KER-012, which is, another ligand trap that's targeting pulmonary, pulmonary hypertension as well as cardiovascular disease.
We completed a phase I study and demonstrated its mechanism of action, and now it's in a phase II trial in PAH, and we'll indicate the timing for last patient being recruited into the study in the first half of this year, so you'll be able to get an idea of when the top-line data could be on the basis of that. And then in the second half of the year, we have a trial, open-label trial, biomarker trial in patients with heart failure, and so we hope to see changes in NT-proBNP in that trial that'll be in the second half of the year. And then our third product candidate, KER-065, is another ligand trap now that's increasing skeletal muscle.
As a consequence of increasing skeletal muscle, you're increasing energy expenditure, and that has opportunity in obesity, both in combination with GLP-1 receptor agonists as well as as monotherapy. That phase I study is ongoing, and we'll have readouts from that in the Q1 of 2025.
Great. Great overview. I want to start off by talking about KER-012, and obviously we're coming up against a potentially very important milestone with the sotatercept PDUFA later this month. Maybe we just take a step back, and you could talk about sort of the design of 012, how you've designed it to be differentiated from sotatercept, and how you think that differentiation will manifest in the clinic.
Yeah. So when we came up with 012, what we were looking for was a molecule that has some of the biology that sotatercept has, namely, okay, right, that it acts to inhibit activins but allows BNP signaling so that you can see a profound effect on bone. That's how we looked for it in preclinical study, which is what sotatercept showed. However, sotatercept had an increase in hemoglobin, red blood cells, that limited the doses that it could be used in those indications. Therefore, you could never dose high enough for treatment of bone, and it became an anemia drug. So with 012, we were looking out of the gate for a molecule that does not increase hemoglobin, right? And that's what 012 phase I data showed, that it does not. Now, therefore, it should have shown all of the biology of sotatercept.
As we have learned with sotatercept, there's additional safety signals associated with that, which are due to, we believe, inhibition of BMPs, BMP-9 being one of them, but other BMPs are also involved. When we were looking for 012, we were looking for a molecule that had the greatest differentiation in terms of inhibiting the activins while preserving BMP signaling. We think that's going to translate to potentially greater efficacy because we can get to higher target engagement, without the hemoglobin changes, but also potentially better efficacy because we're binding less tightly to the BMPs.
Got it. You have the phase I healthy volunteer data in hand. I guess walk us through a little bit of the highlights of that, and how much can we really extrapolate from the healthy volunteer experience into PAH, the disease state in patients with PAH?
Yeah, good questions. So when we started that trial, we were trying to recapitulate what was seen in terms of biomarker changes with sotatercept, but most importantly, no increase in hemoglobin. And we went up to the highest dose, 4.5 mg/kg, three times at a monthly interval of dosing, and we saw no increases in hemoglobin. So that was checkbox checked, okay, right? And then we saw increases in bone mineral density, as well as changes in bone biomarkers that were very consistent with what was observed with sotatercept, but we were seeing those at lower doses than sotatercept produced. So at 1.5 mg/kg, we were seeing equivalent to what was observed with sotatercept at 3 mg/kg, right? So we were seeing this differentiation between activin and BMP binding already translate to the bone tissue.
Like many folks ask me, well, how is this going to play out in PAH? So we went back, okay, to the Healthy Volunteer study and asked the question, can we see changes in parameters that are more relevant to the PAH patients? One of the things that has been demonstrated is that PAH patients have elevated NT-proBNP. So can we actually inhibit and reduce NT-proBNP in a Healthy Volunteer population, right? Well, postmenopausal women have an elevated risk for cardiac disease, so they actually have a slightly elevated NT-proBNP, and we demonstrated from a single dose, at the highest dose, we could reduce NT-proBNP. That actually encourages us to look for other markers, so we look for inflammation and also for changes in fibrotic markers.
Using a proteomics approach, we saw numerous biomarkers that are consistent with reduction in inflammation and reduction in fibrosis, which I think now shows that that biology is pertinent to the PAH patients. Thank you, okay, because you guys kept asking the questions. It forced us to think and mine additional data from that trial. At this point in time, I'm actually very, very encouraged with what we've seen, and I just can't wait to see the readout for TROPOS.
Yeah. Same. Because you mentioned the NT-proBNP in healthy volunteers, and I want to come back and talk about sort of sotatercept and talk about TROPOS, but you do have the heart failure study, and we're looking for biomarker data in a handful of patients in the second part of the year. I guess what, you know, help kind of frame expectations for that? This will be looking at NT-proBNP in patients, right, active disease state. What would you consider, like, clinically meaningful change? What's sort of the change that you're looking for?
Yeah. In postmenopausal women, it's elevated, but it's still below 100 pg/mL, whereas in patients with heart failure, it's going to be multiple hundreds, okay, right, of pg/mL, 500,000, some even much higher. So we're looking to see a reduction in individuals that have these elevated levels of NT-proBNP. Now, when we designed that study and started implementing it, we didn't have the data from our other program, in particular with 050 in MDS patients. And MDS patients, second leading cause of death is cardiac disease, so we went in and looked at to see if there were any patients with elevated levels of NT-proBNP. And indeed, we saw some, and we saw that 050 treatment at 3.75 mg per kg dramatically reduces NT-proBNP from a single dose.
So in some ways, okay, the study that we're doing in heart failure, we've already got all the answers, okay, right, from the various studies that we've conducted. So I was thinking originally that the heart failure study would be an early read to thinking about how to further make the investments in 012. At this point in time, I've got all the data that I need to continue to move that program as rapidly as we can, but we have that study. It'll give us additional data.
Yeah. No, I think it'll be a very interesting data point in the second half of the year. I do want to come back and talk about sotatercept, and I think, most investors see approval as likely. We'll be focused on quality of the label. And there's been emerging evidence, both case studies and data reported out at PVRI in January that's highlighted the risk of bleeding events after treatment with sotatercept. How do you expect sotatercept is going to end up being labeled? I mean, this is one I think a lot of investors speculate around, you know, could there be a black box, or could there be a REMS? I guess what's your sort of base case expectation for sotatercept and how it'll be labeled, and then how do you think this reads through to KER-012?
I think, first of all, I do believe it's going to get approved, okay? There's no question in my mind. Sotatercept will be approved. I think it's what's on the label, and none of us know. I think when PULSAR open-label extension read out, there was the telangiectasias. And a year ago, Merck talked about the telangiectasias as a cosmetic AE. I think in the course of the year that has elapsed, physicians have become much more concerned about it. So I think there's going to have to be an education, how that is in the label, okay, right? We don't know. It could be in the form of a REMS. It could be a black box or just very strong language in the label. We just don't know, but it's going to be there.
How that plays out for us is, well, when you get these bleeding events, what do you do about it, right? Do you take the patients off the drug? Some are now talking about less frequent dosing. Well, if you're dosing less frequently, you're not going to get to that level of target engagement. Therefore, the efficacy is going to be reduced. When you go back again and look at the PULSAR open-label extension, you see that the benefit came in the first 24 weeks. There was no additional benefit reported 18-24 weeks analysis. So maybe, okay, right, if these telangiectasias and bleeding events are occurring, you take the patient off drug. Now, how long do you keep them off drug? Because these telangiectasias are not going to go away overnight, okay? So I think it actually represents an opportunity for us.
One other thought, okay, right, on that front is there is an increase in hemoglobin. How does the increase in hemoglobin relate to bleeding events? Because while an increase of hemoglobin from 13 to 17 is not going to cause heart failure, it is going to increase the viscosity and therefore blood pressure. Through a blood vessel that is already fragile, it's going to increase the likelihood of breakage. So is there a correlation? I'm looking forward to the data being shared in a few weeks' time.
Yep. We'll have the answer soon enough.
Right.
Okay. I want to talk about the Phase II TROPOS study, and you announced this, I think, back in August. Maybe just give us a little bit of overview on the design, and then I know we have a formal enrollment update that you're expecting in the first half of 2024, but just talk qualitatively, like, what you're seeing in terms of enrollment trends when you think about the availability. Sotatercept, does that help you, hurt you? What does that do to TROPOS enrollment?
Okay. So 3 dose levels in the study. We got 1.5, 3, and 4.5 mg per kg. At 4.5, we got maximum target engagement. At 1.5, we got target engagement to the levels of sotatercept, okay? So we think we have the opportunity to be able to see if there's greater efficacy to be achieved in these patients, right? I think that's what the study allows us to do. If, okay, there is no additional efficacy to be gained, you might see a flat dose response, in which case, what's the safety profile? Because you could anticipate, as you go to much higher doses, you could see some safety signal. So this really allows us to optimize that, okay? With respect to enrollment, we were always thinking that sotatercept would be on the market in 2023.
I can tell you Acceleron would have rushed it, okay, right, and made sure of that. But with that, we knew that we would have a significant headwind in the U.S. Therefore, we have a study that has 60 sites, which is larger than most studies with 90 patients. Three-quarters of them are ex-U.S. So we were planning with a headwind that we have not seen. And I think the enthusiasm for the mechanism and the differentiation of the mechanism from sotatercept is well appreciated by the community to the point where I think most physicians are very excited about it.
We hear it in the form that as you're getting ready to open up a site, they're telling us about patients that they have in mind that they want to screen, and then when the site is open, if they say they've got three patients, there's three patients in screening. I think that's pretty exciting.
Yep. Yeah. All right. I want to switch gears and talk about KER-050, and this is, as you alluded to, this is your most advanced program. I want to talk about MDS, and you've guided to engaging with regulators in the first half of the year regarding a potential phase III trial design in MDS. Just maybe give us your latest thoughts on what that trial looks like. Are we targeting, like, what specific patient populations or subpopulations are we targeting, and do you still expect this to be a placebo-controlled study?
Yeah. It's going to be second-line treatment after ESAs. That's the study that we're going to propose to the regulators. So it will be placebo-controlled with the transfusion independence as the primary endpoint. So it will be similar to MEDALIST. In terms of the population, we've seen a very good response in the high transfusion burden patients. Response in low transfusion burden is also there. Our durability of response appears to be really good in the high transfusion burden patients, where you can actually see the big difference. So I think in terms of the label that we want to seek is broad treatment of anemia in MDS patients. So that's going to include RS, non-RS, low transfusion burden, high transfusion burden. Exactly what the population mix is, that's going to be the subject of discussion with the regulators.
Understood. There's a competing agent that has an FDA AdCom scheduled for later this week. There were briefing documents that came out this morning. In looking through those briefing documents, I mean, the agency appears focused on, obviously, safety in this patient population, but also efficacy and, I guess, consistency across all of these various measurements, not just transfusion independence, but quality of life, patient-reported outcomes. I guess when you think about the data that you have in hand for 050 today, you know, just maybe summarize briefly kind of like what you've seen and what gives you confidence that you're going to end up checking those boxes in the phase III.
So I think a couple of things. First of all, we're seeing a response in all patient groups that we're looking at, RS, non-RS, low transfusion, high transfusion burden. And the safety profile is really important in the low-risk patient population. You've got to have a clean safety profile, and 050 to date has been a very, very benign safety profile, not too dissimilar from luspatercept, okay? However, where we are seeing some real benefit there is that for the first time, we reported on the fatigue score at ASH, and we're seeing improvements in the fatigue score that gets above that minimally clinically important difference of three in the patients in our study. So I think for the first time, you're actually seeing a benefit. And prior to that readout, we were hearing of physicians telling us that patients felt better. Describe, okay, what that is, okay, right?
How is it that they're feeling better, right? You're seeing that in the fatigue score. Quality of life, okay, right, is going to be an important instrument. The early signals, okay, with the fatigue score is that we could have a differentiation from these other products.
Got it. Okay. Let's talk about myelofibrosis, and we'll speculate that there may be an update at EHA. You presented data at ASH. Just help kind of frame expectations. What should we be looking for with an updated data set? You said we were starting to reach higher doses and perhaps see a bit more in terms of characterizing the efficacy in myelofibrosis, but help sort of frame expectations ahead of that data readout.
Yeah. So if you recall, Tom, when you had asked me that question in previous years, I said, "Initially, we're going to see response on the cytopenias, the anemia, the thrombocytopenia." And then when we have got patients on treatment for longer than 6 months, and I didn't know what that was going to be, whether it's a year or more, and you're resuming hematopoiesis back in the bone marrow, eventually, you're going to get a reduction in spleen size. So at ASH, we reported hematology parameters on the 3 lowest dose levels through 12 weeks. And in the 2 lowest dose levels, 0.7 and 1.5 mg per kg, we had had patients that had gone beyond that 24 weeks of treatment. Therefore, we could evaluate them for the spleen size as well as symptom score. And we saw reductions in spleen as well as reductions in the symptom score.
That, to me, was a surprise, pleasant. I was not expecting that at those 2 low-dose levels. At the time of the data cutoff, we had patients in the 3 and the 4.5 mg per kg dose. Those patients should now be maturing to 6 months of data, their 6-month endpoint, most of them, so that that will be the mid-year data readout. In addition, we opened up Part 2 in the fourth quarter of last year. Those patients should be getting to 6 months and longer by the end of the year. So there'll be a lot more data in terms of numbers by the end of the year.
That gets to be exciting because we believe that we could be in a position at the end of the year or early into 2025 where we have enough data now to engage with regulators on the design of a phase III trial for myelofibrosis.
Got it. And just in the last five or so minutes, let's switch gears and talk about the newest program, KER-065. Talk a little bit about the mechanism and how you design this compound. And then, obviously, this is an area of increasing investor focus, but how does 065 compare to some of the other activin and myostatin approaches that are either in the clinic or approaching the clinic?
Yeah. I think the first thing is to recognize that in this pathway, there are multiple negative regulators of skeletal muscle. They're myostatin and activin A. While myostatin was identified as first, right, later, it was realized that both of them are at play. In rodents, myostatin levels are high, activin levels are low, and in humans, activin levels are high, myostatin levels are low. So if you inhibit with a myostatin inhibitor, you see much larger effects in rodents in terms of increases in skeletal muscle than you see in primates and in humans, okay? So we believe that the best approaches are to inhibit both activin A and myostatin, and that you can do through either two antibodies, as is Regeneron doing, where they have an activin A and a myostatin antibody.
You can go to an antibody that inhibits the signaling of these ligands in skeletal muscle, which is bimagrumab. It targets the two receptors. Or you can go with a ligand trap. And the advantage of the ligand trap is that in obese individuals, that infiltration of skeletal muscle by fatty tissue means that you're secreting not only activin A from that fat tissue but also activin B, activin C, which all signal through the same receptors. So we think that bimagrumab and 065 have very similar potential for efficacy. Bimagrumab has to be dosed by IV just because of its mechanism. It's targeting a receptor where even a single signaling event leads to the consequences. Therefore, you have to inhibit 99.9% of the receptors in order to see the efficacy, whereas with 065, it's going to be subcutaneous administration. The data with this approach is already there in the literature at Acceleron.
We had the ACE-031, went into healthy volunteers, where you saw about a 5% increase in lean mass, about an 8% decrease in fat mass in postmenopausal women from one month of treatment, either at two-week intervals or a monthly interval. That program was put on clinical hold because of nosebleeds and gum bleeds due to very tight BMP9 binding. That has been dialed out of 065. We know what to expect if this trial was in postmenopausal women. Our trial is in men. In the multiple ascending dose, we have BMIs of 27-33 with the appropriate waist-to-hip ratio, so they are obese individuals or overweight individuals. We're looking to see, are we getting a trend of reductions in fat mass, increases in lean mass, and also anything on the bone axis because we would expect improvements in bone as well.
That should come out about this time next year, Q1.
Q1 2025. Okay. That's great. And just in the last minute that we have, you have a proprietary library of these compounds. I think investors, not that they were surprised to see 065, but I think they were excited to see 065 kind of move forward out of that proprietary library. How are you thinking about advancement of additional compounds into the clinic? You have a lot on your plate now, but how are you thinking about other programs that could move forward over the next 12, 24, 36 months?
Yeah. It's all dependent upon capital. We're in a good place where we got runway into 2027. But we have other molecules, and you've seen us highlight some data with another ligand trap very similar to 065, 034. And so we will be preparing to do safety studies. And at the appropriate time, we will then make a decision whether it's appropriate to take it into the clinic. I think, really, we want to make sure that we're advancing our advanced assets as quickly as possible and not bringing in too many early assets that are distracting and detracting from building value on the advanced asset. It's really important to keep that balance. 065 wasn't identified last year. When we announced that we were going to take it into the clinic, we had already completed chronic tox studies with it, right?
That's how we can continue to build a pipeline, building value by having completed those studies so that when we go into the clinic, we're not limited for the duration of treatment. If you recall, with 050, that was when we were a private company that we took it. We only had short-term treatment. We had three-month tox studies, and therefore, our initial trials were limited to three months of treatment. That's not the case with 065. We can continue to treat forever.
Right. Okay. That's great. Unfortunately, Jaz, we're running up against time, but thank you so much for joining us, and we'll stay tuned to the Keros story. Big year of execution for the company.
Every year is okay, right? This is not different, but it's now at a different place. So it's very exciting.
Absolutely.
I want to thank you once again for giving the opportunity to have this chat.
Absolutely. We'll stay tuned. Thank you.
Thank you.