Great, thanks so much. Thanks everybody for joining us here, second day, Piper Sandler Annual Healthcare Conference. Great to kick off this next session with Keros Therapeutics. Joining us is their CEO, Jas Seehra. Jas, thanks so much for joining. Always a pleasure to talk to you. Obviously, a number of data sets we'll get in a couple of weeks, so a lot to talk to, but maybe I could give you a minute or two, you could introduce Keros, what you guys are doing and what we have to look forward to.
Yeah. Thanks, Joe. Thanks for the opportunity to have this discussion. So Keros is a clinical stage biopharmaceutical company, focused entirely on modulating the TGF-β superfamily of ligands. And as a result, we now have a pipeline of assets that are potentially differentiated. KER-050 is in hematology, and we'll spend some time on that. In phase II trials in MDS and myelofibrosis, KER-012 is a another ligand trap that's in PAH and in heart failure trial. And then we've got KER-065, that is a new asset that we unveiled a few months ago, that increases skeletal muscle and has opportunities in neuromuscular indications, Duchenne being one that we're particularly excited about, but also in broader populations where there is muscle wasting, including the intense excitement around obesity. So with that?
Yeah, that's perfect. Maybe with that, we could start with KER-050, and a couple of the datasets we'll get at ASH. Maybe starting with MDS. I think I've got a number of questions on this update and what we should expect to see. So maybe you could help sort of set the stage, and what the dataset could look like at a high level, what new we might learn about KER-050's profile, and then we could sort of dig into maybe some specifics.
Yeah. So I think we, the last dataset that we reported was at EHA, where we had an April 1 data cutoff, and at that time, we had 59 patients in the safety database. Approximately 22 patients at that time had not received sufficient number of doses to be efficacy evaluable. Now, those patients would have progressed with enough treatment to be efficacy evaluable. So you're going to see, in terms of efficacy, looking at durability of response, and also the number of patients with the response will be increased at ASH. So I think the key thing that we've been sharing over the last few years is the broad population that KER-050 is showing efficacy in. It's showing efficacy in terms of anemia and low transfusion burden, non-transfused patients, as well as in high transfusion burden, RS and non-RS.
I think where we're seeing the differentiation from luspatercept is in the high transfusion burden patients. These are harder to treat patients, therefore, the more difficult to treat patients are the patients that require a drug that has a better, more, is a more powerful drug. Okay, right, and you're seeing that. So one of the presentations will focus on the efficacy data. There's another presentation at ASH that will focus on biomarkers, and we've also been talking about that. Luspatercept was designed not to bind Activin A. Activin A is a driver of inflammation, fibrosis, and bone loss. And therefore, it creates environment in the osteohematopoietic niche that is hostile to normal hematopoiesis. KER-050 binds and inhibits Activin A signaling.
And as a consequence, we've been seeing this response in higher transfusion burden patients, but there's many more aspects of the disease that where Activin A is playing a role. And we've been sharing some of that, that we see a very dramatic decrease in serum ferritin that was not observed with Reblozyl. In the ASH abstracts, we also started talking about bone specific alk phos, right? Increase in bone specific alk phos suggest that you're restoring that environment in the bone, where it is much more of normal bone rather than bone that is undergoing resorption. So we're going to talk about that. We're going to talk about other biomarkers that we're seeing. Patients with MDS die of progression to AML, and, but the second leading cause of death is heart failure. Right? So we're going to be looking at these.
We're also going to be looking at quality of life, so we'll have some data on that. So I think it's a more substantive data set in MDS, and on the basis of this data, we are going to engage with the regulators on the design of the phase III registration trial. So we believe we have a sufficient amount of data.
Yeah. That was great. Maybe I want to sort of follow up on a couple of things you said there, maybe on the sort of durability aspect.
So you mentioned, like, the clear benefit you think you're observing in high transfusion burden patients. I think when we last spoke, there was potentially this view that, you know, you could differentiate from luspatercept in low transfusion burden patient on the durability side of things. Maybe you could sort of speak to that and what's like a median time of transfusion independence that sort of points to longer duration of effect?
Yeah. I think, you know, patients that are earlier on their in their disease journey, that they are going to be the non-transfused and the low transfusion burden patient. The durability of response is going to to see that, you're going to have to treat for much longer period of time. After all, in the COMMANDS data for luspatercept, you see that the ESA response in frontline is 90 weeks. Okay? So you're going to have to treat for many years in order to be able to see it in those patients. However, in the high transfusion burden patients, they quickly, okay, right, stop responding to other drugs, and therefore, they move to transfusions again.
So it's gonna be quick, easier to see it in the higher transfusion burden patients first, but I still believe, okay, right, that in the low transfusion burden patient, you will see an even longer response, but that's gonna take longer treatment decision.
Maybe a weird question, and maybe it doesn't make sense, so but going back to the biomarkers around, like, heart failure, something you've shown with KER-012 in healthy volunteers around BNP. Can we connect some dots from what you ee with KER-050 and MDS and reductions there and what it means for KER-012 and the context of heart failure?
Absolutely. I think these molecules all bind activin, and therefore, activin levels, which are increased in heart failure patients. You're gonna see correlation from one molecule to another. It's gonna be small numbers of patients that in MDS that would have that. It's a greater number of patients in PAH that have a stressed heart. So yes, you are gonna be able to read from the biology that you're seeing in one with one molecule and one indication in a small subset of the population that reads through to the other molecules.
The other thing I wanted to follow up on, of course, is something you mentioned around, you know, now move towards the registrational study. How do you think about balancing, designing a study that allows you to capture the biggest population of low-risk MDS versus designing a study where you could, you know, more definitively show that KER-050 is behaving differently and better than luspatercept?
Yeah. I think we have to go with a strategy for registration as one that tries to capture the broadest population so that you're not pushed into a third line or fourth-line treatment. Okay, so the registration trial will be a placebo-controlled trial, second line after ESAs, broad patient population. Now, it's all about what the makeup of that population is. What are the... We know what the endpoint is going to be. It's gonna be eight weeks for registration, but the durability of the response is important to patients, to the payers. So what are those secondary endpoints, okay, right, that we can have in that study? And then what else, okay, we can incorporate that are biomarkers that could be exploratory or secondary endpoints, right?
But I think to your point, how do you demonstrate the clear differentiation of KER-050? And it's gonna be in additional studies that can be done in parallel to the registration trial.
Got it. Maybe along these lines, how do you guys think about the potential implications around luspatercept use in the ESA-naive population? I was sort of a little surprised that it was captured a broad label and based on RS status. You know, what does that mean, maybe both from a clinical development standpoint, but also maybe longer term commercially?
Yeah. So commercially, okay, right, I think first thing that we keep seeing and from physicians is that they're still gonna use ESAs. There's, you know, 20+ years of experience there. But don't forget that the payers are gonna have a significant role here. And with the five times the difference in cost, I think you're gonna have patients going on ESAs and then going on to Reblozyl. I think what COMMANDS data in newly diagnosed patients does is it gives license to physicians to move patients through Reblozyl sooner than they would have otherwise. So I think you're gonna see the penetration into the ESA to a greater extent, and we've already seen that with luspatercept, that the durational treatment on ESAs has shrunk from about 18 months down to 11 months. Okay, so it'll get a little bit shorter.
In terms of trials, we still believe that you can do a placebo-controlled trial. Many countries in Europe, Reblozyl will not be available, and even in the U.S., not every patient will have access to it, so we think that is not going to hinder.
Well, last maybe question on this topic, and then I want to shift to myelofibrosis. The one thing I've heard around luspatercept, kind of from a safety standpoint, is it has this weird phenomenon of, in some patients, even though their hemoglobin levels go up, they just have, like, excessive fatigue that is tough to deal with. Not sure if you understand why that's the case and whether, you know, there's been any indication of KER-050, you know, maybe avoiding that.
Yeah, we've not seen that. We've actually had reports of patients that are seeing a response but not getting to TI, actually wanting to remain on the drug because they're feeling better. I think the part of the reason could be is that luspatercept was designed not to bind Activin A. Activin A is a potent pro-inflammatory member of the TGF-β pathway. And we all know when you have inflammation, right, you don't feel great, and therefore, while you're addressing the anemia component, you're not addressing the underlying smoldering disease, okay, right, with luspatercept.
Great. So, so maybe with that, we can move to myelofibrosis, another sort of update coming at ASH, maybe similar to the first MDS question. What should we expect to see there, scope of the dataset, dose levels, and what we might learn about KER-050 within the, the context of myelofibrosis?
With KER-050, we've been going through the dose escalation. Last year at ASH, we presented a little bit of data from short-term treatment at the lowest dose, 0.75 mg/kg. Since then, we've dose escalated through all of the dose cohorts, all the way through to 4.5 mg/kg, and opened up part II with a recommended part II dose of 3.75 mg/kg. So it's the same as MDS. So patients that were on treatment at the low dose of 0.75 and 1.5 mg/kg will have had treatment for a longer period of time. Those patients that came in at the 3 mg/kg, they have become efficacy evaluable, but it's short-term treatment, all right, 12 weeks or so and thereabout.
The 4.5 mg/kg is only going to contribute to the safety database, right? There will not be any safety data from that. So you can see hematologic changes at the lower doses through to the 3 mg/kg dose, and then we're looking for changes in symptom scores and spleen volumes, and we reported in the ASH abstract that we're seeing reductions in spleen size, okay, right, as well. This is at the lowest doses. Okay, right. Yeah.
Mechanistically, I guess, what, you know, again, going back to Revlimid and what we've seen from it within the context of myelofibrosis, why should there be opportunity for KER-050 to have monotherapy activity, versus maybe what we saw with luspatercept, where it had to move forward in combination with ruxolitinib?
Yeah. So when you have patients on rux, those patients were making red blood cells in their spleen, and rux reduces the proliferation. But there's still some cells, okay, that can benefit from a TGF-β ligand trap, and therefore, you get some improvements in the anemia. However, in the monotherapy, you've got proliferation of megakaryocyte precursors, right? And now you're looking to reduce the breakdown of those megakaryocyte precursors, and as a consequence, you're going to then start restoring hematopoiesis back in the bone marrow. Therefore, okay, you should be seeing responses on both the anemia and on the thrombocytopenia. And the data that we shared last year, we're seeing changes in erythroid and in megakaryocyte lineages at the low dose.
Ultimately, and I know this is all data dependent, like, what's the easiest path for development in myelofibrosis? Is it a sort of monotherapy in a post-rux patient, or is it simply adding on to ruxolitinib in patients who maybe are not achieving full, full benefit?
Yeah. I'd, I'd think we have to let the data mature, okay? It's a little too early to do that. But it may well be that in patients that don't tolerate rux, the monotherapy, you can demonstrate the broader effects of having improvement in anemia and thrombocytopenia. Remember, these patients have multilineage dysplasia and reductions in spleen size because... or symptomatic score, because you're now seeing a benefit that is purely due to your drug, whereas in the rux, there's always the confounding aspect of rux itself. Yeah, we just have to wait, okay, and let the data speak for itself. It's early days. I think you're going to see data from at, at ASH that shows through the dose escalation. I think 2024 becomes exciting because now you have a lot more patients.
At the more efficacious doses and at the recommended part II dose.
Maybe in this last six or seven minutes or so that we have here, we could touch on KER-012. And maybe start with this question, and it's one I often get of, you know, how do we take the healthy volunteer data that you guys have generated and translate that to what it could potentially mean within the context of PAH and other therapeutic settings? So when you look back at the healthy volunteer data, you know, what can you confidently say about KER-012's profile and what that means within the context of PAH?
Yeah. I think we saw changes in bone biomarkers at the lowest dose that we used. In the bone, activins and GDFs are negative regulators of bone, while the bone morphogenic proteins are positive regulators. That's exactly what happens in the vasculature. Therefore, us seeing target engagement, changes in bone biomarkers at the lowest dose and plateauing out at lower doses than with sotatercept tells us, "Okay, right, that we've got a drug that is more BMP sparing, and therefore, you're able to engage that biology." Now, for many folks, that's a stretch. Okay, right? Bone and vasculature. That was the reason why we went in retrospectively to take all those samples that we got from the healthy volunteer study and do a broad proteomics study, where we looked at changes in other biomarkers.
NT-proBNP, which is a marker of stress on cardiac tissue, was decreased with KER-012. So, and it was observed with sotatercept and with other molecules, including sotatercept in the PAH trial. So there's a direct readout, and it was driven by exactly that question. Well, what are the markers, okay, that gives you confidence that this drug is going to have benefit in PAH?
The phase II trial is now up and running. I'm wondering if you could just sort of give an overall update on how things are progressing, site activation, investigator enthusiasm, and sort of where enrollment is maybe tracking.
Yeah. I think we will provide an update in 2024 as to when the top-line data could be. We've sort of generally talked about the historical examples of PULSAR, which was at 21 months from regulatory approval, and in Gossamer's case, it was 24 months. However, we are seeing a great deal of enthusiasm from the proof of mechanism by with sotatercept. And we're seeing as we're getting ready to open up sites for that site initiation visit, months in advance, the physicians are telling us that they've got particular patients in mind for the trial. I think that's pretty exciting.
Yeah. Maybe connecting back to sotatercept, which you mentioned. How does the potential approval there play a role within the phase II trial? Is there, like, a window right now for you guys to aggressively enroll in, into U.S. sites? How does that all factor in?
I think, you know, we've got U.S. sites, but we've got 60 sites globally. So we think, okay, right, that the enthusiasm from sotatercept on the proof of mechanism is going to drive enrollment both here in the U.S. as well as in rest of the world. But I think the safety signals that have been seen with sotatercept are causing physicians to pause and try to think about what it means. And they're looking for a drug that potentially could have a profile of fewer safety signals. And with the better BMP-sparing properties of KER-012 , that is potential. So I don't think, okay, right, that tailwind that we're seeing is going to turn into a headwind.
Last one on KER-012, the heart failure trial. I would assume that opportunity to get some data there ahead of the phase II PAH study. You know, maybe speak to how you're looking at KER-012 in that setting, some of the data you could generate, and if it... any de-risking that could provide as we look towards PAH later out.
Yeah. I think in that trial, it's a small trial, it's an open label trial, and that trial is getting going in this year, and therefore, we'll be able to have data from that in 2024. Where exactly we'll be able to present that, that's to be debated as yet. You know, we'll be, we've typically presented data at medical meetings, and we'll continue to do that. So in 2024, we will have data from that. What is-- What does that data look like? Well, in these patients, because of the stress on the heart tissue, they've got elevated NT-proBNP. We showed in post-menopausal women who have a slightly elevated NT-proBNP, that we can decrease that with treatment.
Now we're going to patients that have elevated NT-proBNP, similar to what you see in PAH patients, right? So can we reduce that? What's the magnitude of the reduction? That's what we will see. That's the first sort of data set that will come out. But as we did with KER-012 in the healthy volunteer study, we're going to take a broad profile look at other biomarkers. That's going to take time because they're not routinely conducted in the clinical setting, right? NT-proBNP, it's done all the time, so that's going to be the first set of biomarker that will come out.
Perfect. With that, we're out of time. I want to thank Jas for your time and thoughts. Looking forward to the data in a couple of weeks. Thanks, everybody, for joining us, and enjoy the rest of your day here. Thanks so much.