Keros Therapeutics, Inc. (KROS) Earnings Call Transcript & Summary

All right. Good afternoon, everyone. Thanks for joining us here at the SVB Leerink Global Healthcare Conference. My name is Tom Smith. I'm one of the Senior Biotech Analysts here at SVB Leerink, and happy to welcome our next company to the virtual stage, Keros Therapeutics, and their CEO, Jas Seehra. Jas, thanks so much for joining us. Appreciate it.

Thanks, Tom, okay, for that introduction, and thank you for giving us the opportunity to present the exciting story of Keros.

Cheers. Cheers. So just before we jump into it, just a quick reminder for the audience here. If you have any questions for Keros, you can either ask them via the webcast portal or you can sell them directly to me via e-mail. And with that out of the way, Jas, I think you have a couple of slides you're going to use to give us an overview of the company, and then we'll jump right into Q&A.

Okay. Thank you, Tom. So before I go into the slides, okay, but -- the usual disclaimer about forward-looking statements. And what I'm going to tell you, okay, that Keros is harnessing the powerful biology of TGF-Beta superfamily. We are a clinical stage company, okay, that targets the TGF-Beta superfamily to do novel therapeutics. And what's really exciting about this biology is that -- and this has been validated with marketed products, infuse a bone morphogenic protein 2 for spinal infusion Reblozyl, okay, modified active and receptive for treatment of anemia in beta-thalassemia as well as in myelodysplastic syndrome. And really, what we are able to do at Keros is that we're able to leverage our extensive experience in the TGF-Beta superfamily, knowing the structures of these protein, their function and couple that with protein engineering to generate a clinical pipeline of differentiated products. So you see, okay, right, the first 3 product candidates on this slide. 050 is a modified activin receptor IIA ligand trap and is designed to address ineffective hematopoiesis by modulating this pathway. And based upon our Phase I data, we believe that it has the potential to correct multiple cytopenias in patients with ineffective hematopoiesis, that includes MDS and myelofibrosis. Our second product is a small molecule kinase inhibitor targeting another member of the TGF-beta superfamily, activin receptor-like kinase-2. And this is being developed for treatment of anemias where iron imbalance is the underlying cause. And that includes rare genetic diseases, such as iron-refractory iron deficiency anemia to the broad indication of iron deficiency anemia, which includes chronic inflammation. And then there is the opportunity for the same molecule in a very rare genetic disease called fibrodysplasia ossificans, where following injury, skeletal, muscle and tendons turn to bone. And we'll be starting Phase II studies with this program in 2021. And then lastly, okay, right, a preclinical asset that will be entering the clinical data. Another activin receptor ligand trap that is being developed for treatment of disorders associated with bone loss and pulmonary arterial hypertension. And I think what you see here, okay, is the unique position that Keros is in where our discovery pipeline is yielding us multiple product candidates, and we'll continue to build the pipeline in bone, muscle and pulmonary diseases. So this is a pipeline that is shown on our chart, okay, right? And you can see that 050 has completed its Phase I studies. It's in its our first Phase II study in MDS patients, and we'll report the initial data in mid-2021. And we're also starting the Phase II study in myelofibrosis this year. And then 047, which completed its Phase I towards the end of last year, will be initiating 2 Phase II trials, one in IDA and the other in IRIDA in 2021. And then lastly, 012 will enter the clinic in the second half of 2021. So you see a nice progression of assets, okay, advancing through the pipeline and new ones coming in to fill that pipeline. And the anticipated milestones as shown on this slide, where we will be announcing the initial data from our Phase II trial in MDS with 050 in middle of the year. And then starting that myelofibrosis study in 2021, the 2 Phase II trials with 047 in 2021, and we'll be presenting data from the 012 in PAH at a major conference in '21 and starting the clinical trial in the second half of '21. So with that, I want to thank you, okay, for the opportunity, and I'll leave it to you, Tom.

Great. Thanks, Jas, really nice overview. Just -- and you alluded to the mechanism. I want to start with the mechanism for KER-050, and you talked about it a little bit, but maybe if you could just go into a little bit more detail around differentiation between 050, and say, luspatercept and sotatercept, the other ligand traps that most investors are familiar with?

Yes. So I think all 3 molecules are derived from the activin receptors, the difference is that luspatercept the approved product is derived from activin receptor IIB with a single mutation in that, that actually renders it incapable of binding to one ligand activin A. Sotatercept binds to multiple ligands, activin A, activin B, GDF8, GDF11, and those have all been published either in payer review journals or alternatively in the patents from Acceleron. So the different between sotatercept and luspatercept is that luspatercept does not bind activin A. KER-050 binds all the same ligands that sotatercept does, activin A, activin B, GDF8 and GDF11. And therefore, it has some of the same biology that has been observed with sotatercept. Sotatercept, based upon both preclinical and clinical studies that have been published, increases bone mineral density as well as increases red blood cells. Luspatercept, according to all of the published work, only increases red blood cells, does not have effects on the bone axis. 050 actually increases red blood cells, increases bone mineral density and reduces fibrosis. And you see -- begin to see some of the impact, okay, of that, that it is indeed, okay, cause increasing red blood cells in healthy volunteer studies, but also increasing platelets. And that's because there is an interplay talk, okay, right, that's ongoing in the bone marrow between red blood cells and platelet precursors [indiscernible] . So 050 is the first molecule that has demonstrated increases in platelets in healthy volunteers. And we think this really could be a meaningful differentiation for 050, whereby in MDS patients in the low-risk to intermediate risk for developing to AMR. 90% of those patients are anemic. But 30% to 40% of those patients have low platelets and low neutrophil. So the thrombocytopenic or neutropenic. And therefore, an agent that works on multiple images has the potential okay, right, to correct the cytopenias that are in those patients. And some of the first-line treatments in MDS are, in fact, the hypomethylating agents, such as azacitidine, okay, right? They actually cause the relief in anemia, but they cause thrombocytopenia. Therefore, there is a significant unmet need for -- in molecule that actually does treat the thrombocytopenia and patients. So we think that's a real advantage. And then sort of thinking about it in myelofibrosis, in myelofibrosis is the primary defect in the JAK/STAT pathway results in actually overproduction of megakaryocyte precursors that fail to develop to platelets. And megakaryocyte precursors breakdown cause information in the bone marrow and set up that vicious cycle where you don't get red blood cells, you don't get platelets, and then eventually, you end up with extramedullary hematopoiesis. Hematopoiesis in the spleen, which then causes all of the other symptoms, right? And JAK inhibitors really only treat okay, right, the symptoms, okay, right? They're not disease modifiers. So if you could actually allow those megakaryocyte precursors to mature all the way to platelets you actually reduced the inflammatory signals in the bone marrow. Therefore, there's a potential you could reduce the fibrosis, and therefore, okay, right, start having a disease-modifying growth. That, to me, is very, very exciting. And I think that's how 050 gets differentiated from other molecules, okay, right, that are of similar lineage.

Okay. Yes, that was -- appreciate that the distinction there. And I guess, if we could -- I'd like to go through the Phase I data in a little bit of detail. But if we could maybe focus on the erythropoiesis pathway? And one of the potential benefits here is you're acting earlier in the pathway. You think about potential application of that, perhaps you see a broader impact or an impact on anemia and a broader segment of the population. What clinical data Jas or what preclinical data do you have that gives you confidence that 050 is actually acting on the earlier parts of the erythropoiesis pathway? And then similarly, how is that differentiated from either sotatercept or luspatercept?

Yes. I'd think okay, right, it's sporadic or small pieces of data that we will give you that confidence, okay? The first one, okay, right, is actually the preclinical data, where, okay, right, mechanistically, you see what's going on. And what you see, okay, right, is the treatment from a single dose of 050 results in rapid increases in red blood cells. And remember that in the journey of a common myeloid cell, okay, all the way to a red blood cell, in humans, that takes 21 to 28 days, whereas in rodents, it takes about 7 to 10 days. Therefore, okay, if you see increasing very, very quickly, looking, right, that means, okay, that you're actually accelerating the progression of precursor cells that were already almost completing their journey. So preclinically, we see those increases as early as 12 hours. We've only looked at 12 hours. We haven't looked any earlier, but we see those increases. That means, okay, the blood cells almost completed the journey. But then we continue to see that increase to day 14 that we published in our ASH presentation. And when you then look at the clinical data, you see the same thing, right, that you see that there are increases on day 2 in hemoglobin levels, red blood cells and reticulocyte. And those increases then continue. Now the Cmax of the drug in our Phase I study is achieved on day four. So on day 4, you've got the maximum concentration of the drug. And then by day 15, where the drug levels have gone significantly lower because of the half-life of the drug being 12 days, right, you still continue to see increases. In fact, those increases are occurring through day 29. And if you look carefully at the data that I've done, you see, okay, that some of the participants in that study had increases through day 42. That means there are cells, okay, that are coming out into the bloodstream, that started their journey, okay, right, 21, 28 days earlier, okay, right? Because by that point, there's no drug on board, okay, right? So that really does fit, okay, the preclinical data that you're acting at old stages of erythropoiesis. Furthermore, we've seen, okay, right, that with 050, we get increases in erythropoiesis in circulation in rodents. And that, again, means that you're helping the [indiscernible] cells at the earliest stages, which is being committed to go down that urethra in a minute. So I think that -- all of that data tells me that there is potential, okay, right, of this molecule to be working at multiple stages. And in patients with ineffective erythropoiesis, their heterogeneous population, they don't have interruption of these pathways at one spot, okay, right? Otherwise, they would all be very, very homogeneous, and therefore, you were to get a treatment okay, right, and where everybody is responding. And the -- and as a consequence, okay, I think of it as well, you have the potential where Reblozyl, okay, showed really nice data, okay, right, and with respect to increases in a small subset of the MDS patients, these RS patients and they got a response of roughly 40%. Why aren't the other patients who respond, they must have other okay, right, perturbations in the pathway that are preventing, okay, the maturation of red blood cells. So I think, okay, given, okay -- what we see is given, okay, the broad mechanism, there is a potential, okay, that we can treat RS, non-RS and perhaps even have a different response rate.

Okay. Yes. No, I appreciate those comments. I guess shifting to the Phase I results and maybe the Phase II study design here where you're going to have initial dose escalation data in MDS patients in kind of mid-2021. If you could talk a little bit about how you're thinking about dosing and then a little bit about the design of this dose-ranging study and some expectations ahead of the initial data cut? How many patients -- potentially how many dose levels? Obviously, you're still running the study and collecting the data. But what's your sense for what investors should expect ahead of this data readout?

Yes. So we haven't actually shared what our starting dose is, okay, right? But I think if you go back, okay, to our Phase I study, where the lowest dose in the SAD that we detected drug in circulation was the 0.5 mgs per kg. And then the lowest drug that we tested in the MAD was 0.7 in that gives you sort of the range, okay, right, for the starting dose, okay, right? We know that those doses, we got some increases, okay, right, in reticulocytes and increases in red blood cells. Therefore, we were getting target engagement. So think of those sort of 2 numbers as sort of the starting dose. And then after that, okay, right, dose escalation is very simple. It's the usual, okay, the Fibonacci design of a study. So it's the first one, okay, number is 1, 2, 3, 5. So you're dose escalating, okay, right, with the next dose, after the first one would be double, okay, right? And then the next one is triple and so on. So we have 4 cohorts planned in the dose escalation. It's an open-label study, where there are 6 participants in each dose level, 3 RS, 3 non-RS and based upon the safety data, you'll escalate to the next dose. And in the RS -- in each cohort, the patient will receive 4 doses at 28 days interval over a 3-month period, and then be followed for another 3 months. So that's what we -- the study is designed. And what we're looking for in part 1 of that study, okay, right, which is dose escalation, we're looking for signals of activity. Does this drug, okay, show changes in red blood cells count? Does this drug show changes in platelets, okay, right? So that's the lowest level, okay? And then when you start thinking about beyond that, how durable is that effect, okay, right? Is it a few weeks? Is it a month? Is it 2 months, okay, right? Because then that starts giving you information about how you're going to treat the patients, okay? Is it a monthly dosing? Is it less frequent than that? We believe, based upon our Phase I data that we can dose on a monthly schedule. [indiscernible] dose okay on a less frequent mechanism, based upon the -- we don't know, we have to find that out in the patient. Then what are the doses at which we see -- begin to see responses in RS versus non-RS? Are they different, okay, right? And then these are patients. They are okay, right, or already have the disease. Some of them would not have seen any -- would not have received any transfusion, but others would have had received transfusion. What about those that have received transfusions? Are you seeing, okay, the same signals of activity on the red blue on the platelet axis, right? And if you are seeing those okay, right, are some of them having to skip, okay, right, transfusion because they don't need. So you're going to get a lot of data. But it's going to be a little bit of data, okay, right, because you're going to get some patients responding others, okay, right? And that gives you clues about how you want to start thinking about your part 2, which is the dose confirmation of the study, where you have a larger number of patients. So you start to get experience with -- in terms of what your response rates are? And also, okay, right, you're treating now in your open-label extension, and you're getting your experience as to what the durability of the effects are.

Right. Right. Okay. Yes. No, that's really helpful. Just in terms of patient numbers, Jas, and understand that you're still enrolling the study and you don't have perfect visibility into that, but it is an open-label study. Like what's a reasonable expectation for number of patients at this initial data cut?

Tom, we haven't shared that, okay, right? But what I would tell you, okay, right, is that we started the study in the second half of last year. And as I just shared, okay, the actual treatment period of this study and follow-up is 6 months. But prior to being dosed, okay, the patients have to be followed for 2 months so that you get the baseline characteristics for the hematology [indiscernible]. So the patients are in, okay, the study, essentially 8 months, okay, right? So you can imagine, here we are in February that we are moving, okay, right, on this. I can't tell you, okay, right, how many patients we'll have, but I think what I can tell you is we're progressing along that study, okay? It is an open-label study, so we'll share the first of the data in midyear. But we'll continue to give additional updates, okay, right, as we get more data.

Okay, okay. I appreciate that. Maybe we can talk -- you mentioned also initiating the study for 050 in myelofibrosis. Maybe if you can talk to a sense of timing for kicking that off, trial design and then what you're looking for in terms of initial signals within the MS population, how that compares to MDS?

Yes. So I think, first of all, okay, right, we'll be starting that study this year, okay, right? We haven't given any more guidance than that nor have we shared the design of the study, okay, right? But I think previous okay programs have actually hinted at what that study is going to look like. It's going to be study, okay, as a monotherapy in myelofibrosis patients as well as those that are on JAK inhibitors, okay? And again, here, okay, the ability to model it -- platelets, okay, could be a real, real advantage for us. So in those patients okay, right, it's going to have to be a longer treatment, okay, right? Probably 6 months and longer, okay, right? And then the one -- what you're looking for is, are you correcting the anemia? Are you actually correcting the thrombocytopenia in those patients, okay, right? And then as you progress, okay, right, what are the impacts on the bone marrow, okay, right, there? And one of the things, okay, that happens in these patients, okay, right, is that as that [indiscernible] bone marrow is expanding, the inflammatory signals are actually causing bone resorption, okay, right? So you get osteoporosis. Do you, okay, right now begin to see, okay, a reduction in that. You can follow that through bone biomarkers and so on, okay, right? So we'll be looking at all of that. And I think that's where, okay, right, returning that bone marrow to a more normal state has the potential located to have a disease-modifying properties to the molecule.

Okay. Yes. I mean, certainly, a lot of -- I think a lot of potential as a pancytopenia agent, particularly within MF. And so we'll be interested to see what the initial signals look like. Maybe I want to be conscious of time and make sure that we get through -- you guys have a number of programs. KER-047, your ALK2 inhibitor. Maybe if you can talk a little bit about the plans here in IDA and IRIDA. And I guess, also putting it in context with the Phase I data that you announced late last year. Do you feel like you need further dose-ranging here? Or just kind of walk us through what the -- how you're thinking about positioning 047 within these initial indications?

Yes. So I think they'll -- look, the Phase I data was really exciting data because for me, seeing the biology once again translate from rodents to humans is really exciting. And our Phase I data really showed, okay, right, that you can inhibit out to, as a consequence of inhibition, you get reductions in hepcidin, the master regulator of our iron. And immediately, you begin to see changes in seroma, increases in seroma. So you're mobilizing that. But then what was more -- even more satisfying is that, that iron is available for incorporation into red blood cells, where we saw increases in hemoglobin content of reticulocyte, the newly minted red blood cells that are found in circulation. So that was all very exciting, okay, right? So in our multiple ascending dose, our lowest dose was 50 mg dose. And even at the 50 mg dose, we saw decreases in hepcidin from that treatment. And these are normal healthy individuals. They already have low hepcidin, okay? And yet, we saw decreases in that hepcidin. So as we think about going into the IRIDA, into the iron deficiency anemia, some of these patients are going to have much higher levels of hepcidin. So when you have high hepcidin, okay, right, do you actually get a bigger response? Do you actually need lower doses of the drug? So in some ways, okay, right, knowing, okay, that we see changes at 50 tells you, you've got target engagement and you're initiating all of the downstream biology. So I think, okay, right, exploring lower doses in patients that have high hepcidin is warranted for those reasons because you may end up okay, right, seeing similar or bigger effects in patients even at lower doses. And as you well know, the lower your dose, okay, right, the better, okay, right, in the long run, in terms of safety observation. So I think it is warranted that we explore a lower dose, okay, right, in the patient. How many doses, okay, right, I'm not sure, okay, that we need to explore too many doses there. So it is likely, okay, that in the IRIDA patients that will do an open-label study, okay, right? Similar sort of in concept to the MDS study, where you look at dose escalation, so that you can see changes in these red blood cells founded in a patient population that have high hepcidin. And that then actually sort of feeds into what we do in the IDA as well.

Okay. Okay. That makes sense. Maybe if we could switch gears just quickly on KER-012. It's a compound. I've been getting increasing number in down questions around, and we've seen some of the success with sotatercept in PAH. And I think there's obviously increasing enthusiasm around that approach. Maybe if you could just kind of walk us through how 012 is differentiated versus sotatercept? And you've talked about, I guess, filing an IND for 012 later this year, maybe how you're thinking about the initial clinical approach here?

Yes. So look, 012 is based upon the activin receptor IIB where sotatercept is based on the activin receptor IIA. And that doesn't mean too much, okay, right? It's just they have different origins okay, right, in that. But I think where, okay, right, there is a difference is that when we were trying to come up with another molecule, we were looking for a molecule that would increase bone mineral density without increasing red blood cells. Sotatercept has demonstrated, both in preclinically as well as clinically that increases bone mineral density but also increases red blood cells. And in fact, okay, right, the clinical development of that program was in Phase I study, okay, right? It was looking at bone as well as red blood cells. And the red blood cells effects was large enough okay, right, that it became an anemia drug. More recently, it's pivoted to PAH because the biology suggests that could have a benefit. But I think -- and that's where the excitement is because Acceleron did a really nice job of demonstrating, okay, the role of TGF-Beta signaling in PAH. But there is limitations to the sotatercept. And in their Phase II study, they showed, okay, really nice response in terms of the 6-minute walk test, the PVR, which is catalyzing their progression to a Phase III study, but they also saw increases in hemoglobin content, okay, right, and red blood cells. And that increase at the higher dose, the highest dose they used 0.7 was 1.5 grams per deciliter, a mean change of 1.5 grams per deciliter. And their lead okay, right, clinician investigator, did say that she wants to use the higher dose. But I've asked the question, what about patients okay, right, that may need a higher dose? Could you dose them, okay, right? Higher, not with sotatercept. And that I think is where 012 has the advantage, okay, that it does not have that effect on red blood cell access. And we were looking for a molecule from day 1 that would have the effects of sotatercept on bone axis without the red blood cells. And we published that data at ASBMR in September with respect to the bone. And then in the updated S-1, we shared, okay, that we're seeing the activity in the PAH model. So this makes me very excited. And I think given okay that we've got multiple indications to eventually go into, it, again, warrants, okay, us going into a healthy volunteer study, right? Establishing what are the safe doses, okay, and the efficacious dose. The thing here is by having that effect on bone axis, we can measure that. We can measure that through bone biomarkers, right, in circulation. We can measure that part through imaging techniques and so on. So you can actually get a feel for what the drug is doing pharmacologically as well as determining the safety, okay, right? And then that allows you to pivot in all the indications that you want to go in PAH, okay, right, bone indications, such as osteogenesis imperfecta.

Got it. Got it. And so yes, we'll be looking for those initial early signals, it sounds like probably at some point in 2022. All right, Jas. Well, look, we're -- unfortunately, we're up against the time here. But thank you so much for spending the time sharing your insights. And we'll stay tuned here. Certainly, a lot of interest ahead of the initial MDS data coming up in mid-2021. Thank you, everybody, for joining us, and I hope everyone has a great rest of the day.

Thank you, Tom, for the opportunity. Appreciate it.

Bye.