Incyte Corporation (INCY) Earnings Call Transcript & Summary

Welcome to everybody online, in person here in the room. And a big thank you to all of you and specifically to our presenters today for the time and the energy you are putting behind this project. So the data -- we'll be speaking about 989. The data that is presented today is the first clinical data presented with this product. And in fact, I was thinking of the ASH preclinical presentation that was a plenary session like 2 years ago, 2.5 years ago. And what you could wish at the time you will see in the first 50 patients that are treated with ET and what it would look like. And frankly, that's what makes this data so exciting is that it looks very much like what we have seen a few minutes ago. So it will be -- this entire presentation will be about that subject of what the data means and where we are going to go with this project. I just want to say a word before we go into the science about why this -- today is an important day for Incyte as a corporation. We all know we have been facing questions about the long-term growth of our Jakafi franchise beyond 2029, and that's an appropriate question to have in mind. And as you know, Jakafi franchise is made of GVHD and MPN. In GVHD, we have the launch of Niktimvo that is taking place now as we speak. And we think it will be the growth driver for that part of the franchise for the long term. And frankly, 989, what we have been discussing this morning at the late-breaking abstract session is what's going to drive the growth of our MPN franchise beyond the sort of the Jakafi franchise itself. So it is an important day for patients who are suffering from MPN. It's an important day because it's a new approach for their treatment, but it's also an important day for the company because it's a result of a research program that we have executed over the past years in many ways. So we will be going through a number of presentations, and I will let Pablo introduce the scientific program for the session. Thank you very much.

Thank you, Herve, and good morning, everyone. Good afternoon here in Milan. We have a really busy agenda for you today. So we're going to get right to it. We're going to start with an overview of essential thrombocythemia, which is the disease we're going to focus on today with the data from 989. Claire Harrison is going to give us an overview. Then Dr. Nangalia is going to talk about the biology of the CALR mutation and why this is so important, and it has really -- when it was described first a little over 10 years ago, it started to change the way we try to develop new medicines for this set of diseases. Patrick Mayes, our own Head of Discovery, Biology at Incyte will talk about the mechanism of action of 989, critical to understand some of the data that you're going to see. And Dr. John Mascarenhas will talk about the data from 989 in the Phase I study that he presented earlier today, the late-break sessions. Hopefully, we'll have time for Q&A after that. Let me spend a couple of minutes here on the introduction to our speakers. Dr. Claire Harrison is a Professor of myeloproliferative neoplasms and Deputy Chief Medical Officer at Guy's and St Thomas’ Hospital in London. She co-founded MPN Voice about 20 years ago and is a trustee of Blood Cancer UK. She sits on several European hematology EHA boards, very relevant to the meeting today and is Deputy Editor in Chief of HemaSphere. The next speaker, external speaker would be Dr. Jyoti Nangalia, who discovered the CALR mutation and its importance in MPNs and published a paper in New England Journal in 2013. She has continued to make landmark scientific contributions across this set of diseases and provides finalized predictions outcomes in the MPN -- for MPNs. Her team identified other genetic drivers of myeloproliferative neoplasms, and she's also a group leader at the Sanger Institute and Cambridge Stem Cell Institute. And Dr. John Mascarenhas, Professor of Medicine at the Icahn School of Medicine at Mount Sinai, Director of the Center of Excellence for Blood Cancers and Myeloid Disorders, and a member of the Tisch Cancer Institute, where he directs the adult leukemia program, and he leads clinical investigations with the MPN disorders program. He's been a principal investigator and presenter of a lot of recent clinical investigations in MPNs and other related areas. So with that, I'm going to hand it over to Dr. Claire Harrison. She's going to tell us about ET, a disease overview and some of the important points to remember as we start thinking about the potential role that 989 can play in these patients going forward. Claire?

Hello, everyone. It's my great pleasure to be here today to introduce you to this disease area and to give you some insights as a clinician who's looked after patients with these diseases for more than 25 years. As many of you will be aware, the MPN family consists of 2 more perhaps indolent diseases, although as I think I'll show you, they're not so indolent, ET, polycythemia vera and then myelofibrosis. Focusing in specifically on ET. This is a relatively rare disease. The incidence is about 1 in 100,000, but the prevalence is at least 30 per 100,000. The disease has a bimodal distribution. The median age of presentation is -- for patients in their 60s, but many of the clinicians here today would be able to describe to you that we also have a peak in younger individuals, particularly young females. And I'm going to tell you the story of one of my patients later. Main clinical presentation is thrombosis and hemorrhage, but the risk for the patients and the thing that they're most worried about is disease progression. And this is a time-related event and disease progression is principally to myelofibrosis, which we do have some treatments for, but none of them are particularly curative, as you all know, although there's a work in progress. And acute leukemia is also a risk for these patients. Zooming in, in particular, on mutant CALR positive ET. This represents around 30% or so of our ET patient population. What do they look like? So I'm summarizing that for you on this slide. They tend to be younger. They have a higher risk of MF transformation. That's what this graph is showing. This is myelofibrosis-free survival. These are the CALR type 1, and these are the CALR type 2. My colleagues will explain what a type 1 and a type 2 mutation is. But just focusing in on, these are the patients in my ET population who are most likely to develop myelofibrosis, and they do so quite quickly. In this disease, these are pertinent points for you when you see the results later, a higher VAF tends to be the case for CALR patients compared to JAK2-positive ET, which is around 50% of the population. And specifically in the CALR population, a higher VAF is known to be associated with a higher risk of disease progression, thrombosis and poorer outcomes. These patients also tend to have more indication for starting treatment. So they tend to have a higher platelet count. That tends to be our reason for initiating treatment for these patients. And they also, as I'll show you in a subsequent slide, in some data generated by us in a study that we started in 1997, tend to be less likely to respond to treatments. If I look at a patient with ET in my clinic, or John does or Jyoti does or any clinician treating these diseases, these boxes here, these are the things we're trying to focus on. We're trying to minimize the impact of the disease, control the risk of thrombosis and hemorrhage and do so in a safe way, allow them to become pregnant if that's something they want to do. But we don't have, as you'll see, anything at present that we know for these patients can prevent disease progression, can restore normal hematopoiesis and by any traction-able means, reduce the clone size, which is something that we would perhaps be defining as disease modification. So just to show you what the current treatment landscape looks like for a high-risk patient, generally, that would be defined as either someone who's over 60, someone who's had a thrombosis or someone who has a platelet count of over 1,000 to 1,500, which is where the indication for treatment for CALR comes in. These patients, we give these very old-fashioned drugs to hydroxyurea or interferon, or in the second-line setting, a drug called anagrelide, and you will have seen some data at the EHA meeting comparing these two therapies. These treatments are used by us in day-to-day practice with these three aims: Control the blood count, reduce the risk of thrombosis and reduce some symptoms. There's no evidence at present that they modify disease. Interferon, we have been using for decades in this disease setting. We never say to a patient that's going to modify their disease. I told you that CALR-positive patients tend to respond less well to standard therapies. This is data that we published before Jyoti described the CALR mutation. This is in the PT-1 study, which we started in 1997 in the U.K. And this is response rates to patients with hydroxyurea or hydroxycarbamide, platelet counts, which tends to be our primary target. And here, you can see the JAK2 negative patients running a higher platelet count than the JAK2 positive ones. So we needed more doses of hydroxyurea, and we noted more cases of resistance or intolerance to disease. I can't show you the specific data for CALR patients in this setting, but I've summarized some of the available literature. So there is still unmet need for ET patients. We haven't had a new therapy for these patients since the approval of anagrelide in 2005 in Europe. Resistance and intolerance to available treatments, I'm showing you the data here. These treatments do not have an impact on long-term disease outcomes. And indeed, there is evidence that hydroxycarbamide or hydroxyurea may well increase the risk of leukemia, and anagrelide is certainly associated with higher risk of myelofibrosis. We showed that in the PT-1 study. Anagrelide interestingly, is also associated with VAF increases for CALR. So I think this is a fair summary for our current treatment landscape. No current treatment offers a cure, evidence of disease modification or consistent change in key pathological features, VAF or resolution of marrow changes. So now I want to tell you a story of a patient who I first met in 1997. And I think this story illustrates the statement I made on the last slide that available treatments don't really modify disease course. She was 17 when I first met her. This is her blood count. We did a biopsy. It was consistent with ET. We did some molecular workup. But in fact, in 1997, we had nothing to do. During the course of her disease, she developed a knee injury actually while skiing in the states. She was put on anagrelide by colleagues in the Mayo Clinic to manage surgery because the platelet count was high. And at that time, anagrelide was believed to be safe and was frequently used for younger patients. But subsequently, the PT-1 study and other studies showed that it had these risks associated with it. Later, she plans to get pregnant. At that time, her platelet count was 1,600 plus. We treated her with interferon alpha to reduce the risk of adverse pregnancy outcome, but she had a very poor experience with that drug, had a miscarriage and because of her poor experience, has subsequently consistently refused to have interferon. Later, we found that she was indeed CALR positive. And around 20 years after diagnosis, she developed myelofibrosis. That's actually quite a long time. I saw a patient just last week who -- she's 30, she's got 2 children and she transformed to post-ETMF with CALR-positive disease within 5 years. So it's very variable. So does reducing VAF matter for these patients. The only data we really have for CALR at the moment is this increase in VAF with anagrelide, which we know is associated with fibrosis. In an academic study, we took actually ironically another Incyte asset, ruxolitinib. And in the U.K., we performed an academic study. Jyoti was very heavily involved in this. We recruited 180 patients, and we treated them with ruxolitinib. And in this study, we showed that molecular response, which was just a 50% reduction in VAF. But by the way, the results are the same if the threshold was 75% reduction or 25% reduction. By 1 year, event-free survival and event-free survival is thrombosis, hemorrhage, transformation or death. I think we'll all agree those are relevant outcomes. And later, Jyoti and team in this publication, which we published in JCO, we were able to show for the first time that an endpoint in MPN correlated with overall survival advantage. So you can ask Jyoti the question about difference biologically between JAK2-positive disease and CALR-positive disease, but I think there's a good rationale for why a reduction in VAF, that might be viewed as relatively modest, may have a disease implication. So in conclusion, and I'll speed up a little bit, I've shown you the distinct phenotypic profile for these patients, the limitations of our current therapies, unmet need for these patients and that there is emerging evidence that reducing the clone size even to a modest extent can correlate with good outcomes for these patients. And I believe that's because clones -- these clones are more indolent. They tend to increase more slowly. Anyway, thank you very much for your attention. And at this point, I think, Jyoti, I'm handing over to you.

Afternoon, everyone. Thank you, Claire, for that talk and also to Incyte for having me here today. I'm going to talk a bit about the biology of a CALR mutation and its relevance in MPNs to set the scene for the clinical data to follow. So I still remember the day in July 2013 when first stumbled across the CALR mutation. Initially, it was confusion and intrigue and about 8 hours later, after growing excitement, I e-mailed my PhD. supervisors at the time at about 3 in the morning. It was new. The JAK2 mutation had been discovered in 2005. And really since then, we did not have a molecular basis for at least half of patients with ET and MF to try and understand what was driving their disease. And it was -- these stories were published in the New England Journal of Medicine in December 2013, and it was really exciting for me in 2022 when Edimara and the Incyte team stood up and presented the preclinical data on the first mutant CALR targeting antibody and as published in Blood soon after. So what is the CALR mutation and what does it do? Along the top here is the normal protein sequence of CALR. So if you take the CALR protein and stretch it out into a string, you start with one end here called the N-terminus and you go down to here called the C-terminus, the two different ends of the protein. And just note here that the last four amino acids of the protein are KDEL. KDEL is a very important part of normal CALR. And what does CALR do normally? So we have DNA and our DNA encodes for genes. And if you want to switch on a gene in a particular cell, you make a copy of that DNA into RNA and that RNA forms a template for making a protein. Once you've made your protein, your protein is just a long string, and you need to fold it into a particular complex shape for it to do its function. And CALR is involved in helping that long string of proteins fold into whatever shape they need to be in a place in your cell called the endoplasmic reticulum. And KDEL anchors CALR to the endoplasmic reticulum. It keeps it there so that as proteins traffic through, it helps it fold and then those proteins go on their merry way to do whatever job they do in the cell. And if you look at KDEL across humans, mice, zebrafish, [indiscernible] KDEL is conserved. This is the sequence of CALR across species. It's preserved within the animal kingdom. And what was really exciting is when you look across all the mutations of CALR at the time, this was the most common. This was the second most common, and there are a whole bunch of other things. What was really quite remarkable is that when you -- I did this manually, working out what they did, the KDEL was lost in all of them, and there was this novel C-terminus sequence to the protein. And here is the structure of the protein now folded. And you can see that there's a novel -- there's a new tail. So CALR has got itself a new tail. Now why is that important? So as I already mentioned, proteins go through the endoplasmic reticulum here, the ER, which is the sort of web around your nucleus. And CALR is normally there, and it helps these proteins fold. But what it's doing weirdly is that mutant CALR specifically binds to the receptor for thrombopoietin. Thrombopoietin is a hormone that signals to our blood stem cells to make more platelets and to make megakaryocytes. And it itself as a receptor needs to go through the ER to get folded into its right configuration before it goes into the cell membrane ready to receive signals from TPO. Mutant CALR binds the receptor for TPO in the ER. And because it's no longer docked to the ER because it's lost KDEL, it goes onto the cell surface with the receptor for TPO and not only brings it to the cell surface but activates it because it's continuously locked. So mutant CALR with TPOR in combination become this force of nature that then leads to this continuous signal in the cell to make platelets. And that's what drives ET. And when that disease evolves genetically to acquire more mutations, that's what then drives myelofibrosis. So the key thing here is wild-type CALR is involved in normal cellular homeostasis, but mutant CALR has this particular predilection for the receptor of TPO that then single-handedly drives this disease. Even though this mutation, this protein is found in all of our cells, it drives this disease in hematopoietic stem cells. And then that link to TPO becomes even more important when you're measuring its activity in your cells. Remember, every one of our stem cells that is part of the ET clone has a mutation in CALR within its DNA. But whether that mutation is active in doing anything is entirely dependent on whether that cell is a cell that normally makes the receptor for TPO. So this is a map here where we've basically taken all the different types of cells you have in your bone marrow, NK cells, T cells, B cells, these are lymphoid cells. CD34 is a marker for your stem cells, that rare population of cells that make all of our blood cells and the myeloid/erythroid lineage, which is so typically elevated in MPNs, and you'll see here that mutant CALR is specifically expressed in these compartments. So when we measure VAF, we are measuring it at the DNA level, which many cells have. But when you're looking at where it's active, it's only active in certain cell types. And it's only then pathogenic when those cell types also express TPO. So here are the stem cells at the top of the hematopoietic tree. These stem cells sit at the top of the pyramid and make all the different blood cells. TPOR is expressed in these stem cells as is mutant CALR. So this is fortunately the target that you want in MPNs, and it's the target that both expresses TPO and expresses CALR. And then TPO and CALR are expressed down this myeloid lineage all the way down to the megakaryocytes and the platelets. But of note, these cells here will have mutant CALR in their DNA, but it won't be pathogenic. So if you then look at these cells here, which were not highlighted in the slide before, these are the cells that are the daughter cells of your mutant stem cells. They do have mutant CALR, but they aren't really contributing to the pathogenicity of it. And as these cells disappear, if you could get rid of them, these will go down quickly because they'll see the antibody, and these will eventually disappear once their parents disappear. So these might lag. So if you look here, the early clinical indicators that one could have with a drug that targets the mutant CALR specifically in the cells that it's expressing TPOR with, you'll expect a reduction in platelets, you might expect a reduction in megakaryocytes because there are many mutated CALR megakaryocytes in the bone marrow, and a reduction in the stem cell compartment and you'd measure that as a reduction of number of cells that both express mutant CALR and TPO. And then these cells here, where they may contribute to your DNA readout of mutant CALR, they will go down more slowly as their parents get eliminated, these stem cells. So in conclusions, MPNs are cancers that arrive after the acquisition of a driver mutation within your hematopoietic stem cell and mutations in CALR that occur in those stem cells can drive ET and MF. A small fraction of cells that both express the mutant CALR and the TPO receptor will then drive the disease and be the recipients of targeted therapy. And then the cells that don't express TPOR will hopefully be removed later. For me, in particular, we're very interested in my lab about how these MPNs grow, and what the merits might be of reducing the burden of disease. And the way I see all MPNs and their growth over life is really like a compound interest on your bank account. So these are slow-growing cancers. There may be a 5% to 10% interest, and that goes on for 20, 30, 40 years, and you end up with millions to billions of cells that are responsible for your disease. If you could reduce that compound interest rate from 20% to 10%, I'm sure many people in the room could recognize the impact of that in 20 years' time. So if you could reduce not only the burden of cells, there's less cells waiting for potential evolution to myelofibrosis, but if you could also reduce the rate at which the overall clone is growing or indeed regress it, then that could push back disease by decades potentially, which is what the last point makes. So now over to John Mascarenhas -- sorry, over to Patrick Mayes, apologies to talk about the novel mutant CALR antibody.

Thank you, Dr. Nangalia. So I'm going to spend the next few moments just to speak or provide an overview of our mutant CALR monoclonal antibody, INCA033989. 989 is a first-in-class agent, which targets the mutant CALR oncogene. It is a fully human Fc-silent IgG1, which binds with equal affinity or high affinity to both type 1 and type 2 mutant forms of calreticulin,, and upon binding leads to the potent inhibition of the constitutive JAK/STAT signaling that you just heard about, downstream of mutant CALR interaction with TPOR. And the net effect here is that you have antagonism of mutant signal while having no effect on normal cytokine signaling through TPO, TPOR. So to demonstrate this selectivity, we isolated CD34 hematopoietic stem and progenitor cells from either MPN patients or human healthy donors. And you can see from the upper left-hand schematic, this is what was described mutant CALR interaction with TPOR leads to constitutive signal through the JAK/STAT pathway. This provides the oncogenic properties to the MPN cell, whereas in a normal HSPC, cytokine is required to bind the cytokine receptor in order to engage and initiate signaling downstream. So if we look then in the context of 989 treatment in the center graph, what you see is the potent and complete inhibition of the mutant CALR induced JAK/STAT signaling shown in blue, while 989 has no effect on the cytokine or TPO-induced signaling through JAK/STAT shown in green. And this is quite distinct from other therapies used to treat MPNs, where there is no selectivity between mutant-induced JAK/STAT signaling and wild-type or TPO-induced JAK/STAT signaling in an HSPC. And this provides -- or this creates an issue wherein you cannot address complete inhibition of JAK/STAT signaling in a mutant cell because of the dose-limiting toxicities associated with wild-type JAK/STAT inhibition in a normal cell. So this exquisite selectivity of 989 is directly linked to the disease-modifying potential of this agent. As you heard from Dr. Nangalia, mutations in CALR arise in stem cells, and this provides a clonal advantage of the mutant stem cell versus wild type. So over time, this provides in the bone marrow of patients, the opportunity for mutant cells to become the dominant clone in that person. However, what we know about ET is that the average mutant CALR VAF is about 30% to 40%. And what this means is that there is a meaningful reservoir of wild-type HSPCs that still exist in these patients. So if you think about the context of therapy with 989 as shown on the right, the antibody will bind to and eliminate the mutant form of the HSPCs while sparing activity against the wild type, allowing for expansion of those wild-type clones, eventually, over time, leading to the counterbalance of decrease in mutant, increase in wild-type and a decrease in mutant CALR VAF over time. So to demonstrate this directly, we isolated CD34 HSPCs from either MPN patients with a CALR mutation or human healthy donors. These cells can then be cultured in an ex vivo liquid culture [ effects ] of 989 on the survival and proliferation of these cells. And this is shown in the graph on the right, where you can see the effects of 989 resulting in dose-dependent elimination of HSPCs in a mutant CALR MPN patient sample, while 989 has no effect on HSPCs from a healthy donor in this system. So this is important because it shows direct activity against the disease initiating sustaining cells in MPN. So using a similar system, we can isolate CD34 cells and culture ex vivo in a cytokine milieu, which biases towards megakaryocytes. Here, we can show the direct effect of 989 on megakaryocytes in this system. You can see from the data on the right, in an MPN patient with a CALR mutation, we see very significant effects of 989 towards megakaryocytes in this patient's sample, whereas 989 has no effect on a JAK2-V617F megakaryocyte patient. This is important because, as you heard, ET is characterized by the overabundance of both megakaryocytes and platelets, which are derived from megakaryocytes. So this demonstrates the ability to modify the pathological endpoints of this disease. And finally, we have tested 989 in a mutant CALR transgenic mouse model. And in this model, we can knock in mutant CALR into the hematopoietic cells of these mice using an MX1-cre recombinase system. Following knock-in, the mice are allowed to age for 4 months, which allows for development of MPN disease, which, as you can see, results in extensive thrombocytosis in these mice, anemia and megakaryocyte hyperplasia in the bone marrow. We can treat with 989 for 12 weeks and upon treatment, look at disease endpoints. And what you can see from the graphs in the middle is significant effects on normalization of both platelet counts and white blood cell counts, as shown by the blue bars on the top graphs as well as reversal of anemia as shown by hematocrit and red blood cell below. If you look then in the bone marrow with the HNE plots on the far right, you can see a normalization of megakaryocyte cell numbers, both a decrease in number as well as a normalization in the morphology of those cells. So all in all, showing the disease-modifying activity of 989 in this mutant CALR preclinical model. So in summary, we've developed a high-affinity selective monoclonal antibody, which antagonizes mutant CALR function. This is selective for the constitutive JAK/STAT signaling while sparing activity through normal cytokine-induced signaling. As I showed you, this has direct inhibitory effects on the disease, initiating and sustaining cells within MPN as well as megakaryocytes, which are the precursors which derive platelets from and is how these diseases are identified. And then as I showed you, in the mutant CALR conditional knock-in model, having transformational activity of the antibody, leading to platelet normalization, reversal of anemia and elimination of megakaryocyte hyperplasia in this model. So overall, the preclinical safety and pharmacology profile of 989 makes this a very promising candidate for clinical testing in mutant-CALR positive MPN patient samples. So I'll stop here and hand the microphone to Dr. Mascarenhas.

Okay. I want to thank Incyte for inviting me to join today to review with you the results that were just presented as a late-breaking abstract with 989. And for me, it's very exciting. And what I'm going to do with this talk is personalize it a little bit more to give you a sense of my own personal -- and it's not just mine, I would say, it's shared by the investigators who are treating patients, impression of the data beyond the podium presentation that we just gave because I really do think that this has the potential to be transformative in the field. So this is the schema of the Phase I study. And you'll see that it is a study that is done both in the U.S. and ex U.S., but we're presenting the joint data. Today, we're just focusing in on the ET population, but you'll notice that there are MF patients, both monotherapy and in combination with ruxolitinib that are being treated, and that data will be presented at a later time. But today, we're really focused on the ET cohort in a drug that's given intravenously every 2 weeks in a patient population that is confirmed to have ET by WHO criteria, that's deemed high risk, in need of therapy. So these are patients who are at risk for thrombosis and progression. They're over 60 years of age. They've either had a thrombosis, they have high -- or a hemorrhage, a high platelet count maybe with Acquired Von Willebrand and they're in need of better therapy. By definition, they have to have thrombocytosis, so a platelet count greater than 450,000, but they could be on anagrelide or hydroxyurea, which is a standard of care for most patients worldwide, including the U.S., but are probably not succeeding to come on the study, meaning that they're on some dose, it's probably limited by toxicity, which is very common and/or not achieving its goal at the maximally tolerated dose. So this is really representing a group of patients that are in need. And I point that out because it was not hard to recruit to this study for that very reason. So once you open up a study, you have to figure out which patients are in need and will they come onto a first-in-human Phase I study. I've done a lot of studies. This was a very easy study to recruit to as a first in-human because I think it represented a group of patients that were in need and were looking for a novel immunotherapeutic-based treatment. The endpoint is safety and tolerability, characterizing the safety profile and of course, trying to identify dose-limiting toxicities. And as you'll see, we didn't have DLTs nor did we hit an MTD. So what will be our recommended Phase II dose will be, we'll need to determine. And then, of course, looking at secondary endpoints, response rates. So response rates here for the purpose of this talk are going to be normalization of the platelet count, the white count. And then most intriguingly, and I would say, most impressively are some of the biomarkers I'll show you that would suggest that this is not simply reducing platelet counts. We have drugs that reduce platelet counts to some extent, but this is likely modifying the disease. As Jyoti pointed out, that could have a much more greater impact and profound impact on these patients over time. These are the 49 patients. So these are the 49 patients that were recruited at the time of data cutoff. From 24 milligrams, which is the starting dose in the U.S., all the way up to 2,500 milligrams across multiple centers. This was really a global effort, I should point out, too. It was very important. This wasn't one center generating data. This was many centers, many phone calls, many toxicity reviews and a lot of confidence in what we're showing here today. This is the patient population that you would see and would expect. Median age in these diseases is usually in the 60s. Here, the median age was 60. And the time from diagnosis was 7 years. I think that's also an important point. These are chronic diseases. These patients sometimes start with therapy that don't carry them the distance and need additional therapies. And I'll show you that although on the right, hydroxyurea was the most common therapy, you will notice that the number doesn't add up to 100% because patients do end up cycling through therapies, and that becomes a problem. They go from one therapy to the next therapy in search of a therapy that will both be tolerable, control the accounts and potentially modify the disease, and hydroxyurea and anagrelide for sure, don't hold that promise. So there's clear unmet need. I think this table summarizes some of that unmet need by the types of patients we're enrolling. You will notice that 57% of patients are type 1. That's the more common type and the rest were either type 2 or other. And the median VAF was around 30% with a very high platelet count. So again, this was a patient population in need. These are not patients that have polycythemia vera, which you may be more accustomed to seeing or myelofibrosis. So the white counts weren't super high. Their spleens weren't super big. And their TSS may not be super high. But I will tell you from dealing with a lot of ET patients, they do have symptoms and there's a lot of anxiety that doesn't come across in tables like this when you live with a chronic disease. So therapies that can effectively control the disease relieve a certain degree of anxiety. And I will tell you that, that also makes for a much shorter office visit. So when you have controlled the disease and the patient is no longer anxious, they don't spend 3 hours sort of telling you about all the different complaints that they have, which are repeated. And I've noticed with giving this drug, the office visits are actually shorter because patients are enjoying the benefit and the anxiety of that platelet count comes down considerably, and that's not a trivial aspect. So here is the safety data, which was, I think, really beautiful. It was really beautiful in the sense that the treatment related -- so overall, any treatment-emergent adverse event, 85%, but treatment related, deemed by the investigator in a Phase I, a lot of this was early on when we didn't know what to expect. It was about 61%. Grade 3 was about 1/3 of the patients, less than 1/3 of patients and serious events were 3, which I've outlined here. Asymptomatic lipase increase happened very early on in a 24-milligram dose. We know that's not an effective dose. And we know that this occurred in other patients, but seem to have no clinical sequela. There were no cases of pancreatitis. There were no radiographic changes. It was transient. Often, we would check it, and Claire and I have made this comment before, we don't normally check lipase in ET patients. So I'm going to start doing it now because out of curiosity, I wonder if lipase is elevated [ transily ] for reasons I don't know. But we didn't see any correlation with the drug. In terms of we would see the lipase go up before the dose and then it would come down sometimes even after the dose was given. So it didn't seem to be clinically impactful. This one patient, I did want to make this talk somewhat personal, just to alleviate any concerns. If you see visceral venous thrombosis, a very common complication in MPN, splanchnic vein thrombosis. It could be Budd-Chiari, could be portal vein, splanchnic vein. We see this frequently. In this individual, which was treated at our center, low dose, 24 milligrams by cycle 5. She came in after like a flu-like symptom, which might have been the precipitating event and had extensive splanchnic vein thrombosis requiring hospitalization and anticoagulation. And she had a history of DVT. She also had a lot of risk factors for thrombosis, including obesity. So she had many explainable reasons why this could occur and very, very unlikely related to the drug, just to be very clear. The diverticulitis also happened in a patient who had recurrent diverticulitis, also not thought to be drug related. So sorry -- so dose reductions to treatment adverse events, one, infusion interruptions, none and no DLTs in a Phase I study, very, very clean. You see the toxicity here, I can assure you that most of these -- and this is regardless of attribution, of course, most of these toxicities are toxicities that are commonly seen in the clinic in ET patients, except for the lipase, which I pointed out, which again was transient and without any significant clinical sequela. So this was a very clean study. It was one of the few studies I participated in and I've participated in many, where I felt very confident putting patients on in a first in-human and letting them know that my concern about toxicity was very minimal and my hope for efficacy was quite high. And that's what we saw. And we saw it very obviously. So I'm highlighting here the change in platelet counts over time. This is grouped by dose at 24 to 250 milligrams and then 400 to 2,500 milligrams to make the point that in each group, we saw very dramatic reductions in platelet counts, near normalization of platelet count, even at the lower end of the dosing range within the first month of therapy. It was even more impressive in the higher range. So the rapidity in which you see it is remarkable. And then the durability of that normalization is remarkable. And we did not see thrombocytopenia, which is key because what we don't want to do is try to control counts to the detriment of inducing the alternative problem of low platelets. So any concern of that was wiped away when we saw this data. It was very satisfying to see platelet counts. I mean, I had patients literally cry in the room because they had never seen a normal platelet count on them, and it was just so anxiety relieving. And that's -- it sounds maybe unimportant, but it was actually a very big deal. So 31 patients did come on with either hydroxyurea or anagrelide. And I think really importantly, 65% of them were able to come off. Come off because they don't need the hydroxyurea or anagrelide. The 989 study drug was clearly controlling the platelet count in the absence of that concomitant cytoreductive therapy. This plot is a swimmer spot to maybe give you a little bit more granularity and help understand what we're really looking at. So this is the 24 to 250-milligram treated patients. Here is the 400 to 2,500. So the doses are listed here. And the responses down here are color coded. So this dark blue are the patients who had CR. The light blue are the PRs and then the base are the stable diseases. The triangles are the type 1s, the more frequent ones and then the circles of the type 2 and the others. So you can quickly see that even at 70 milligrams, and this was just random -- wasn't -- we didn't try to just put type 1s at this dose, but it just happened to be. You get this very quick and durable control even at 70 milligrams. So you look at this and you say, "Well okay, but maybe not as durable or as impressive in the type 2s." But then you go to this side of the screen, and you start to see that at higher doses, you're starting to see these responses with the type 2 patients as well. Particularly here, we're seeing that response as well. So I do think that there's a dose dependency. You probably need lower doses with type 1, higher doses with type 2. But I would say that across the board, patients with ET that have a mutant CALR protein enjoy some degree of response rapidly and safely with this compound. And only one patient again, discontinued treatment. All the other patients continue to receive every 2-week therapy. This was really gratifying to see. We don't often check mutant CALR VAF in our patients who aren't on trial because most of our therapies don't meaningfully change it, and it's not something that would normally be done. But in the context of a clinical trial where you're asking the question, is this truly stem cell directed? You want to see some reduction in this VAF as a surrogate marker of disease burden. And I think at a very early time point -- this is within 6 months. This is from the peripheral blood within 6 months of receiving therapy. These asterisks show the patients who only got up to 3 months, 3 cycles. So this is very short for a long-term disease where a median of 7 years before they got on the study, within months, we're seeing reductions in the CALR VAF. This is shaded by dose, so the darker the shade, the higher the dose. Most of these patients here we're at 750 milligrams. And this does include type 1 and type 2 patients. So you can get a sense that there's darker blue as you move to the right. And of course, there's also less time for the patients with darker blue that have the higher dose. So we need more time to see even further reductions in this biomarker. All 18 molecular responders also achieved a hematologic response of CR, PR, it was mostly CRs. So correlating very nicely this change in biomarker with the clinical outcome measure of platelet control. And then from the peripheral blood in 7 patients, single cell RNA sequencing was done. And by immunophenotype, this was shown in a previous slide, the HSPCs here and the myeloid/erythroid lineage derived cells down here in gold. These are the cells that are expressing mutant CALR in theory with thrombopoietin receptor. This is the targeted cell population for 989. And you can see the very significant and within a short amount of time -- here are the 4 patients that we're highlighting at the different doses and pointing out that this patient here was a type 2 patient, you can see significant reductions in the CALR VAF in the whole blood, but it's driven by reductions in this population here, the stem cell population. So it's an anti-clonal, anti-stem cell-directed therapy. And with it, you have concomitant increase in the wild-type or normal cells that are CALR wild-type. And then from the bone marrow, additional biomarker data that really, I think, is very satisfying to see with immunohistochemical staining for mutant CALR in yellow, blue are the mutant CALR negative megakaryocytes. After 6 cycles of therapy, you see a more normal profile of megakaryocytes with distancing of these cells and more prominent normal blue megakaryocytes. And again, on the right, the proportion of total MKs in the bone marrow go down, again, driven primarily by these mutant reduction in mutant cells with a concomitant increase in non-mutated megakaryocytes, what I would say is normal megakaryocytes. So bone marrow evidence of disease modification to complement molecular evidence from the peripheral blood. So in conclusion, in this 2 studies combined, 989 monotherapy, I think anyone would agree, very well tolerated, very satisfying to see that in the ET population of patients who are in need of therapy, relapsed/refractory after cytoreductive therapy. We didn't see a DLT. We didn't even hit an MTD. We only had 1 patient dropped off for reasons that I think are unrelated to the study drug. And it is quite clear that you get a rapid and sustained hematologic response with this agent and probably more so with higher doses. And then this biomarker evidence of reducing the clone is very satisfying for a physician who is looking to not just control counts for a thrombotic purpose but change the natural tendency of this disease to be progressive and maybe even transform to myelofibrosis or acute leukemia. So these findings are of significant note. And we look forward to sort of expanding the doses to continue to look at efficacy and follow safety. But I do think that this represents a very novel, a very simple, a very safe approach to treating mutant CALR ET patients. Thanks.

Thank you, John. Okay. I'm going to spend a few minutes now recapping some of what you've seen. But before we do that, it's very rare in drug development to see what we just witnessed over the past 10, 12 years from the -- to basically discover a new biology in 2013, reporting a new driver mutation for an entire group of diseases. We -- then designing a medicine specifically to address that problem, which is what 989 is, followed by a clinical experiment that basically replicates exactly what the preclinical data predicted we would see because everything that John just summarized is exactly what the preclinical data for 989 predicted we would see in terms of normalization of blood counts, reduction of mutated megakaryocytes and CD34 positive in the bone marrow and the peripheral blood. So it's a great story, and I hope you share our excitement that we have at Incyte with this data and what we think this new medicine can do for patients. So real quick, what did we learn today? Well, 989 in this patient population was very well tolerated. 1 out of 49 patients discontinued. It was a patient at 24 milligrams, the very first cohort. Dr. Mascarenhas described his impressions of that patient. It seems like it was a complication of the disease, unrelated to the drug. So very reassuring at this point in the clinical development of 989, the safety data that we have. Normalization of platelet counts. And I think it's a very important distinction between reducing platelets, which a lot of medicines can do and normalizing platelets. What you see in the platelet count graph that John showed is that the platelet count drops. And when it gets to the normal level, it stabilizes. That's consistent with translational biology that he showed. But that's very different from having to keep the platelets there by constantly adjusting the dose of hydroxyurea or an anagrelide. Patients normalize, they don't just drop, very different. We saw rapid and sustained reductions in VAF in most patients despite the fact that we have short follow-up. As you saw in Slide 42, the patients at the highest dose cohorts have relatively short follow-up. And despite that, so we believe that the VAF data will continue to look better and better over time because of the continued treatment of those patients. And then very important to put the whole biology story together is the reduction in mutant CALR-positive megakaryocytes in the bone marrow and the reduction in mutant CALR-positive, CD34-positive cells in peripheral blood that really show that 989 is doing more than normalizing counts. It's basically addressing the disease at its core. So what are the next steps for 989? We wanted to give you an idea of what's happening behind the scenes that what you're going to hear more about over the next several months. Our goal based on this data, and we are in the process of already expanding certain dose cohorts to understand better what the dose is for future development. That work is already ongoing. Our goal is to start pivotal trials in ET by early 2026. We appreciate the urgency of this, the importance of bringing true medicine that can change the natural history of the disease for ET patients and our job now is to do this as quickly as we can. We will present, as we committed, data in myelofibrosis patients before the end of this year. That work is ongoing. The combination with ruxolitinib is ongoing. We mentioned before that we thought it was very important for the myelofibrosis data to have combination data with rux. Rux in myelofibrosis, as you know, improves survival and having that data together was very important before we disclose it, and we'll do that before the end of the year. And we will accelerate the development of 989 as a single agent and in combination with rux in patients with MF. We announced early this morning, we established a collaboration with QIAGEN to develop a co-diagnostic across MPNs, initial focus in CALR mutations. So we are ready for that, and we're moving forward with our partner now. And very importantly, the development of the subcu formulation is already ongoing. We believe that initially, the intravenous formulation every 2 weeks is acceptable. We'll hear from our KOLs, but we're already working on a subcutaneous formulation. So then what is the scope of this -- of the patients or the impact that 999 can have in patients with ET. And there's a simple way to look at it, which you see summarized in the slide. If you take the entire population of patients with ET, both in the U.S. and the EU28, you divide that by 25%, which is approximately the frequency of CALR mutations, and then you divide them in two groups broadly, one group that basically is managed with watch and wait, which is about 1/3 to 1/4 of the patients and a group that is actively treated with cytoreductive therapy today. It's about 40,000 patients, give or take. Just to put that in perspective, the total number of patients in the U.S. treated with Jakafi in 2024 across all indications were 28,000 patients. So this gives you an idea how many patients could potentially benefit from 989 once it's available to them. Just to mention real quick, our commitment and Incyte for patients with MPN, as leaders in this space, continues. Our goal is to develop medicines for every single patient with myeloproliferative neoplasms. We're not done with 989. We introduced a T-cell engager bispecific. The data was presented today at the EHA conference, and it will enter the clinic very soon. And of course, many of you know the V617F inhibitor for the other most frequent driving mutation in MPNs. So just to repeat, our goal at Incyte is to have a solution for every single patient with a myeloproliferative neoplasm in the future. And let me just close with something I showed at the beginning of the year, which was a very rich set of catalysts that we committed to you to deliver in 2025. And if you look at the first half, it's coming to a close that we have, I believe, done an excellent job delivering on those catalysts on time and across a range of indications that is in the process of how we defined at the beginning of the year to transform our pipeline. So with that, let me stop, and we'll be happy to take questions from the room and online. Thank you.

[Operator Instructions]

Steve Willey from Stifel. Great data. I noticed in the cell types that harbor mutant CALR, it also looks like there's some positive NK cells and plasmacytoid dendritic cells. So just curious if you're depleting PDCs, if you would be concerned about antiviral immunity risk? And would you have to maybe prophylax these patients if you are indeed depleting that cell type?

Can you hear me? Yes. Now you can. Yes, that's interesting. Clinically, we do not see that these patients have anything beyond the sort of myeloproliferative phenotype. So we don't see any NK or dendritic cell phenotype. So that aspect of the expression profile needs to be looked at. For example, we need to look to see whether those cells actually express TPOR or not because that map that was shown is purely of mutant CALR expressing cells as opposed to cells that also express TPOR. So you can imagine that if they don't express TPOR, then the mutant CALR expression may not do anything. So that's work to be done. And at the moment, there's not a clinical indication as to why that might be a concern, but it's something to be checked.

I think that's correct, right? The key point here is that there's cells that carry the mutant form of CALR, but it's nonfunctional, right? So even though you saw positivity of mutant CALR on the UMAP within other cell types, the key slide that Dr. Nangalia showed where those cells will survive as long as the parent making that cell survives. So when you ablate or eliminate the mutant clone upstream, those cells will eventually disappear as well. But it's a slower process. It will happen across the natural half-life of that cell population. So even though CALR is there, you see it visualized in that UMAP, it's nonfunctional in those cells.

And then maybe just a quick question regarding potential Phase III trial design. There's obviously a couple of analogs that we've seen where you could run, I guess, kind of a post-hydroxyurea trial, just look at platelet normalization. I believe there's also a Phase III study of another drug going head-to-head against hydroxyurea where you'd have to look at events, I'm assuming -- and that would be a much longer study. So just curious as to how you're currently thinking about what the development pathway might look like here.

Sure. So we have -- as you mentioned, actually, as you summarized, we have a couple of different options. We could test 989 in patients previously untreated, or we could test it in patients that have received prior cytoreductive therapy, whether they either were inadequately controlled or they were intolerant. And as I think John mentioned, there's plenty of those patients, unfortunately, to enroll those trials. It's a little bit early for me to comment on the specific design. I think that some of the designs you mentioned are pretty straightforward. Their ELN criteria for hematologic response is well established, something that some of our competitors have used in the past. However, I think it's important to understand that this medicine is different from everything else that has been tested in this disease. The VAF reduction that we saw today in a very short period of time is much more pronounced, and we expect will continue to deepen over time than what has been seen with other drug candidates in this disease. That tells me that there's room to have a conversation with health authorities about looking at endpoints in a different way. We'll have those conversations in the near future. And depending on that, we'll design those studies appropriately. But it's a little bit too soon to commit to one path or another. We could develop this in both first and second-line ET and with traditional endpoints or with newer endpoints incorporating VAF in some way. Any other questions in the room?

So we'll take some questions online, Pablo. Next question is from Brian Abrahams, RBC. It looks like there's modestly higher proportion of nonresponders in the type 2, particularly at lower doses. For the company, would you anticipate different doses to be applicable for different mutation types?

So as I mentioned in my closing remarks, we are in the process of expanding certain dose levels to understand the efficacy in a larger group of patients. And I think it's very important when one looks at the data, particularly if you look at Slide 42, that the 1,500 and 2,500-milligram cohorts have very short follow-up. So to conclude that a certain mutation needs a certain dose or not, I think it's too early to decide that. What we've seen, to be clear, is responses in type 1 and non-type 1 patients. I think that's clear from the data that John summarized. And we believe that higher doses with longer follow-up will clarify a little bit what the design of future studies should be. For now, our commitment is obviously to develop 989 in all patients with ET regardless of mutation type. And we think that the data that we've seen, including the translational data, supports the role of 989 across the spectrum of the disease.

And a follow-up for Claire, John and Jyoti. Given the differences in prognosis and possible potency, how would you anticipate potentially using 989 in type 1 versus non-type 1 patients?

I mean, I think as per the answer to the last question, we still don't really know whether there really is a difference. And for me, the slide that showed the changes in the hematopoietic cell compartment, there were patients with type 1 and type 2. There was a type 2 patient there who responded beautifully at quite a low dose. So I think we don't know yet. I think it's interesting to note that type 1 patients are the ones that have the higher risk of myelofibrosis. But I've treated patients with this disease for nearly 30 years. And I've got 3 patients on this study. I've got 8 patients waiting to go on this study. My patients are keen to be on the study. They're keen even if they have to come every 2 weeks for a 1-hour infusion. And if it becomes a subcut preparation, that's just the same as giving interferon effectively. So I don't think we would treat them any differently at the present time. Thank you.

Next question is from Ash Verma, UBS. So looking at the Q2 weekly or Q2W dosing, do you think that this is a barrier for patients? This is for the KOLs.

No. I don't think it's a barrier. And I think what Claire pointed out is exactly right. I think if it was a barrier, we wouldn't have put 49 patients on like that. If it was a barrier, we wouldn't have a list that we fight over to get our patients on. I have physician patients with mutant CALR that are coming in to get treated, knowing it's a Q2 week. I think this data, if anything, reinforces the value despite a Q2-week infusion. I think if we saw a lot of infusion-related reactions or toxicity, that might be a different issue. But it's a very well-tolerated infusion. As Claire pointed out, perhaps if it was a subcutaneous delivered agent, it would make it even slightly more attractive than that. But no, the answer is no.

So a few questions coming from Jess Fye from JPMorgan. What is the hypothesized mechanism behind the lipase increase? And do you believe that this is on target or off target?

I'm going to ask Patrick to comment about the hypothetical mechanism, although we -- I'm not sure we have a hypothesis yet. But I want to emphasize the lipase increases were asymptomatic. They did not require dose adjustments, dose reductions. None of them led to dose discontinuation. So there were laboratory observations without any -- and they resolved spontaneously on drug. I think we need to understand that. No dose dependency, no discontinuations and they resolved spontaneously on drug and no symptoms of any related illness.

We just had a quick discussion actually. So Jyoti and I were discussing that, too. So lipase, Jyoti and I both work in hospitals that have Epic. We have a lot of pre-embedded tests for our patients. And we started seeing that actually in many MPN patients, lipase goes up and down. Maybe it's related to the platelet count, maybe it's something else. But for my patients, it's not meaningful. And I'll hand to Patrick in a minute, but I don't think there's really a clear mechanism unless it's some antibody interference we were conjecturing the other day with an assay, but it's totally asymptomatic. Patrick?

Completely agree. I mean it's not related with kind of the stage of dosing. It's not related to dose cycle. I think it's potentially something to do with the biology of the disease, and it's being measured here, right? There's extramedullary hematopoiesis happening in various tissues throughout the body. If those cells are responding in a meaningful way to this therapy, you may see blips like this. So I think it's in line with the points that have been made so far.

Follow-up question. Can you compare and contrast the profile we're seeing with 989 to what has been shown with other competitors such as bomedemstat?

So I think totally different. Bomedemstat, which is an LSD1 inhibitor is a drug that has to be dosed to platelet count. It's actually a more complex -- so it's a very interesting drug. It's a rational approach. But I would say completely different, different therapeutic, different delivery and different expectations. If you were to ask me, ET is primarily treated in the community practice, like in the U.S., community practitioners probably treat 2/3 of ET patients. What you want to do for a disease like this is have a therapy that's simple, simple to deliver without a lot of need for dose adjustments or calculations like that. And what's really nice about 989 is it's an infusion basically probably going to be at the same dose at intervals that don't require one to dose adjust. And that's attractive when you're trying to avoid toxicity and maintain efficacy in the community setting. So to me, that separates this drug from almost all the other drugs that we use in the field.

Maybe I could also add having treated a lot of patients with bomedemstat, they get dysgeusia. They also get arthralgia. And there's no evidence at the present that, that drug changes anything in terms of disease pathology. There's no reduction in VAF, there's no normalization of the bone marrow. But don't get me wrong, it represents an important alternative for these patients who don't have other options, but it's not disease targeted and it's not well tolerated.

The only other thing that I would add is that I have never seen a drug in a Phase I study target the origin of the disease, the stem cell clone and reduced VAF. In the graph that we showed, the waterfall plot of reduction in VAF, you don't do those kinds of graphs in Phase I studies. You hope for something like that in a Phase III. So it's completely different in how it targets the origin of the disease.

So one last follow-up for our KOLs. Putting aside the VAF changes, can you talk about the clinical significance of a PR as it relates to platelets? Can you describe how the 600,000 threshold may be associated with reduced rates of thromboembolic events and/or transformation? Or should we focus on CR and normalization of platelets?

I need to think about that. So I'm one of the authors on the ELN response criteria. So I would say that those response criteria are developed on the basis of expert opinion, which is the worst level of evidence. They're generally -- I mean, they're worthwhile, but they're generally developed after a meeting like this, in a room where we all vote consecutively. And if we don't reach consensus, we can't leave the room. So it's reasonable, but the first ever randomized study that was done in ET was done not far away from here in Bergamo. And in that study, randomizing patients to reducing the platelet count to less than 600,000 versus not, produced a significant reduction in events, and we have not yet bettered that with another treatment. So I think it's reasonable to say less than 600,000. There's no evidence that 400,000 versus 600,000 is actually better in ET. In PV, which is JAK2 driven, we just have some emerging data that might be the case. But I think the biology of CALR disease is different. John, do you want to add?

I totally agree. I'll just add the other point. If the question is, is 600,000 meaning getting -- and it is for some patients. So one of the phenomena that you see with thrombocytosis can be twofold. One are microvascular complications like headaches and brain fog and things like that, that do improve when you reduce the platelet count even below 600,000. So what's not captured here is sometimes the patients feel better on with a lower platelet count. And then obviously, for some of the patients that we had that had really extreme thrombocytosis with an Acquired von Willebrand, particularly menstruating women that might have very heavy periods or people who have to go for surgery, you want to get that platelet count lower. So getting under 600,000 usually removes the Acquired von Willebrand, usually resolves the headaches that plague these patients sometimes for years before the diagnosis is made. So even beyond the fact that there is Italian data demonstrating reduction in thrombosis, there's also a lot of quality of life and practicality issues with bringing the platelet count below 600,000. But to be honest, I like it under 400,000. So I think that...

Next question is from Malcolm Hoffman, BMO. Given the data we've seen so far in ET, what aspect of this data makes you most confident for the upcoming MF data later this year? What should our expectations be going into later years?

So let me tell you the way I look at it, and I was asking Dr. Nangalia to maybe comment as the expert in this mutation. They're very different diseases, ET and MF, but the molecular basis here is the same in terms of the -- what mutation in the CALR and the mutated protein does in terms of binding to TPO receptor, going to a cell surface and have driving an oncogenic signal. That's the same phenomenon. So whatever preconception one had about the probability of success in MF before the data that we saw today, particularly translational data that shows the impact on the CD34-positive population and on mutated megakaryocytes, whatever preconception one has, has to be -- that probably has to go up because of the data we just saw. That's at least my opinion. And maybe Dr. Nangalia...

I mean I would say actually the data for clonal reduction in MF is perhaps closer to the clinic. So John has shown with imetelstat, we've shown with navitoclax, a 10% to 20% reduction in VAF actually reads out to overall survival advantage for MF patients. And this year, there was a nice New England Journal paper showing the only other treatment we have for MF that cures patients, but it's very toxic, its transplantation and clearance of the clone after transplant also led to benefit. But a more modest reduction, which may become even deeper over time is also likely to be significant. And I think both of us are already also treating patients with MF and have got very good experience. I don't know whether either of you want to add anything.

I would add from a biological perspective, myelofibrosis is also a stem cell-driven disease. It will -- 30% of patients will have a CALR mutation, but many of those patients with MF have now accumulated additional driver mutations, making the clone a much more faster-growing clone and therefore, associated with far greater complications in disease burden. So if in myelofibrosis, you could also target the stem cell clone to significant levels like we're potentially seeing here with the ET data, that would be particularly exciting. And there's no reason to believe that, that won't happen provided that the CALR and the TPO are on the cell surface. The second thing I would be interested in seeing is the trajectory of reduction because, again, in those situations, I'm very interested in, one, how many mutant cells you have and the disease burden; but two, how fast is that clone growing. So if you could again reduce the number of cells and reduce how quickly they are growing, then you might be on a path and a journey towards protection from your disease. So I'd be interested in reduction of the stem cell clone in the MF data and trajectory of reduction, however slow or fast that is, I think as long as you're heading in the right direction, that will be a bonus.

I'd make one additional point. I think the genetic evidence is clear. Even in the setting of a secondary mutation that may make the clone more aggressive, those cells are still fully dependent upon CALR, that initial driver mutation. So that's very evident. So the ability to antagonize that, showing the clinical evidence of that in ET that should translate fully to MF based upon the genetic evidence.

Next question is from Derek Archila, Wells Fargo. This is for Pablo and Patrick. Which patients may be better suited for the monoclonal antibody versus the bispecific?

So the answer to that question, first of all, will depend on more data than we have so far. But it also depends very importantly on the T-cell engager. Not every T-cell engager is the same. We have designed our T-cell engager in a very specific way with a CALR arm that binds to a different epitope than 989. And what that means is it can potentially address the needs of patients that, for some reason, don't respond to 989 or that progress after 989 therapy, if that indeed happens, which obviously will take more time to determine. So that's the thought process behind the T-cell engager that we designed at Incyte. Other T-cell engagers have differences with that. So I can't generalize that answer, but focused on the one that we're introducing in the clinic, that's the plan to be designed to treat patients with a different -- that progress or do not respond to 989. I don't know, Patrick, if you want to comment further?

Next question is from Paul Jeng, Guggenheim. How are you thinking about long-term dosing strategy for mutant CALR in ET once initial platelet normalization and molecular responses are observed. Is there potential to explore a lower maintenance dose or different dosing frequency? And are there any differences in how might you think about dosing moving forward in ET versus MF?

Three great questions. Long-term dosing strategy, lower maintenance dose or frequency or difference between ET and MF. And obviously, we haven't reviewed the MF data, so I can't comment on that. And for the other two, we need time. We need to follow these patients for a longer period of time. I think it's important to remember what we're trying to achieve here. What we're trying to provide patients with MPN with 989 is a path to cure. This is not just about normalizing platelets. It's a path to a cure. That requires profound VAF responses. So we'll continue to treat these patients in the studies, and we'll continue with the higher doses to see if we can get to that point. And then we'll have to discuss with our KOLs what the next steps instead of testing this frequent dosing or not are appropriate at a certain point in the future.

Next question is from Kripa Devarakonda. What are the time lines for the development of the subcu? And are you doing this in-house or in collaboration?

So I'm not going to provide time lines today, but development of subcu is being done. So the formulation development is done in-house and then the potential need for a device like an auto-injector, for example, will be done with a collaborator that -- those conversations are already ongoing. And we're looking for every possible way to accelerate development of the subcu formulation. We understand the importance in this population.

Next question from Jess from JPMorgan. Did you see any difference in response between mCALR exon 1 versus exon 2 patients?

Type 1 versus type 2. So I'll refer back to the preclinical data that we've publicly disclosed, right? You can refer to the Blood paper. This is an active drug against type 2 patient samples in the work that we've done. I mentioned the binding affinity. This antibody binds with very high affinity to both type 1 and type 2. There is a subtle difference in binding 3.5-fold, but still a single-digit nanomolar binder to type 2. So this binds to and is active towards type 2 samples. We collected preclinically 8 type 2 patient cell -- patient samples, tested those in an ex vivo culture system, 6 of the 8 responded very well at the mid-dose that they were tested. This concentration relates to the dose range that we're covering in the study outlined here. So this is an active drug against type 2. I think all the points raised around dose intensity, duration of treatment, et cetera, I think, is what we're going to have to just follow and see how it plays out there.

Next question from Brian Abrahams, RBC. Looking at the safety profile, could you elaborate a bit more on the rates of TEAEs for the therapeutic dose level greater than 400 milligrams?

I don't have the breakdown here for the different dose level. I would emphasize some of the comments I made earlier. The only drug discontinuation happened at 24 milligrams. There was no dose-dependent increase in adverse events. Patients continued on drug. There were no dose reductions. And as we mentioned a few times, only one patient discontinued drug in a Phase I study. Maybe John and Claire from your direct experience with patients, different dose levels, you can comment. But so far, we have not seen any trend that seems to point to a dose dependency on any of the adverse events.

No. And I could -- to even add a little bit more color to that, even in this trial, there is intra-patient dose escalation allowed, which allows us to understand in a given individual, whether we would see a change in toxicity profile as one increases the dose. And even in patients who increased the dose, I didn't see any new treatment-emergent adverse events in those patients. So I think that's a pretty good test within a given patient that there really wasn't a dose-dependent TEAE profile. I mean this was -- I've said it before, and I'll say again, this is really well tolerated. Even the treatment adverse events that you see there are classic clinical trial, a patient comes in, reports something, the CRCs record it necessarily, attribution is provided, but those are mostly grade 1, 2, things that often can be explained by the disease or concurrent illnesses likely unrelated to 898 (sic) [ 989 ].

I mean in Europe, we've been using very high doses. I have a patient on the top dose and my patients are doing well. That patient that's on the top dose, walks 15 flights of stairs to get to the infusion facility. Fatigue is a symptom that 90% of MPN patients complain of. So it's really not worse for these patients. I think it's actually surprisingly very well tolerated personally.

Next question coming from Kripa Devarakonda from Truist. It's a 2-part question for our KOLs. What do you think is the relevant significance of symptom improvement? For example, in MF, there's a lot of focus on TSS50. Would you say there's similar contribution focus in ET?

So there's -- in general, there's less symptomatology in ET than MF, which is a much more inflammatory-driven disease, although some patients with ET clearly do, and some of those are vasomotor symptoms. You didn't need a specific TSS score to get on to this study. So it makes it a little bit challenging to formally evaluate and report in a dose escalation study, what the symptom improvements are here. I will just tell you anecdotally from the patients I treated, we definitely didn't see worsening of symptoms. If anything, for a lot of the symptoms, particularly headache-related symptoms, they got better.

I don't really have anything to add other than to say, I think interferon recently got approved in ET without a symptom endpoint. It's much less prevalent in myelofibrosis -- than in myelofibrosis. And yes, many of the symptoms do relate to the height of the platelet count. But I think we shouldn't underestimate. If you remember the patient story I told you, the emotional and psychological burden of living with a chronic incurable disease and not having an available treatment that can address that disease, that leads to a kind of chronic morbidity or chronic stress state for a patient. So slightly unrelated, but I just want to put that point across.

Yes, I feel compelled to bring this point up again because it never comes out in these kind of slides. But actually, what I witnessed were patients who were more reengaged in life actually in ways that I had not seen them before. So there's something about reducing the platelet count that provides a certain degree of confidence in going out in the world, doing things that they probably were afraid to do or were depressed about. So there was a huge psychological component of this that does not read out on those slides nicely.

The only thing I'd like to add is telling patients you aren't going to do anything about their platelet count because their vascular risk is low and they're young and that they have to live with it is a very big demotivator. Telling them that you can offer them something that can reduce their disease burden potentially in the future, a path to cure is incredibly motivating. That's why stem cell transplants are an option. They don't improve your symptoms, but they offer you a chance of cure. So making people feel better, but being able to prolong their survival and reduce their risk of transformation is a powerful motivator for using a drug in a condition.

Second part of the question and the last question online. Beyond platelet count and reduction in VAF, what would you see as additional metrics that strengthen the potential of disease-modifying activity?

I was really hoping disease modification would come up. It has kind of beautiful symmetry for me. Last year, I closed EHA in the presidential session with a very difficult question. Are we ready to disease-modify MPN? And there's kind of a beautiful symmetry for me to be now talking today and hearing the data beautifully presented by John. I think, okay, controlling the platelet count, but reducing the VAF and then seeing those really powerful images of the marrow morphology with the yellow CALR megakaryocytes disappearing and the new, the non-mutant megakaryocytes reappearing and the changes in the progenitor cells. I think that's really powerful indication of disease modification. We also showed that with the MAJIC-PV data actually, that the progenitor cells change. I think that's really powerful. I'll hand over to my colleagues to also make a comment.

I couldn't agree with you more. For someone who -- I spent a lot of time with the pathologists looking at the bone marrows of every patient I see and seeing those changes in the bone marrow to me, would signify that there is a fundamental difference in the pathophysiology of the disease. I have to believe that, that ultimately pays off not just from a prettier good-looking marrow, but a better outcome for the patients.

Any questions in the room? Okay. With that, we'll close the meeting. Thank you, everybody. Thank you everybody online for attending on a very early Sunday morning in the U.S. And we look forward to keeping you updated on the progress of this program in the next few months. Thanks again.