Geron Corporation (GERN) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by and welcome to Geron's Key Opinion Leader Conference Call. [Operator Instructions] I would now like to hand the conference over to your moderator for today, Suzanne Messere, Head of Investor Relations. Ms. Messere, please go ahead.

Thanks, Jack. Welcome to Geron's KOL event to discuss recent data presented at the virtual edition of the 25th Annual European Hematology Annual Congress. I am joined today by Geron's Chief Medical Officer, Dr. Aleksandra Rizo; Dr. Valeria Santini; Dr. John Mascarenhas; Dr. Rami Komrokji; and Geron's Chairman and CEO, Dr. John Scarlett. Before we begin, please note that this presentation and question-and-answer session will contain forward-looking statements relating to Geron's plans, expectations, time lines, beliefs, statements of potentiality and projections. These include, without limitation, those regarding that imetelstat has potential disease-modifying activity and that Geron has a plan to open for enrollment a Phase III clinical trial in intermediate to or high-risk myelofibrosis in the first quarter of 2021. These statements involve risks and uncertainties that can cause actual results to differ materially from those in such forward-looking statements. These risks and uncertainties include, without limitation, risks and uncertainties related to whether the COVID-19 pandemic slows or prohibits the company's ability to open for enrollment, the planned Phase III clinical trial in Intermediate-2 or high-risk MF in the first quarter of 2021, whether regulatory authorities permit the further development of imetelstat on a timely basis or at all without any clinical hold and whether imetelstat demonstrates disease-modifying activity in clinical trials. Detailed information on the above risks and uncertainties and additional risks, uncertainties and factors that could cause actual results to differ materially from those in the forward-looking statements are explained under the heading Risk Factors in Geron's quarterly report on Form 10-Q for the quarter ended March 31, 2020, filed with the SEC. Undue reliance should not be placed on forward-looking statements, which speak only as of the date they are made, and the facts and assumptions underlying the forward-looking statements may change. And now I'd like to hand the call over to Aleksandra, who will review today's agenda and get us started with our first introduction. Aleksandra?

Sure. Thanks, Suzanne, and thank you all for joining us today. We're so glad we could connect with you in our new virtual format. As we've announced, we have 4 abstracts accepted for presentation at this year's EHA conference. The data from the abstract suggests potential disease-modifying activity of imetelstat in the 2 indications that are our current focus, low-risk MDS and myelofibrosis. The recent data from the Phase II part of IMerge, which continues to show meaningful and durable transfusion independence will be presented first. Of particular note, we're reporting a 1-year TI or transfusion independence rate for the first time. Then 3 separate analysis from IMbark will be presented. And of utmost importance, the updated overall survival analysis will show correlation of the potential improvement in overall survival with other clinical benefits, such as fibrosis. We expect these results to drive further interest of investigators, which will promote enrollment for the ongoing Phase III clinical trial in lower-risk MDS and the planned Phase III study in refractory myelofibrosis. After each speaker, as already mentioned, we will have a Q&A session, and we encourage the invited analysts to participate. To close out today's event, I will provide an overview of the trial design for our planned Phase III trial in refractory MF. And then Chip, our CEO, will provide closing remarks. Now I'd like to start by introducing our first speaker. Dr. Santini will be reprising the MDS presentation. And I'm sure a lot of you will know her. She's one of the most prominent and renowned investigators in the MDS field. She's currently Associate Professor of Hematology at the University of Florence in Italy. Dr. Santini has hands-on experience with imetelstat and even continued to enroll patients through the COVID pandemic period nowadays. I've worked with Dr. Santini for the past, I believe 10 years or so, and it has been nothing but pleasure to hear her advices, discuss patients and continue to brainstorm on protocols. With that, Dr. Santini, would you like to start?

Yes. Thank you, Aleksandra, for the very nice introduction. And I think going for the first slide, and I will introduce what was presented at EHA, the virtual EHA 25th meeting. Now MDS are a very different and heterogeneous group of disease dominated by the presence of anemia in the majority of cases. This anemia can be extremely severe but can be treated in the beginning with the use of erythropoietic-stimulating agents. The problem being that the majority of patients become transfusion-dependent and because they lose response to ESA. Now what we need to have for this increasing population that is the majority of MDS patients is a treatment that avoids transfusion and increase transfusion independence with a durable, of course, duration of response. And we also want to have agents that target both ring sideroblast positive and ring sideroblast negative, being these 2 the characteristics indicated by WHO classification as a different type of lower-risk MDS patients. And of course, our goal would be to get disease-modifying therapy to prolong survival. I'm sorry, I don't see the slides any longer. I don't know what happened. Next slide, please. So this is the study that I'm going to present, next slide, that was an international study in which imetelstat was used for lower-risk MDS patients. You see imetelstat is a first-in-class telomerase inhibitor and blocks telomerase activity, which normally is increased in MDS, in myelodysplastic cells and progenitors. Next slide. So what I'm going to talk about is the Phase II study that is completed and that is a single-arm open-label study. As I mentioned, for relapsed/refractory lower-risk MDS that received treatment with ESA or were not suitable for treatment or not eligible for treatment with ESA. Imetelstat was given 7.5 milligram per kilogram IV every month and 38 patients were treated in this study. The patients had to be, as I mentioned ESA refractory or not eligible because of endogenous EPO measured higher than 500 units liter, but they also shouldn't have received treatment with HMA or lenalidomide. On top of that, they need to have a transfusion dependency measured in the 16 weeks before screening and the transfusion need had to be higher than 4 units. The primary endpoint being the achievement of 8-week transfusion independence. As a key secondary endpoint, we had a 24-week transfusion independence, duration of this transfusion independent and hematological improvement. I will talk at the end about the Phase III currently ongoing and enrolling. Next slide. So this is a study that has been conceived with a very long median follow-up. As you may see, there are 2 years of follow-up. The median duration of treatment for old patient was 8.5 months, and the number of cycles was 9. So this 30 -- next slide, these 38 patients were elderly, median age was 71.5 years. The striking thing is that the patients enrolled in this study had a very high transfusion burden with a median of 8 transfusion -- 8 units of red blood cells in 8 weeks. And as you may see, the majority of them, 84% had more than 6 units in the 8 weeks at baseline. That means that this was a group of patients with quite important burden of disease and high burden of transfusion. The majority of patients as it happens very frequently in the study with ESA-resistant patient was formed by refractory sorry, of course, refractory to ESA, but ring sideroblast MDS patients and almost all of them received ESA and very -- well, 1/3 had highest endogenous EPO level higher than 500 units. Next slide. So the patients are still ongoing, treated, 9 out of 38, 76% stopped treatment for several reasons, some for lack of efficacy, some for the adverse event. But the median follow-up, of course, it's the reason why we see so many stopping treatment. Next one. Now the results are, in my opinion, extremely interesting because the 8-week transfusion independence was achieved -- reached for 42% of these patients. And the duration of transfusion independence is remarkably long with 88 weeks of response with the rise in -- and once -- more than 1/3 of the patients higher than 3 grams deciliter of hemoglobin. So the duration of response indicates the stability and the depth of these responses. At 24 weeks, transfusion-independent patients were still 32%, and there was almost a 30% -- 29% of patients who were still transfusion independent after 1 year of treatment with longest transfusion independent of 2.7 years. Next slide. Now when the hematological improvement, which is certainly another very important point that is the decrease of transfusion burden in -- measured always in 8 weeks. We see that according to IWG 2006, hematological improvement was present in 68% of the patients. And the duration of this hematological improvement-erythroid was 92.7 weeks. So extremely long as well. But then we measured with more stringent criteria of response, the ones of the IWG 2018 group criteria, you see that major response was achieved for 16-week transfusion independent. So as I mentioned, very stringent, 37% of the patients achieved this response. And more than half of the patients had decrease in transfusion higher than 50% in 16 weeks. So these are data that are really very promising and very intriguing. Next slide. Now when we look at the duration of response, as I mentioned before, you see that in this plot, 29% of the patients are still transfusion free at 1 year. And as I mentioned, many of these patients have been treated for a very long period with 1 transfusion -- with transfusion-free period of 2.7 years. What is very striking is also that the great majority of the responders had a substantial rise in hemoglobin, 3 grams/deciliter. And as you may see in this dark green bar, these are the patients who have achieved and maintained response in terms of transfusion independence longer than 1 year. Next slide. Now if you look across the different patient subgroup according to WHO category or IPSS-R, you see that there are not so many differences in the rate of response to imetelstat. What is to be noted is that both ring sideroblast positive and negative patients do respond. The interesting factor is that the patients who had a very high transfusion burden, so once again, the median was 8, but here it's more than 6 units. You see that these patients do respond as well as the ones who have a lower transfusion burden. IPSS low or intermediate doesn't really matter as well as the level of endogenous serum EPO. So very, very active all across the subtypes of MDS. Next slide. Now the adverse event observed during imetelstat treatment were mainly due to hematological adverse event, a little bit of myelosuppression, thrombocytopenia, neutropenia were reversible. And here, you see that some of them were grade 3, 4 in around half of the patients. There are some patients who also presented anemia. Just to remind, these are the -- all the patients, responsive and nonresponsive ones. And anemia is, of course, the characteristics of the patients undergoing therapy with imetelstat. So that means that the patients not responding will have anemia as grade 3 or 4 as the majority of them. Now as I mentioned, all the cytopenias were reversible. You see in light blue, 10% of the patients will not recover from grade 3/4 cytopenias within the 4 weeks. That means that they resolved cytopenias after 4 weeks or they have been observed for a shorter time than 4 weeks. Next slide. Now the nonhematological adverse event were really not significant and not very frequent. I want to stress the fact that the liver toxicity was very mild, only very few grade 3 for adverse events regarding liver hepatic function. Next one. Now the activity of imetelstat, as I mentioned in the beginning, is due to the inhibition of telomeres. And on target activity was demonstrated for several patients. The 50% reduction in telomeres activity was seen in a good proportion of patients. And the reduction of hTERT expression correlates with the response. So the patients who responded were the ones who had higher reduction in hTERT expression. Meaning that the on-target activity of imetelstat is the one that is really determining response in terms of increase of hemoglobin. Next one. To note, imetelstat has a very important effect on the variant allele frequency of some mutation present in this patient. Now as you noticed, the majority of patients included in the study were ring sideroblast positive ones, meaning that the majority of them bear the SF3B1 mutation. Now the 11 patients who responded and were mutated had a reduction in the variant allele frequency of SF3B1. And the greater the reduction of the mutated clone the longer was the transfusion independence. This is something extremely significant, in my opinion, because that means that the drug is acting directly on the dysplastic clone, decreasing the burden of the clone itself. So this is something that not is common in active drugs. But I would say, almost exclusive of this imetelstat study. Next slide. I'd like to conclude just by stressing and reminding the data. The data indicated 42% of the patients treated with imetelstat achieved at 8-week transfusion independence rate and 68 of them had a hematological improvement erythroid. The duration of both transfusion independency and hematological improvement erythroid was almost 2 years, 20 months and 21 months. And then all the patients with MDS included in the study seem to be responsive to this treatment, especially with the ones -- to stress the ones with high transfusion burden and very high transfusion burden. Finally, the data showing on-target activity, reduction on SF3B1 mutation by treatment and the length of the response it's indicating a potential disease-modifying activity of imetelstat. Evaluating adverse event. There were no new safety signals. The cytopenias induced by imetelstat were reversible and were the most frequent adverse event. There is a trial that is ongoing. It's a Phase III double-blind randomization 2:1. Next slide, with this currently enrolling study, we will be able to consolidate the data I just show you. We have been busy, and we are busy, including patients that are a population similar to the previous one. So non-del(5q), lower-risk MDS with transfusion dependency, more than 4 units in 8 weeks and HMA and lenalidomide naive, here as well the primary endpoint is 8-week transfusion independence and the key secondary, again, 24 weeks transfusion independent. Now the -- I think that we learned a lot from the Phase II study, and we are very eager to see the final results of the current Phase III study. With this, I wish to go to the last slide and thank all the people who cooperated, the patients. My patients are very collaborative, and they are very eager to participate through these studies and all the collaborators around the world. And I thank you for your attention.

Thank you, Dr. Santini, for that nice presentation and overview. So then, we will have some Q&A now, I believe, right?

[Operator Instructions] Chad Messer with Needham.

Great, good morning, at least for me here. Very interesting that there's a nice subset of patients that are responding so well and for so long. Maybe 2 questions on that. One, can you put into context other treatments and things that are used, luspatercept, for example. Are patients responding that long to other treatments? And it is also very interesting that there was some correlation with hTERT reduction. Did you look at baseline hTERT? In other words, is that a possible prognostic indicator for who may respond to this?

Now to answer your first question, of course, there are many studies focusing on this subtype of patients. So the big unmet need of these patients, the lower-risk patients with transfusion dependence. They live very -- they live longer, which is nice, but they need to find something that releases them from transfusion. And as far as my knowledge, there are not many studies, and I would say, no studies that have shown such a long duration of transfusion independence. Sporadic long responders are found in many studies. I leaded a study with lenalidomide several years ago. And there sporadic patients with long transfusion-independence can be seen as well as, for instance, in the luspatercept trial. But in this case, there is a good proportion of patients that do respond for very long time. I mean, 30% -- 1/3 of the patients are still responsive at 1 year. So this is something quite indicative of the activity of the drug. And your second question was the on-target evaluation. Now as far as I know, but this maybe it's more preclinical. There haven't been prospective studies evaluating hTERT before treatment and then using it as a biomarker of response. Certainly, the signal we received here with 50% reduction of hTERT in the patients who do respond is quite interesting. So I think the next step would be the one you suggested to use it as a biomarker of -- possible biomarker of response to the drug.

If I may add there to Valeria. So since I'm -- I always do both, MDS and myelofibrosis. For the first question, I think as clinicians we consider a transfusion independency of 6 months meaningful for the patients, like the duration. And that's probably the bar that we look at for drug approval. For your question, for example, with all the current available therapies after ESA with lenalidomide and non-del(5q) typically it comes to somewhere around 48 weeks, transfusion independency. Where the luspatercept, the median was like 30 weeks, even with hypomethylating agents, it's -- all of them are really in the less than 1 year. So 88 weeks really looks very interesting and exciting.

Yes. As I mentioned, it's sporadic, and then we are used to have something below -- well, below 6 months usually. And which is what we consider as Rami just said, we consider significant because if you get few weeks of transfusion independence for the patient is really not relevant.

Tom Shrader with BTIG.

Thank you for the update. I'm a little curious about your thoughts in the real world, how do you use a drug like this? If you look at Slide 16, there are a couple of patients there that had a huge number of transfusions and then got pretty remarkable benefit from the drug. How do you decide who to keep on? Is it clear patients that are deriving nothing? I'm just -- it's really an interesting curve that they're almost a year with looks like no benefit and then the drug started to work. So how would you use it in the real world?

Well, it is also very much up to the patient. What I learned over the years is that with MDS, you don't have to be in a hurry. You very often with several drugs, several types of drugs, you need to wait to see a response. Now clearly, if the patient is progressing or is increasing transfusion needs, of course, that's another story. But when patients do maintain a stable transfusion dependency, and then you may keep them with a drug that is not giving so many adverse events then I think that's the way to go and wait. When is a different story. When is a critical question, I would say. I have a patient just now. I can tell you that slight borderline responsive, and he wanted to keep on being treated. And this is what we will do because probably it's a late responder. And it's a little bit the same story also for ESAs. Some patients do respond very late. We, at present, don't know exactly why this is happening, but I suppose that it's something that we should elaborate on. So in real life, what I would do is that I would surely stop the patients who either have increased transfusion need or absolutely no effect at all or even adverse events, but for the other ones, I would keep, if possible, and if the circumstances allows it for at least 4 to 6 months treatment.

And the drug is benign enough that you think most patients would be okay with that, waiting half a year?

What I -- in my experience, I didn't get many important adverse event patients. Now this is a trial. So it's a little bit different situation. In real life patients it's difficult to say, but some patients would be okay. If they know that the drug can give them an advantage in a substantial proportion of cases. If you explain in the beginning to the patient that they should come once a month to get this IV infusion because 50% or 60% of patients have had an advantage with a median -- as you see, with a median time that is -- so that is not immediate, then I suppose they will stay for it.

Okay. If I may add again, sorry. I think if you look at the median time to response was 8 weeks, and yes, there are like some late responders. But in reality, most of the time on a clinical trial or real life, we'll probably be seeing some clinical benefits, that there is some reduction that will make you continue the treatment. The other thing to keep in mind, any time we are applying a disease-modifying agent technically or like from examples of lenalidomide or hypomethylating agents, we always see worsening of the cytopenias actually in the first few cycles and then the improvement. In deletion 5q where lenalidomide is the standard of care, actually, cytopenia on therapy upfront predicts response. So upfront, you expect sometimes if there is a drug that's disease-modifying to have some cytopenia, less extent we see that with ESAs, for example. But I would say onset -- the median time of 8 weeks is very reasonable. Probably what we do at that 4 or 6 months, if those patients were starting to see some reduction in the transfusions, we continue. And that's why you see some patients became transfusion independent completely almost at 40 week.

Okay. And again, this is true for all the drugs for MDS, almost all of them.

[Operator Instructions] Stephen Willey with Stifel.

Thanks for the update. So maybe just a follow-up to Tom's question, and hopefully, it doesn't sound like the same question, but just curious as to how you think about the difference between achieving transfusion independence and this notion of "clinical benefit." And I guess when you treat these patients clinically are you treating with the intention of achieving transfusion independence? Or is a reduction in transfusion burden and some kind of erythroid response sufficient enough for some patients. And I guess I asked the question because I know in the luspatercept data set, they showed very low transfusion independence rates in patients who had very high transfusion burdens but I believe when they updated their longer-term data, they showed a much higher proportion of "clinical benefit" occurring in these higher transfusion burden patients at baseline. So I guess I was just trying to figure out to what extent either of those things matters more importantly to you as a clinician.

Well, these patients here had -- the patients on imetelstat trial had really high transfusion burden, as you saw. So these 8 units in 8 weeks means that this patient had one transfusion weekly, which is a lot. With the imetelstat, therefore, you have severely transfused, I would say, a patient group who gets transfusion independence in 40%, 42%. So that is really an achievement. Now is this clinical significant for the patient? Indeed, it is because if you talk to patients who are transfused weekly, their dream is to get rid of the transfusion and be free for a substantial period of time. Now here, they had almost 2 years of transfusion independence. So that means that heavily transfused ones stopped coming to the hospitals to be transfused. So to me, this is really meaningful. Now going back to the hematological improvement that here is nearly 70%. Again, these patients may delay their transfusion, so they may delay their dependency on the hospital care and on transfusion, they would feel, I suppose, much better. At least this is what I hear from my patients that instead of coming once a week they come every 2 weeks or once a month that would be extremely, extremely significant for them. What is important is to talk to the patients and hear to them. And once again, to echo what was said before, to me, it's indicating disease-modifying activity. If I see that the duration of response is substantially long. So these patients have a decrease in the, I would say, the burden of the disease, indeed. So that's something I would call clinical significant. Talking also from the point of view of the patients. For me, from a more scientific and biological point of view, it's quite meaningful for the patients for practical reason and for their well being and quality of life.

Okay. That's helpful. So I guess if we see a replication of these results in the Phase III and you have a patient who presents to you in the clinic with a high baseline transfusional burden, they're RS positive, you start them on this drug? Or do you start them on luspatercept?

That's a difficult question to answer, but I would say that if the patient has the -- as usual it is high platelets and the -- no neutropenia, I would probably try to start with imetelstat. And luspatercept is a very good drug. I would rather have luspatercept for patients who have lower transfusion burden and for very high transfusion burden would go for imetelstat. If these results are like that, you see the median transfusion burden in the luspatercept trial was 5 units and here it is 8. So there is a bit of difference. So I would go for lower burden for luspatercept and for higher burden imetelstat with normal count or near-normal counts.

That's very helpful. And then maybe just one question for the company. You've disclosed median duration of therapy from this Phase II. But I'm giving -- I'm assuming just given the long duration of responses here that something like mean duration of therapy is probably a more important statistic, is that information that you have at all?

Yes. Aleksandra here. We do not disclose that yet. But it's a good suggestion. I think it's something we can certainly update. And the other points we were discussing even today with Dr. Santini were just like the median duration of treatment just for the responders as well. So these are a couple of things that we will be updating shortly and shared, will be of interest to understand that.

There are no further questions at this time.

Okay. Then I would like to thank Dr. Santini so much for taking the questions as well. And I would like now to move to our next speaker. That will be Dr. Mascarenhas, a prominent myelofibrosis expert whom -- actually do not know if his introduction is needed. He is currently Associate Professor of Internal Medicine at the Icahn School of Medicine at Mount Sinai, New York. He has been involved in the development of imetelstat, from the inception I believe from the Phase II IMbark study. Most recently over the past year or so, we have been working closely with him to evaluate all of our myelofibrosis data and discuss the design of the Phase III trial in myelofibrosis that we expect to start next year. He will have 2 presentations today, and his first one will be on the biomarker analysis from the Phase II IMerge portion of the study. John, would you like to take it from here?

Thanks, Aleksandra. I hope everyone can hear me clearly. So my name is John Mascarenhas. I think I've met some of you previously when I've had the opportunity to present and I'm honored to be able to come back again, and I've really enjoyed working with the folks at Geron. So I hope I convey my excitement on behalf of our coinvestigators, of the development of imetelstat for myelofibrosis because I really think this is the potential to be a game changer and I'm saying that as an investigator, obviously not as a Geron employee. So please understand what I'm trying to convey to you is sincere excitement about this drug. And I was excited to hear Dr. Santini present in MDS, and it makes obvious sense because there's so much biologic and clinical overlap in MDS that you would see potential for disease course modification improvement in an important end point like anemia in low-risk patients. But what I'm going to show you today is in a related myeloid malignancy, myelofibrosis, the potential impact for a very urgent unmet need, which is survival, which we've talked about previously. But what I'm hoping to do is show you the link to biomarkers nontarget activity that makes this even more tangible and even more exciting and I think brings it even closer to moving this forward. So if you can give me the next slide, please. So for those of you who just want to refresh, myelofibrosis is a clonal hematopoietic and progenitor stem cell malignancy. It is very closely linked to mutations that you see across myeloid malignancies, including MDS and AML. It involves malignant megakaryocytes that elaborate cytokines that induce fibrosis and scarring in the bone morrow, and ultimately impaired hematopoiesis. So anemia and thrombocytopenia can be prominent features of advanced myelofibrosis as well as splenomegaly and other areas of organomegaly due to extramedullary hematopoiesis from inappropriate trafficking of hematopoietic stem cells that leave the protective bone marrow niche and set up hematopoiesis in other organs. So this is often a major manifestation, an issue related to myelofibrosis, in which JAK inhibitors have been very successful in addressing spleen and symptoms, which are thought to be related to cytokine overproduction by the malignant clone, and this includes fevers, weight loss, night sweats and a myriad of other symptoms. There are -- there have been many attempts to address the fibrosis. And the name of the diagnosis is myelofibrosis. So trying to address the phenomena of bone marrow failure that's related to the fibrosis has also been a translational focus of research. For patients with Intermediate-2 and high-risk myelofibrosis, in the U.S., we have 2 options that are available commercially, ruxolitinib, fedratinib. Ruxolitnib since 2011, fedratinib since August of 2019. Both are excellent drugs, they address spleen and symptom. But I would argue that, unfortunately, that's somewhat limited in what they do. They don't really change the clonal nature of the disease. There really is very limited data that would suggest they changed the natural history of the disease. And the painful truth is that by 3 years, about half the patients have discontinued therapy. And by 5 years, over 75% of patients are off therapy. So the effect of JAK inhibitors that we have right now, whether they're commercially approved or in clinical testing is often limited in duration of therapy. And when patients fail ruxolitinib, multiple groups have shown this in analysis, the outcome is quite dismal with the median survival of approximately 1 year to 1.5 years. And right now, there's really no approved therapy that effectively changes that. Let me restate that. There's no approved therapy. There is no experimental therapy that improves that. And that's why this drug in its indication is really quite exciting. And again, I'm hoping I'm conveying that to you effectively. So there's unmet need. Therapies that can improve survival in patients who fail JAK inhibitors is really what we're looking for in the clinic because although improving spleen and symptom is important, when you fail a JAK inhibitor, many of the patients I have want to live longer. That's the primary goal. They don't want to die from the disease. Many of them may not be able to go to transplant, which is definitive therapy. And even those that want to go to transplant, often need to be debulked of their disease process because there's a concept that patients that go into transplant with less disease are more likely to enjoy the benefits of transplant on the back end. So there's clear need for disease-modifying therapy. And I think that, that speaks to the fact that we didn't -- need a non-JAK inhibitor mechanism of action therapy such as telomerase inhibition. You can see the stats on this slide. I think they're underestimated because I don't think we capture the actual incidence and prevalence of myelofibrosis well. But the estimations are that over 35,000 patients in the world have intermediate to high-risk myelofibrosis by the International Prognostic Scoring System. And in the U.S. alone, there's probably over 12,500 patients, with an incidence of approximately 2,000 patients diagnosed a year in this high-risk category. Next slide. So I know you -- everyone on this call is familiar with the IMbark study, which was evaluating imetelstat in patients with intermediate or high-risk myelofibrosis who had failed ruxolitinib previously, and this is the study schema for that study. Patients were randomized to the study drug, either at the high end of the dosing schedule at 9.4 milligrams per kilogram every 3 weeks or the low dose, 4.7 milligrams per kilogram every 3 weeks. And the co-primary endpoint was spleen and symptom. The secondary endpoints were IWG responses and overall survival and then exploratory endpoints for cytogenetic and molecular responses. And of course, just to be clear, relapsed/refractory patients were defined as patients who had been on a JAK inhibitor previously, had worsening splenomegaly related abdominal pain at any time after the start of the inhibitor, and had no reduction in spleen volume after 12 weeks or worsening splenomegaly at any time after the start of JAK inhibitor therapy documented either by nadir and then increased by 25% by imaging or often more commonly, in the community, increase in palpable spleen on exam. Next slide. I think it's not needed for me to explain to you now what Imetelstat is, as everyone here knows what it is. We are looking, and you'll see in these presentations, and I think it will speak to a question that was asked previously about -- in MDS, about perhaps biomarkers that show were hitting the target and biomarkers that could actually have the potential to predict response. And those biomarkers of interest are going to be telomerase activity itself, hTERT expression level, and then particularly to MF would be driver mutation, allele burden changes, for example, JAK2 V617F as this is a marker, a surrogate marker for disease burden since we don't have imaging like RECIST criteria for solid tumors. Short telomere length, high level of telomeres activity and high expression of hTERT levels correlates with higher risk and disease progression and shorter overall survival in patients with myeloid malignancies. So cells with short telomeres, high levels of telomerase activity and high levels of hTERT expression represent the best target for treatment with a telomeres inhibitor like imetelstat. And the optimal pharmacodynamic effect of imetelstat has been defined as greater than 50% -- 50% or greater reduction in the telomeres activity or hTERT level from baseline and is correlated with antitumor activity in preclinical PK/PD study. So the objectives of this study to evaluate the on-target PD effect of the drug in relationship to dose as well as exposure in MF patients treated on the study to assess the correlation of the optimal PD effect with symptom or spleen response, in my opinion, more importantly with overall survival and explore the association between baseline telomere length, hTERT expression level and clinical benefits as well as evaluate the changes in allele burden, and again as a surrogate marker for changing the disease activity. Next slide. So here, we're first looking at dose-dependent and exposure dependent PD effect. So if you concentrate first on the left-hand slide, what we're showing here is the percent of patients who achieved at least a 50% reduction either in telomeres activity, which is the area on the left, the 2 bars on the left or hTERT level reduction on the right. And you can see it's color coded. So blue is for the low dose and yellow is for the high dose 9.4 milligrams per kilogram, as you recall that's the active dose that we were going to move forward into the Phase III study, and in both cases, for both pharmacodynamic on-target activity, the high dose arm achieved the PD effect of both telomerase activity reduction, hTERT level and a greater proportion of patients that was statistically significant in both cases. If you look to the right, we're looking at exposure-dependent pharmacodynamic effects. Again, this is the percent of patients who have at least a 50% reduction in hTERT levels. This -- on the first 2 bars are looking at cycle 1 AUC area under the curve, and it's stratified by low exposure versus high exposure and to the right, second one, Cmax. So basically, what we're showing here is that a higher proportion of patients that had high exposure rates achieved higher PD effect in terms of hTERT attenuation. So this was very -- to me, this was very important to show because this shows that we are hitting the intended target and that it is directly related to the amount of exposure of the drug, very importantly. Next slide. Okay. But how does that correlate to the clinical response. So what we're showing here is that if you look, first, concentrate, please, on the left, so we're looking at the spleen and symptom improvement. Again, this is the percent of patients who achieved optimal PD effect. PD effect, again, is defined as in the responders, the spleen on the left and then symptom on the right. So the patients who had a spleen response, that's 35% spleen volume reduction, 100% of those patients achieved the PD effect versus 76% of the patients who did not. Nearly 100%, 91% of patients who had symptom response that's a 50% symptom improvement by the TSS, also achieved optimal PD effect, again suggesting that there's a correlation between hitting the target and getting those 2 responses. But what I think is much more important, which I want you to look at on the right is the survival curve. So this is the patients who achieved optimal PD effect, had a better overall survival. And that is shown with these 2 curves. So if you look at the survival of those patients who achieved the PD effect, it was 27.2 months versus those who didn't, it was 18.3 months with a hazard ratio of 0.54. So this was -- to me, in this presentation this is probably the most important and most telling slide because here, we're showing that it is not by chance, and it's not random that imetelstat was associated with an improved overall survival because it's hitting the target, and those patients that are hitting the target are more likely to enjoy that survival benefit. Next slide. And then does it affect the disease in a meaningful way? And this is an attempt to show by correlation with varying allele fraction of the driver mutations. And what I want you to look at in the table, and it's really showing the same thing, but look at the table first. And what you will see is, if you look at those patients who achieved a 25% or greater decrease in a driver mutation. This covers about 90% of patients with myelofibrosis that have driver mutations and are used as a marker for disease modification. That, in each case, the higher dose arm had a greater proportion of patients that achieved it, but these are very small numbers. So if you scan down and say, well, of any driver mutation, what was the differential in effect. And you can see that 42% of patients who had any driver mutation achieved at least a 25% decrease in allele burden versus only 5.6%, one patient in the low dose arm. So correlating dose with effect on the malignant clone, reduction in the malignant clone. And if you recall back from the original paper in New England Journal of Medicine by Dr. Tefferi and his colleagues, there were even cases where he treated them perhaps more intensively with this drug of complete molecular responses. And we do not see that with any therapy other than transplant in myelofibrosis, where you can actually induce a molecular response of that nature. So this still -- to me, this still speaks to the fact that we are modulating it, we're hitting the malignant clones, not just killing off late progenitors. And the graph on the right just graphically shows you what I'm also telling you. So next slide, please. So what about patients if you segregate them by telomere length. Since it is a telomerase inhibitor, can this be a predictor of response, and this came up in the discussion previously with MDS. So this is on the left, looking at the percent of patients who achieved the response and then segregated by having either a shorter telomere length or a longer telomere length. And again, it's color-coded by dose. And I would ask you to pay attention to the 9.4 milligram per kilogram bars, that's the active dose. It was a higher proportion of patients, achieved a response with shorter telomeres. And that's true for spleen as well as symptom. And then if you look at the right, if you look at baseline hTERT level, a higher proportion of patients achieved a response both in spleen and symptom if they had higher hTERT levels. Again, suggesting that there is a baseline biomarker that also can predict or associate with response in terms of these clinical endpoints. Next slide. But I would still argue, spleen and symptom is nice, but we're interested in overall survival in these patients who are unfortunately doomed to do poorly. And this is really where imetelstat stands apart from the other therapies that you can see presented at the EHA meeting. So here, I would ask you to focus in on the right at the 9.4 milligram per kilogram arm, this is segregated by baseline telomere length in which the survival, although perhaps not statistically significant is longer and that curve sits apart from the patients who had longer telomeres lengths at baseline. Again, trying to associate a biomarker for a mechanism of action of this drug with an important endpoint such as survival. Next slide. So in conclusion, imetelstat, here, the data shows achieves dose and exposure-dependent reduction of telomeres activity and hTERT expression level thus demonstrating on-target mechanism of action, very important, particularly for those, I don't mean anyone on this call, but those naysayers who may say, well, this was a study initially that didn't hit its endpoint to spleen and symptom and the survival benefit could have been random, I would argue it's not random. It's not random at all because we're showing you that there's on-target activity. Achieving optimal PD effect, as measured either by greater than 50% reduction in telomerase activity or hTERT level in patients treated with imetelstat is correlated with clinical responses and longer overall survival. So this really validates the preclinical PD findings, so that's to me very satisfying. Significant dose-dependent reduction in the variant allele fractions of JAK 2, CALR and MPL mutations were observed. This would indicate that imetelstat is actually modifying the disease at the basis, at the STEM and progenitor cell level, very important because I believe that's why you see the survival benefit. I believe that's why even if you don't continue the drug indefinitely, and you have a median exposure of 8 months, 32 weeks, you've got a survival that approaches almost 30 months. Treatment with the 9.4 milligram per kilogram imetelstat dose improved clinical outcomes in patients with short telomeres and high hTERT expression levels. And again, that to me is very important because that again suggests that we're hitting what we should be hitting. And the results are consistent with telomeres biology in cancer cells. And this really provides very important evidence for on-target mechanism of action through telomeres inhibition. And this is the first clinical report to systematically evaluate the mechanism of action based on PD effect of imetelstat and its relationship to exposure and clinical benefits. And in my humble opinion, very biased humble opinion, I was really surprised this didn't get an oral abstract. And if you look at the oral abstracts in the myelofibrosis, I think that they were modest, in my opinion, and this was much more exciting and addressed an unmet need that was unfortunately ignored by the symposium leaders that decided on who was going to present. But that's just my opinion. I think that's it for me.

Thanks so much, John. I suggest we move to the Q&A session. I'm sure there will be a couple of questions for Dr. Mascarenhas as well.

[Operator Instructions] Tom Shrader with BTIG.

I have only one question. How do you use the drug? And what I'm getting at is of the patients on JAK inhibitors, how many of them are on JAK inhibitors simply because there's nothing else to put them on? Are they all getting benefit kind of till the end? Or would a drug like this increase your rate of taking people off a drug like ruxolitinib to get them the survival benefit. And as a slight follow-up, does fedratinib fit anywhere in that equation?

Okay. So thank you. I think that's a really important question. And Aleksandra and I, who spent a lot of time discussing this and planning the Phase III and sort of thinking this through have touched on these important points. So right now, patients almost across the board with Intermediate-2 and high-risk and even Intermediate-1 in some lower-risk patients are getting JAK inhibitors mainly driven by spleen and symptom, but in some cases, because they have the diagnosis, and there are no other approved therapies and in the community that just might be prescribed because it's prescribed, because it's available. I think the reality is, when you don't have other available options, you tend to lead patients on that therapy as long as possible until they have "failed it." And that's -- that definition is very different for different patients depending on the clinical context and the treating physician's perception and the patient's perception. In a clinical trial, obviously, you have to be quite rigorous and uniform in how you define things. But in clinical practice, I would venture to believe that if, for example, the Phase III study, which I'm very optimistic, will be a positive study. And if this drug was then available clinically, I don't think that the thought process for many patients or physicians for that matter, would be to wait for patients to really fall apart, which is what happens eventually. Like, for example, I saw a patient yesterday, who's been on ruxolitinib now for 6 years. And he really had a terrific response for 4.5 years of those 6 years. But his spleen is progressing, his symptoms are progressing. His counts are declining. I can see blast circulating. His disease is getting worse. He's either going to go to transplant or he's going to go on a clinical trial. If imetelstat was available, it would undoubtedly be my choice, even if he was going to go through a transplant, it would be a bridge to transplant. Even in the absence of a transplant, it would be the drug. But if I saw a patient starting to fail, if I had a clinically -- a commercially available drug, I wouldn't wait for them to fail worse because we know in this disease the longer you wait, the harder it is to salvage the situation and improve the outcome. So I think the clinical use would be likely broader than the clinical trial criteria. I hope that answers the question.

Does fedratinib matter? Do you believe the treatment data?

Yes. I think fedratinib matters in the sense that at least in the U.S. right now, it is available agent. The data really from JAKARTA-2 is -- supports the use of giving fedratinib as a second-line JAK inhibitor treatment for patients who are losing response to ruxolitinib. But I think it is a drug -- my feeling is fedratinib is a good drug. And perhaps if fedratinib got out before ruxolitinib, fedratinib would be first and ruxolitinib may be second. There are differences in tolerability. There's more GI toxicity with fedratinib that has to be considered when giving it to a patient. But I think the decision-making is going to be one of what is the treatment goal in any individual patient. So if I had an 85-year-old patient, where the treatment goal was to address maybe discomfort in the left upper quadrant, some splenomegaly and early satiety, that may be a fedratinib patient to use. But if I had, for example, the guy who I saw yesterday, a pretty robust 63-year-old, and now as I get older, I don't see 63 as old, I would treat that patient with imetelstat because the goal there is to try to have a more profound effect on the disease. One could always use or leave fedratinib for a later time point. I think that was one of the interesting things that we saw from the IMbark study was you don't need to necessarily get imetelstat endlessly. So for example, and Dr. Santini will appreciate this as well as Rami, when you give azacitidine for MDS, you continue to treat through the response and maintain the response. I'm not convinced you need to do that with imetelstat in myelofibrosis, at least for the endpoint of survival. So one could imagine that if -- for example, if cytopenias became an issue, and I think they're manageable with imetelstat. And two, if anyone is going to ask me the question, I'm going to answer it now. It is a drug that can be delivered in the community because these are therapies that community hematologists would be comfortable with on a 3-week to 4-week basis and manage the cytopenias. But let's say, cytopenias were to become problematic for any given individual, one could imagine then going on to fedratinib, an oral drug after that, potentially enjoying the benefit of the survival benefit and disease course modification benefit but then go on to get imetelstat -- to go on to get fedratinib. So I think fedratinib would be used in certain situations. But I think for many of the patients where survival is really the focus of ruxolitinib failure, I think imetelstat would be the choice.

Charles Duncan with Cantor Fitzgerald.

I had a couple of questions for Dr. Mascarenhas. Very interesting data for target engagement that supports differentiated mechanistic rationale for the activity of imetelstat but I'm wondering if you would see these biomarkers as useful in the clinic in the future for predicted response? Or is that even needed in 2 different patient cohorts that you outlined that may be best used or exposed to imetelstat? Those being patients who failed JAK inhibitors or those in which you're trying to debulk before transplant. Is this data that you'd like to see translated into, say, prognostic indicator?

That's an interesting question. I think to me, the data is most important because it supports the idea that the drug is hitting the target of interest and that we're correlating that with outcome measures. I think in -- if I were to see this drug approved and commercially used, I'm not actually sure I would want to advocate for biomarker-driven use of the drug because my fear would be, you may miss the opportunity to actually have an effect -- beneficial effect in patients that may not have the "biomarker of interest" because although the association seems to be more robust with short telomeres or hTERT expression, there is still the potential to have benefit from this drug, even if you don't have those clear biomarkers. So I think it's great in the context of developing the drug and proving a point. But I think a lot more work would have to be done in feeling comfortable as a commercial assay to drive use of the drug. So again, to make a parallel, perhaps unfairly in MDS. If -- we don't really use methylation status to determine whether we're going to treat with azacitidine. We don't use DNMT3A mutation status. We treat irrespective of that. So I don't think it would be incorporated in a commercial setting. But I think the Phase III study would also give us a lot more information, confidence and insight in those -- in the use or utility of those. And I don't -- sorry, just to answer the last part. If I were to segregate the patients who were simply trying to improve outcome versus a bridge. I don't think -- I don't necessarily think there would be a difference or utility incorporating biomarkers in those 2 settings.

So it seems like pretty provocative data has been shown thus far in IMbark. And with a differentiated mechanism and now this biomarker data that suggests that it's clearly the result of the drug. I guess I'm wondering why you think it wasn't selected for an oral presentation. Is there any bias against the mechanism or anything?

No. No, I think -- no, I don't think there's any bias. I think I would venture to guess, it's the same reason for any meeting for any abstract in which politics and friendships and global politics perhaps play a role. I don't think it's anything personal against Geron or the drug. Perhaps the selection committee was overwhelmed by COVID at the time. So it's too bad for them to -- that it wasn't an oral abstract. But most importantly, I'm forward-looking into getting the Phase III out there because I think that's where it's going to go. And I can tell you that there's true excitement regardless to whether it was an oral abstract. I've gotten a lot of phone calls and e-mails from physicians who are following this, actually even patients in the community that go online. And actually, it's interesting in the -- I presume the MDS community is the same. There's a lot of patient engagement, patient advocacy. So there's a lot of excitement in the patient community and the physician community, obviously, in the investor community to see this move forward. So despite the snub, which we won't take personally, I think that the abstract was well received.

Okay. So the debate, it seems like in the community regarding the value of the clinical outcome being improved survival versus symptom improvement is not really the driver in this case.

No I don't think so. And just to emphasize, we had a meeting, a patient-driven meeting with the FDA. I think it was either September, October of last year, in where stakeholders from pharmaceutical companies, clinical investigators, patients and the FDA were present. And the discussion -- a lot of the discussion evolved around -- revolved around what is meaningful to patients and what should be meaningful outcomes, particularly as it relates to JAK inhibitor failure. And if you go back to the transcripts of that meeting, what's really important to take away is that although spleen and symptom are important, they were endpoints that were created years ago, and I remember it well because it was created during the development of the registration study for COMFORT, and it stuck. And for JAK inhibitor-based therapy, it makes sense. But if you are now failing a JAK inhibitor, spleen doesn't always tell the whole picture for any given individual or even symptom burden for that matter. But survival is important and patients will tell you that. And physicians, obviously, want their patients to live longer. So to me, the debate really -- there really isn't much of a debate. I think patients want to live longer. And just to lastly reemphasize is that I don't mean to underestimate the benefit to spleen and symptom by imetelstat because 30% of patients hit that endpoint actually of 50% improvement, that's not shabby second line, even for a JAK inhibitor. And although 10% of patients hit the 35% spleen volume reduction, I think 25% or more hit 10% more reduction. And we now know that there's nothing magical about 35%. Even if you reduce the spleen by a small amount, patients feel benefit from that. So there is spleen and symptom benefit, but it's the survival benefit to me that's much more exciting.

Andrew D'Silva with B. Riley.

I just have one. Just really curious if you thought the enrollment criteria for IMbark favored patients with shorter baseline TL? Or is there a way to stratify the Phase III population anyway or have a patient population that more heavily favors a shorter length patient population? And then I was also just curious, was there any correlation with triple-negative patients in them having a shorter baseline TL?

So I think that the -- what's really important to look at for the population that went on to IMbark was these were truly ill patients. This was not a cherry-picked population. These were patients who had, a majority of which had a high-molecular risk mutation who were anemic, who had high DIPSS scores and these are the patients who don't do well. So I think that they went on the study because particularly at that point, there was a lot of excitement that this was a non-JAK inhibitor-based therapy and that the results of the original pilot study were very fresh in people's minds. So there was really a lot of interest in taking these patients that were very ill after RUX's failure, putting them on the drug. Now whether that by itself selects for patients that are going to be enriched for short telomeres, I don't know. I don't know the answer to that. I'm not sure. So I won't even guess with that. But I'll leave it to Aleksandra or other folks to speak about the association of telomere length and triple-negative because the triple-negative patients, if you recall, were really -- they're the bad players. We knew that before the study started. They have a median survival of about 3 years. They particularly do poorly, and they were also enriched in this study. And I think they were enriched only because those are the patients who do poorly and come off ruxolitinib more frequently and come off doing very poorly. And those patients did particularly well with this drug, in which we have no therapies necessarily for JAK2 -- for JAK2 wild-type, MPN wild-type, calreticulin wild-type patients. So that triple-negative population is also an area of interest that Aleksandra may want to speak about in terms of further clinical evaluation development and perhaps the association with telomere length.

Yes. Thanks, Dr. Mascarenhas. I would suggest that maybe we defer the triple-negative question and the length of telomere for after the presentation from Dr. Komrokji. If that's okay.

Okay.

Yes. No, that works for me.

Andrew, that will be okay? Okay. Sure.

Stephen Willey with Stifel.

Just a quick follow-up. So I guess, Dr. Mascarenhas, what's your thought on this notion of treating suboptimal RUX responders with add-on therapy, right? You gave the example of the patient who came in and they started to deteriorate at the 4.5 years. Is that the kind of patient that you would think about add-on therapy to RUX to try to salvage? Or do you think that, that patient is just too far along in this course of disease to be salvaged? And is this notion of suboptimal response? Is that something that you see playing out of the clinic within a year post RUX initiation? Like, is that for early salvage patients or late, like I asked before?

So yes, we do see suboptimal. So suboptimal really means you're not hitting "end points" that were regulatory end points. And it's a continuum. So you have some patients that are truly suboptimal. They still have a significant symptom burden. They still have a large spleen, and it's very unsatisfying for the patient and the physician. And those are patients in which I would say would fall in the camp of either switching therapy or adding on a therapy to see if you can salvage the response. And that adding on, one could envision, and one would have to, obviously, do a study to justify this, as one could envision adding imetelstat to ruxolitinib and considering that there may be overlapping myelosuppression, but that would be a very interesting combination. And right now, there are a number of different trials with either PI3 kinase inhibitors, BET inhibitors and BCL2 inhibitors, trying to capitalize on patients who may not meet the regulatory endpoints. But within those types of patients, there are definitely patients you see. And you see them within 6 months that have a totally inadequate response to ruxolitinib. And those are the patients where I'm seeing them. I start to already discuss and prep them and set expectations that if we're not having success, usually by 3 months, but definitely by 6 months with RUX, we have to decide for that patient if it's a quick move to transplant, if they're transplant eligible, if it's a quick move to an add-on strategy with an experimental drug or even a commercial drug, with typically experimental, or if this study allowed it would be a quick move to imetelstat. And I think it would be patient specific. It would be context dependent in terms of what's actually driving the patient's suboptimal response. But you do see those patients. And for a point of clarification, I think Aleksandra is going to speak to this when she discusses the Phase III study is we are looking -- forward-looking, we're looking at patients who are refractory, that are truly refractory. Because those patients, we think, are not really going to benefit from another JAK inhibitor. And that we could debate that. But I think that the idea is that we're taking the worst of the worst. The patients who don't really have benefit from JAK inhibitor saying rather than trying to add something to it or rather than trying to switch them for another JAK inhibitor in which the expectations would be low for a JAK inhibitor-refractory patient, we are going to try to give them a non-JAK inhibitor-based therapy because these patients are destined to do poorly.

Chad, Needham.

Thank you, Dr. Mascarenhas, for your presentation and insights and rest assured that your enthusiasm for imetelstat in MF is coming through loud and clear. I think we're asking the right clinical question and focusing on JAK failures with this drug right now, but just wondering about your thoughts on the prospect. Given that we have a differentiated mechanism of action that's disease-modifying whether ultimately, some point out in the future, you would see a use for imetelstat in earlier line patients, perhaps in combination with the JAK inhibitor? Or JAK inhibitors do work very well for a lot of patients, do you think they're too entrenched for that to ever happen?

No, I think there could be a world in which, let's say, if the Phase III study is a positive study, the drug is available commercially, even in the absence, not that I'm advocating for it, but even in the absence of the label, I would bet that in certain patients, you're going to see imetelstat use even earlier in the disease course, particularly in patients where the physician -- and I know, for example, Rami sees a lot of MDS and MF patients, where there's comfort level and understanding of the potential for imetelstat to address anemia in MDS and to me, there's a significant overlap in biology and even clinical features of the 2 diseases. Although there are clear distinct differences, there is similarities. So one could imagine evaluating this drug, and I would advocate for evaluating it formally in a clinical trial earlier on in the disease course because anemia in MF is also an unmet need in which luspatercept is also being explored and has Phase II data for 30% anemia responses in combination with ruxolitinib. But the majority of the patients don't get that response, and many of the patients lose the response eventually. And luspatercept doesn't affect the disease probably beyond erythropoiesis maturation. So I could see this drug being used earlier on. So for example, patients who present, maybe 25% of whom present for therapy and they're JAK inhibitor naive, in those patients, although there's data that ruxolitinib can afford those patients spleen and symptom benefit, we know often that the anemia gets worse. So one could envision a situation where the drug could be introduced earlier in the disease process.

There are no further questions for Dr. Mascarenhas at this time.

Thank you, Dr. Mascarenhas for now. We'll come back to you again with the last presentation that we have planned for today. But I'll suggest that we now move to our third speaker, who is yet another renowned expert in hematologic myeloid malignancies. You may remember Dr. Komrokji from last year KOL, just like Dr. Mascarenhas from the one in 2018. But Dr. Komrokji was presenting the collaboration project on the RWD analysis from the IMbark study last year. He is the section head of the leukemia and MDS at the Moffitt Cancer Center, it is one of the leading cancer centers in the U.S. We continue to work with him on multiple avenues in both myelofibrosis and myelodysplastic syndromes. He's representing the triple-negative abstract that we have for EHA this year. Rami, would you like to take over?

Sure. Thank you for the kind introduction. I'm glad to be back again. I was hoping it will be in New York live, but this is actually working nice. I'm going to focus particularly on the triple-negative. This question already came. So at this abstract, in EHA, the goal was to look at the subset of patients that are triple-negative. So this is from the same study, from the study looking at patients with triple-negative disease. Next please. I think Dr. Mascarenhas set the stage very nicely about the disease background, the imetelstat. What I'd like to cover is really a little bit more about that triple-negative patients. So that's a term we currently use for patients that lack evidence of any phenotype-driver mutation, the JAK2, the calreticulin or MPL. So we call them triple-negative. And that subset of patients have a worse outcome, unfortunately, with median overall survival reported in the range of 2.5 to 3 years and our database 30-month survival from time of diagnosis actually for those patients. There is a little bit of challenge in making the diagnosis in those patients. Sometimes they are missed. Sometimes patients with MDS with fibrosis or MDS MPN, there is a lot of overlap. So that group is really challenging in terms of diagnosis and management. As discussed, also allogeneic stem cell transplant is the only potentially curative option. However, it seems also that those patients that are triple-negative have worse outcome with transplant. So this is clearly a group with unmet need in terms of improving the overall survival for those patients as well as the response. Also SCARS data from JAK2 inhibitors, they don't do well with the JAK2 inhibitors, like the responses could be less, the duration of response could be less as well. So -- and in this aspect, we try to focus on the triple-negative patients and to look at the responses with imetelstat among that group. Next, please. And this is the study design already discussed. The only thing I would like to highlight or add probably that many of the studies you look at have included patients that are intolerant to JAK2 inhibitors, which in my mind, sometimes you see higher responses in those groups because that intolerance a little bit is more subjective. Those were truly relapsed or refractory patients. And that goes back to the question that was just asked. I think of patients into groups of primary refractory, no clinical benefit at all; loss of response, like -- or secondary failure; or intolerant. So those studies with imetelstat would really including patients that were either primary refractory or relapsed. And every time you include the patients that are intolerant, you diluted -- like, the responses would probably look higher. Probably those are the ones that are can be sometimes with optimal response. Next. So when you look at the study here, if you look at all the patients included, around 25% of the patients or 26 patients out of 105 patients were what we call triple-negative. So they lack JAK2, calreticulin or MPL mutation. And obviously, this is a little bit higher than what's reported in general in myelofibrosis, but that reflects the point Dr. Mascarenhas reported that the study was enriched by patients actually that are higher risk are not doing well in general. Next, please. So if you look at the response rates and to focus on the 9.4 milligram per kg dosing, if you look at the spleen response among patients that were triple-negative, it was around 19% compared to actually to 7% in the nontriple-negative. And the symptom improvement was seen in half of those patients with triple-negative. Again, so there were more clinical activities, spleen and symptom response in patients with triple-negative disease. Next. And what's very interesting to me is looking at the survival among those patients with triple-negative. So if you focus again on the 9.4-milligram dosing, the median overall survival for those patients with triple-negative after JAK2 failure was 36 months. And that was statistically better than the nontriple-negative, which was 24.6 months. Just to remind you, again, that even in the nontriple-negative, this is an improvement. So the benchmark we are looking at is really somewhere around 14 to 18 months survival after JAK2 failure. The nontriple-negative seems better, but what's interesting is that also the triple-negative is more. I just mentioned to you, typically, we expect 2- to 3-year survival with the triple-negative from time of diagnosis, not JAK2 failure. And our database is around 30-month survival from time of diagnosis. So this, to me, looks interesting, that the triple-negative even did better than nontriple-negative. Keeping in mind, again, that the nontriple-negative did better than the historical outcomes after JAK2 failure. Next, please. And in line to what was mentioned, that we think of fibrosis as a surrogate marker for the underlying biology of the disease. And there has not been many agents that change the fibrosis on the bone morrow and some of the new models, like the MIPSS70+ incorporated the degree of the fibrosis as part of -- as a prognostic factor. So when you look at those patients that have triple-negative disease, you can see that the grade 3, which is the most severe or extensive grade of fibrosis, the rate was higher at baseline, 94% versus 66%. But when you look at the improvement, which is defined in the study as a grade or more, 50% of the patients in the triple-negative compared to 39% had a grade or more improvement in the fibrosis. So again, we rarely see fibrosis improvement anecdotally with the JAK2 or like in small subsets with the JAK2 inhibitors after a very long follow-up, you see some improvement. But we tend to think of it as a surrogate marker for changing the underlying natural history of the disease or the surrogate marker of biological response for the treatment. Next, please. And again, this was covered, but when you look particularly at patients that are triple-negative, you do observe that those patients had shorter telomere length at baseline, 79% versus 46% as well as higher level of hTERT at baseline. Again, maybe explaining a little bit more why we are seeing higher responses or better overall survival because, as Dr. Mascarenhas just nicely showed that there is correlation with the PD effect with the overall survival in those patients. Next, please. So I think, and in conclusion, that we see in this study that the patients that were triple-negative did much better than we expect after JAK2 inhibitors with the imetelstat. The study included around 24% of patients that were triple-negative. We see a survival signal of 36 months, again, compared to 2 or 3 years from baseline of diagnosis. This is -- was after JAK2 inhibitor. We see higher responses among patients that were triple-negative correlation with some higher grade of fibrosis, shorter telomere length and higher hTERT expression at baseline. And some improvement in the fibrosis among those patients. So definitely, it was on further investigation and looking at that subset of patients that actually probably may be driving less benefit from the JAK2 inhibitors from the get going. So I'll stop here, and I'd be happy to take any questions.

[Operator Instructions] Andrew D'Silva with B. Riley.

You did answer my last question there at the end. But I'm actually just curious, so in IMbark, it was noted that it was a little bit more heavily weighted to triple-negative patients than necessarily the typical MF population would be. As you look at the enrollment criteria now for the Phase III trial, it is slightly different than IMbark. Could you maybe discuss your thoughts on that? And if you think it would even more heavily favor triple-negative patients than IMbark would have?

Right. So I think that it will be part of discussion of the Phase III design, and it's not going to be obviously exclusive for the triple-negative. So it will involve -- include those patients. And in general, if you look at all myelofibrosis patients, the triple-negative patients are around 10% of the myelofibrosis, and the study was 24%, partly, as we mentioned, because probably, those patients don't do well on JAK2 inhibitors. There is really not much data on the responses with the JAK2 inhibitors in those patients or the duration of response. As I mentioned, the benchmark, the survival for those patients is somewhere around 2 to 3 years from time of diagnosis. I can tell you in our database that the responses to those patients with JAK2 inhibitors are less than the nontriple-negative. And the duration of response also could be less. So I think naturally, anytime we're going to look at refractory or relapse JAK2 population, we are going to see some enrichment. I don't think we should restrict the study to that group particularly, because I think there will be a survival benefit probably also in the nontriple-negative, and that's the majority of the patients. I do think that the triple-negative could be an opportunity to think, answering an earlier question, moving this earlier even in the course of the disease. Let's say, the group that has only 2-year survival advantage. So you could think in the future of looking at those patients upfront, particularly if the responses to the JAK2 inhibitors are low. And in my experience, those patients also tend to have more cytopenia that sometimes precludes use of JAK2 inhibitors. So I would envision down the road a study in triple-negative patients as an upfront strategy even before JAK2 inhibitor. But I think, we'll cover a little bit more of the design of the Phase III. And by nature, there will be always some -- a little bit more enrichment than the baseline 10% in those studies because those patients don't do well with the current standard of therapy at all.

I was just wondering could John, because you'll be the PI of the study and also has been working very closely with us in the design, if you -- John, you have anything to add to this question maybe.

I mean, I absolutely agree with Rami, and I just want to stress the point that what was, at least for me, a somewhat unexpected finding was that this really poor prognostic molecularly defined group actually did well with this therapy, but not to suggest, as Rami pointed out, that the nontriple-negative patients didn't. So the Phase III study will take refractory patients in which my assumption, as Rami put out, it will incorporate or include triple-negative patients. But the benefit of this drug, I think, is irrespective of the mutational status of the patient. But encouragingly in the worst of the worst, you get responses that are utterly unexpected and without any other obvious therapies. So I would look forward to seeing what that subset that gets enrolled in the Phase III study does in terms of their survival outcome. But I'm looking forward to enrolling patients regardless of their mutational status, that meet that refractory definition that Aleksandra will go over at the end of this discussion because it's really those patients that really are this unmet need that are probably the worst players and are probably most likely to see the benefit, particularly in the randomized fashion of the study drug.

[Operator Instructions] Stephen Willey, Stifel.

So maybe just to follow up or to clarify, and I apologize if I missed this, but -- so the triple-negative genetic signature that was shown in the presentation and the correlation to shorter telomeres and higher hTERT expression, is that a consequence -- is that a fundamental consequence of mutational burden in these patients? Or is that just a consequence of the fact that there was a small number of these patients in the Phase II?

I don't think we know exactly. I think what we know is that, obviously, those patients lack the phenotypic driver mutations, they tend to have worse outcome. It seems that there is some clinical features for them. They tend to be more cytopenic, more anemias, more leukocytosis. They do have more severe fibrosis on the bone marrow. And on this study, I think this is the first time that we actually -- people have looked at the baseline telomere length and telomerase activity to my knowledge in that subset and it seems that they do have like shorter telomere length, higher hTERT activity and that correlates with the data that Dr. Mascarenhas presented, showing that there is some good correlation between hitting the target PD and the overall survival. So it makes sense that this population will drive more survival advantage because it seems that maybe the telomere length and hTERT activity is one of the main biologic derivative of the disease in the triple-negative. And to be honest, in my mind, that group is sometimes very difficult to diagnose. That will mean that the implication for this for me could extend beyond this because we have a large group of patients, we call MDS with fibrosis that really share a lot of features of this triple-negative. That -- actually, some people think that triple-negative should be a separate entity by itself. There is a group of patients that have MDS MPN with some fibrosis. And I think it will be important that we go and start looking at those patients, MDS with fibrosis. So you look at the telomere activity and the hTERT, and that may open the door for a group that also has unmet need for treatment, MDS MPN, the same. So for me, that's the interesting part. That category is evolving. But now we are starting to see that there could be different biology there. At least on this study, when you look at the baseline mutations other than the phenotype driver mutation, there was no difference in the triple-negative versus the nontriple-negative. But there is not much there, and we are trying to look at that actually as we speak, so.

Okay. That's helpful. And then just lastly, do you have information on the duration of prior RUX therapy in these triple-negative patients prior to enrolling into IMbark?

I don't recall the number, but it was the same. Probably, there was no difference in the baseline characteristics between the triple-negative and the non. I don't remember the exact duration, but it would probably be somewhere 2 years something plus. But Aleksandra can correct me, but there was no difference between triple-negative and nontriple-negative in the duration of exposure.

Right. So it was approximately 23 months for both triple-negative and the nontriple-negative patients of the study. So yes, Rami, you're right.

Great. I think that helps to put some of the OS data you're seeing in perspective as well.

Valeria, you were trying to ask something. I don't think we heard you. So -- I think the voice has gone.

Dr. Santini, are you on mute maybe? We cannot hear you. Maybe the moderator can unmute Dr. Santini. No?

Hello. Do you hear me?

There you go. Yes.

Okay. Okay. No, it's just that I agree completely with what Dr. Rami just said. We have urgent need and a very unmet need with -- in MDS with fibrosis. These patients are more than we thought about. So -- and these patients share a lot of features with MF, as it has just been said, but we do not have any therapy really active in this subtype of patients. So HMAs are not really active when you have high degree of fibrosis. So we are left with almost nothing. And therefore, imetelstat, especially for the data we just saw, it's highly suggestive of working in MDS with fibrosis as well. I would be really happy to use in these patients and see how the burden of disease and how the clonality goes down in this very, very difficult subtypes of patients. So very well. Think about it, Geron. It's really something we do not have any trial going on for. Thanks.

Thank you, Dr. Santini for that commentary as well.

Charles Duncan with Cantor Fitzgerald.

This question maybe for Aleksandra or going back to Dr. Mascarenhas as PI for the Phase III trial. I guess I'm wondering if you would look for in the conduct of the study or a statistical analysis plan there to be at least a secondary analysis that would pull out the triple-negative patients from the overall cohort. And then what percentage of the patients in the sample would you anticipate to be triple-negative? I imagine it's consistent with population generally, but what would that be?

Yes. I think the percentage of triple-negative is going to be the same, probably 25% or so or 30%, which will also still allow us for it to look at the survival among that group. As I mentioned, as baseline upfront, newly diagnosed patients is 10%, but any study after JAK2, it would be higher, so.

And I can just add to that, that we would definitely prespecify that analysis in the statistical analysis plan for the study to look specifically for the triple-negative patients within the overall study population of refractory MF.

Okay. That's helpful. So Dr. Mascarenhas, you assume that it's post JAK inhibitor, it may be around 25% or so?

Yes. I agree with Rami. I think we're going to see similar numbers in the Phase III study because these are the patients that unfortunately don't fully enjoy the benefit of JAK inhibition and are more likely to "fail" or be refractory and are in need of a disease-modifying therapy. So my guess is it will be very similar. Just to reemphasize, when we enrolled to IMbark, we weren't enrolling based on molecular features. So that was really just what came through the door. And I suspect it will be the same population going into the Phase III.

There are no further questions at this time.

All right. Thank you. Then John, I'll hand it back to you, again, to give an update on the overall survival data and the correlations with other clinical benefits from IMerge -- sorry, from IMbark, I apologize, for the MS study.

Thanks, Aleksandra. So I'm aware of time, so I'll keep this moving. So here, we're going to talk about favorable overall survival that correlates with other clinical benefits, which I tried to allude to in the discussion previously. Next slide. Everyone here knows the study schema. There's really nothing more to say about the study schema. Next slide. Everyone knows the drug is imetelstat, and we're excited that we're hitting -- we seem to be hitting the target that seems to correlate with disease outcomes that are important, like spleen symptom and overall survival. We know that patients do poorly when they fail ruxolitinib. And we know that the improvement in overall survival is seen with the 9.4 milligram per kilogram on. And it is clearly better than what you would see from real-world data controls for multiple studies, including Dr. Komrokji's study. So here, we're going to evaluate the association between overall survival and spleen volume reduction at week 24 as well as total symptom score reduction and fibrosis improvement. Next slide. So here, we're showing the overall survival at the active dose, 9.4 milligrams per kilogram. This is the updated analysis. And you can see that the median overall survival, when you compare the 2 arms, in the low dose is 19.9. I still think that's better than what you would see normally with patients who come off ruxolitinib versus 28.1 months in patients who got the active 9.4 milligram per kilogram. If you look at landmark time points, like 12 months, the survival difference is 78.6% versus 84% and it -- perhaps even more impressively at 2 years is 42% in the low arm versus 57.9%. So again, if you were to sensor, and we did this in the initial analysis, we used sensor for patients who went on to get transplant or other subsequent therapies, including JAK inhibitors, or patients who were on low dose who had the opportunity to dose escalate to 9.4 milligram per kilogram, it didn't change the fact that the survival benefit was different -- was improved with patients who got the high dose. Next slide. So here, we're pairing up everything. And the point of this slide is to show -- is to impress upon you that there is a benefit across the board when looking at any one of these rows in the high dose versus the low dose. So if you're looking at survival, again, 19.9 versus 28.1 months. If you're looking at symptom response, 6.3% versus 32% symptom improvement. And again, I'll just reiterate, the 32% gives us 50% reduction is quite good actually for second line, even for a JAK inhibitor. So that's not to be minimized. And then spleen responses were not there at the low-dose arm, but in 10% of the patients, they had a 35% spleen volume response. The progression-free survival was also better, 14.8 months versus 20.7 months. And then if you were to use strict IWG criteria for clinical improvement, which includes symptom, anemia and spleen, it was 16.7% versus 25.4%. If you look at transfusion independence, not the primary endpoint or goal in this study in myelofibrosis of 12 weeks or longer, it was 14.3% versus 25% and then reduction in bone marrow fibrosis, which is -- which was done centrally is at least 1 grade reduction we've seen in 20% versus 43% in the high dose. And then importantly, and I stressed in the previous presentation, allele burden reduction of 42% in the high dose versus 5% in the low dose. So in every aspect, no matter how you cut, slice or dice the data, there seems to be a dose-dependent benefit in the high dose. Next slide. If you look at overall survival and you look at the effect of degree of fibrosis improvement. So going from at least 1 -- having at least 1 grade reduction in fibrosis versus a worsening of fibrosis, there was an improvement in survival, reduction in hazard ratio, for death. That was statistically significant. And although not statistically significant, still there was a trend to improve survival in those patients who had stability in fibrosis versus worsening of fibrosis. Again, making the point and what Rami has shown that affecting fibrosis of biomarker just like affecting JAK2 mutation, allele burden also can correlate with improved -- correlates with improved outcome of overall survival. Next slide. If you look at symptom improvement on the top, if you have a 50% or greater symptom improvement, again, appears to correlate with improvement in survival. And that also makes sense to me because symptoms is also mediated by cytokine expression and patients who feel better tend to do better overall. And whether you have a spleen improvement, and here, it's broken down by degree of spleen improvement, and you can see there's a stepwise correlation, improvement in correlation with survival. If you have a 35% versus not versus the 20% versus not versus even a 10% versus not, you have the plot on the left side of one which would suggest that there's a survival benefit in all these cases. Next slide. If you look at clinical variables that we know are associated with core prognosis, so for example, having a higher risk DIPSS score, having higher neutrophil -- baseline neutrophil count, having anemia versus not having anemia, having a lower platelet count versus a higher platelet count, being transfusion-dependent versus not being transfusion dependent, what kind of response you had to a JAK inhibitor previous -- prior to imetelstat across the board, it favors, you have a survival benefit, irrespective of those clinical variables that are associated with core outcomes. So it wasn't like there was 1 patient characteristic that was driving the response with imetelstat. It was seen nicely throughout all the different clinical variables. And I think that slide is a very important one to consider because it gives me confidence going forward in the Phase III study that we're not missing some subset of patients that the Phase III will exclude for some unknown reason that this is going to be -- this should be a benefit that you're going to see no matter how you look at the patients or how you divide them by risk. Next slide. So in conclusion, imetelstat shows dose-related improvement in survival in patients who relapsed or refractory JAK inhibitor, that survival benefit is observed and is supported by trends of correlations with other clinical benefits like spleen and symptom, like allele burden reduction and fibrosis reduction. And with a median follow-up of 41.7 months, the median overall survival is 28.1 months in the 9.4 milligram per kilogram arm versus 19.9 months for the 4.7. I'll just remind you again, 12 to 14 months in multiple historical control, so much better. And among 57 patients across both treatment arms that had matching bone marrow samples, 20% -- 20 patients, which is 35% had 1 degree or better bone marrow fibrosis improvement while on study and had a significant longer overall survival than those who had worsening, again, correlating biomarker with an important outcome, like survival. And a similar trend was seen in patients who had stable disease versus in worsening fibrosis. Patients who achieved symptom and spleen response in 24 weeks also showed a trend for longer overall survival compared to patients who did not. And pretreatment DIPSS score, ECOG Performance Status, transfusion dependence, response to the last JAK inhibitor, baseline neutrophil count, lower hemoglobin and platelet values, all poor prognostic variables correlated with increased risk of death. But didn't drive -- not one of these particularly drove the outcomes that was seen across the board. And that's the end of this presentation.

Do we have any questions for Dr. Mascarenhas?

[Operator Instructions] Tom Shrader with BTIG.

You made a comment that kind of piqued my interest. Do you value more data at the lower dose, so you have some confidence if you have to back off the dose in a patient?

Well, I think -- so if you look at spleen, symptom and survival, the high dose wins on all 3 endpoints. But what has never escaped me, as you're seeing, is that even at the low dose, you are going to get some on-target activity engagement and there does seem to be a survival benefit that is better than what you see with anything else or patients who historically come off ruxolitinib, it would suggest that if you run into cytopenias, if you have to dose modify, it gives me some degree of confidence that there's still activity that could be obtained and/or we have data that patients who started on lows can go up to the high dose and enjoy benefit as well. So I suspect that there's some degree of activity for -- even with the low dose, but clearly, the 9.4 is the winning dose. And if one were to think it as a controlled study where the control was the low dose, which I actually don't think it's an inactive control, I always thought that one could almost look at IMbark as a randomized controlled study in some ways because the low dose could be the control arm. And there was a clear difference in survival, but there was probably some survival benefit with the low dose. So yes, the answer to your question, and my long-winded answer is, yes.

There are no further questions at this time.

Okay. I suggest we go once more through the planned Phase III trial and the study design, I think that's also important. So what you see here on this slide is the scheme, a schema for the study, which will be led by Dr. John Mascarenhas, as discussed, but we're also joining forces with Dr. Verstovsek who will be also a co-PI on the study. It is a study that will be conducted in refractory myelofibrosis patients and these are defined as patients that have inadequate spleen or symptom response after being treated with a JAK inhibitor for at least 6 months. And within the 6 months, they have had a treatment with an optimal dose of a JAK inhibitor. The patient will be randomized to receive either imetelstat at 9.4 milligrams per kilogram on a 3 weekly basis. And we see the dose that John and actually Rami were talking that we are taking forward based on the data we have from the Phase II study. The control arm will be best available therapy, which will have to be predefined prior to the study entry for each of the patients. And it is important that it will be excluding JAK inhibitors. The importance of this is the fact that we will be able to hopefully closely match to the historical data or, for example, the data that we have from the RWD analysis where the expected outcome for patients in this setting that have discontinued JAK inhibitors between, I know, is approximately 14 months. The study is designed such that it will have an interim analysis. And then obviously, a final analysis for a primary endpoint of overall survival, both the interim and the final analysis will be looking at an overall survival. They are event-driven, obviously. So it is, at the times that you're seeing here in terms of readouts, our estimated timings. The -- and with the assumptions that we've put forward, right, for the overall survival, we expect to have approximately 320 patients enrolled. John, anything that you would like to add maybe from your point for the study design?

The only thing I would add is that this was really, I think, really well thought out. And I think the patient population is the right patient population. I think the control arm is the right control arm and this was discussed with the FDA, and I was really happy to see that they totally agreed with this because this is a patient population that has -- it is refractory to JAK inhibitor. And although one could imagine maybe getting a JAK inhibitor after imetelstat at some point, I would much rather try to treat them with a non-JAK inhibitor-based therapy. And I think the dose we're going with is the right dose because what -- maybe I didn't stress before is it was the more active dose that won on all 3 endpoints, spleen, symptom and survival, but also there wasn't really more toxicity that led to worsening clinical events with it. So from an efficacy and safety perspective, this is the dose. So I think I'm excited with the study design because I think it's the right study design. And it's the first study we've seen where survival is the endpoint, which is really the unmet need.

Thank you for that. Are there any questions?

We have a question from Charles Duncan with Cantor Fitzgerald.

A quick one on the conduct of the study and then on the interim analysis. First of all, with regard to the study conduct, do you accommodate any dose holidays? Or how will you accommodate any dose holidays or reductions during the course of the drug exposure?

So there will be -- and I guess, this is a question for me, Aleksandra. So we will have a clearly defined dose modification scheme, which will be -- it's, I would say, driven from the knowledge and experience that we have from the IMbark study. So obviously, we will try to -- or we are taking into account and trying to have patients on treatment as long as they can tolerate and they see efficacy from the -- or benefit from the trial treatment.

Yes. So I guess what I'm saying is, over the course of the administration, would you ever anticipate patients being exposed to the lower dose, as kind of referenced in the previous presentation?

I understand now your question. Actually, in the myelofibrosis study, the starting dose is 9.4, and we will allow 2 dose reductions. So the lowest dose that the patients will be receiving on the study is 6 milligrams per kilogram, actually patients will not need to go down to 4.7. So we'll clearly have achievement with an active dose of the drug on the study.

Okay. That's helpful. And then with regard to the interim analysis. I think we've discussed this before, but I can't remember. And that is what is the approximate or the actual number of events that you're looking for to trigger that? And then could you share with us either the specific number or kind of what you would look for out of that interim analysis in terms of superiority. I imagine since it's an interim analysis, it requires very high bar in terms of the P value. But can you give us some additional details?

Sure. I can do that. So I mean, for the final analysis, right, and typical for a study with an overall survival as an endpoint, you would like to have more than 50% of the patients died on the study, right? So you would like to have more than 50% of the event to occur when you perform the final analysis. Moving to the interim analysis. So we want to make sure that we have enough of these events such that we have enough power to detect the difference. So at the interim analysis, we plan to have approximately 70% of the events that are required for the final analysis. And so still a good number of events will be, right, needed for the interim analysis. Of course -- go ahead.

So my math is 7x5, 36% -- around 36%, 35-or-so percent of the total events that you'd expect.

Right.

Okay. Sorry, again.

No, no, that's okay. That's okay. And I mean in terms of a hazard ratio, right, of a P value, we would expect to have, again, what is important is to have statistically significant difference between the 2 arms. And again, we have taken the hazard ratio of 0.6 in terms of assumptions, right? But it doesn't mean that we necessarily need to hit a hazard ratio of 0.6 to have a statistically significant difference between the treatment arm that's -- to have an indication for a successful study.

Stephen Willey with Stifel.

Just curious with respect to the secondary endpoint of spleen response. Is that going to be prespecified as an SVR 35%? I know that there's some attempt here, I think, from a regulatory perspective to try to bring that threshold down to maybe an SVR 20% or maybe an SVR 10%. So just curious how you're going to be prespecifying that as a secondary.

Sure. I mean, a spleen response, as we now know of and it's per the IWG criteria, it's more than 35% in volume reduction. So this will be a second -- I mean this will be a secondary endpoint. But needless to say, we have prespecified analysis with different cutoffs of 20% as well as 10% because we believe that's very important, especially in the refractory setting, as John was alluding for. I mean we did have 37% of the patients who had the 10% reduction in SVR. So we are definitely looking in different cutoffs.

I will now turn the call back over to Dr. John Scarlett for final remarks.

Well, thanks, everyone. As we close this event on behalf of the company and all of our employees, I'd like to sincerely thank Dr. Santini, Dr. Mascarenhas and Dr. Komrokji for their presentations of what at least we found to be extremely exciting data coming out of this year's EHA. All 3 of these really outstanding clinicians and investigators have made significant contributions to the development of imetelstat, and we look forward very much to working closely with each of them and their colleagues in the future. Needless to say, these data presented today continue to encourage us that imetelstat has the potential to be a very important treatment option for patients with the disorders described by our presenters. As the sponsor of the drug, we remain committed to developing imetelstat in myeloid heme malignancies, and we now have really all the pieces in place to move forward. We have a novel drug and unique target. As described today, it's showing very meaningful clinical activity in indications where there are significant unmet needs. We have 2 Phase III trials, one that's ongoing in lower-risk MDS and one in the refractory MF, which will be in the start-up phase, and we hope to begin enrollment in the first quarter of 2021. And finally, I have to say we have a very experienced team who can collaborate with these outstanding clinicians in executing all aspects of our drug development. And now we have as well the financial resources to reach significant value inflection points for patients, investigators and stockholders. So we really look forward eagerly to the coming year. And I want to thank everyone again for participating in this event, and I hope everyone has a great day. Thank you.

This concludes the call for today. We thank you for your participation. You may now disconnect.