Curis, Inc. (CRIS) Earnings Call Transcript & Summary

Good morning and welcome to Curis' virtual KOL event, discussing updated data for CA-4948 in AML/MDS, presented at EHA. [Operator Instructions] Please note, this event is being recorded. I would now like to turn the conference call over to Curis' Chief Financial Officer, Bill Steinkrauss. Please go ahead.

Thank you, and welcome to Curis' KOL event to discuss the updated clinical data from the Phase I/II study of CA-4948 in AML and MDS, as presented at the European Hematology Association's 2021 Virtual Congress today. Before we begin, I would like to remind everyone to go to the Investors section of our website at www.curis.com to find our data release and the slides from this presentation. I would also like to remind everyone that during the call, we will be making forward-looking statements, which are based on our current expectations and beliefs. These statements are subject to certain risks and uncertainties and actual results may differ materially. For additional details, please see our SEC filings. Joining me on today's call are Jim Dentzer, President and Chief Executive Officer; and Bob Martell, our Head of R&D. Along with special guests, Dr. Guillermo Garcia-Manero, Chief of the section of Myelodysplastic Syndromes within the Department of Leukemia at the University of Texas, MD Anderson Cancer Center, and a lead investigator on the study. We will be available for a question-and-answer period at the end of the call. I'd now like to turn the call over to Jim. Jim?

Thank you, Bill. We're excited to be here today to discuss updated clinical data on our first-in-class IRAK4 inhibitor, CA-4948. In December 2020, we presented encouraging preliminary data showing marrow blast reductions in 6 of 6 patients, with 2 of the patients achieving objective responses. Earlier today at the EHA conference, we presented new data, including an update for evaluable patients with elevated blast counts at baseline. This update showed marrow blast reductions in 10 of 12 patients with 5 of those patients achieving objective responses as well as early indications of partial or full heme recovery. This is exciting news on 2 fronts. It increases our confidence in our long-term commercial strategy to develop CA-4948 in combination with other agents in earlier lines of treatment. And second, it increases our confidence in our regulatory strategy with 2 paths for potential accelerated approval in monotherapy. The first path is in late-line patients with spliceosome mutations. The second path is in late-line patients with FLT3 mutations. In both cases, genomics analysis indicates that CA-4948 is not simply helping patients achieve an objective response, it is fundamentally modifying their disease. Given that there are no currently approved therapies for these patient subgroups, we are excited about the potential to address this sorely unmet need. In short, we believe these data further support CA-4948's potential to be a novel, first-in-class and game-changing therapy across the entire spectrum of AML and MDS. With that, I'd like to turn the call over to Dr. -- excuse me, Dr. Garcia-Manero, who will walk us through some of the key details from today's EHA presentation. Guillermo?

Thank you, Jim, very much for this opportunity and for the audience to being here. I understand this presentation was released today. My name is Guillermo Garcia-Manero MD Anderson Cancer Center in Houston. Next slide, please. So I'm going to walk through the slides that were pre-recorded for the EHA meeting this year, and then I think the question period will be actually next week. So let's go through this presentation. This is, again, a Phase I dose escalation trial of a novel oral IRAK4 inhibitor CA-4948 for patients with AML or myelodysplastic syndrome, again, an interim report. And I'm giving you this data also on behalf of my collaborators in North America and in Europe. Next slide, please. These are my disclosures. Next slide. So Interleukin-1 receptor associated kinase 4, or IRAK4, plays an essential role in Toll-like receptor and Interleukin-1 receptor signaling pathways. And this is actually shown on the right side of the slide how this TLR signaling goes downstream into NF-kappaB activation, a fundamental aspect of leukemogenesis. These pathways are frequently deregulated in non-Hodgkin's lymphoma and, of course, AML and myelodysplastic syndromes. It is also important to realize that there is an oncogenic form of IRAK4 that we call L for long isoform, that is frequently driven by spliceosome mutations that is preferentially expressed in over 50% of patients with AML or MDS. And then activated IRAK4 has been identified as a driver of adaptive resistance in acute myelogenous leukemia. Next slide, please. So CA-4948 is a novel small molecule IRAK4 inhibitor. You see the structure on the right. It is a first-in-class inhibitor. It inhibits hematologic malignancies that are driven by overactivity of this Toll-like receptor IL-1 pathway, which, as we said earlier, depends on IRAK4. Of note, CA-4948 also inhibits FLT3 [ mutate ] at AML, both in vitro and in vivo. The data indicated a high binding affinity to IRAK4, around 23 nanomolar and 2 FLT3, as you see there 31 nanomolar. Preclinical and clinical events showed no myelosuppression DLTs, excellent oral viability and a dose-dependent PK with clear pharmacodynamic correlates. And as safe RP2D dose of around 300 milligrams PO BID was identified in patients with relapsed or refractory non-Hodgkin's lymphoma. Next slide. The next couple of slides showed some of the preclinical data supporting the use of this compounding human clinical trial. So on the left here, you see an experiment where IRAK4 is basically blocked using an shRNA type of approach. And you see basically that there is a very significant reduction on leukemia clones. On the right, you see kind of the same experiment performed with CA-4948. On top, you see control; on the bottom, you see with the actual compound. And if you focus on the box on the bottom right on that flow diagram, you see a very significant reduction on leukemic blast, and this is data from patient-derived xenografts. Next slide, please. This is additional data. First, on the far left, you actually see some data suggesting that perhaps there is differential level of activity for cell lines that are expressed in a long isoform of IRAK4. You see that, for instance, comparing normal CD34 on the top left and then follow the graph on the bottom right, you see THP-1, TF1, MDSL. On the middle cartoon, you see, for instance, the effect of CA-4948 in maintaining normal spleen size that is a surrogate for leukemia burden in this experimental xenograft models, using the THP1 cell line that actually contains the IRAK4-L isoform. And then you see the spleen size of the control animal. And then on the right, what happens to the spleen size, that is basically controlled once those animals are actually [ treated with ] CA-4948. And then on the right, you see kind of a prevention type of experiment for those animals that were xenografted where actually there is a significant less amount of engraftment for those patients that were treated with CA-4948. On the bottom of this graph also, you see the structure of the long versus short isoforms of these genes. Next slide, please. So with all this information, we designed this Phase I multicenter single-arm clinical trial of CA-4948 as monotherapy in patients with AML or high risk MDS. The drug was given continuously on a 28-day cycle. You see the registration number in clinicaltrials.gov. We plan to study these 4 doses ranging from 200 to 500 milligrams orally, twice a day. As in any Phase I trial, the primary objective was to determine the safety and RP2D dose. Secondary objectives were pharmacokinetic analysis. And then to have a hint on the initial efficacy, including overall response rates for those patients that were evaluable in the study. And then exploratory objectives, of course, pharmacodynamics and biomarker analysis related to the mechanism of action of this compound. Next slide, please. This is the patient eligibility for the study. So inclusion was high-risk MDS or AML. Patients had to be at least 19 -- 18 years of age with, an acceptable performance by ECOG Score. And then all patients had to be relapsed/refractory, failing at least one line of standard therapy. We excluded patients with acute promyelocytic leukemia, patients with active CNS leukemia involvement. We did not allow patients with blastic phase of chronic myelogenous leukemia. And then at least 60 days from prior allogeneic stem cell transplant or presence of active significant graft versus host disease. Next slide, please. So at the meeting, we presented data on the first 22 subjects. You see the distribution by dose, level 3 at 200 BID, 6 at 300 BID, 10 at 400 BID and 3 at 500 BID. As you see in this consort diagram, at the end, we were able to present data on 17 patients that were evaluable for response. And the cut-off of this data was April 30. You see also the demographics of the patients, and I think this is a very important data to understand the rest of the presentation. So the median age of these patients were 74 years of age, with a patient as old as 87. Half of the patients had AML, half of the patients have myelodysplastic syndrome. And then what I think is noteworthy is that 64% of these patients received azacitidine, 32% decitabine, 14% cytarabine. And then very importantly, 10 of these patients or 45% received venetoclax. So this is a group of patients with highly pretreated disease, mainly consisting of hypomethylating agents with or without BCL2 inhibitor, venetoclax. You there see the distribution of cytogenetics and then some of the hints in terms of molecular alterations included FLT3, and splice mutations such as SF3B1. Next, please. So the first step is to evaluate the toxicity of this compound. This is looking at treatment-related adverse events in at least 2 patients. And this table shows this data by the dose level in different cohorts. So as you can see in the table, a 200 milligrams BID. There was no grade 3 or 4 toxicity. At 300 BID, the same. We did not observe in grade 3 or 4, toxicity. At the 400 BID, there was a patient with presyncope. It was unclear, this was related to study drug. But at 500 BID, there were 2 patients with DLTS, one with grade 3 rhabdomyolysis and 1 with grade 3 syncope. All of this resolve very quickly without sequelae. Next, please. In terms of the PK profile of this drug on this patient population, the half-life was around 6 hours, was rapidly absorbed with maximum plasma concentrations observed between 0.5 to 3 hours post dosing. CA-4948 exposure levels were not altered in the presence of a strong CYP450 inhibitors, such as antifungal azoles. And we also saw a dose proportional exposure with minimal or no accumulation with continuous BID administration. And you see the pharmacokinetic profile on the graph on the right. Next, please. This is a data indicating basically a swimmers plot of treatment duration and patient response. So on the column on the left, you see the disease, AML versus high risk MDS. Next, second column, you see the response that these patients achieved. The colors indicate a different cohort of those level and then whether they are still on a study or not. So basically, you can see that at the time of this presentation, there were already patients on this compound that had achieved a CRI or a stable disease that were over 200 days on a study. Next slide, please. As was mentioned earlier by Jim, we -- so in this early Phase I clinical trial, significant reduction of marrow blast in 10 out of the 12 patients that were evaluated here. And we divided these plots, whether they were AML or high risk MDS. So on the left, you see the kind of trajectory in terms of blast percentage on those patients with acute myelogenous leukemia. On the intermediate cartoon, those with high-risk MDS. And then you have basically the data presentation, actually with the exact numbers per cohort level at 200, 300 and so forth and with the actual number of blast. Next, please. I believe, Jim -- and I apologize about this, that I may have actually skipped because of small -- the actual responses. Do you mind going back? I apologize about this. Go back. Here. I'm sorry, on the short/long side, I missed the key part. So if you look at the right, we compute a best response. So on top, you have the high-risk MDS, 9 patients and then 8 patients with AML. And then again, in this early Phase I trial in patients with high-risk MDS, 1 achieved a complete response, 2 patients achieved a marrow CR, 4 patients had a stable disease, and 2 patients had progression of the disease. In those patients with AML, we documented patient with a CRi, 1 patient with a PR, 3 patients with a stable disease and 3 patients with a progressive disease. Sorry about this. We can continue now. Thank you very much. So this shows the blast reduction. And then here, we give a little bit of a deeper dive in terms of these durable responses that we have seen in this high-risk group of patients, and it gives you an idea in some detail in terms of the cytogenetic and molecular characteristics. I'm not going to go through all of them. I believe Dr. Martell is going to do a little bit of this later. But it's interesting, for instance, to highlight patient #1, patient with therapy-related high-risk myelodysplastic syndrome. That's for the mutations that this patient had with an U2AF1 mutation, one prior line of therapy with azacitidine, achieving marrow CR with this compound. Second patient with secondary AML with an SF3B1 mutation. Previously treated with decitabine that is 8 months on therapy with a CRi MRD negative. And then if you go to the bottom of this table, patients with AML, complex cytogenetics, multiple mutations, including FLT3 mutation treated with decitabine, venetoclax and gilteritinib that achieved at the time of this presentation, a PR. Next slide. Another issue that is -- that we wanted to understand was the dynamics of hematological improvement of serving patients treated on this trial. And whether these blast reductions were associated with hematological recovery. And then you see here that those patients that had a response such as MCR or CRi, MRD. Actually, this was associated with recovery of platelets and neutrophils, as you would like to see. Next slide. And then finally, data that I think is also quite important suggest that not only we are seeing a blast reduction, but also a decrease in the molecular tumor burden of these patients. So these are from a subset of patients where we had data -- several endpoints. Looking, for instance, at the BAF of particular mutations in the context of therapy. So for instance, you can see a reduction on the load on WT1, RUNX1, maybe SF3B1. And then a very interesting phenomenon is that you see kind of a normalization of the ratio of the long versus short ratio. So on the top in red, you see how this ratio should look in a normal CD34 positive cell. And then in a particular patient, how these ratios are controlled. So this is interesting and important data indicating a molecular effect. And again, in an early Phase I study, where you see blast reduction and this kind of molecular reductions, this is of quite interest. Next slide. So in summary, and to conclude my presentation, I think that this -- so far, this data can be summarized as shown here. So oral CA-4948 monotherapy is safe and well tolerated, up to doses of 400 milligrams PO BID. We document dose proportional exposure with minimal or no accumulation with continuous administration. There is clear anticancer, antileukemia activity in patients with relapsed/refractory AML or high-risk MDS patients that were heavily pretreated. And of now, 3 of 3 evaluable patients with an IRAK4-related spliceosome mutation achieved a marrow CR or better. And all patients with objective response also showed some sign of hematological recovery. We are continuing the study. We are continuing monotherapy molecular-defined groups, for instance, those with spliceosome mutation or FLT3 mutation. And we are developing combinations with hypomethylating agents and BCL2 inhibitors. And again, I want to thank everybody that was part of the study, including investigators, staff and the patients and families that were part of this. Thank you very much for your attention. And then, I guess, I will pass it to Dr. Martell.

Thank you, Guillermo. Appreciate it. Can everyone hear me?

Yes.

All right. So I would like to take a few minutes and add a little bit more color and perhaps some detail to some of the observations that Guillermo just described. First of all, I think there are several things that really help us inform thinking about our upcoming clinical development plan. We have significantly more information now compared to what we had in December when we presented at the ASH presentation. In particular, we've seen now that is pretty broad-based reduction in tumor burden, suggesting disease modification across the broad population. In fact, the majority of patients were seeing blast reduction. And this is in the context of an really excellent safety profile. We've seen this in our lymphoma trial and are seeing this here as well with really no significant hematologic toxicity, which, in our view, makes this great potential combination agent. And as we've described, IRAK4 seems to be a fundamental driver of many cases of AML and MDS. In fact, it seems to be produced in excess in over half of patients here. And interestingly, finally, along these lines, we've shown synergy. I'm going to talk about all of these things in a minute and sort of add some additional detail to it. But this gets to our broad development plan where we feel that the significant majority of patients with AML and MDS may be eligible here to drive benefit from this drug in combination with other agents. And we see this as an excellent combination agent with fundamental disease-modifying activity and an excellent safety profile. In addition, Jim mentioned, the data that Guillermo just described, has identified 2 specific genetically-defined populations where we may be able to seek more rapid regulatory approval initially as a monotherapy. So what I'd like to do is start by describing 3 case studies that illustrate some of these 3 points that I just mentioned in terms of regulatory strategy. One is a broad registrational path in combination with other agents and, two, specific paths. So let me start. Next slide. With this, this is a patient with a spliceosome mutation, SF3B1. Now as you may know, spliceosome mutations have been seen for many years in AML and in MDS. They come up quite commonly. They actually are maintained all the way from early disease into late disease, even as the patient progresses and retain these mutations. No one really knew what they were causing and what their importance was in the disease until about 1.5 years ago. And several comment researchers discovered that the primary cancer-causing effect of these -- or a couple of these spliceosome mutations actually was the creation of the oncogenic long-form of IRAK4. Now this is a genetically defined patient population and as you know, IRAK4 is the direct target of CA-4948. And also, as Guillermo mentioned, we presented the first 3 patients that we've enrolled onto this particular study that had spliceosome mutations, all 3 of them achieved a marrow CR or better. So let me go through this particular patient. This particular patient had secondary AML, was treated at the 300 twice daily dose level. In addition to the SF3B1, he also had a RUNX1, and a WT1 mutation. Now this patient, as you can see on the right, achieved a rapid blast response that was maintained, quite striking disease control. The exciting thing here was the fact that we also observed a significant deletion of the malignant clones. So basically, this patient became MRD-negative and suggesting disease -- clear disease modification. And supporting that is a couple of other experiments where we were able to look at the variance allele frequency during treatment and showed that this was significantly reduced for the mutations that we're tracking. In addition, we monitored IRAK4 long/short ratio and showed that during study that this went below 1, so quite impressive disease modification, a really nice example of a genetically defined disease population. And actually, a population where there's already a companion diagnostic available in foundation and luminal panels, they test for this, and this is commonly known. So we see this as a great opportunity for potential monotherapy rapid registrational path. The second patient that I'd love to describe is a patient with a FLT3 mutation. So this is a really interesting population. There are already a couple of FLT3 inhibitors approved, but it turns out that these patients develop steps to these FLT3 inhibitors. And one of the primary mechanisms of resistance is adaptive resistance that basically hijacks the innate immune signaling pathways and is dependent on IRAK4 in this resistant scenario. And so an IRAK4 inhibitor is an obvious choice for patients with FLT3 who may develop resistance. And in fact, as you know, our drug hits both IRAK4 and FLT3, which is a very interesting combination. So let me go into this particular patient. This patient had previously received decitabine and venetoclax, ultimately relapsed following that. Then they came on to gilteritinib and hydroxyurea. And the patient was completely resistant to this, basically refractory, did not even respond to the leading FLT3 inhibitor currently on the market. You can see that their mutation status is quite complex and adverse. When they came on to this study, you can see on the far right, they achieved a very rapid blast reduction from 60% all the way down to 6%, and eventually down to 5%, I believe. The most exciting thing was eradication of the FLT3 mutation. So this was no longer identified, again, suggesting really strong disease-modifying effect of this drug in this population. And again, this supports our second potential rapid regulatory path as a monotherapy, and we plan to pursue this as well as the spliceosome population in our current expansions in our clinical trial. And then switching to the third patient, example. So I mentioned some key aspects that Guillermo highlighted. Generally, we've seen very strong blast reduction in disease modification. In fact, almost the significant majority of patients had some blast reduction on study with some great examples of reduction of long/short ratios as well for the IRAK4. We've also demonstrated a great safety profile and again, reminding you that IRAK4 appears to be a fundamental driver of disease across MDS and AML. So in this particular patient, this patient had been heavily pretreated, multiple lines of 7 and 3 over their treatment course. A variety -- kind of a complex mutation status. This patient did have some blast reduction. It didn't quite get all the way to the normal range. But really interestingly, this patient has been maintained on this with stable disease but has seen a steady improvement in the hematologic parameters, with their platelets steadily improving, neutrophils steadily improving as well. So this is -- I think it's really a strong support for even when we don't achieve a full surrogate that might be going towards a regulatory endpoint by itself, we think that this drives a great foundation with evidence of heme improvement and blast reduction and great safety profile that could go into combinations. And that's why we're starting some combinations coming up in the clinic. The study combining this with venetoclax and/or azacitidine is starting imminently. And I'm going to talk in a minute about our preclinical data that also supports that. If you could go to the next slide. So before I get to that, I'd like to go into our thinking about the dosing for this drug. And just sort of announced that we've finally determined what we're going to go forward with our recommended Phase II dose. So next slide. And this is based on a variety of parameters, including pharmacokinetics, pharmacodynamics, safety and efficacy. So let me start with the PK/PD profile. So you can see on the left that these are excellent PK exposure curves. We've seen that it's nicely dose-dependent and fairly tight PK. And here, I'm assuming PK in the range, it's relevant to this particular study, from 200 to 400. And you can see that the trough exposure on the right of the left-hand graph levels off at around ranging from 2.5 to 5.9 micromolar over this dosing range. And if you correlate that with the right-hand graph, this correlates with target inhibition ranging from 97% to 98% at the 2.5 to 5.9 micromolar range. In particular, focusing on the 300-milligram twice daily dosing. Again, this achieves that target inhibition continuously even at the lowest doses when you're dosing twice daily like this. We also saw that going to 400 did not really add significant additional effect on the target compared to the 300 twice-daily dosing. Next slide. So based on this and other factors that I'm going to go through now, we chose a 300-milligram twice-daily dosing as our recommended dose for leukemia, AML and MDS. And this actually is the same dose that we've gone forward with for our lymphoma trials and also our low-risk MDS study. And just to summarize, the PK supports the twice daily dosing. And again, this achieves excellent trough levels of exposure that is 98% -- correlating with 98% inhibition of the target. And the safety has been excellent at this dose level. There were no dose-limiting toxicities. And also just here, we've had patients on for, as I mentioned, 7 to 8 months with no cumulative toxicity. And if we go back to the lymphoma trial where we've had patients on for over 2 years, again, no cumulative toxicity. And actually, on that particular study, it also suggests that the efficacy seen is maintained over significant periods of time. And then down in the bottom bullet, at this particular dose level, 4 patients had blast counts at the beginning that were evaluable for dose reduction. And all 4 of those patients achieved a reduction in their blast count. So clear signs of efficacy at this dose level with one of the patients -- actually, the patient that I showed earlier, having CRi along with negative MRD. And so based on these data, we've chosen 300 twice daily and now are rapidly moving forward with this dose as our chosen dose for some of those studies that I mentioned earlier. Next slide -- And next slide, let me just spend the last couple of minutes talking about some preclinical work that we've performed and is present as part of poster presentations. And this is preclinical data that support the 4948 combination with azacitidine and venetoclax. So if you look at the left-hand graph, you can see that the combination of 4948 plus azacitidine, basically completely eliminates tumor growth in this THP-1 cell line. This particular cell line has increased levels of IRAK4-L. And then -- and in fact, this is superior to the venetoclax/azacitidine combination in this model. And then if we go to the triplet combination, it not only stops growth, but basically strong cell killing with this triplet combination, suggesting very strong synergy. On the right-hand graph, again, a different cell line OCI-AML2. In this particular cell line, again, the 4948/venetoclax combination, in this case, completely stops cell growth in these cells. And likewise, the triplet combination causes profound cell killing. So in summary, this is really great support for moving forward with our combinations, again, given the great safety profile. Both of these drugs, azacitidine and venetoclax, are strongly myelosuppressive and this is a dose-limiting issue with these. So we now have a drug that shows clear synergy here that does not add additive myelotoxicity to these regimens. So thank you for allowing me to present this. We are very excited to move forward, and let me hand it over to Jim.

Thank you, Bob. With each new update, the continuous and consistent stream of data generated by CA-4948 continues to impress us, last count reductions in 10 of 12 patients; 5 objective responses, including all 3 spliceosome patients; the complete eradication of FLT3 mutation in a patient who is unresponsive to gilteritinib, the leading FDA-approved FLT3 inhibitor; and finally, the genomic analysis that demonstrates the disease-modifying activity of this scientifically novel first-in-class therapeutic. As you can imagine, excitement is running very high. With that, let's open the call for questions. Operator?

[Operator Instructions] The first question comes from Alethia Young from Cantor Fitzgerald.

Can you guys hear me?

Yes.

Congrats on the data. A couple of questions from me. I'll just ask a couple, a time, and then maybe ask 2 more. One, I guess, I just wanted to hear your perspectives on your confidence level of activity in both MDS and AML. It seems like you got responses in both, but just wanted to kind of get maybe the doctor's perspective and the team's perspective. And then I just wanted to also talk about as well, do you guys feel like it's a dose relationship? Or is it more about this like IRAK4 long-form expression or the ratio that drives like a better response? And then I have a couple more.

Yes, it's probably a better question for Bob. Bob?

Yes. So I think -- good question. So in terms of the long/short ratio, based on the preclinical work, we saw -- and Guillermo presented that earlier that very long excess of long-form is associated in preclinical models with better activity. And that's -- and if you sort of extrapolate that out, the preclinical work suggested that the ratio of 1.25 excess of IRAK4-long would be driving sensitivity to this drug. Interestingly, that may not correlate directly in humans, and which was so exciting that we saw the blast reduction in the significant majority of patients. Now are the patients with very much excess IRAK4-long, the ones who are responding the best, say, getting the marrow CRs and things like that. We do know that these specific spliceosome mutations definitely drive overexpression of IRAK4-L. And we've seen clearly that as we presented the first 3 patients with these spliceosome mutations did achieve CR -- marrow CR or better. And so we -- and that's why we chose this as a potential monotherapy approach because we feel that in monotherapy, clearly, those patients are a genetically defined population that clearly benefits from blast reduction and striking blast reduction in disease modification.

Then the MDS, AML, like confidence of that going well?

Yes. So -- yes, again, I think it's a little bit early to say for sure. But as you saw, half of the patients on the study were AML and half were MDS. And you saw that while some of the MDS patients do have a blast count that is less than 5%. And so for those particular patients, we're looking for things like reduction in the variant allele frequency perhaps or heme improvement, but the majority of those patients actually also had some level of blast elevation above normal. And you can see that -- again, the majority of MDS patients also had really striking and significant blast reduction at the same -- kind of the same level as the patients with AML. So we think that across both populations. And we do know that across both populations, the majority of patients have excess IRAK4-L for a variety of reasons. Some of those are for spliceosome mutations and some are other reasons for the IRAK4-L that have yet to be determined, but we do know that the majority of both AML and MDS patients have this excess IRAK4-L.

And then just looking at the durability that you've seen in early days. I mean do you feel like those patients who have the spliceosome mutation, like that could be instructive of like this kind of 5-, 6-month durability? And is that something that you think is robust enough to kind move forward in monotherapy with a fast pathway to market? And then maybe the second part of that question is just like what are the next steps from here as far as like what we will expect at ASH? And then also how you're thinking about kind of clinical development from here on out?

Well, I think it's early days for the spliceosome mutations. And clearly, these 3 patients did have striking responses on the treatment. The challenge with this patient population is it's an extremely poor population in terms of survival. These patients tend to not live very long. And I think Guillermo can speak to this because he treats these patients all the time in terms of their survival is extremely short, prognosis is very poor. And so the durability that we've seen and the patients that we've shown so far is quite striking. Guillermo, maybe you can comment on some of the challenges that we see in the clinic with this population who has previously seen these HMAs and are there therapies that you utilize that you can rely on? Or is there a need for new?

No, this is a very important group of patients so the standard of care for most patients with MDS and now majority of all the patients with AML is a hypomethylating agent. So the survival after first-line therapy in a high-risk patient with HMA failure is around 4 to 6 months, and this is being validated in prospective clinical trials. And then for patients with AML, particularly those treated with a combination of an HMA plus BCL2 inhibitor, we've shown at ASH, I believe, last year also by our group. It's actually a few weeks. So this is a really unfortunate clinical situation, and there is no standard of care. So that swimmer plot, in my opinion, again, in the first kind of path in a Phase I clinical trial is actually, I think, quite significant and encouraging. So we are excited to continue to be part of this clinical trial.

Awesome. And then maybe just Jim and team, can you talk about the clinical plans and what to expect at ASH as well?

Yes. So we kind of alluded to it already, but our clinical development plan, I would say we divide that into 2 general categories. One category is where we see this sort of long-term commercial plan, we see that this is a disease-modifying agent that seems to have activity in the majority of patients, whether they're AML or MDS. And we think that it will go very well as a combination partner with agents that might be cytotoxic themselves, an so we're not adding to that. So we see the broad commercial strategy is adding earlier line to -- across the spectrum, AML and MDS. And we're just starting the initial steps on that with this venetoclax and aza combination that I talked about. But also, we also saw the opportunities for very rapid registrational initial approval in a couple of genetically defined populations. So again, we're launching those studies. So those are going to be going on in parallel. I'm hoping to get initial registration there as a monotherapy. But then broadly, for our broader commercial strategy, we feel that this has application across the spectrum. And I didn't mention, but we do have an investigator-initiated trial by Uwe Platzbecker in Europe, who's one of the leading experts in MDS and low-risk MDS. So he's studying low-risk MDS as a monotherapy there. And this is a population where this is wide open for therapies that may benefit these patients.

Let me add into that, Bob. So as we think through the strategy, the first step for where we wanted to go is to make that transfer to get into the expansion. And that was, of course, setting the Phase II dose, which is what we did today. And the key there was we needed to make sure that 300 milligrams, which looked to be the optimal dose for efficacy and safety in lymphoma and of course, as we've mentioned, that's also being used in low-risk MDS, we wanted to make sure that, that was the same thing true here. And the answer is, yes, that we think we've got a very wide safety window at 300 milligrams. We haven't had a single DLT in any of the studies, which is terrific. So we know that this is broadly safe. We look at the inhibition levels and we are getting a 98% target inhibition. We actually get the same -- roughly the same inhibition at 200 and 400, a little better with 300. But even going to 400, there was no difference. 300 is the optimal dose for getting target inhibition. And of course, we're getting great responses as we go through. The efficacy that we're seeing at 300 milligrams, especially to Bob and Guillermo's point. In a setting where the standard of care is supportive care, we really like those odds. So now that we've declared our Phase III dose, very safe, no tox, very effective, we're getting responses. And in fact, we're getting the best responses as a monotherapy in exactly the setting where we'd hope to, which is the spliceosome patients and the FLT3 patients, our clinical strategy is going to reflect those data. And of course, we're hoping for more consistency going forward. So hope for a rapid approval path in either spliceosome patients or FLT3 patients. And then look for the very broad earlier line therapy in combination with aza/ven, which is supported by the preclinical data, but we're about to jump into expansion now. But that key is, today was that pivot point of determining the dose that was widely safe, widely effective and affected exactly where we hoped it would be.

Our next question comes from Justin Walsh from B. Riley Securities.

I think to start, any thoughts on the possible mechanistic explanations for the responses that we've seen in the non-spliceosome, non-FLT3 mutant patients. And do you think that, that's all the IRAK4-L expression? Or is there something else going on there?

Yes. It's all -- so this is Jim. Thank you, Justin. Yes, it's -- we do have the 1 patient, right, 1 patient out of 22 that has a FLT3 mutation. So really, it's IRAK4. I think the conclusion here is just reinforcing the conclusion that we had at ASH. And that is going into ASH, all of the lab data and, of course, that seminal article in Nature in 2019 had implied that half of the patients were going to see a reduction in blast counts. Of course, what we saw when we went into patients was that it was much more broad-based than that. And I think the conclusion that we've taken from that is that IRAK4-L is simply a much more powerful driver in humans, in patients that it appeared to be in the lab. And sometimes it works the other way in some drugs in some cases. This one just happened to work, frankly, in our favor. That the data in humans is simply much better. So yes, it's clearly, in our view, an impact of hitting the driver of disease, which is IRAK4.

Got it. Next question. So I noticed that in the presentation, you noted that one patient who'd achieved a stable disease went on to receive transplant. I'm wondering if we can get any more details on this patient? Was it an AML or MDS patient? And I know outcomes if you don't have a complete response, tend to not be great with post-transplant. But I'm wondering if we have any details on that patient status as well.

Yes. It's a better question for Bob. Bob?

Yes. I don't have all the specific details right in front of me, but I believe that patient was an AML patient who basically was on study. He did not progress and essentially had stable disease and had the opportunity to move on to transplant. And really, that's all I can say right now. We don't have the long-term outcomes yet.

Okay. And I'll just do one more before jumping back in the queue. I know that we'll get a lot more -- a lot of these questions will be answered when we go through the combination studies and really look at the drug in more patients in the different populations. But I'm wondering with maybe Dr. Garcia-Manero can take this. Based on your experience with the drug so far, where do you see this likely fitting into the treatment paradigm in both AML and MDS? Do you think it's the type of thing that is likely to be towards the front line in combinations or even used after transplant. Just what are your thoughts on -- like with the limited amount of data we've seen so far?

That's difficult to answer, but I'm going to give you my perspective, and I would like to dissociate my statements. First of all, this pathway, it's actually really important in leukemia and actually other diseases like lymphomas, et cetera. So this pathway of Toll-like receptor innate immunity activation, NF-kappaB activation is really important. And as Bob was saying, targeting in low-risk MDS makes total sense. We've published data before we knew [ at ] Curis and there were IRAK4 inhibitors, that actually proved that concept. So I think that if you are interested on targeting this particular pathway, a molecule like this particular compound could have actually a very broad spectrum of potential activities as a single agent or in combination. I mean I'm not saying that a single-agent is going to have the answer for everything, but it could have a potential role in different situations, both in MDS and in AML. The study -- the drug needs to be studied in combinations for sure. So I think there is a lot of potential, but we need to do a lot more studies. And some of these were actually suggested by you and your colleagues. So that is one part of this, like scientifically, IRAK4 is a very important target. And the second aspect is what we see from this Phase I trial, I think it's exciting. I mean seeing blast reduction. Some of these patients going into CR. Even those patients, I think you were mentioning that went to transplant that had a stable disease for an amount of time, that's something that we don't very frequently see in this kind of patient population in your prototypical Phase I clinical trial in leukemia and MDS, particularly in this context of HMA failure. So I see that this profile of the safety, hitting the target, seeing some early indication of activity at this level is quite exciting. And then we were part of this trial because we're very interested on modulating this IRAK4 pathway. So I think this is going to be a very interesting study to continue to be part. And I think this data is very exciting at this point.

Our next question comes from Yale Jen from Laidlaw & Company.

Can you hear me?

Now we can. Excellent.

It seems the IRAK4 long-form is -- has been reduced. What do you think the direct mechanism of that is? Was that at this specifically at pricing level or other mechanism?

It's your, Bob.

Well, we think that the likely cause of the reduction of IRAK4-L, for example, is essentially, the killing of those blast population, is that what you're asking? I mean, ultimately, the IRAK4-L is signaling NF-kappaB, which produces a variety of growth stimulating factors as well as anti-apoptotic factors. So once that is shut down, these AML cells likely undergo apoptosis.

Okay. Great. And maybe one more question here is there's a heme recovery. Again, what do you think the -- are reasons behind it?

There's probably a couple of reasons. One, obviously, is as we reduce the blast, ultimately reducing the blasts. The blasts are creating a bone marrow environment with abnormal cells and unable to function normally and produce normal hematologic growth. Once we reduce those blasts, the marrow obviously will take time to recover its normal function. But once those are reduced, less crowding out of the marrow, allowing normal marrow cells to grow and ultimately proliferate and differentiate and produce normal hematologic cells. The other factor is that, as you know, IRAK4 itself, and in particular, IRAK4-long is creating a highly inflammatory microenvironment within the marrow. And really, that's one of the core pathologic -- pathophysiologic problems here and why IRAK4-long starting way at early stages of disease, and then existing throughout the duration is problematic partly because it's causing this inflammatory microenvironment. And that itself is likely to be inhibiting growth of the myelopoiesis. And so suppressing that inflammatory microenvironment by inhibiting IRAK4, which can directly do that, may also be a component of why the recovery is coming. For example, even in that patient who didn't get complete clearance of their blasts, they got clear improvements in their heme parameters.

Okay. Great. And maybe the last question here is that in patient case #1, the blast level seems increased to -- toward the end. What do you think about that? Would that be something more transit and that could be returned back to a lower level? Or how do you see that?

So the blast count -- so in that particular patient, the blast count went down fairly dramatically. In fact, all the way to 0 at one point. The increase back up could be due to a variety of reasons. In fact, that's still in the normal range. So it may be normal blast repopulating the marrow. So we don't know at this point yet. The fact that you've had complete elimination of the genetic clone with the MRD-negative, a strong evidence for disease modification. And so when you have blast counts that are in the normal range, those may be normal blast as well.

Our next question comes from Edward White from H.C. Wainwright & Co.

Congratulations on the data. Just a couple of questions. First, when thinking about the triplet combination, how should we be thinking about the timing of that study and perhaps the combination -- triplet combination protocol? Is this something that you're going to do concurrently with the doublet studies? Or will it be a separate study? And then in that study, would you look at doublet versus triplet?

Yes. I think the sequence will be, we're going to look at doublet first. So we're going to be running 4 sets of patient populations at the same time. So of course, the low-risk MDS is ongoing in Germany. In the accelerated approval path for monotherapy, we're going to have the FLT3 and the spliceosome path. And then to your point, the combination in earlier line of aza and ven is going to also be kicking off right now. And, of course, presuming that those go where we expect they would go, and they show great data. After that, then, of course, we would look to -- try the triplet as well. But I think first things first, let's do the doublets and let's make sure that all the data is consistent. So I think the key is, as I say, today is a big pivot point for us in the study. We've been able to put our stake in the ground at 300 milligrams. And at 300 milligrams, we think we've got a dose that is very safe, no DLTs, very effective. We're getting responses exactly where we expected to get them. Duration looks great, small and early days, but duration looks great. And I think we've identified all the populations we want to go to, combo therapy, monotherapy and then very early-stage disease as well. And we're going to move very aggressively over these next months and quarters of bringing those patients into study and frankly, seeing where this drug can go.

Thanks, Jim. And you just said you're moving very aggressively. So I'm just wondering if you have any thoughts on timelines, particularly in the -- when we can see more monotherapy data at the recommended Phase II dose. Will we be seeing more patients at ASH? And when will we be seeing or maybe get your thoughts on the number of patients that you will need to -- for regulatory filings?

Yes. Those are great questions. Obviously, we're going to be -- now that we've got our Phase II dose, we're going to be looking to add as many patients in at 300 milligrams as we can across that spectrum. I think it might be a little premature for me to guess on what enrollment rates look like. But I think it's reasonable to expect that at ASH, there's going to be a lot of interest in knowing where we stand, so we'd love to be able to give an update on those patients then. And in terms of the -- how quickly can we go to the FDA and have a discussion, that's an open question. I think both -- the 2 areas where we had that discussion first, in my view, are going to be this spliceosome population and the FLT3 population. But it's still early days in those. We've got the 3 spliceosome patients, we got the 1 FLT3 patient. I think we need to see more. Obviously, getting the kind of hit rate we're getting, which is we've got responses in all of them. That's terrific. But I think we're going to need a few more patients in both of those buckets before we go talk to the FDA. But I think stay tuned. We're going to have several opportunities to talk between now and ASH. We're happy to give updates as we go. But I'd say, let's turn our attention to executing on getting more of those patients in at 300, and we're very encouraged that we're in a good place right now, getting a few more patients in at this very effective and safe dose could be very exciting over the next few months and quarters.

Our last question comes from Soumit Roy from JonesTrading.

I'm just trying to understand if you have any thoughts on why the AML patients are responding a little less than the MDS patients despite most of the AMLs, I'm looking at the spider plot looking curve are on 300-milligram and above. And is there a concern that you are having a narrower therapeutic index given the upper limit is probably 400 BID or you're comfortable looking at the PK curves?

Yes. So let me start and then I'm going to ask Bob to chip in. So I'll go in reverse. I think the safety index is, frankly, fantastic. That we have, at 300 milligrams, a dose that has been dosed in several studies across multiple years and we don't have a single DLT. So it's terrific. We do know that we saw some safety signals above the recommended dose. But all of those safety signals were easily identified and easily resolved and there, frankly, weren't very many of them. In this case, for example, we know we saw [indiscernible] 13% of the patients at 400, but our dose is 300. So yes, is there -- are there small numbers of patients if you dose beyond the recommended dose that you might have a safety issue, I think that's true. Sure. But at 300 milligrams, which is our dose, that's where every patient is going to be going forward. We don't have any history of issues. So I think we've got a very broad and predictable safety profile. In terms of efficacy, yes, it's interesting that all of the patients on that spider plot that were at 300 milligrams saw blast reductions. The only 2 that were increased were at 400. I think that goes to show that when you look at the slide that Bob presented on PK/PD, the inhibition percentages were 97%, 98%, 98%. There really was very little difference in inhibition between those 3 levels and no difference between 300 and 400 in terms of target inhibition. I think the bigger difference now moving forward as to who's going to see efficacy as well what we're seeing. So if you're in one of these groups, if you have a spliceosome mutation or if you have a FLT3 mutation, those are going to be the patients that benefit the most on monotherapy. And that's really the way we thought it would work. And then for the other patients, like our patient case study #3, that patient has not yet achieved a response, but clearly is seeing benefit from being on drug. That's an ideal candidate for somebody who earlier line in treatment would go on combo therapy. Bob, would you like to add to that?

No, I think you covered it fairly well, Jim. Again, would endorse the strong safety profile and clear efficacy at 300. Even 400 in the AML study was below the MTD. So I think we have a really nice safety window here.

And one last question. Can you give us some bookends for the prevalence of these spliceosome mutations in MDS population or even the WT1, RUNX1 mutations in MDS? And does that prevalence increase or decrease with lines of therapy?

Yes, Bob?

So I can comment and maybe ask Guillermo as well if -- your experience. But the -- so the spliceosome mutations across the broader AML/MDS population, at least, these -- the 2 that are directly linked to IRAK4-L, the SF3B1 and the U2AF1, represent approximately 20% of all patients, so it's a subset. There's another subset of patients that also have IRAK4-L where the cause has not been yet identified. It may be epigenetic changes, it may be other mutations that have yet to be directly linked, but across the broader population. And that tends to be a little bit more focused in the MDS. So I would say roughly 10% -- or I'm sorry, 30% of the MDS broader population has spliceosome mutations, whereas in the AML, it's probably more closer to 10% of that population. But what's been found is that these spliceosome mutations, they're seeing very commonly even at early stages of disease, suggesting it's a foundational mutation, foundational process. And studies that have looked at patients who progress from MDS to AML, for example, tend to retain that mutation. Again, suggesting it's a fundamental foundational event that the tumor itself is essentially requiring during that transition process from MDS to AML. Guillermo, do you want to comment on your experience with seeing these various mutations in patients.

No, you summarized this very well. For instance, the SF3B1 mutation is the most common mutation in myelodysplastic syndrome, and it occurs both in high-risk and low-risk disease. And what Bob said in terms of the dynamics of these mutations during this progression is accepted. And there's actually quite of genomic data out there really showing that. So these are not rare events. They're actually probably one of the most relevant events in both MDS and AML.

Before we wrap, I just wanted to go back on one of Justin's questions. The patient who'd transferred to stem cell therapy was a high-risk MDS patient as opposed to AML. And it should be identified on the swimmers plot. Jim?

This concludes our question-and-answer session. I would like to turn the conference back over to the company's President and Chief Executive Officer, James Dentzer, for any closing remarks.

Thank you, operator, and thank you, everyone, for participating in today's call. And as always, thank you to the patients and families participating in our clinical trials, to our team at Curis for their hard work and commitment and to our partners at Aurigene, ImmuNext and the NCI for their ongoing help and support. We look forward to updating you all again soon. Operator?

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.