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

Good morning, and welcome to the Curis January Clinical Update Conference Call. [Operator Instructions] Please note, this event is being recorded. I would now like to turn the conference over to James Dentzer, President and CEO of Curis. Please go ahead.

Thank you, operator, and welcome to the Curis' January clinical update call. Bob, can you move to the next slide? Before we begin, I would 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. Next slide, please. I'm Jim Dentzer, President and CEO of Curis. Joining me on today's call are Dr. Bob Martell, our Head of R&D; and Dr. Daniel DeAngelo, a clinical investigator in our IRAK4 study and the Chief of the Leukemia division at Dana-Farber. As many of you know, Dr. DeAngelo is a renowned medical oncologist, Chief of the Leukemia Division at Dana-Farber and also a Professor of Medicine at Harvard Medical School. He specializes in treating leukemia and myelodysplasia and has participated in numerous clinical studies with over 300 publications. We are honored to have Dr. DeAngelo on our call today, and thank you for joining us. Next slide, Bob. So with that, let's get started. Good morning, everyone. It's my pleasure to welcome you to the call. And I'd like to start the discussion with a brief overview of today's exciting news. A year ago at ASH, we presented clinical data on 2 patients with spliceosome or FLT3 mutation. 6 months ago at EHA, we presented data on 4 patients with spliceosome or FLT3 mutation. As confirmation of the IRAK4 thesis began to bear out and the clinical efficacy data began to roll in, excitement and enrollment kicked into high gear, and today, we are pleased to present data on 13 patients with spliceosome or FLT3 mutation. In the first part of our presentation, we'll walk through our clinical progress with CA-4948, our first-in-class IRAK4 inhibitor. And today's data confirm and extend the exciting data we saw 6 months ago. We have a drug that is demonstrating clear clinical efficacy as a single agent in genetically defined populations. Even better, the added potential benefit of also hitting FLT3 appears to confer the additional advantage we were hoping to see. And with CI-8993, our first-in-class VISTA antagonist, we're thrilled to be able to say today that we have begun dosing -- I'm sorry, we have been successful in clearing both the 0.15 and 0.3 mgs per kg dose levels, and have now begun dosing at 0.6 milligrams per kilogram. We're especially excited to announce that we have also begun to see the PD effects that we hope to see, indicating that the multiple anticancer mechanisms associated with VISTA are being activated. This is obviously a very exciting day for Curis, for our investigators and especially for our patients. With that, I'll ask Bob to walk us through the data. Bob?

Thank you, Jim. I'm very excited to provide this update on CA-4948. Just as a preference -- a preface, I just wanted you to note that the slide is -- slide gets a little bit delayed, so I'll do my best to kind of keep track with moving the slides forward. So since our last data update in June, we've more than tripled the number of patients with genetically defined disease, who we are evaluating for monotherapy efficacy with CA-4948. As you will see, the data continued to support consistent anticancer activity and an effective safety profile. Discoveries have revealed paradigm-changing insights in the cancer biology of AML and MDS, highlighting IRAK4 as a fundamental driver of disease and directly linking IRAK4 to foundational spliceosome mutations frequently found in AML and MDS. CA-4948 is the leading IRAK4 inhibitor in clinical development for cancer. It was designed to hit IRAK4 selectively and potently. In addition, this drug has activity against a wide spectrum of FLT3 mutations. With this in mind, we have evaluated CA-4948 in AML and MDS. The study determined the recommended Phase II dose and safety profile in an unselected population as well as efficacy in 3 genetically defined subpopulation. Initial enrollment was open regardless of genetic mutations. And as of December, we had enrolled 49 patients for whom we'll present safety data today. 14 of these patients had genetically defined disease with either lysosome mutation or FLT3 mutations. All of those that won were evaluable for monotherapy efficacy, which we will also present today. For these patients, our enrollment cutoff was in September in order to allow for at least 2 on-study disease evaluation. The safety profile of CA-4948 is very clean. The drug is taken orally twice daily. Dose levels on this study range from 200 milligrams up to 500 milligrams, and the recommended Phase II dose was determined to be 300 milligrams. Adverse events have been readily reversible and with minimal sequela or complications. This profile, along with the lack of dose-limiting myelosuppression is very favorable for monotherapy relative to other agents used to treat these diseases. CA-4948 is also a promising candidate for combinations because of its nonoverlapping safety profile. I'd like to actually ask Dr. DeAngelo, who we're very fortunate to have on the call today, to briefly comment on this profile for patients fighting AML and MDS.

Yes, happy to, and thank you very much, and good morning, everyone. I think the safety profile here is really astounding. I just want to make sure that people are aware that this is a very heavily pretreated relapsed/refractory population of patients with acute myeloid leukemia with a median survival of 3 to 6 months. And so seeing a non-myelosuppressive agent here with this safety profile is really remarkable, calling out the rhabdomyolysis, this was a transient, very easily recognizable and treatable and reversible side effect that can be mitigated just with the dose reduction. So not a huge issue in this relapsed/refractory population and a very encouraging safety profile.

Thanks, Dan. I appreciate that commentary. So let me now switch gears and focus on the efficacy population. These populations include patients with lysosome-mutated AML, patients with spliceosome-mutated MDS and patients with FLT3-mutated AML. In this next slide coming up shows these different populations and some characteristics. Interestingly, 2 of the 3 patients with FLT3-mutated AML also had lysosome mutation. And so for completeness, we've included them in both AML subpopulations. The patient demographics here illustrate an elderly population with unfavorable disease characteristics, poor prognosis and no good therapeutic options. All of the patients in the Curis study in these targeted efficacy populations have received prior hypomethylating agents, a situation where no existing therapies are effective. We observed strong anticancer activity across all 3 populations. In the slide in a moment that you'll see in patients with lysosome-mutated AML, 2 of the 5 patients achieved a CR or CRh, and both of those patients actually achieved an MRD-negative status as well. In the patients with spliceosome-mutated MDS, 4 of the 7 patients achieved a marrow CR, one of whom was actually able to proceed on to stem cell transplant. And lastly, in the FLT3-mutated AML population, overall, 2 of the 3 patients with FLT3-mutated disease experienced eradication of the FLT3-mutated clone, demonstrating strong anticancer activities. Let me give more details on each of these 3 targeted populations. First, I'd like to talk about patients with spliceosome-mutated AML. As I mentioned earlier, this is an extremely difficult population to treat and this slide illustrates why. The treatment algorithms on the right show that once a patient has progressed following initial therapy, nothing worked and the outcomes are dismal with survival in this AML population of 2 to 4 months, illustrated in the graph on the bottom right. All of the patients on the Curis study fit into this category. Compounding this poor outlook are additional challenges related to spliceosome-mutated disease, which are highlighted on the left. These mutations cause a chronic inflammatory microenvironment within the marrow. Chronic inflammation is known to cause persistent impairment of normal hematopoiesis. This has been illustrated in clinical studies demonstrating much worse ability to achieve ER with full hematologic recovery in patients with spliceosome-mutated disease compared to patients without spliceosome-mutated disease. Despite these challenges, CA-4948 exhibited strong anticancer activity, along with significant hematologic recovery, providing hope that this drug will give much needed benefit or much needed therapy for these patients in significant need. As I mentioned on the prior slide, these patients don't live very long. And this slide shows the most commonly used agent in relapsed or refractory setting based on market data. Efficacy outcomes are extremely poor and these treatments are limited by serious toxicity issues, and they have the added burden for patients of requiring IV or subcutaneous administration. While in early days, we believe that CA-4948 compares favorably to these treatments. This slide highlights the impressive efficacy seen by the 5 patients with spliceosome-mutated AML on our study. The reason this is so striking is that these patients all had very poor prognosis disease. They all had complex disease that was considered intermediate or adverse risk. They had all been previously treated with the hypomethylating agents. These patients were in a situation where nothing worked and their life expectancy is only a couple of months. Despite this, 4 out of the 5 patients on this study for blast reduction, including 2 patients who achieved significant hematologic improvement, one with a full CR, and the other with a CRh and partial hematologic recovery. Both of these patients achieved an MRD-negative status and had treatment duration of more than 6 months. Briefly, I'd like to point out that the first patient on this chart actually met criteria for both CRi and CRh. We had utilized the CRi designation in a prior presentation, and here we present the CRh since this is a more commonly accepted criteria. I'd also like to point out that this patient came off study after 8 months, still with blast in the normal range. We're representing 7-month duration of treatment here because their last dose was after 7 months. The patient had opted for a pause in treatment due to an AE with the intention of subsequently resuming treatment. But in the end, the patient and physician decided to continue to just monitor the patient off treatment where the patient actually came off study with blast in the normal range at that point. The second patient achieved a full CR. This patient started the study with platelets in the 10,000 range, neutrophils near 0 and requiring frequent red blood cell and platelet transfusions. This is a very parallel clinical situation. During treatment with CA-4948, this patient experienced full hematologic recovery for CR, and no longer required transfusion. Dr. DeAngelo may be able to give a little bit more color on this in a few minutes. I'm going to now switch to the second genetically defined population patients with spliceosome-mutated MDS. Here, again, the outlook is not good with patients surviving a median of only 4 to 6 months once they progressed following a hypomethylating agent. This is illustrated on the right. Confounding this issue is the chronic marrow inflammation caused by spliceosome mutations that I mentioned earlier. This may prevent hematologic recovery independent of disease control since chronic marrow inflammation is known to cause persistent and enduring impairment of normal bone marrow function. All of the patients with MDS on the Curis study fall into this category and no effective therapies exist today for them. This is a clear unmet need. Again, on this next slide, I'm sharing market information on the most commonly used agent in the relapsed or refractory setting. Efficacy outcomes are extremely poor and these treatments are all limited by serious toxicity and require IV or subcutaneous administration. The next slide I'll present will show that despite these challenges, the efficacy we saw with CA-4948 in this population is excellent. Here are the details of the 7 patients with spliceosome-mutated MDS. Note, the high disease complexity and significant prior therapy, including, again, all of these patients having received prior HMA. Despite this, 5 of these 7 patients had some reduction in marrow blast cells, 4 of whom actually achieved a marrow CR, reverting their marrow blast on [indiscernible] normal. And one of these patients was actually even able to proceed on to a potentially curative stem cell transplant. Finally, I'll switch and provide details on the third genetically defined population with FLT3-mutated AML. Before I get into this, though, I'd like to review some published literature that highlights compelling synergy between IRAK4 and FLT3 inhibition. It's known that IRAK4 activity dramatically increases following treatment with a FLT3 inhibitor, both in preclinical experiment as well as in patients treated with the leading FLT3 inhibitor, Gilteritinib. IRAK4 is thought to be a key mechanism of resistance to FLT3 inhibition. Upon further investigation of this observation, it was found that concomitant treatment with both an IRAK inhibitor and a FLT3 inhibitor caused synergistic cell killing that even persisted long after the drugs were [ renewed ]. Researchers also found that in a mouse model on the right, a dual IRAK FLT3 inhibitor was dramatically superior to the FLT3 inhibitor, resulting in survival of most of the mice. With this in mind, the dual targeting of IRAK4 and FLT3 by CA-4948 is very promising for patients with FLT3 mutations, which represents approximately 30% of all patients with AML. Therapies used in this relapsed/refractory setting, again, are ineffective, toxic and often difficult to administer. This represents another significant unmet medical need for patients. This slide shows data from 3 patients treated with CA-4948 who also had a FLT3-mutated disease. 2 of which also had a spliceosome mutation, which I mentioned earlier. All 3 of these patients had some blast reduction. And as you can see, 2 out of these 3 had quite sharp reductions in blast counts. And in fact, both of these patients saw complete eradication of their FLT3 mutations. Interestingly, 1 of these 2 patients with striking blast reduction did not have a spliceosome mutation in their disease and their cancer was previously completely refractory to Gilteritinib, a leading FLT3 inhibitor. This is a really nice example of the clear contribution of targeting IRAK4 in this setting. Let me finish by summarizing in this section a very favorable safety profile as well as data from 3 genetically defined patient populations where oral CA-4948 achieved significant single-agent efficacy. This is particularly impressive given the current therapies do not work in these patients. I'd like to ask now Dr. Daniel DeAngelo to provide his perspective on the data and provide a view of these patients' challenges in a clinical situation. Dan, thanks for joining.

Sure. Thank you very much. And just to provide a high-level view, in a Phase I study, small number of patients, but we're seeing here in 3 genetically defined groups of patients, the spliceosome-mutated acute myeloid leukemia, spliceosome-mutated myelodysplastic syndrome and FLT3-mutated AML, we're seeing huge reductions in terms of bone marrow disease, including some CRs. And in fact, just to go through the data, 4 out of the 5 patients who had spliceosome-mutated AML achieved a blast reduction, including 2 CR/CRhs with significant hematologic improvement. 5 of the 7 patients with spliceosome-mutated MDS achieved blast reductions, including 4 patients achieving a marrow CR. This is really astounding given the fact that these patients who are refractory to hypomethylating agents have a really limited 3- to 6-month median survival. And then furthermore, all 3 of the patients with FLT3-mutated AML had a blast reduction, including 2 patients who went into a genetically negative marrow status, including 1 of my patients, which is really just profound. My patient who had myelodysplastic syndrome and then transformed into acute myeloid leukemia was treated with venetoclax and azacitidine. He's an 80-year-old gentleman, had a nice response, but then progressed. And at the time of progression with venetoclax-based therapy, responses are low and poor and duration of remission, if achieved, is very low. And my patient was able to achieve a remission after only a month-or-so of therapy. He's transfusion independent. And he's actually back to work. He plays in -- he's 80, but he plays in a jazz band, and he has 3 or 4 gigs a week. And it's just remarkable how well he's doing. He just started his eighth month of therapy, so I think this shows that there is some really bonafide important responses, clinically important responses in these 3 genetically defined areas of disease. As I've already alluded to, my impression is that this oral drug is well tolerated with a manageable safety profile. The reversible rhabdomyolysis is not an impediment to therapy. And although small numbers of patients, as we've already alluded to, this clearly supports further development, both in the single agent as well as in combination therapy in these 3 genetically defined areas.

Thanks, Dan. That was really helpful for perspective, and we appreciate that. So let me finish up with this slide before I move on to 8993. This slide crystallizes some of the key points of our leading IRAK4 inhibitor, CA-4948. Moving forward, we will expand our experience with monotherapy in these genetically defined populations. We intend to discuss further expanding data with the FDA in the first half of this year with the goal of defining a rapid regulatory path to bring this novel therapy to patients in dire need. We've also initiated a study combining CA-4948 with azacitidine and with venetoclax, and we expect to present data from this study later this year. Now let me switch gears. I'm really excited to present our data on CI-8993, our monoclonal antibody targeting a checkpoint called VISTA. This is the first time presenting clinical data with CI-8993, and the findings are quite compelling. As we have been able to overcome safety challenges that slowed earlier development by Janssen. In addition, I'll share some very compelling pharmacokinetic and pharmacodynamic biomarker data. It demonstrate our ability to affect multiple anticancer immune mechanisms in patients. Before I get into the detail about CI-8993, I'd like to provide some perspective on how VISTA is differentiated as a cancer target relative to other checkpoints. Because of VISTA's localization on a variety of immune cells, targeting it affects numerous cancer immune mechanism, many of which are simply not addressed by targeting PD-1, CTLA-4 or other checkpoints. I've outlined 5 distinct cancer-related immune mechanisms here on this slide, where VISTA plays key roles. Let me briefly describe each. I'll start with mechanism 5 on the right, expanding and transforming activated and exhausted T-cells into effector cells. The primary actions of pembrolizumab, nivolumab, atezolizumab, ipilimumab and many other checkpoint inhibitors can largely be lumped into addressing this mechanism. This also has independent contribution to this mechanism. However, in order for these agents to function, they require activated T-cells. And actually, the majority of T-cells remain in an inactive quiescence state, and they're held here primarily by VISTA, which is illustrated in mechanism 3 on this diagram. Targeting VISTA releases T-cells from quiescence and thus enabling them to be activated by other checkpoint inhibitors. Another VISTA-related mechanism is myeloid-derived suppressor cells, or MDSCs, highlighted by mechanism 1 on the left. Targeting VISTA can transform these MDSCs into activated tumor-fighting macrophages. In mechanism 2, targeting VISTA potently activates natural killer cells or NK cells. These -- this is an effect that other checkpoint inhibitors don't have, to my knowledge. Another unique aspect of targeting VISTA is enhancement of antigen presentation by APC, represented by mechanism 4 on this figure. Finally, I'd like to emphasize the broad potential applicability of CI-8993 on the far right. We've listed indications where checkpoint inhibitors are currently approved. Targeting VISTA offers significant opportunities to expand the effective cancer immunotherapeutics because it is involved in multiple cancer immune mechanisms not addressed currently by approved checkpoint inhibitors. Interestingly, data are already emerging for a key resistant mechanism to checkpoint inhibitors. In some of these indications, it is, in fact, increased expression of VISTA. With that in mind, we've brought CI-8993, the first VISTA monoclonal antibody into the clinic. Our protocol incorporates a brief desensitization for patients to mitigate the on-target transient CRS. The oncology community is now much more familiar with the management of CRS with new guidelines and expanded use of agents like tocilizumab. This is a Phase I dose-escalating study, which started at 0.15 milligrams per kilogram, which was the highest dose that Janssen was able to safely administer during their study. In the Curis study, patients were safely managed at that dose level without dose-limiting toxicity. In addition, we were able to escalate to the 0.3 milligram per kg dose level, again, without dose-limiting toxicity. And we just recently initiated dosing at the 0.6 milligram per kilogram dose level. The safety profile so far has been excellent. These patients are mainly experiencing low-grade nondose-limiting symptoms of CRS as you can see on this slide. In the next slide, it's quite interesting in terms of the pharmacokinetic data we're seeing in these patients. One thing that Janssen noted when studying this drug several years ago was a sink effect when the drug was given to patients at very low dose levels. What this means is that with the low dose levels, the antibody is quickly soaked up like a sponge, resulting in limited antibody detection in the circulation. Curis reasons that if we could saturate that sponge with our desensitization regimen and with higher doses, we can overcome this thing, allowing much better antibody exposure. And this is exactly what is happening. For example, if you look at the graph on the right, when we double the dose from 0.15 to 0.3 mg per kg, the Cmax actually increased by almost tenfold. Then when we doubled the dose further to 0.6 mgs per kg, the Cmax increased by another eightfold. You can also see that after several doses by day 29, the exposure is further increasing. The ASC exposure is also disproportionately increased as we increase dose. These data suggest that we are, in fact, saturating the sink that limited the exposure from the prior Janssen experience. The graph on the left is plotted on a log scale and shows really nice PK curves across these dose levels. In these next 2 slides, I'm going to show some really interesting pharmacodynamic and biomarker data. Recall that the mechanisms that we discussed earlier, a couple of slides ago, many of which differentiate CI-8993 from other checkpoint inhibitors. It looks like CI-8993 is having a clear impact on these mechanisms in patients. Using immunophenotyping, we observe a significant reduction in those tumor -- in those immune-suppressing MDSCs, which are represented by mechanism 1 in this figure. We're also seeing a dramatic activation of NK cells which represent a key first responder to attack and kill cancer cells. Now you may notice that by 2 weeks after drug dosing, the effect is starting to return closer to baseline. This is represented by the pre-third dose time point from this graph. This is exactly where the disproportionately high PK exposure at higher doses that I mentioned on the prior slide may come into play. With higher doses, we hope to see sustained activation of these anticancer immune mechanisms. So on this next slide, I'm showing 2 other anticancer immune mechanisms affected by CI-8993 in these patients. Antigen presentation is significantly increased by the 0.3 mg per kg dose level. This helps the immune system recognize and attack the cancer. Additionally, T-cell factors are released, which play roles in T-cell expansion and transformation into effector cells. Overall, hitting these pharmacodynamic endpoint and these data provide proof of concept that we are not only hitting this, but we're also affecting multiple VISTA-mediated anticancer immune mechanism. So let me briefly summarize our studies with our VISTA antagonist, CI-8993. We have shown that CRS can be successfully managed. It looks like we are able to overcome this PK sink with very nice exposure as we increase dose. Finally, the pharmacokinetic findings suggest we hit multiple immune-enhancing mechanisms in patients that clearly differentiates CI-8993 from other checkpoint inhibitors. Let me now hand it over to Jim. Thank you.

Thanks, Bob. I'd like to close our discussion today with a few key points. First, this latest look at our clinical progress with CA-4948 confirms and extends the exciting data seen to date. In specific genetic population, patients whose survival is measured in months and for whom existing treatments simply don't work, we're seeing potentially game-changing results. These are still early days, but in all 3 populations tested, populations of patients who are all relapsed/refractory after receiving HMA therapy, we are seeing clinical responses. The efficacy of 4948 compared to existing treatments looks terrific. Second, our positive safety profile is a clear advantage. While as much as 50% of the AML/MDS population, those patients whose disease is characterized by either spliceosome or FLT3 mutation may take this drug as monotherapy. The other 50% may want to add 4948 to a combination therapy regimen. This is where the safety profile becomes so important. We believe 4948 is well positioned to become the combination therapy addition of choice, and we look forward to exploring that potential in our combination study this year. And finally, we have VISTA. Today, we're thrilled to be able to say that our clinical strategy has been successful. The new administration procedures we implemented have enabled us to safely dose escalate past the doses that tripped up Janssen. We cleared both 0.15 and 0.3 mgs per kg and have recently begun dosing at 0.6 mgs per kg. In addition, our PD data indicate early signs that key anticancer mechanisms associated with VISTA are being activated. So I said at the beginning of the call, this is obviously a very exciting day for Curis for our investigators and especially for our patients. With that, we can open the call for questions. Operator?

[Operator Instructions] And our first question will come from Alethia Young of Cantor.

Congrats on the progress this morning. A couple. One, when you look at kind of the relapsed/refractory MDS patients, I mean, I know they're very unfortunately beat up. I mean they got marrow CRs. Would you ever expect some to ever have -- move to full CR? Or have full recovery hematologic is the first question. The second question is just a little bit broader on this asset as well on 4948 as far as going kind of what would be needed or what are you kind of looking for is for the amount of people in your conversations with the FDA over this first half of the year or so? How many people do you kind of think that may be needed to kind of compel or so effect? Because it seems like in tough populations, you are showing efficacy. And then the third question is on 8993. And you've obviously moved further and went to 0.6 mg per kg. Do you have any kind of hypothesis on where you think the dose is that you might start to see like kind of efficacy or responses? Or kind of maybe you can give us a flavor kind of just based on what you know so far and what you now have a Janssen molecule, kind of how far we might be away from that if you had to hypothesize?

Yes. Bob, actually, these are probably better for you.

I think, Alethia, I appreciate those questions. So your first question was regarding the patients with high-risk MDS that we presented today, 4 of which, as you mentioned, had a marrow CR, which is -- represents a really strong control of the cancer. And so what we oftentimes want to point out is that cancer effect -- anticancer effect and then what's required subsequently is the normal remaining marrow needs to be able to recover the comp. We think that these patients are still ongoing and we'll still monitor for their heme recovery. I think it's also important to point out the extremely short survival of this patient population. So normally, these patients are known to survive only 4 to 6 months. And the patients on our study who achieved these marrow responses are continuing on for significant periods of time. When we -- sort of merging into your second question regarding our thinking about FDA, let me talk a little bit about the strategy overall. We kind of look at this as a molecularly defined approach where patients with spliceosome mutation is a genetically defined indication, kind of like MSi. So as we take these groups of patients, patients with spliceosome-mutated AML, patients with spliceosome-mutated MDS to the FDA, we will make the proposal that this broader population where we're seeing consistent efficacy all the way across the spectrum of disease. As you know, the spectrum disease ranges from MDS all the way up to AML. We're seeing consistent kinds of efficacy here, anticancer activity and make an argument that this is an approach that could allow approval of this product in a rapid and even possibly single-agent, single-arm approach. And I think I'd like to actually pause for a second and ask Dan DeAngelo, who's joined us today, to kind of give his perspective on the clinical utility and the sort of the clinical scenario that these patients are experiencing. And then I'll get back to your question on 8993 in a minute.

Yes. Thanks. I don't have too much to add, but I just want to reiterate that in these patients who are heavily pretreated, achieving a full count recovery that is a full CR and at least in the MDS patient population is unusual. If you look at most of the data, these are marrow reductions in terms of their blast count for appropriate patients trying to bridge them to a transplant which is really all you need. And that's been the big Achilles' heel in these patients is trying to get them in a safe way to get their disease under control with respect to their blast burden. That seems to be the most important impediment to survival in terms of treating of these refractory MDS patients. In terms of the AML patients, again, it's the same thing. These are heavily relapsed/refractory patients. It's not atypical to see patients who have marrow responses without full count recovery. In that situation, it depends on the background and what they've been treated for and what the goal of therapy is. In our clinical practice, achievement of a marrow CR in terms of trying to bridge a patient is essential. The patient that I described to you, my own patient is too old for transplant. But then in that particular situation, you can see how this gentleman has had an 8-month remission, back to work, transfusion independent, doing remarkably well. So if that theme holds out, then it may be a moot issue in terms of the bridge to transplant issue that I brought up.

Yes. Thanks, Dan. And just to further answer that question, Alethia, you were asking about how many patients we would want to take to discuss to the FDA. We think that, as we mentioned before, we'd like to target 10 to 20 patients with a specific spliceosome mutations as a number of patients that we think would be a viable number of patients to have a meaningful discussion with the FDA. And then your last question was related to CI-8993, and sort of wondering about what dose level we might expect to be clinical efficacy here. So let me answer that in a couple of ways. So first of all, we were actually really excited to see such a nice pharmacodynamic activity at the dose levels that we've explored so far, which suggests we're actually getting into dose levels where this could translate into anticancer activity. One thing I mentioned was related to the pharmacokinetics. It does look like we're able to overcome or start to overcome this sink. And that's where this kind of gets interesting and so I think we're sort of scratching the surface there. And as we continue on at the 0.6 dose level and potentially higher dose levels, we'll see where that exposure goes and pharmacodynamic data goes. Based on preclinical projections, translating to preclinical models, we were anticipating between 0.5 and 2 mgs per kg as a dose range where efficacy was seen. But obviously, clinical experience is -- never can be directly linked to preclinical work. So we remain to see exactly where we'll see efficacy. We hope that in the coming year, we'll be able to present more data on that.

The next question comes from Dane Leone of Raymond James.

Congratulations on the progress with both programs. So I'd like to probably focus my question to Dr. DeAngelo. In terms of the data that you see today, obviously defined by the 2 spliceosome mutations for the patients enrolled across AML and MDS. What do you think, from your perspective and your institution, makes the most sense to move forward with development? Do you see the best path forward right now as a monotherapy agent in development kind of like what IDH1/2 inhibitors were developed initially? And do you see the patient population molecularly defined receptive to the FDA and the clinical community like IDH1/2 or potentially what being under development with menin inhibitor class? And then finally, from what you understand of spliceosome mutations across these different patient groups, what do you think the actual addressable patient population would be at your institution for this type of therapy, either in a monotherapy setting or potential combination therapy, obviously, pending that clinical data?

Yes. So those are great questions and hard to be too specific without more data. So just to compare -- to draw a comparison with the IDH1/2 inhibitors. In de novo treatment, you see a really marked response. In the relapsed/refractory setting, a lot of it depends upon how heavily pretreated they were. And that gets to the question from the other gentleman who had asked whether, at least in the MDS patient population, what would you expect? In these Phase I studies, as you know, these patients have had -- are heavily pretreated. And so you're not going to -- it's hard to expect a lot from a single agent therapy. Having said that, the fact that we're seeing such high biologic responses in both the MDS and AML with regards to the spliceosome mutations is very encouraging. And I think that you can see a couple of pathways forward in the relapsed/refractory setting because right now, this represents about 25% to 30% of our AML/MDS patient population for which we really have no good therapies at all. And I think if a single agent activity is able to see who knows a 30% response rate, that may be good enough. And then we'll have to see what happens in the combination therapies. Of course, in the combinations that have been outlined, a lot of these patients have already had that. And so that may be an opportunity to see patients who are less heavily pretreated. It's hard to be too specific and answer your question more thoroughly without getting more data. But you can see this moving forward -- you can imagine it moving forward either as a single agent or in combination and less heavily pretreated patients with regards to the spliceosome mutation. With regards to the FLT3, we -- there's already a busy field with Gilteritinib and we're seeing this agent having responses even in those patients who have relapsed disease or resistant disease. And so that may be an opportunity to get an approval as a single agent in that particular disease setting. So different strategies for both different groups and opportunities both in the de novo as well as in the relapsed/refractory with respect to the spliceosome mutation MDS and AML. Really, it's kind of an open game because we have nothing here to compare with. I hope that addressed your questions.

The next question comes from Ed White of H.C. Wainright.

So first of all, I just want to ask a question on the FLT3 responder, I guess the best responder was the one that also had spliceosome mutation. Do you have to tease that data out when talking to the FDA when looking at the FLT3? And also, do you have any data on what the patient population could be for FLT3 and spliceosome mutation, the patient population?

So Bob, you're probably the best, yes.

Yes. Let me start to answer that. So the question regarding sort of teasing out what's driving this activity, the FLT3 versus the targeting of IRAK4. It's a really important point. I think, at this point, we don't know for sure exactly which contribution of both agents, but I can kind of add some light to that. So we do know that targeting IRAK4 is clearly an important event here. So if I go all the way back to some of our preclinical work and preclinical work done by others as well, we know that sort of hitting IRAK4 specifically, for example, with CRISPR/Cas9 or shRNA that inhibiting specifically IRAK4 is able to essentially reverse that cancer phenotype and really inhibit and kill cancer cells in, in vivo and in vitro model. We also know the clear contribution of IRAK4 based on the patient who did have a FLT3 mutation, but was refractory to Gilteritinib. So clearly, in this case, also the IRAK4 was playing a major contribution. I think in Dr. DeAngelo's patient, this patient had both FLT3 and IRAK4. We know that there's a -- based on the preclinical data that I showed, nice synergy between hitting IRAK4 and FLT3. So in this case, there may have been a contribution from both. We'll obviously have to discuss that point with the FDA and understand what they would like to see. Nevertheless, we think that they will be interested in all of these populations regardless of prevalence of FLT3 mutations versus spliceosome mutations. Regarding the broader prevalence, obviously, there is some overlap in our data alone. We see that 2 of the patients with FLT3 mutation also had the spliceosome mutations. Spliceosome mutations tend to occur early in the evolution of disease. It's a foundational mutation, which often remains present as the disease progresses from MDS to AML. You'll notice that many of the patients with AML in our study had transformed AML, meaning that they had transformed from earlier phase of the disease from MDS.

Okay. And then on the VISTA program for 8993, how should we be thinking about the next data set and the next doses that will be used? Will you continue to double? Are there other doses that you're thinking of? When can we expect to see some of that data from those higher potentially efficacious doses that you're talking about?

Well, I mentioned that we're going to be expanding our development with this drug, expect to present data within the coming years. So we haven't specifically identified a venue yet to present that, but we'll continue to dose escalate. This is a dose-escalating study where we're able to dose escalate until the maximum tolerated dose. And so we'll continue to do that and present the data at a venue that we'll announce at a later time.

The next question comes from Soumit Roy of Jones Research.

Congratulations on the robust data. One question possibly for Dr. DeAngelo. When you're mentioning the mCR status of these MDS patients, the ones who did not get stem cell transplant, was it due to age and health? Or is any other factor stopping them from getting the transplantation? And the second, the -- on the VISTA program, could you remind us the desensitizing regimen you were mentioning, Bob? And would you be collecting any paired biopsy of these patients?

So I don't -- so I haven't reviewed all the primary data, and I don't have all the data on the MDS and AML patients with spliceosome with regards to their age and what was the decision in terms to proceed or not proceed to transplant, so I'm not going to be able to answer your question specifically. But oftentimes, these are older patients and much sicker patients. And so the ability in these relapsed/refractory patients to proceed to a transplant is not trivial. So I just don't know the details of the patients ever presented. Sorry.

Yes. As you know, MDS tends to be a disease of the elderly. And so ultimately, it's a decision that the investigator and the patient make and is often based on a variety of factors, including age and overall comorbidities and health at that time. Your question regarding the VISTA program, so we had some -- when we brought this program in, we are very interested in how to overcome some of the challenges that Janssen had experienced, one of which was managing this CRS. And so a number of preclinical experiments were done in a model of CRS preclinically that showed that we could essentially mitigate and reduce the level of CRS in these models by starting out with some lower doses over a short period of time. And then once the higher doses were reached, the CRS was mitigated. And so we reasoned that we could put a strategy like that into our protocol. And essentially, this allows the investigator to give initially low doses of 8993 over a brief period of about a week-or-so prior to giving the full dose of 8993. And as you can see from the data, this has worked very well. We did not see dose-limiting toxicity. And in fact, it's quite striking that the exposure, as I mentioned, basically going from 0.15 to 0.6, the exposure increased 78-fold. And so despite that higher exposure, we're not seeing the dose-limiting CRS. And we think that, that's due to our regimen of desensitization. We've also incorporated some steroids into that dosing regimen initially. We obviously have tissue data from all patients before they start the study. As I mentioned, we're getting immunophenotyping data during the study. So that helps us understand the immunophenotype within the patient circulation. And we have the option for tissue biopsies, although that's not a mandated event on the protocol during the study.

The next question comes from I-Eh Jen of Laidlaw & Co.

I'm sorry, I was late. A lot of questions has been answered as well. Just one quick one. In terms of the FLT3 data, so far, that's very encouraging. So what's the next step in terms of in this particular cohort? Do you anticipate expand cohort to get to bottom of it to see whether this will be a cohort that to be pursued more aggressively?

Yes. We do plan to expand the opportunity for patients with FLT3 mutation to come on study. So this is 1 of the 3 genetically defined populations that we talked about today. Our initial thinking in presenting our data to the FDA will be focused on spliceosome. But there are some interesting data with these FLT3 patients. And as we build that database, obviously, we'll be talking about some of those patients who had the FLT3 when we talk to the FDA on the spliceosome data. But separately, there may be opportunities, and we do intend to expand our FLT3 experience as well.

Again, congrats on a very outstanding data.

Thank you, I-Eh.

This concludes our question-and-answer session. I would like to turn the conference back over to 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 the 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. And of course, a very special thank you to Dr. DeAngelo, both for his leadership in our AML and MDS study and for joining us today. We look forward to updating you again soon. Operator?

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