Aptose Biosciences Inc. (APS) Earnings Call Transcript & Summary

Good afternoon, everyone. According to the name badge, I'm Dr. William Rice. I'm the Chairman and CEO of Aptose Biosciences. And I want to welcome you here this afternoon and also thank you for joining us for this KOL event both all of you who are present today as well as those who are listening remotely. I also want to, before we get into the activities, introduce everybody to the members of the management team that are here today. You see, Mr. Greg Chow, over by the door, Chief Financial Officer; Dr. Joti Marango, Chief Business Officer, also over by the door; and here, we have Dr. Rafael Bejar, our Chief Medical Officer. So hopefully, I have the opportunity to engage all these individuals, okay? So finally, let's talk about or remind you of why we're here today. So I would -- I suspect, on many occasions, you've heard me talk about our small-molecule kinase inhibitor, known as CG-806. We've also -- we refer to it as 806. This molecule is a highly -- has a highly unique kinase inhibitory profile, and that unique profile then positions it for development across a broad span of hematologic malignancies, including lymphoid malignancies as well as myeloid malignancies. So 806 has garnered a bit of notoriety recently because of its role as a noncovalent or reversible BTK inhibitor being developed for the B-cell malignancies. In fact, 806 is in a clinical trial currently for patients that are being treated with CLL as well as non-Hodgkin's lymphomas, and we're continuing to dose-escalate the patients on that trial. We're thrilled about that trial and the fact that 806 is a reversible BTK inhibitor. But it's also very important to remind you that 806 is far more than just a reversible BTK inhibitor. And so what we want to do today is to bring awareness, shine a spotlight on the fact that 806 is also going to be developed for the myeloid malignancies. In particularly, when we talk about myeloid malignancies, we're talking about acute myeloid leukemia or AML. So today, what we've done is we've assembled a stellar group of key opinion leaders. What they're going to do is give you a quick overview on AML, the disease itself, the current therapies, the needs of the patients going forward as well as how 806 might address the needs of those patients. So rather than stealing all the thunder, what I'm going to do is I'm going to ask our Chief Medical Officer, Dr. Rafael Bejar, to serve as the moderator today, and he'll introduce our KOLs. So thank you so much. Thank you for being here.

Thanks, Bill, and it's great to have you all here in the room and to those of you who are on the webcast, welcome. I'm glad you could spend your lunch hour with us learning a little bit more about AML and how CG-806 might fit in that landscape. We have a great series of speakers today that are going to do most of the talking. The first is Dr. Eytan Stein from Memorial Sloan Kettering. He is a hematologist oncologist, treats patients with AML and has played pivotal roles in developing targeted therapies in AML, including taking drugs all the way through to approval. He'll tell us a little bit about AML in the age of targeted therapies, how has treatment for this disease changed in the last few years, where we've seen a lot of drug approvals and a lot of changes in our therapeutic approach. The second speaker will be Dr. Aaron Goldberg, also from Memorial Sloan Kettering, another expert in leukemias, and he'll be speaking more in the area of FLT3 inhibitors, specifically, how has that landscape changed and what are the emerging challenges that we see now that we have used those drugs in the clinic. And finally, we'll have Dr. Druker, who has kindly come all the way over from the West Coast to speak with us a little bit about his impressions. So I'm sure you all know who Dr. Druker is, but he has been the pioneer in establishing tyrosine kinase inhibition in cancer, not just in hematology but in cancer in general. And I look forward to hearing his perspective on how CG-806 compares to other molecules he's worked with and what he sees as the potential use for this drug in the future. So without further ado, since we have a lot of slides to get through, I'm going to start with inviting Dr. Eytan Stein to come up and tell us about AML in the age of targeted therapies.

Okay. Thank you very much. Thanks, Dr. Bejar. So we're going to be talking a little bit about treatment of acute myeloid leukemia in the age of targeted therapies. These are my disclosures. So all of you know that what we have discovered over the past 10 to 15 years is that acute myeloid leukemia, while it has the same phenotype of having a black and myeloid differentiation at the myeloblast stage of development, it's really driven by many, many different molecular and cytogenetic alterations that are depicted here in this slide put together by Hartmut Dohner in the European leukemia net criteria for the diagnosis of acute myeloid leukemia. And it's becoming clear that, in order to target this disease, you really have to pick off the individual genetic events that are happening, such as FLT3 inhibitors, IDH inhibitors that we're going to talk about in a second, to make any real progress in the treatment of this malignancy. So my colleague, Dr. Goldberg, is going to talk much more extensively about FLT3 inhibitors. I'm just going to say very briefly, the FLT3 ITDs are very, very common and FLT3 tyrosine kinase domain mutations, together, they occur in about 30% of patients with acute myeloid leukemia. There have been multiple attempts to target FLT3, and having a FLT3 mutation, an ITD or a TKD, despite what the European leukemia net says that maybe some of these patients might be favorable risk disease, I would still consider them all unfavorable risk disease who are very high risk of relapse and then death. You all know about the Phase III RATIFY trial, which was a randomized controlled trial comparing the combination of a moderate FLT3 inhibitor called midostaurin in combination with chemotherapy to placebo with chemotherapy with the key outcome of this trial being overall survival. And at 5 years, there was a 7% overall survival benefit with the multi-kinase inhibitor that also targets FLT3 midostaurin compared to placebo. The second thing that Dr. Goldberg is going to talk about but I want to mention very quickly is the use of targeted FLT3 inhibition, relapsed and refractory acute myeloid leukemia. So this was a randomized controlled trial published a few months ago in the New England Journal of Medicine by my friend and colleague, Dr. Alexander (Sasha) Perl, randomizing patients with relapsed and refractory FLT3 positive acute myeloid leukemia to salvage chemotherapy, which you can see on the bottom after the 2:1 or the targeted FLT3 inhibitor gilteritinib. And the results of that trial show that there is a overall survival benefit with the median overall survival with gilteritinib of 9.3 months versus 5.6 months with salvage chemotherapy, making gilteritinib, really, I think, the standard of care right now in patients with relapsed and refractory AML. So what about other targeted therapies that we can use. I've worked a lot, like Dr. Bejar said, on targeted inhibition of IDH1 and IDH2. I find this very interesting because IDH1 and IDH2 inhibitors function as differentiation agents. So what does that mean? So what that means is that when you have a mutation in IDH2, which is found in the mitochondria, or IDH1 in the red, which is found in the cytoplasm, the end product of the normal reaction, which is alpha-ketoglutarate, gets converted to what we call an oncometabolite beta-hydroxyglutarate, and increased levels of beta-hydroxyglutarate lead to a freezing of the cell in the undifferentiated state in the myeloblast stage of development. And then if you target mutant IDH, you lower intracellular levels of beta-hydroxyglutarate, allowing the cells to then differentiate and then getting rid of the disease. So this acts as a differentiation agent rather than acting as a cytotoxic. Interestingly, gilteritinib, the FLT3 inhibitor, can also act as a differentiation agent rather than acting as a cytotoxic. So there are a couple of different IDH inhibitors, one of them that has been approved by the FDA is the IDH2 inhibitor, enasidenib, and relapsed and refractory AML, based on this clinical study over here, showing then overall response rate of 40% in patients with relapsed and refractory AML got 100 milligrams of enasidenib once a day and a complete remission rate of about 20%. And then when you look at this single-arm trial, the median overall survival, the median overall survival in all patients who are treated was 8.8 months. Obviously, the patients who had achieved a complete remission, do much better. Those patients have an overall survival -- median overall survival of 23 months. Now what happens if you take IDH2 inhibitors and combine them with standard of care and enasidenib for newly diagnosed patients who are unfit for intensive chemotherapy. So this was a randomized trial presented at ASH this year by Courtney Dinardo, and forget this part over here. This is the part that's important over here, where patients with newly diagnosed IDH2 mutant AML were randomized to receive azacitidine alone or azacitidine with the IDH2 inhibitor enasidenib. Interestingly, the overall response rate is dramatically higher, 71% versus 42% in patients who received the combination. Similarly, the CR rate is much higher, 53% versus 12%. However, the overall survival is equivalent, okay? So that might be, if you look at the event-free survival curves here, the combination appears to be numerically better than giving azacitidine alone. And what we know is that some of these patients who received azacitidine alone, specifically 7 patients, they actually -- when they did not respond to azacitidine monotherapy, they crossed over and were able to receive enasidenib sequentially, and that might be what led to the equivalent overall survival. But I think it does raise the question, when you're thinking about combined therapy with IDH inhibitors or any targeted therapy with azacitidine, it raises the question, should you be giving these drugs together? Or can you be giving them sequentially and get the same or perhaps even a better effect. What about the IDH1 inhibitor, ivosidenib? So similarly, in a clinical trial of patients with relapsed and refractory acute myeloid leukemia with an IDH1 mutation, the overall response rate was quite high at 41.6%. The rate of CR was 21.6%. And then this new category we have called CR with partial hematologic recovery in combination with CR was 30.4%. Similar to the IDH2 inhibitor, the median overall survival that's depicted in black was 9 months, 8.8 months for the IDH2 inhibitor, and patients who achieved a CR or CRH depicted in blue at the top, had a median overall survival of 18.8 months. Okay. So moving on to different targets, targets that might be available in all patients with acute myeloid leukemia, not just a genetically mutated target. There's been a lot of activity about inhibiting BCL-2, not only in CLL but also in AML. So you all know the BCL-2 is an anti-apoptotic protein. So you get overexpression of BCL-2, cells live. You inhibit BCL-2, the malignant cells are supposed to die. And when you combine hypomethylating agents with venetoclax and newly diagnosed patients unfit for intensive chemotherapy, you get these really dramatic rates of CR and CRi of about 70%. And those rates of remission really persist, whether you have intermediate risk disease or unfavorable risk disease -- poor risk disease. So here, you're getting 60% remission rates in patients with poor risk cytogenetics. And what that translates into is median overall survival in all patients treated of 17.5 months. And this is regardless of what dose of venetoclax these patients got. I think what's important to note when you look at these slides, which are shown everywhere, is that just because the rates of complete remission are high, the rules still stand, meaning unfavorable risk patients continue to do poorly. And why is that? They all go into remission, but the unfavorable risk patients relapse very, very quickly. So if you have a p53 mutation, we're getting these p53 mutant patients into remission now, but they're relapsing 2 months after going into remission, while the NPM1 mutant patients, which will be favorable risk, those patients may have very, very prolonged duration of response. And if you look at the sort of the survival curves, the poor risk disease continues to do bad, the good risk disease continues to do well. So BCL-2 inhibition doesn't alter the underlying biology of how you think a patient would do. It doesn't turn bad-risk disease into good-risk disease. It's still bad-risk disease. Okay. Finally, I want to talk about this trial here because I do think this is potentially practice changing. We'll just go back over here. This is the first randomized controlled trial to show an overall survival benefit with maintenance therapy in acute myeloid leukemia. There was actually a randomized controlled trial with a histamine agonist called Ceplene that showed a progression-free survival benefit, but this is a trial taking patients in complete remission, older than the age of 55, but a performance status of 0 to 3 intermediate or poor-risk cytogenetics who did not have an allogeneic bone marrow transplant. And then randomize those patients to receive CC-486, which is an oral formulation of azacitidine or randomized them to receive placebo. Patients who had a CR continued CC-486. Patients who had sort of an early relapse were able to increase the dose of CC-486, and patients who had a real relapse had to stop treatment, no matter if they were getting placebo or CC-486. And this is really, really interesting. You can see that there is an overall survival benefit for those patients who received CC-486, 25 months versus 15 months in these patients. And I do think for patients who don't have an allogeneic bone marrow transplant, I think maintenance therapy is really going to be a real thing, either with CC-486 or in the context of patients who have FLT3 mutations getting a FLT3 inhibitor, either if they don't have a transplant, without a transplant or post-transplant for those patients who have had an allogeneic bone marrow transplant. With that, I'm going to move it on to Dr. Goldberg. Thank you very much.

All right. Thanks so much, Eytan. And thanks to Raf and the organizers for this opportunity. So I'm going to talk more specifically about FLT3 inhibition in the context of acute myeloid leukemia, recent advances and also emerging challenges that Eytan and I see in the clinic all the time. These are for my disclosures. So I think just backing up, it's important to remember that even now in this era of novel and targeted therapy, AML is a deadly disease. Unfortunately, the vast majority of patients that Eytan and I see in the clinic will die of this blood cancer. It can be diagnosed at any age, but in -- the median age of diagnosis is 68 years old. It can progress rapidly. It can be very difficult to treat. It's actually the second most commonly diagnosed leukemia just after CLL with 21,000 people diagnosed with AML this year, but it's not so prevalent. And the reason for that is because so many people die of the disease. So with 10,920 deaths approximately from AML occurring this year and a 5-year survival rate even now of about 28.3%. We see that AML is a deadly disease even in so-called favorable risk subtypes. This data is a little bit older, but the curves don't look, unfortunately, all that different even now. Favorable-risk patients, a 55% chance of overall survival doesn't sound so very favorable when you're a patient, and it's obviously much worse if your intermediate or unfavorable risk. And that's, of course, based upon the cytogenetics and molecular abnormalities at diagnosis. And this is particularly true when you look at older adults. And again, we talked about the median age of diagnosis is 68 years old. Many of the benefits that we've seen, at least until recently, in acute myeloid leukemia have gone to younger patients. This is just looking at survival curves in MRC trials in England from the '70s up to the 2000s. And you can see that, over time, younger patients seem to have done better. Not so much with older patients, and a lot of that may just be due to advances in supportive care, antifungal medications, ICUs, patients who were able to survive a therapy more -- better, but they weren't actually getting biologically directed treatment. FLT3 activation is actually the most common abnormality in AML, backing up to what Eytan showed you, that slide, in the modern era, what we would like to do ideally is target the underlying biology of the disease. So this is actually looking at genomic sequencing of over 1,500 patients. This is from Elli Papaemmanuil at Sloan Kettering. And what she found is that there are a variety -- new heterogeneity of mutations in AML. One really stands out and that's FLT3, which is tyrosine kinase that's found mutated in 25% to 37% of patients. So really about 1/3. FLT3 is a tyrosine -- a receptor tyrosine kinase. There are 2 in general types of mutations in FLT3 in AML patients. There's these ITD, internal tandem duplications, in the juxta-membrane domain. That's in about 25% to 30% of patients. This really confers a high relapse risk and, unfortunately, makes patients really less sensitive to intensive chemotherapy and is clearly a poor prognostic marker. FLT3-TKD, tyrosine kinase domain mutations, about 5% to 10% of patients. The prognostic significance, I would say, is still controversial in the context of chemotherapy, but they certainly can be a mechanism of resistance to FLT3 inhibitors, in particular, TKD mutations confer resistance, for example, to quizartinib, which we'll talk about. Here's the data showing that FLT3-ITD mutations, in particular, are poor prognostic markers in AML. Back in 2002, these patients who received intensive chemotherapy, the ITD mutation patients, obviously, did the worst, and then D835, perhaps, really worse than FLT3 -- non-FLT3-mutated patients. The worst outcomes were actually seen in patients who were triple mutated for NPM1, DMNT3A and FLT3-ITD, which is actually the most common triplet occurring mutation in all of AML. So in terms of the landscape of currently available FLT3 inhibitors that's summarized here on this slide, there are 3 FLT3 inhibitors that are really approved worldwide, only 2 in the United States. So midostaurin, so I really think of this as a dirty drug, right? It's a multi -- truly multi-targeted but really dirty tyrosine kinase inhibitor. It's given orally. All these drugs are oral, twice daily, FDA-approved in 2017. It's approved only in the context of newly diagnosed patients and only in combination with chemotherapy. It has been tested, of course, as monotherapy in relapsed and refractory AML and has very poor response rates but does seem to provide some overall survival benefit for which we'll talk about. It targets both FLT3-ITD and TKD mutations, and there are some adverse effects, particularly GI toxicity. Gilteritinib is the only FDA-approved FLT3 inhibitor in the context of relapsed and refractory AML with FLT3 mutations, more recently approved. It's approved as monotherapy. So as a pill alone, targets both ITD and TKD and also has its own set of side effects, although I would say, overall, well tolerated, but we certainly can see LFT abnormalities, CK elevation and other rare toxicities, including myopericarditis, which I've seen. Quizartinib. So this is an interesting story. So quizartinib is approved in Japan on the basis of data, which I'll talk about, but not FDA-approved in the United States, also tested in relapsed and refractory AML, approved in Japan as monotherapy and really only targets FLT3-ITD. Does not target the FLT3-TKD mutations, which really serve as a key mechanism of resistance and adverse effects really are QT prolongation in particular. So Eytan already showed you the RATIFY study, which led to the FDA approval of midostaurin. These were fit patients up to the age of 59. And remember, the median age of diagnosis of AML is 68. So these are younger patients, right, than even the median age of diagnosis, younger patients with FLT3-ITD or TKD-mutated AML with a greater than 5% FLT3 allelic frequency who were randomized to receive intensive chemotherapy plus midostaurin intensive chemotherapy plus a placebo. If they achieved a complete remission, they could go on to get maintenance. And they could actually also go and just receive allogeneic stem cell transplants. I'll use the pointer now. The remission rates on this study were superior in the midostaurin group but overall relatively similar, 59% versus 54%. And when you look at the median overall survival, this looks like an enormous difference, 74 months versus 25 months, but it's really because that's where the curves separate. When you look at the actual landmark of 4 years, the difference in overall survival is about 7%. So we agree that this is the new standard of care for patients who are fit for intensive chemotherapy with newly diagnosed AML or FLT3 mutation. And of course, as we say in the clinic, you'd rather be on the blue curve, right? But at the same time, this is not, I would say, as dramatic and advanced as we would like, and there's certainly room for more effective FLT3 inhibition in the upfront setting. So in terms of limitations, as I spoke about, this was a younger group of patients. The median age of diagnosis is 68, a marginal benefit with overall survival of 51% and still half the patients dying of their disease. Midostaurin is a dirty TKI and among the least potent FLT3 inhibitors, and there are more potent FLT3 inhibitors, including crenolanib and gilteritinib, that are being combined with chemotherapy and that are very promising Phase II studies with Phase III randomized studies that are ongoing. So turning our attention really to, I think, one of the most interesting topics, which is relapsed and refractory AML, and particularly the stories on quizartinib and gilteritinib. So backing up, relapsed and refractory AML patients, this is a different patient population than what we just saw in RATIFY. These are patients who've already received prior therapy. So they've, for example, gone into remission then relapsed, or they were just refractory to upfront therapy and never even went into remission. And of course, they have a very poor prognosis, particularly -- that's really across the board, but particularly for the FLT3 mutation. So this is looking at patients with a FLT3 mutation who are either relapsed or refractory and relapsed within 6 months in the Cephalon 204 trial. And this is a John Hopkins cohort, a similar group of patients, and we're looking at median overall survivals of 3 to 6 months. This is a terrible, obviously, survival rate, and it's going to be devastating, of course, as a patient, if your leukemia comes back, and they're really, up until recently, particularly the FLT3 mutation, there really wasn't much to offer these patients with intensive chemotherapy, CR rates of about 12%. So this brings us to the story of quizartinib. So quizartinib is a FLT3 inhibitor that specifically targets FLT3-ITD. It does not target FLT3-TKD mutations. And this was developed by Daiichi Sankyo, and it was tested, in this case, in a randomized Phase III study now, most recently, in 367 patients with FLT3-ITD positive AML who were relapsed or -- within 6 months or refractory into intensive chemotherapy. They were randomized 2:1 to receive either quizartinib, which again, this is a pill, a pill alone versus a variety of chemotherapy backbones, a nonintensive low-dose cytarabine regimen or intensive salvage chemotherapy. If they responded, they could go on to receive allogeneic stem cell transplant and can continue maintenance quizartinib. And they were randomized 2:1. 245 patients receiving quizartinib, 120 receiving salvage chemotherapy. 32% of patients in the quizartinib did receive hematopoietic stem cell transplants after treatment, only 11.5% in salvage chemotherapy. And then this is -- one thing that caught the FDA's attention for those patients who were randomized to salvage chemotherapy, 23% did not receive treatment but 1.6% received treatment with quizartinib. Why is that? So -- but as a clinician, that's not totally surprising to me. Eytan and I were talking about this in the -- after the ODAC meeting. If my patients randomized received salvage chemotherapy, and they're really not fit for that. What are you -- I mean, how are they going to even go onto that study or maybe they might say, you know what, there's another study of a FLT3 inhibitor down the road. I'm going to go do that. This doesn't seem like a good idea. So it's not -- and at some level, surprising, but it is profound imbalance, obviously, in the arms, which caught the attention of the FDA. So looking at the response rates, the response rates were superior, for sure, in quizartinib, the overall response rate of 69%, even looking at a complete -- composite complete remission rate of 48% versus 27%. And there was a modest but statistically significant benefit in terms of overall survival with a mean overall survival of 6.2 months versus 4.7 months. And again, this is modest. And really, because of the very modest nature of this benefit and because of also the adverse effects that were seen in terms of QT prolongation and because of the imbalance in the arms and concerns about the quality of the data, the FDA actually recommended in their the ODAC meeting not to approve quizartinib in the United States. But based upon the same data, it is approved in Japan. One of the reasons also that we see the response rates are probably so modest is, again, quizartinib only hit FLT3-ITD and not TKD mutations. And this is, I think, really beautiful work by Cathy Smith and colleagues. It was published in Nature, looking at specific mechanisms of resistance, and a lot of them really converge upon emerging mutations in FLT3-TKD, tyrosine kinase domain mutations, that came up in the setting of FLT3 inhibition. So FLT3 inhibition is important and was a clear driver, but when you gave a targeted therapy that only hit one specific target in FLT3, there were clonal evolution and emergence of mutations that conferred resistance. So it validates FLT3 inhibition, but there were, in particular, multiple mechanisms of resistance, but in particular, D835 TKD mutations. There are FLT3 inhibitors that are currently undergoing development that are active against FLT3-TKD mutations and, in particular, gilteritinib. So Eytan mentioned, this is the Phase III ADMIRAL study. This is a study that, again, similar to quizartinib, randomized patients with relapsed and refractory AML to receive either oral targeted therapy with gilteritinib versus salvage chemotherapy. They're randomized 2:1. And the primary endpoints really were overall survival as well as CR and CRH rate with other secondary endpoints of event-free survival. And in terms of the salvage options, again, also low-dose cytarabine, azacitidine or MEC or FLAG-IDA. These are intensive salvage regimens. And based upon the data from the study, gilteritinib is now approved, both in Japan and in the United States. And it's really, of course, they're the only FLT3 inhibitor approved in the United States in the context of relapsed/refractory AML and as monotherapy. So I think what's so striking about this already is that, if you do target the biology of the disease, you can actually provide benefit even over multi-chemotherapy drugs, right, multi-agent chemotherapy. So that's striking. In terms of the remission rates, 54% versus 22% in terms of the composite complete remission rate, 26% of patients being able to be bridged to allogeneic stem cell transplantation. And the survival curves are different. A mean overall survival of 9.3 months versus 5.6 months. If you look at a landmark of 12 months overall survival, 37% of patients still being alive on the gilteritinib arm. Again, an oral target therapy alone versus salvage chemotherapy, only 17% of patients. And this survival benefit is really what led to the FDA approval of gilteritinib. And if you look at the patients who do the best, it's really those patients who are able to be bridged to allogeneic stem cell transplant. What I think is really clear is that gilteritinib alone is not a cure. I don't sell this to my patients as a cure. I don't say we're going to give you this pill, and we're going to cure of your leukemia. I'm hoping this will provide some benefit to you and prolong your life and give you a good quality of life, maybe not have you to have transfusions, for example, for a time. If they're candidates for allogeneic stem cell transplantation, then yes, I'm trying to actually bridge them to allogeneic stem cell transplant, which could be a potential cure. But even in the context of stem cell transplantation, you still see that many and, really, most of these patients eventually are progressing and dying from either complications of the transplant or of relapse of their disease. Outcomes do seem to be better, again, validating the concept of FLT3 inhibition, and those are patients who resume gilteritinib. But of course, there are confounders, which is those patients who resume gilteritinib, obviously, were healthy enough to do so, for example. But going back to the overall survival curves, although I think gilteritinib is an incredibly important advance, this is not curative therapy. And eventually, everyone in this therapy appears to be progressing basically. Over time, if you don't bridge the patient to allogeneic stem cell transplantation, you're not going to be curing the patients of their AML. And I see these patients all the time in the clinic, and we'll talk about the mechanisms of resistance. But this, in my mind, represents an urgent clinical need for these patients. There's a really important tool out there, but it's not enough. So when you look at the mechanisms of gilteritinib resistance, and this is, I think, a really interesting paper in cancer discovery by McMahon and colleagues. So these are oncoprints. So we're looking at here, these are just different patients, 41 patients that were enrolled in the study. And so those are the different columns, different patients. The rows are different types of mutations. So you can see in blue, all the patients had different FLT3 mutations. They either had FLT3-ITD mutations to the top or FLT3-TKD D835 mutations here in the second row, and other mutations are indicated below. If you look at the time of progression on gilteritinib, the landscape -- the genomic landscape, in this case, looks different, at least in many patients. We see this all the time in the clinic. AML is a polyclonal disease in many patients and most patients, really, at diagnosis, in particular, in response to therapy. And so in some cases, you simply see that the gilteritinib is effectively treating the FLT3 mutated clone, but then other clones at AML just grow out. And so you see a loss of FLT3. In many cases, you see that those other clones actually grow out with NRAS or RAS mutations. RAS mutations, in general, was the story of this paper. These are mutations that really confer a proliferative capacity to the AML, and this seems to be a primary mechanism of resistance to the targeted therapy with gilteritinib. And it seems to grow out in a large number of these patients and confirm resistance enforcement of the therapy because it's not targeted by gilteritinib. There's also this gatekeeper mutation FLT3 691L, which wasn't present at diagnosis but then emerged basically on treatment with gilteritinib that seems to not be hit by gilteritinib. There's also a cytogenetic evolution that [ crit ] the carry type and chromosomal complexity increases over time. And then there are other mechanisms. If we made a whole genome wide oncoprint, I'm sure there would be other mutations that we would see that are simply not captured on these next-gen sequencing panels that were used. But there are other mechanisms of resistance. But what you can see interestingly, most of these patients still have the target FLT3-ITD and TKD, but yet they are resistant to gilteritinib. And what about venetoclax? I think venetoclax is certainly one of the most exciting advances in AML treatment, not as a monotherapy but really given in combination with either hypomethylating agents or low-dose cytarabine approved in the context of newly diagnosed AML patients. And at least upfront, the response rates with FLT3 mutation seem to be close, basically to non-FLT3 mutated patients. But over time, again, FLT3-ITD mutation seems to serve as a mechanism of relapse. This was a paper from Courtney Dinardo. They're just published in Blood, looking at mechanisms of clonal evolution in response to patients who are treated upfront with venetoclax combinations. And you can see that patients with FLT3-ITD are diagnosed with a lower [indiscernible] frequency, over time, can go up. In this patient, actually, when they achieved remission, all of a sudden, everything is undetectable, and that's great. But then at the time of relapse, it appears to be FLT3-ITD. It's a driver of this particular clonal relapse. Similarly here, and down here, even though you didn't have any detectable FLT3-ITD at diagnosis, over time at relapse, it became a strong driver of the clone. So I think the story here with FLT3 is that outcomes have improved with our better understanding of the biology of the disease. But most patients with AML, particularly FLT3 mutated AML, still die of their disease. FDA-approved combinations include midostaurin, which has given really and approved only in the context of newly diagnosed AML for patients who are fit to receive intensive chemotherapy in combination with chemotherapy. Gilteritinib is the only FDA-approved FLT3 inhibitor which is approved as monotherapy in the context of relapsed and refractory AML. Quizartinib has similar patient population, is not approved in the United States but approved in Japan and, again, only hits FLT3-ITD and not TKD. Current FLT3 inhibitors however are not curative. As we saw, there are multiple mechanisms of resistance. And I see this -- we see this in the clinic all the time. Even responding patients will eventually progress. The mean overall survival is still only 9.3 months. And I think the development of FLT3 inhibitors for FLT3-mutated AML patients, particularly who fail gilteritinib, represents an urgent clinical need. I see these patients all the time. And what I would -- if I could make a list sort of of characteristics of a drug for -- that would be an ideal drug for gilteritinib failures, I would want a drug that's a potent FLT3 inhibitor, highly potent, it should target both FLT3-ITD and TKD. It should also target multiple mechanisms of FLT3 resistance. I want it to be active also in the context of RAS mutations because that's a clear mechanism of resistance to gilteritinib, and ideally, it was synergized with venetoclax. And that's, of course, the reason that we are here is that I think that CG-806 is that drug. That's why Eytan and I are so excited to participate in this upcoming study. So with that, I would like to thank you for your attention.

That's great. I want to thank Dr. Stein and Goldberg for their presentation, and I think that they've done a lot of the heavy lifting today. My role here is now to just tell you a little bit about what we know about CG-806. And why we are excited about putting it in the clinic for AML. I don't know if you guys know, but I joined Aptose about a month ago. Prior to that, I was working as a physician scientist at UCSD, where I have a lab that focuses on myeloid malignancies, MDS in particular, and a clinic where I see patients with the same disease. So my understanding of the myeloid disorders and the need in the space was one of the things that actually drew me to join the company in the first place was that I said this is an incredibly promising agent. It really is very different than some of the other things that are out there. As you know, it is not just an oral FLT3 inhibitor. It also has activity as a reversible BTK inhibitor. And that makes it very unique. It's really a first-in-class in that way. Not only does it inhibit these enzymes, but it also inhibits the wild-type and mutant forms of both BTK and FLT3 that gives us broader applicability in patients that either have or don't have the mutations might still be candidates for treatment. And as I'll show you, it targets more than just these 2 kinases that it really does target other very important kinases that confer additional signaling to the cells that they might be able to use as a mechanism of resistance. And fortunately, at least as far as we can tell with all the preclinical data that we've been able to gather, that it appears to be very safe. It spares those kinases that we believe are associated with more toxicity and have given other drugs in this class difficulty in clinical studies. I'll give you a quick update on CLL, where we are CLL and non-Hodgkin's lymphoma because we're in the clinic with CG-806 in those patients. And I'll tell you how that is going to inform our ability to open the AML trial. So no talk about a kinase inhibitor is complete without the kino map. I'm sure you've seen this thousands of time, but what I want to point out is that this is not like midostaurin. This is not a drug that hits kinases all throughout the map. It really hits kinases in restricted clusters. So we refer to this as a cluster-selective kinase inhibitor. You can see it has activity in this cluster labeled BTK. This is the TEC cluster. Interestingly, it doesn't actually inhibit TEC itself, which is one of the toxicity-associated kinases. It has activity against TRK and has activity in the FLT3 kinase domains, including neighboring kinases that I think are very important for AML, in particular, like CSF1R and PDGFR-alpha. So why is it important to hit more than one kinase when you're going after a target. As Dr. Goldberg showed you, there are multiple mechanisms of escape that cells can use, either by mutating the target itself or by mutating neighboring pathways or up-regulating neighboring pathways that allow the cell to bypass that targeted protein. So in AML, we think of FLT3 as being the target in CLL and non-Hodgkin's lymphomas. We think of BTK as being the target, but the reality is that all of these often coexist in the same cell, whether it's AML or CLL that we're talking about. And in particular, if you inhibit FLT3, you can see that all the other pathways that are available to the cell can crosstalk and activate those downstream signaling genes, essentially bypassing the target that you've hit with a very selective inhibitor. So CG-806 has the good fortune of actually targeting multiple of these receptors, starting with the surface receptor tyrosine kinases as well as intracellular receptors, intracellular kinases that can activate downstream pathways. So these are where CG-806 directly inhibits these enzymes. But as a consequence, it also down-regulates the activity of all of these downstream signaling molecules, as I'll show you in the next slide, and ultimately, down-regulates the important survival signals that get transcribed to get up-regulated in the setting of cancer. So my thesis adviser once said that if you show more than one western blot on a slide, you have to apologize. So a thousand apologies, but I think that this actually is useful. What I'm trying to show here is that there are multiple different AML cell lines that we've looked at where we've treated them with nanomolar concentrations of CG-806 and looked at what happens, not only to the enzymes that we target directly, like FLT3 and BTK, but also those other downstream signaling molecules that get phosphorylated when they -- that signaling pathway is in its active state. And you can see that, even with these low doses in these in vitro studies, you get loss of those phosphoprotein signals, like loss of Phospho-Akt, loss of Phospho STAT5, loss of Phospho-Sic, loss of pS6K and, ultimately, down-regulation of some of these very important signal transcriptional profiles, like those driven by c-Myc. And these are AML cell lines. Yet, you can still see that BTK has activity in the untreated state. And then when you treat them, that activity goes away. So where are we with the development of this drug? As I mentioned, we are in Phase I clinical trials for CLL and non-Hodgkin's lymphoma. We, again, had the good fortune of having a fairly clean preclinical safety signal. And we're allowed to do rapid-dose escalation, where we could put it in single patients at the lowest doses and eventually get to a cohort where we need to put 3 patients on at a time. And so far, we have completed Dose Level 1. We have completed Dose Level 2. And we have fully enrolled Dose Level 3. This is important because this is going to inform how we carry out and when we carry out our AML study. Essentially, we didn't feel like it was appropriate to treat AML patients who have very short expected overall survivals in the absence of therapy with drugs that we knew were going to be subtherapeutic. We really wanted to do our dose escalation in a population that was going to be more tolerant of taking subtherapeutic doses and then be able to translate what we think is going to be an effective dose into the clinic with AML at the start. So what have we learned from the B-cell study so far? So far, we have evidence of safety that the drug does not seem to cause myelosuppression in these patients that we treated, including some that have very compromised bone marrow function to begin with. We haven't seen any SAEs or any dose-limiting toxicities. And we have seen evidence of engagement with a target. I would call this pharmacologic activity. Where we see down-regulation of Phospho-BTK, both in the cells from treated patients themselves and in cells that are exposed to the plasma from treated patients where the drug is actually in circulation. We have a hint of clinical activity as well. But as I'll show you, we don't necessarily equate Phospho-BTK inhibition with the therapeutic response, that those 2 things need to be -- need to occur, but the dose is going to be very important. And fortunately, we have well-behaved steady-state oral pharmacokinetics, which I'll dive into here. So these are the PK curves from our first and second patient. We've presented this before. You can see that, with our first patient, we achieved a steady dose level with twice-a-day dosing. Our second patient at 300 milligrams twice a day actually achieved quite high levels. They're about 0.6 to 1 micromolar. These are levels that we think are just below where we would expect it to see some therapeutic responses. And as I told you already, we do see some inhibition of the target pathways even at that dose level. What we hope to do is get up into the 1 to 2, 3, 4 micromolar range. This is where, based on preclinical data and our experience with other drugs, we expect that we are going to see on-target inhibition and the potential for efficacy. So this is where we want to get to. We're hopeful that, that third dose cohort that we fully enrolled will bring us there that we'll see the PK data from those patients and make us feel comfortable that when we go into AML, we'll have a potentially therapeutic dose, but if that isn't the case, we'll have the opportunity to dose escalate in the next cohort and learn whether that's the appropriate dose. Regardless, when we do get to AML, we are going to be wanting to dose-escalate the most patients as well to make sure that we're treating them as effectively as we can. Skip this slide real quick. Now I want to share some of the preclinical data that really drove us to consider AML as a primary indication for CG-806. The first is obvious is that it inhibits FLT3-ITD extraordinarily well. With 800 picomolar, we reach an IC50 for this drug. That compares favorably or even better than many of the other drugs that are in development that you've heard about today. But in particular, we have activity against the tyrosine kinase domain mutation that Dr. Golberg referred to. Sometimes, this occurs in de novo patients, but often it occurs in emerging resistance in patients that are treated with drugs that don't adequately cover these mutations. And if you take cell lines that are driven by this mutation, you can see that CG-806 is extraordinarily effective at killing those cell lines, almost a thousandfold better than drugs like quizartinib, at killing these TKD mutation-driven cells, giving us hope that at least for this mechanism of resistance that we're shutting that door for patients that have FLT3-ITD. Now that is obviously in a petri dish where we're exposing cell lines to drug. What happens when you actually go into animal models. This is a very typical AML model when you get MV4-11 cells injected under the skin and allowed to grow over time. This is a cell line that's driven by FLT3-ITD. And you can see that the vehicle-treated mice die within short order. And as we go to increasing doses of CG-806, the mice live longer and longer. And at the 2 highest dose levels, the mice are essentially cured of this disease. We didn't -- at the time that we had to eventually sack the mice, they were still alive and healthy. And importantly, we didn't see anything that would be concerning for a signal of toxicity in these animals. There wasn't a preferential loss of weight. If anything, the animals that were treated looked healthier, obviously, than the ones that were treated with vehicle. But I want to do a slightly deeper dive on these last 2 dose levels because I think it's also instructive and informs how we're going to do our clinical study. So this is the same data shown in a slightly different way. The black shows animals that were treated with vehicle, and it shows that the tumor growth occurs fairly quickly. This is during the 28-day dosing period, where mice were given CG-806 twice a day. And you can see for the animals in the 2 highest dose levels, that completely suppressed tumor growth. These animals didn't show any tumor progression. You stop dosing after 28 days, you can let the animals room and see what happens. And for some of the animals, they were outright cured. 5 of the 11 in this middle dose range and 10 of the 11 in the highest dose range didn't show any recurrence of disease. However, 6 of these did, and one of these did. We let those tumors grow out again and then retreated these mice, this time with the highest dose level, and we were able to cure all of the mice that had recurrent disease. Essentially, we didn't kill them on that first round. So with 2 cycles -- 1 or 2 cycles of therapy, we eliminated the tumor from 22 mice without any sign of effective toxicity. And I think that, that gives us a lot of promise about what we're going to be able to do in a more complicated in vivo system. Now of course, that is what cell line model, and we're fully aware of all the limitations that those can have and how good they can look in practice and then maybe how they don't necessarily bow on theory. So I want to tell you about one other animal study that we were able to do in collaboration with folks at MD Anderson. This is where we take patient cells. This comes from a patient who had AML with FLT3-ITD, who was treated with sorafenib, a drug that's active against the FLT3-ITD plus azacitidine, and he actually went into a CR after one cycle of therapy. When that patient relapsed after 3 cycles, they had acquired a D835 mutation and now had a double mutant kinase. These cells are extracted from the patient and put in the petri dish, exposed to CG-806, where we saw dose-dependent killing of these cells. However, we did not see that with quizartinib as expected, since again, quizartinib doesn't target that 835 mutation. When we look at these cells in a mouse model where we do a PDX xenograft model, we allow the tumors to grow and then we treat the animals in this time with once-a-day, weekday-only dosing due to some limitations of the lab doing the studies. But you can see that the tumor grows pretty rapidly in those mice that are untreated. But in those treated mice, the tumor growth is almost completely suppressed. But we have substantial reduction in leukemia cell burden, and these CG-806 treated animals that have been engrafted with, not just an AML mutant cell line but with a patient FLT3-ITD mutant cell that has been resistant to prior therapy with a FLT3 inhibitor. So I'm going to end there and allow our last speaker, Dr. Druker, to come up and tell us more about his experience, not just in one scenario but looking at a multitude of patient-derived cells in the clinics as he's had the opportunity to work with this quite extensively. And I look forward to hearing his perspective. Thanks.

Thank you. Thank you, Raf, and thank you all for being here. I'm just going to walk you through some of the data that we've been generating over the last several years on CG-806. But as Bill reminds me, we're the old-timers here, and I've been involved in the development of kinase inhibitors for quite some time. As you've heard, there are over 500 kinase in the kinome, and mutated kinases can lead to cancer. There are now 50 kinase inhibitors approved for use in the U.S. And again, I'm old enough to remember when kinases were not a druggable target. And despite that, we were able to develop the first kinase inhibitor [indiscernible] for chronic myeloid leukemia, and it's been on the market since 2001. What's been interesting about the evolution of kinase inhibitors over the last several decades is that imatinib was actually relatively selective. It inhibited 3, maybe 4 kinases in the kinome. And when the second generation came along, it was sort of a midostaurins and sorafenibs and sunitinibs, which inhibited pretty much everything in the kinome. The view was let's inhibit kinase into the clinic. And we saw lots and lots of toxicity with those drugs and have been unable to effectively combine them. So in the industry we were bind to, let's make them really specific, single targets. And what happened with those is we rapidly saw resistance, as you saw with the FLT3. And now we're coming along with a notion, as you've seen with CG-806 is that what if we start to target clusters of kinases, try to understand which kinases to avoid which cause toxicity, try to shut down some ancillary resistant pathways and that's really where we are in the current era in generation with CG-806. I don't expect you to read anything on this slide. This is just a slide that Bill and I've put together to show you that -- this is data from my laboratory. We've been working with CG 806 for quite some time. We presented at numerous conferences over the past several years and are working on a publication of this data to summarize. But I just want to zoom in on one of the most recent bits of data we published, which is looking at patient-derived cells. So over the last decade, my laboratory has been collecting patient samples from patients with leukemia and incubating those cells in vitro with a variety of drugs from kinase inhibitors to the favorite -- flavor of the day in terms of the drugs that are available. We've -- I put on this slide a variety of the different kinase -- FLT3 kinase inhibitors. In red, you see sensitivity. And what's remarkable about CG-806 is you see that virtually every single one of our patient samples, whether they're FLT3-ITD mutation positive or have the ITD or FLT3 negative, but they express FLT3, all are sensitive. And along here, you can see that other drugs like crenolanib and midostaurin, they have holes in their coverage, not every single patient sample sensitive. So there's probably 2 questions you're going to ask. First is what does this do against normal cells? Is this just nonspecific toxicity? The answer is when we incubate 806 with normal cells, we see no toxicity. This is specific for leukemia cells. The second thing you're going to ask is, well, how does this predict how these drugs will perform in a patient. And I can tell you that we can't always tell that. That's what clinical trials are for. But the real point of this slide is, I'm using this as a comparator. It's the best FLT3 inhibitor we've seen compared to any other FLT3 inhibitor that's already out there. So that gives us some confidence when we go to clinic. Yet these other drugs work against FLT3, that this will work, but it also should work against some of the FLT3 mutation negative patients. So then you're going to ask -- as you heard from Aaron and Eytan, who gave fabulous set-ups for I want to talk about next, which is how does it work when patients have other mutations because everybody is going to have FLT3 with something else, whether it be a p53, IDH, NPM1, I'm just going to walk you through a few examples. When you look at either NPM1 or the FLT3 allelic ratio, we see that, as you sort of track to the right here, with high FLT3-ITD, these patients are extremely sensitive, and that's regardless of whether they have MPN1 mutations or not. If you then look at IDH, if you have IDH mutations, you're actually extraordinarily sensitive particularly IDH1 mutant patients are extremely sensitive to 806. But more importantly, p53, which is the worst prognostic factor for any patient with AML, again, wild-type mutant, similar sensitivity in terms of ability of CG-806 to kill these patient samples. But as you heard from Aaron, one of the other things we look at is RAS and whether you're RAS wild-type or RAS mutated, you're equivalently sensitive to CG-806. So the view would be that, if we combined or we used this drug in patients with RAS mutations, we might be able to avoid some of the resistance that's seen with other FLT3 mutations. In addition, you heard from Aaron, what about venetoclax, and it combines extremely well. You can see here's the average sensitivity of venetoclax, average sensitivity of CG-806 and significantly synergy with a combination of CG-806 and venetoclax. Now what's interesting as well is, when we've looked at venetoclax resistance, oftentimes, patients with p53 mutations are resistant to venetoclax. And one of the reasons they're resistant is they up-regulate TRK kinase. And you heard from the profiling that 806 inhibits TRK activity. So our view is there actually may be some mechanistic reasons for why this combination might be extremely effective. And you've also heard, we know that single agents aren't going to get us to the goal of curing more patients. We need to get to these combinations, and we think this would be an ideal combination to strive for relatively quickly. So just to summarize. This has an extremely favorable preclinical profile, strong efficacy and safety, and we think it might be able to avoid some of the resistant -- emergence of resistance. It's hitting a lot of the operative pathways but doesn't hit some of the ones that we think would compromise safety. And our -- my view is that this is one of the best kinase inhibitors we've seen in my lab in quite some time, at least preclinically. Of course, we've got to get to the clinical studies. But as I said, in the comparative analysis, it looks extremely promising. So our view is highly promising drug, more than just a FLT3 inhibitor. We're eagerly awaiting more of the data from the CLL studies. It inhibits some of the really critical signaling pathways that might lead to resistance. And as you heard, it's safe and well tolerated to-date. So again, I want to thank you for your attention, and I'll turn this now back to Rafael. Thank you.

Thank you, Dr. Druker. Before we get to the question-and-answer session, we'll have plenty of time for that. I just want to take some elements from the talks that we heard earlier and put them together to just make some take-on points. So this is actually another way to look at the data from the Beat AML program that was actually run out of OHSU and Brian Druker's lab collaboration with Jeff Tyner. And what we are able to do with this data is we're able to look at certain types of AMLs, for example, in this case, I'm showing NRAS mutant AMLs and ask, are there drugs that have particular differences in sensitivity to this agent. So for example, drugs on this side of the curve are drugs that are going to take higher doses to kill NRAS cells than they do to non-NRAS mutant cells. And here are drugs that are maybe a little bit more potent. They are able to kill NRAS mutant cells better than they kill NRAS wild-type cells. And you can see there's a lot of drugs on this right-hand side. And all of the common FLT3 inhibitors are here. So sorafenib, sutent, quizartinib, midostaurin, gilteritinib. This suggests that these drugs have less activity against NRAS mutant disease. If we look at the drugs themselves, we can then ask what mutations are likely to require higher doses of these drugs to kill in this in vitro study. And you can see, for quizartinib, not surprisingly, it is NRAS, KRAS and TP53. For gilteritinib, it is NRAS as well. Now they're also very highly specific for FLT3-ITD mutants, which is not surprising. That's what they target. And I want to frame this in 2 ways. One is that, obviously, these are important. We've hammered on this a couple of times in the talk today that NRAS is an important resistant mechanism that we hope we have activity against, but these drugs really are FLT3 inhibitors. And then at least for gilteritinib, the indication is for FLT3 mutant patients. However, if we look more broadly, when we look at activity, both in FLT3 mutant or in NRAS mutant or NRAS wild-type, what it can show you is that, for NRAS wild-type AML cells shown here in red, when we do this assay side-by-side with these other agents, we see that CG-806 has a fairly low mean IC50 for killing these cells that a large number of these cells are readily killed at concentrations that we think we can actually achieve in vivo. When you look at the NRAS mutant cells, there is a slight difference in the amount of CG-806 required to kill those cells, but that difference is not significant. In contrast, for gilteritinib and for quizartinib, the difference between killing all the cells and killing the NRAS mutant cells, there is a significant difference. It takes substantially more drug to kill the NRAS mutant cells than it does to kill the NRAS wild-type cells. And for midostaurin, which as you heard is not as potent of FLT3 inhibitor, actually has difficulty with both. But if this is the range of concentrations that we're looking for in patients that we think is likely to be effective, those ranges can be readily achieved for killing NRAS wild-type cells. They can be readily achieved for killing the vast majority of NRAS mutant cells for CG-806. That is not the case for gilteritinib, quizartinib and midostaurin, where many of these patients' samples lie well above those concentrations that we're likely to achieve in patients. So again, giving us hope that we're at least shutting down maybe even another pathway of resistance that could emerge after treatment with this drug. Here it's just highlighted a little bit differently, where we're showing them broken out by group. The NRAS mutant cells for CG-806 have a mean IC50 of about 1 micromolar. That's substantially higher for the other 3 tyrosine kinase inhibitors that were tested in parallel. So to summarize all the talks that we heard today, there clearly is a medical need to control relapsed/refractory AML. And even an opportunity, I think to go upfront with some of these agents that may be more powerful and more effective than midostaurin. So the fact that we have multiple agents now online for AML hasn't really removed the need to treat these patients as most of these therapies are not curative. And we need therapies that are able to treat particularly those unfit patients that may not benefit or may not be able to tolerate high-dose chemotherapy. CG-806 has an appropriate profile. It targets all forms of FLT3, including the resistance mutations and even the wild-type form as well. We are close to reaching -- achieving PK doses that we think are going to be therapeutic in AML patients, and we're planning to submit that IND quickly so we can get that clinical trial up and running. And it has -- it's very attractive from a clinical standpoint, but it's also attractive from a commercial standpoint in that it has activity against a validated target. FLT3 is a validated target for which there are other approved drugs in the market. Our safety profile really gives us a lot of combinatorial optionality. If you are bringing a toxic drug into combination, it may not tolerate the combination well, but if you take a drug that is very safe and effective and you combine it with another drug, then you are going to be less concerned about combinatorial toxicity. In fact, we could imagine this as a potential cornerstone of therapy. This is an oral agent. There are other oral agents that it might pair with very nicely. We already discussed venetoclax. I could even imagine some of the oral hypomethylating agents, like the ones that Dr. Stein mentioned as potential partners for this drug in the future. And ultimately, the fact that we're not limited to the FLT3 mutant patients and that we're not designing our initial AML study to be limited by genotype, meaning that we're unlikely to require a genomic biomarker when we eventually get to market. So the takeaway point I want to talk about also is, what is the low-hanging fruit? So we are going to want to put this drug into a broad category of AML patients. We're not going to want to restrict it based on genotype or other factors. But there are some patients who I think have the greatest need. As Dr. Goldberg mentioned, those are those patients that are failures after the effective FDA-approved therapies that we have today. And CG-806 checks all those boxes that Dr. Goldberg mentioned. And the last point of his slide, it's a potent inhibitor that inhibits all immune forms of the kinase, targets multiple mechanisms of potential resistance, including those RAS mutations has the ability to synergize with venetoclax and potentially other drugs. So we're going to stop now for questions and answers. I'm happy to hear your thoughts. We can always answer your questions here at Aptos, but I would encourage you to take advantage of our experts that are here with us today. So while you're thinking of your questions, I want to thank Dr. Stein, Dr. Golberg and Dr. Druker for coming out here today and spending time with us and sharing their thoughts and insights on this drug. After hearing their talks, I'm even more enthusiastic than I was before and that I admit. Having heard these kinds of talks very often, I learn something new every time. So appreciate you guys coming out and doing that.

Tyler Van Buren from Piper Sandler. I guess the first question I had is, in the animal models, you noted that the CG-806 was very well tolerated even at the higher doses, and there was no weight loss. Can you speak to perhaps the gilteritinib experience and if there was any weight loss or kind of tox observed in early animal models and just, maybe as a follow-up to that, with gilteritinib, do you expect, since it's better tolerated than midostaurin, for the physicians on the panel, do you expect to use it front line over time in combination with chemo?

Do you guys want to take the latter question?

Yes, I think we can take the clinical question for sure. I would say that, overall, it is a well-tolerated drug as monotherapy. I mean some patients, I would say, do experience fatigue on the drug. It's hard to say if that's really an AE, adverse effect, or not. We do see LFT [indiscernible]. We see sometimes CK elevations. We have to check the CK over time to make sure that's not going up, but it doesn't significantly really have QT prolongation. It's overall, I think, a well-tolerated drug. I think the challenge really is that, over time, patients develop resistance, of course, to it, and that's why there's such an urgent clinical need. I can't speak to the mouse data, unfortunately, in terms of how is...

So I'll speak to it briefly. So if you look at the -- some of the other FLT3 inhibitors, in particular, quizartinib, gilteritinib, it turns out what was observed in mice and dogs was equivalent to what was found in humans, for instance, the quizartinib, the QTc prolongation, the cardiotoxicity there, it was seen in mice early on, and it was hoped that, in humans, it would not manifest itself at such low concentrations, but it turned out to be dose limiting. I think the same is true well -- not QTc prolongation. But gilteritinib also, it definitely manifested certain toxicities at higher dose levels in mice, but they clearly are able to achieve doses in humans now that are affected -- that are effective, but you can't continue to dose-escalate further because, if you could, you might be able to avoid some of the drug resistance. But as was shown here, further in RAS mutation, they just can't go high enough in order to achieve those levels. And it's driven by avoidance of toxicity. So typically, with these kinase inhibitors, what you see in animal models correlates very well with what you see in humans, generally speaking.

And I think what I would say also as a follow-up to that is in terms of gilteritinib, the half-life of the drug is very long. We're talking like 5 to 7 days, basically. So when we do, for example, a wash -- when we have to do a washout of gilteritinib prior to enrolling someone with 5 half lives, that's a long time. So you're -- the washout true to here, your trial is photo. But it's -- yes, exactly. So that can be a problem if we do observe any toxicity because the drug sticks around for a long time.

And the second question was just related to the dose for AML. The IC50 data at the end in resistant clones was pretty interesting, showing the 1 to 5 Predicted micromolar doses. So for cohort 3, if you see, I guess, is it just simply come down to observing a plasma concentration within that range to have the confidence to go to the FDA to start AML? Or do you want to see secondary measures of activity in those patients as opposed to going to a cohort 4 and waiting to see that data.

That's a great question. I think we want to see doses in patients that we think are going to be effective in AML. We don't necessarily need to see -- wait for responses to occur in patients in that cohort. Those patients are in the B-cell study. They often take much longer to show responses than an AML patient might or you expect responses relatively quickly after starting therapy. So we are going to hope that we achieve those levels. And this third dose cohort that are obviously going to continue to dose escalate in this population until we find the right dose for the AML patients.

Yes, let me add just a bit to that. So Dr. Druker mentioned we're the old guys in the room. We'll call it elder statesman, if that's okay with you. So something he told me, I think it's about 2 decades ago, is you have to have the right drug for the right patient at the right dose. And it sounds so simple, but it's true. So we need to make sure that if Dose Level 3 here in these B-cell malignancy patients, we think we have the right drug for the right patients to go into AML, but we need to make sure we have the right dose. If Dose Level 3, if we feel confident we're in that micromolar range steady state and we can affect the patients with AML, then we'll try to go there as long as we have a good safety profile. But we don't want to make an error, an unforced error. If it turns out that we believe we should go to Dose Level 4, we will. So in effect, what we're doing is our dose escalation for AML but in these B-cell cancer patients. So don't be frustrated if we say we're going to go to Dose Level 4 before we go into AML. Just know that we're going to be taking a higher dose and to higher exposure into that. So that's the way -- we're very data driven. So that's the way I look at it.

John Newman from Canaccord. My question is regarding the experience that the physicians have had in terms of selecting patients for AML studies in general. One of the points that Dr. Stein made was different patients with different types of mutations have different prognoses, and sometimes those poor prognoses patients progress very, very quickly. So it's a question for the physicians and the company. When you think about selecting patients for these relapsed/refractory studies, what are the things that give you the signal this would be a good patient to put into the study versus other things that say to you, you know this patient is progressing too fast or there are just other things here this patient really shouldn't go into this trial.

So I don't think it's much beyond what you might think, which is that patients who have a relatively good performance status of 0 to 2, patients who aren't clearly having rapid doubling of their white count quickly over time. I think the benefit of targeted FLT3 inhibition, though, at least in a FLT3 mutant patient, is that those patients often have rapid leukocytosis. So you might not want to put them on some trials, but when you're giving them a targeted FLT3 inhibitor, what that does is that shuts off the leukocytosis when it works very, very quickly. So I think in some measures, drugs that are -- that specifically target FLT3 and target RAS are able to -- you actually have more patients you could potentially put on those studies because you're shutting off the pathway that leads to a proliferative advantage of those cells.

I would just mention, when the trial expands to AML, it won't gate necessarily on FLT3-ITD or mutation only because it may have activity broadly in AML. So that's a bit of an advantage in that, as you heard from Eytan, that sometimes the FLT3-ITD patients do have a very rapid doubling, whereas, some of the other patients without FLT3-ITDs will be a little bit less rapidly progressive. But we think this compound might have activity broadly in AML. That's part of what the clinical trials will tease out. And part of the problem in any of these AML studies, you heard it this is an incredibly heterogeneous disease, but the hope is that this could have broad activity regardless of mutation status, but that has to be tested.

Yes, I agree completely. I think the other opportunity potentially for this drug, which we see in the clinic is in the setting of venetoclax failures. So venetoclax, I think it's fair to say, has revolutionized really the front line treatment of AML, particularly because it can be given in combination with hypomethylating agent or low-dose cytarabine to patients who are not fit to receive intensive chemotherapy who are older, for example, or have comorbidities. And the response rates to that combination are quite good upfront, right? So like 70% complete remissions, complete remissions be blood count recovery. But over time, patients do tend to progress. And when we look at the outcomes, this was just presented by MD Anderson actually at ASH to -- for patients who were on venetoclax, for example, who are either primary refractory to it or who failed, you have an overall survival of about 2 or 3 months. So it's really a -- it's devastating at that point. And as you saw in one of the last slides that I presented, FLT3-ITD mutations do appear to be one of the mechanisms that can -- yes, and NRAS, of course, as well. Both are key mechanisms of resistance to venetoclax. And both of those pathways are targetable, it seems, preclinically by CG-806, which is why I think we're really excited. We see -- we were just talking actually before about what are we going to do for some of these patients right now. This trial is not yet open. I wish it were because that's how you put them on this trial tomorrow.

So I have a couple of questions. First of all, obviously, in this disease, I mean, the progression is very, very rapid. So if you have an effective drug, it would be interesting to me to sort of hear literally within that -- from what I understand, within a week, 2 weeks, 3 weeks, you would already see quite different things happening between the 2 groups of patients. So if you could just talk a little bit about if this were effective, what you would be seeing almost in realtime between this 806 group and the placebo or the other group. And my other question is in terms of dosing, I mean, where is the -- I have no idea why 450, 600, 750, I mean, can you talk a little bit about why you would want to necessarily limit the dose or how you would identify the optimal dose in terms of the trial.

I mean, I guess I can speak to what one might expect sort of clinically, and I'll let people handle some of the dosing questions. But so clinically -- first of all, there won't be a placebo on this Phase I because the technical intention, of course, of the Phase I study is really to test safety, although the reason we're giving to our patients because we hope it's going to be effective. So there won't be a placebo-control arm in the Phase I. But what I would say is what I would expect to see is what we've seen when you have an effective FLT3 inhibitor, if you have a patient with a very high white count, a hyperleukocytosis with a large percentage of blasts, you're in the peripheral blood, for example, even an effective drug, you'll see that count drop even within the first several weeks, even sometimes in the first several days, you can see that, that peripheral blast percentage drop. It can take time, though. I mean the median time to response is still -- best response is several months on gilteritinib, for example. So it will take time. If you look in the bone marrow to achieve a full, for example, a complete remission, which would be defined by less than 5% blasts in the bone marrow and also an improvement in the other peripheral blood cell counts like a platelets above 100 neutrophils above 1. So what we're really trying to achieve is an improvement in hematologic parameters, a reduction in the white blood cell -- the bad white blood cell count, the blast percentage, and restoration of normal hematologic parameters. And you could do it pretty rapidly.

For the patients with a very poor prognosis, would you expect to see very much more rapid changes relative to what you would expect in their baseline from that perspective?

I think you would hope to see changes relatively quickly if the drug is effective within the first several weeks to a month -- at least some changes.

It's very different for B-cell malignancies.

Correct. Yes. And it's different from drug to drug. It's different from B-cell malignancies, it's true. It's fair to say. And I think Eytan can speak to this more, though, depending upon the agent, sometimes it can take time for best responses. IDH inhibitors, for example, can take really many months to see excellent responses.

Yes, just speaking from the experience of IDH inhibitors, whether -- if I'm understanding your question correctly. Whether you had an IDH mutant patient who had unfavorable risk cytogenetics, or whether they had favorable risk cytogenetics because both occur. Those patients responded at the same rate. So there was no real difference. If you're unfavorable, if the drug was working, the drug was working as quickly as it worked in a favorable risk patient. Yes.

Yes. But I guess for Aaron and Eytan, one of the advantages of AML is that patients -- some patients will have a pretty high peripheral blast count. And how -- my impression has been that inpatient to a high peripheral blast count, the FLT3 inhibitors can lower that pretty quickly and oftentimes within a week or less. I just wanted to clarify that, my limited exposure to the clinic is correct.

So even if you did a bone marrow biopsy on that patient, you probably wouldn't see, of course -- but in the peripheral blood, I completely agree. Yes, you can already see responses.

So I was going to dive into the dose question, which is a really interesting question that we faced for a long time in targeted therapies, which is, should you identify a maximally tolerated dose or not. And I'm going to come down on the side of, yes, I think you should because think back to the slides about the FLT3 inhibitors and the doses required to inhibit RAS. If one of the FLT3 inhibitors was able to be tolerated at those 2 to 5 micromolar, they'd be working, and you wouldn't see resistance. But they're not tolerable at those doses, so you can't use them. With 806, there may be some resistance that's dose dependent. And if you escalate to it safely and capture some of those patients that might be resistant, I would suggest that you'd want to do that. Of course, you wouldn't -- if you get to 10 micromolar and you're giving 500 pills a day, you might come to a sort of an impractical dose. But in general, my view is that you want to try to see what the maximum tolerated dose would be because there may be other indications where that drug might then be effective. But at the same time, once you're achieving sort of the 1 to 3 micromolar, you know that's where your effective dose range is going to be. You need to expand your cohorts pretty quickly to see how effective your drug is going to be in sort of a Phase Ib, Phase II setting.

And to be clear, we will continue to dose escalate in the AML cohort, even though we're starting at a dose that we think is going to be effective, exactly to that point that Dr. Druker made.

Matt Biegler, Oppenheimer. Dr. Bejar, I think you mentioned it. I just had a quick clarification. So eligibility for the AML trial, you will be enrolling both FLT3 mutant as well as wild-type patients?

That's exactly right.

Okay. And what about MDS patients?

So we're going to start with AML patients, in particular. This is where we have most of the preclinical data. I think, ultimately, we will want to expand to see if higher-risk MDS patients, patients that are close to being at the threshold for acute leukemia, might have activity, but this is something that is for consideration down the road. Our plans are to open in AML.

Got it. And that does kind of tie into the second question because we've obviously -- the topic today has been AML, but what is the drug's kind of activity in MDS? Do you see kind of the same level of potency against primary MDS samples if you have tested them as versus AML? And is there a market opportunity for a safe drug in MDS?

So as I mentioned earlier, my background is primarily in MDS. I think one of the challenges we have in the MDS field is that it is very hard to grow cells from patients who have a disease where the cells don't grow well in them in the first place. So we don't have a lot of great models like the similar models that we have for B-AML, where we can take those cells, put them in a petri dish, expose them to drugs and really extrapolate what that means for patients. So that has been challenging for MDS. There's also not the kind of experience in AML -- in MDS using kinase inhibitors, but this is a disease that's primarily treated by what we consider to be epigenetic therapy, hypomethylating agents for the most part, or now some targeted therapies where patients have very specific mutations like IDH1 and IDH2. Venetoclax is really a revolution of what we do in MDS as well. But again, there, that was borne out in the clinic. That wasn't borne out by doing preclinical testing of the kind that you saw here.

Just a quick addition to that. So as Raf said, we don't have cell lines that really represent properly MDS patients. However, the drug 806 has been tested against a number of patient-derived MDS samples and that Dr. Druker's -- his lab also did those patient studies. And that was done single agent with venetoclax, single agent and our drug in drug combination. So we do have those data. And we've shown that it's active, and it's -- and we showed that it works well in combination with venetoclax in the patient-derived samples from MDS patients. But again, first in AML, then MDS. It's a more complex situation. We hope we get there. But no guarantees.

Akash, Wolfe Research. So on -- just stepping back, I think ArQule showed its first response, kind of at the 1 nanomolar range. And you talked about your therapeutic threshold, somewhere between 1 to 4. So can you dive into what's really -- what makes you confident that -- from your animal data that your threshold is a bit higher than maybe some of the other reversible BTKs? And then how do you kind of take that therapeutic threshold window and apply it to AML. And then maybe also just dig into kind of the importance of constantly suppressing the targets. And maybe why go to a bid dosing.

Regarding the 531 compound -- the ArQule 531 compound, you had said 1 nanometer is actually 1 micromolar. So yes. And so they had spoken about this overtly. They were saying they were trying to target the 1 to 4 micromolar range that they felt that was the range that they need to hit to truly have activity. Yes, they did start seeing some activity at the 1 micromolar range. But again, that's the range that we're talking about to. We're talking about 1 to 3, 1 to 4 micromolar. We're trying to hit that same range. There is also another, since you're talking about the reverse of BTK inhibitors, the 305 molecule. And so they are also trying to hit the 1 to 3 -- 1 to 5 micromolar range, too. So it's in the same range of potency in terms of the exposure level. Now all of them may have a different dose because they all have different rates of absorption percent -- fraction of absorption or bioavailability, but we're all trying to hit about the same area, and we hope we hit that range, that dose range safely in the 1 to 3 micromolar. We may be able to dose-escalate above that, and that'll be great. So that addresses that concern. And what was the other one in terms of BID dosing? Many of the animal models that we had performed with 806, were done once-a-day dosing. But we knew that as we got into our GLP talks that we would need to present to the FDA. The FDA wants to see toxicity. They want to know what is your target organ for toxicity. So when we went into our GLP talks, we decided we're going to dose twice a day because that certainly would get enough drug in to drive toxicity. So we did all our GLP talk studies, twice-a-day dosing, 28 days, and at the maximum feasible dose in animals, and we still got no toxicities. And those were the data. And we were clearly achieving well into the micromolar range. It was well tolerated, but we did not see any adverse events in animals. So we decided to carry that on into the clinic for twice-a-day dosing, and that's the reason because, in cancer, as what was said earlier, you want to hit those cells as hard as you can, maintain the pressure. At some point in the future, we may go back to once-a-day dosing. But right now, we want to hit them as hard as we can as long as we're still safe.

[indiscernible] Maybe a question for Dr. Druker. So in the preclinical model, you saw very low resistance to 806, differently from other FLT3 inhibitors. So I was wondering, have you any clues on what kind of mechanism or genetic driver can lead to better resistance?

Yes. We have quite a bit of data analytics that looks exactly at those questions. And we have not identified any consistent theme in terms of what drives resistance to each of the individual FLT3 inhibitors. Part of it is that we're -- even with 1,000 samples, we -- there's so much genetic heterogeneity that it's really hard to get to statistical significance when we start to do the bioinformatics because the numbers of the combination in mutations are quite large. So I don't have a clear view of what the resistance mechanisms are. And Raf showed some of what we're calling these volcano plots, so we can -- which look at sensitive versus resistance. And there's actually very few genes that reach statistical significance with some of these drugs.

So just to add to that, the possibility is not only -- so with a drug like this, you would expect to have to see multiple mutations occur simultaneous because we're hitting multiple pathways. But there are also mechanisms that cells have that they can pump the drug out of the cell. And so that's one way you could possibly get drug resistance. We tested for that. It's not a substrate for [ peak lack of ] protein, but the cell can up-regulate certain types of pumps that can pump these out. So that's one potential mechanism, but we haven't seen it yet.

Matt Cross from JonesTrading. Just had 2 quick ones. One was, you've discussed CG-806 as a very agnostic inhibitor. And I think lately, the company has received a lot of attention within the B-cell malignancy space and the potential there given M&A and just developments in that landscape. But I guess, what I'm curious to know and get your feedback on is whether maybe within the FLT3 and the relapsed/refractory AML market, an agnostic approach is even more in demand than something like CLL, where most of the kind of new single-targeted agents are targeting pathways that are not mutation driven and are more broadly survival or proliferation related. So just feedback on kind of the demand for -- between the physicians are looking for these kind of more agnostic approaches in AML and also the investment market.

Do you guys want to comment from a clinical standpoint, if that's something that's differentiating for you?

I mean, I think, very simply, we're looking for the most effective therapies, whether they're targeted. And that, obviously, if the market has a FLT3 inhibitor or not, we just want to make sure that it's actually the specific patient in front of me responds. I think what Eytan and I do, I think, in the clinic, is we always -- we think who is this patient? What can -- what are the goals of therapy? What can they tolerate? How old are they? What's their organ function? What's their performance status? And then what is their molecular mutation landscape? And can I use a specific targeted agent that might already be available that they may have a response to an IDH inhibitor, for example. But if they fail that, then what other options are out there. And so I think there is an enormous opportunity for a more effective FLT3 inhibitor, for example, that could target gilteritinib failures. But to your question also, I mean, if this agent preclinically could have activity even in non-FLT3 mutations, that would be outstanding, p53, for example, I mean, these patients have the poorest prognosis. So I think -- and that's really what the point of the early phase clinical trial is all about, is sort of to see. We're not going to restrict to specifically FLT3-mutated patients. We're going to ask -- ask your question, who is the patient most likely to benefit from this, and that's what future clinical trials will target.

So from a "marketing" perspective, we call it a FLT3 BTK inhibitor. People understand FLT3 and its application to AML. People understand BTK and its application of the B-cell malignancies, but it's more than that. So remember, that's the first thing I said today. It's -- yes, it's a reversible BTK inhibitor, but it's so much more than that. So it targets multiple pathways, key oncogenic pathways simultaneously. But how do you get that message out to everybody with a single word. So yes, we inhibit FLT3 and all mutant wild-type and mutant forms that we've tested it. Same is true for BTK. But it's those other pathways that we're hitting are just as important. Because if you're just hitting, I don't care if you get all forms of FLT3, the cell will escape. AML will escape. And so it's like whack-a-mole, yes. FLT3 is one of the whack-a-moles, but you have to hit all multiple and simultaneously. That wasn't so good at marketing, but it's true.

The second one was just on your other kind of lead asset that wasn't really discussed today, 253. And now that you're getting kind of on the cusp of moving CG-806 into testing in AML, just if you could give any kind of comments on the positioning for 253 at the same time, also running in AML and MDS and how those 2 may kind of interplay as you move both of them forward kind of in lockstep.

So 253, we haven't spoken about it that much. But under the radar screen, we're continuing to dose-escalate patients on -- for 253, we're in dose cohort 4 now with -- so we dose this molecule once weekly. We've clearly seen 75% to 80% reductions in c-Myc. So it suppresses the expression of c-Myc. That's in patients, both in AML and MDS patients. We're measuring the pharmacokinetics of that drug. We're clearly getting coverage over 3 or 4 days, but we want to dose escalate and maintain coverage over a week, if we can. So we hope as we dose escalate, we'll get good coverage. We'll get Myc inhibition and we'll see effectiveness in the AML MDS patient population. But now that we know it's a Myc inhibitor, we might be able to take it into other indications, too. But for now, we're progressing with that molecule. And very few people ask about it. So we haven't said much. But again, under the radar screen, we believe there's something there. So we're continuing to dose-escalate.

Any other questions? One moment.

Greg Renza from RBC Capital Markets. Just quickly for the panel, I just want to hear your thoughts about the -- just the aspirational context of combinations of 806 with some of the treatments out there or in development that you think are potentially most interesting to you.

We showed some examples of that in the talks. I think that venetoclax has a drug that spans both B-cell and lymphoid and myeloid indications similar to CG-806 might be a great partner because we could consider using it in combination in both those scenarios. But there certainly are other drugs that I think could work well with CG-806 in AML, specifically, for thinking about their -- we talked about the need for therapies for those unfit patients and that the standard of care for these individuals now, I think, has really become a hypomethylating agents plus venetoclax, yet there -- that isn't a cure, either. So if we can either potentiate that therapy or combine with an HMA has another alternative for these individuals. Again, agnostic to their mutation status, that might be another opportunity for pairing. So this is, again, aspirational thinking down the road, but I think we might have opportunities for multiple pairings, where eventually, the treatment of AML may look more like the treatment of myeloma does today, where we're not doing single agents in serial use. We're actually combining them upfront to get our most effective, most deep, most durable response at the get-go.

As you may know, we're involved in the Beat AML trial, sponsor of Leukemia & Lymphoma Society, where we're treating newly diagnosed patients over 60 with a variety of currently single in some combinations. I'd very much like to see 806 with venetoclax for newly diagnosed, untreated patients and quickly. One of the things that we've seen in our data is that venetoclax has an unusual property that sort of makes everything better. And on its own, it's not that great, but you add it to other things like 806, azacitidine, and it does much better. But again, even in our combinations, and I didn't show the heat maps for those, 806, venetoclax is a huge winner, and we're also looking at venetoclax, so again, there's some really interesting data, but I'd like to get this upfront as quickly as possible. And hopefully, the company will support that.

There are other really interesting drugs out there that I think that could be potential partners, and this is just -- it's something we haven't delved into too much, but there are interesting drugs, including the 47 has a very different approach for treating MDS and AML. [indiscernible] 246 is a very interesting compound that I think could synergize with some of these approaches. So I think there are multiple options for putting drugs together.

I've got another question. Can you talk a little bit about the clinical trial in terms of -- I know you're testing it just broadly, but how it'll be set up or structured in terms of testing different subgroups and sort of the whole constellation of what you need to try and identify in terms of patients' effect, different mutations. Can you sort of just put that in context a little bit in terms of a trial and what that's about.

I think -- so there are obviously many different ways that one could go about a trial like this. I think what we'd like to do is really get into AML quickly with a dose that's effective. And to do that, we're not going to be very restrictive about what kind of patients come on, relapsed/refractory AML patients, regardless of mutation status, are going be the types of patients we want to go into early. What we learned from that experience will influence what we do down the road, how we design expansion cohorts, whether we look at a particular target population. But I think we have the advantage of actually being able to recruit more patients more quickly by staying broad at the very beginning. We are actually working with sites and asking for their input into how -- what the protocol is going to look like. So we're not really in a position to talk exactly about what the details of the trial are going to be, but that's the general philosophy that we're taking into this study.

So I'm also going to have to note that Dr. Bejar and I have a hard stop right now. We have a plane to catch. So we're going to have to sneak out. So Greg, do you want to...

Thank you very much, everyone, for coming.

Yes. Thank you.