Introduction

DR LOVE: Good morning, everyone. I’m Neil Love from Research To Practice, and welcome to “Consensus or Controversy,” as today our clinical investigator faculty will provide some perspectives on the current and future management of acute myeloid leukemia.

We’re really thrilled to have this faculty here with us today. From the far side of the podium here we have Drs Tara Lin, Sasha Perl, Harry Erba, Amir Fathi and Dr Eytan Stein. And Eytan, I really want to thank you for working with us to develop a survey that we did of the faculty and some additional investigators, Drs Cortes, DiNardo and Dr Wang, and we’ll show you the results of the survey as we go through this meeting today.

As always, we will be discussing the use of unapproved agents and regimens so please consult the package inserts for the various products for more information.

I want to thank all the docs who helped prepare these programs we’re about to launch right now. At lot of these docs appear on videos over the next 2 weeks, so we had quite a few mainly general medical oncologists trying to help us understand what we need to talk about that’ll be most relevant. We’re very appreciative to all the faculty members who are joining us for all of these programs.

So let’s get into AML. There’s a lot to talk about. There’s going to be a lot more after this meeting. We’ll try to talk a little bit about that. Here’s where we’re heading. We’re going to talk about up-front therapy initially in patients without targetable mutations who are older, and then get into younger, then we’ll talk about FLT3, IDH and of course the menin inhibitors, the newest entry. Very, very exciting there as well.

Up-Front Therapy for Older Patients with Acute Myeloid Leukemia (AML) — Dr Lin

DR LOVE: So we’re going to get started. Tara, you prepared some thoughts on up-front treatment of older patients.

DR LIN: Thank you so much, and I’m glad to be with all of you guys early this morning. I’m going to be talking about up-front therapy for older patients with AML.

A couple things that we’ll cover in a very quick 10 minutes is what does it mean to be old and have AML and how do we determine fitness, particularly in the context of therapy choice.

We know that disease biology and having all of that initial diagnostic information is really key to determining the optimal treatment, how do we start incorporating novel agents into what we do and how can we be sure that we’re always thinking about our patient and their goals and including them in a shared decision-making process. There’s a question, right? What if intensive therapy isn’t better than less intensive therapy, and I think in some types we may see that it’s not. And then can we adjust the way that we think about clinical trials and design them so that they’re less focused on specific ages but really think about the patient in a more holistic context in terms of fitness.

So what is older AML? I was very surprised when I was looking at the latest ASH guidelines for older adults that it starts at age 55. I don’t know if you remember the Golden Girls, but these women were all in their 50s, right? And at the time when I watched that show I thought that was old. Now for context all of the cast of Friends is also in their 50s and older than the Golden Girls would have been and would fall into these guidelines for older adults. So I think this concept of age as a number isn’t really — it doesn’t really hold water anymore.

When we talk about our trials of less intensive therapy there was an automatic eligibility criteria, and if you were age 75 or older you could get in. And I think age — we need to move away from this idea of age itself as the determining factor for what kind of therapy we offer someone but think about fitness more globally.

Fitness for intensive therapy or for a specific therapy can really help us guide treatment choice, predict how well someone is going to tolerate treatment. When we think about trials of novel agents or novel combinations it's important, right, that eligibility criteria captures a relatively uniform population of patients so that we can understand efficacy in less fit groups, as well as toxicity and tolerance, right? How many people had to discontinue treatment because it impaired their fitness and quality of life. And I really — if there’s 1 message I can leave you with, I’ll have several, is really encourage people to take the time to perform a more comprehensive assessment.

Back in the day, at least in my training, we always talked about age, performance status, which we often assessed by asking just a couple of questions about how much time someone spent in bed, could they go to the grocery store on their own, and then their medical comorbidities.

The Ferrara criteria that have been used for more than 10 years as we look at trials for — of less intensive therapies, trials for patients that we would have traditionally called unfit, include this age over 75 but also medical comorbidities like CHF or cirrhosis.

How should we assess fitness going forward? There’s this whole concept of the geriatric assessment. Our entire population, right, is getting older, our cancer patients are getting older, our leukemia patients are older, and we really need to take the time that it takes to do a more comprehensive assessment. Ask questions about functional capacity. Have it be a dynamic assessment. Obviously, it’s really important to do when you’re choosing that initial up-front treatment, but what about when you’re thinking about changing treatment or in the setting of disease relapse? It’s time to do a full comprehensive assessment again. It can better serve our patients when it’s more specific and thought about in the context of specific treatments. And we need to move away from this concept of it being that person is fit, that person is unfit, so we give treatment X or we give treatment Y.

This is a really nice review that was published earlier this year in Blood Advances from the European LeukemiaNet talking the importance of the fitness assessment. And one of the major points that they make is that we’ve all agreed as a leukemia field that it’s very important to wait for that initial diagnostic information to come back. We’re not treating people on the same day that they get diagnosed based on a wet bone marrow read that you went to the basement to do at 6 PM that night, right? That’s how we used to do it. Now we’re waiting for a lot of information to come back. So you do have the time to do a comprehensive fitness and geriatric assessment for our older patients, and this can really help guide not only their initial treatment but treatments along the way.

They provide a table of some of the validated tools that can be used as part of a comprehensive geriatric assessment. More questions about how well someone can independently perform their activities of daily living, a mini mental status exam, the get up and go test. All of these things don’t take very long to do and just need to become a part of our routine practice and also be incorporated into how we are considering eligibility for clinical trials and capturing data in terms of tolerance for treatments in our clinical trials as well.

These are the NCCN Guidelines, and I do want to take a couple of minutes and go over how these guidelines would work and what the recommendations would be. And importantly the NCCN Guidelines are not divided by age. They are divided by intensive induction eligible or not, and there’s not any guidance necessarily from the NCCN on how we decide that. And so I think based on what we just talked about, right, it needs to be more than just age.

But if you have an older patient who is intensive induction eligible, and they have a core binding factor leukemia, the guidelines are going to recommend that they get intensive therapy with 7 + 3, and the guidelines recommend the addition of gemtuzumab, because very often, even if these patients are older with a favorable-risk leukemia, you can be cured with intensive therapy. Now there’s a big difference between a 55-year-old and a 75-year-old in terms of their age of their organs and different things like that, but the decision for intensive therapy can’t just be made on age alone anymore.

When you think about favorable-risk leukemias or intermediate-risk disease for patients who are fit, and I hate using that word, but we don’t have another — we don’t have another vocabulary for that we’re trying to capture here, but the recommendation would be for 7 + 3. I think for older patients with intermediate-risk disease there can be some question of whether or not you would want to give them a ven/HMA-based regimen instead, but there could certainly be patients who are kind of the younger old patients where this would be a time to include them in shared decision making. For patients with a FLT3 mutation who are going to receive treatment with intensive induction you want to give them a FLT3 inhibitor, such as either quizartinib or midostaurin.

For patients with a therapy-related AML CPX-351 has data suggesting that you get prolonged overall survival over 7 + 3. And then you look at our poor-risk patients, so even those who are fit for intensive induction the guidelines say that there’s no preferred regimen and that a clinical trial would be recommended.

When we talk about unfit patients this is either going to be an HMA-based regimen for almost everyone with the exception of patients who have an IDH mutation, where you may consider an aza/ivosidenib regimen.

I want to pivot and talk about long-term data from VIALE-A, which led to the approval of venetoclax and azacitidine as the standard of care for older, unfit patients. You can see at long-term follow up the median overall survival was preserved. For patients with IDH1 or 2 mutations they had even a greater benefit than the entire cohort. And those patients who were MRD-positive and got the combination had the longest median OS at 34 months. Patients who got azacitidine alone and became MRD-negative had the next longest overall survival, really underscoring the importance of achieving MRD negativity.

When we look at adverse events from the long-term follow up it’s what you would expect from a leukemia study, with febrile neutropenia being higher in the ven/aza arm, but the rates of documented infection like pneumonia and sepsis being similar.

This is a nice summary slide for how we use aza and venetoclax with the caveat being that the FDA label — the FDA approval was for 28 days of venetoclax, but many of us feel that this is really too myelosuppressive and are doing an early bone marrow biopsy, sometimes around day 21, and if there’s blast clearance at that time pausing the venetoclax and giving time for count recovery before proceeding further.

When we talk about managing common toxicities there is a risk of tumor lysis. It is not common, but it is not zero, and when you see it, as we were talking about earlier, it can be bad, and so you want to remember this for patients. Use your typical tumor lysis prevention and treatment strategies. Cytopenias are much more common. It’s important to dose reduce your venetoclax in the context of the antifungal medication that you’re using. Use G-CSF when your blasts have cleared. Delay your next cycle until you’ve had full count recovery, and then if the patient is still taking a long time you can reduce your dose of venetoclax or space out your cycles. When you think about infections you want to use appropriate prophylaxis, and because many of these patients are getting the treatment outpatient and for long periods of time they may be far from your center, and so we like for all of our patients to have written instructions that they can take with them to an outside emergency department, should they present with fever, in order to get prompt management with broad-spectrum antibiotics.

I want to spend a couple of minutes talking about oral decitabine-cedazuridine. This is an oral HMA that can be used, and in early studies it was shown to have an equivalent PK profile as IV decitabine, which led to this data that was presented at ASCO and EHA earlier this year looking at a combination of the oral decitabine-cedazuridine, which I will call dec-C, plus venetoclax in newly diagnosed patients. Patients got 28 days of venetoclax, including a ramp up, and they were treated with 5 days of decitabine given orally.

The baseline characteristics are shown here. The pharmacokinetics are shown here. And there were no concerning signals from this data. And the overall response rate, the CR rate, was 46.5%. The median time to remission was 2.4 months. And what I think is really interesting to note is the duration of complete remission, with 75% of those who achieved a CR remaining in CR at 12 months.

Dose reductions with venetoclax were common in later cycles, as you would expect. Grade 3 AEs were what you would expect, as well, febrile neutropenia and cytopenias.

This is a rapidly changing AML treatment landscape. I would argue that our traditional concepts of fit and unfit are no longer appropriate. Comprehensive fitness assessment and disease biology are essential to guide our initial treatment decisions. The data that we looked at with oral therapy is certainly encouraging. I have some concerns about compliance just thinking about patients taking 2 oral treatment agents plus all of their prophylaxis, especially in the up-front setting or during their first cycle. And I would say when we’re thinking about our older patients you want to take the time, right, take the time to understand their disease biology, their global fitness. Always still consider them for transplant and always still consider them for clinical trials. Thanks.

DR LOVE: Thanks, Tara.

So this is the way the survey results appear. We’ll see how many of these we can get through. You can check out — if we don’t get through all of them you can check them out afterwards. And Harry, it’s no surprise that HMA/venetoclax, aza/ven in particular, has become the standard of care in patients not eligible for intensive therapy. Tara, I saw that you referred to this — I always love the “how I treat” series. I checked out this paper you had, the slide. This is an incredible paper. There’s a lot of stuff in here.

Harry, I was thinking back to when I think it was Dan Pollyea was — I was doing a program with him in 2019, I think, or 2020. He started talking about aza/ven. I’m like woah! And then we started presenting it to general oncologists. They’re like can we actually do that. And now here we are. So many considerations right now and so many questions come up about this. And Eytan, again, helped us design this survey.

So one thing I was kind of curious about, Harry. I was kind of expecting people to be a little bit more pro all oral here. Tara just showed the data presented at ASCO. Any thoughts about whether or not you can use oral decitabine in an HMA/venetoclax kind of strategy, Harry, and how does that play out?

DR ERBA: Yeah. I think it’s definitely possible. It’s just not possible in most of my patients that I see with initial therapy. They often show up, and they’re sick, they have complications of AML, they may have a high white count. You don’t really know them yet. You don’t know if they’re going to be reliable. They don’t live close to our center, and the outlying community hospitals may not be comfortable taking care of a patient during initial induction therapy.

So as I laid out there, there are multiple reasons why I have not yet found a patient that I would start with an all-oral regimen. And then if you do prescribe an all-oral regimen it’s really hard to keep that patient in the hospital if you’re not giving anything parenterally. So I think it is an option for patients after they’ve received initial induction, they’ve gotten into a remission. Although there’s no data on this I think it would be very reasonable to switch to an all-oral regimen in a patient who you know is going to be compliant.

DR LOVE: And actually I can see several people gave the exact same response.

So Sasha, of course one of the big issues, ever since the beginning here and continues to be an issue, is how long do you give venetoclax in the first cycle, when do you do the first bone marrow biopsy. Generally speaking, if you look at how the faculty approaches it they give the ven for 21 to 28 days, do the marrow again at 21 to 28. Can you talk a little bit about that first cycle and at the end of that first cycle how you manage these patients?

DR PERL: By day 21 some of these patients we can already start seeing a response, platelets are already rising, and almost ironically that’s when we often see the neutrophils drop. And so we want to know is there a reason for the neutrophils dropping outside of drug toxicity. Is this a patient you want to stop therapy in and allow for marrow recovery, potentially give G-CSF, as Tara mentioned, if you see blast clearance in the marrow? And that’s why we want to check a marrow around then. So our practice — there’s been debate on 14 days of ven, 28 days of ven. There’s actually a randomized study, the Beat AML, is doing to answer this.

In our practice we just draw the line right in the middle at day 21, and that’s worked for us. We do the marrow, then we stop the ven, then we see where patients are doing. If there’s no blasts we give G. If there are blasts, we start the second cycle the next week.

DR LOVE: So Eytan, I was also curious about toxicities. Of course the cytopenias everybody’s aware of, and you just heard 1 approach to that. But we asked about other toxicities, and I was kind of surprised to see a couple people, including this paper, the review article, bring up the issue of TLS. We talk about TLS — and we’re going to do a program on CLL here in this room, the next program. Of course that’s a huge issue. There’s a system you apply. And I’ve really never heard very much about TLS in AML, and now I’ve started reading about it. So what’s the deal? Do you see TLS with venetoclax?

DR STEIN: It’s very rare with venetoclax and azacitidine in newly diagnosed AML ps. What you worry about is those patients who have a white count that’s high or a very, very proliferative acute myeloid leukemia. That’s the type of patient where you might see a tumor lysis syndrome. We also see something which isn’t exactly tumor lysis, but we can see in these monocytic leukemia patients sometimes a CRS-like syndrome that can be quite severe and needs to be identified immediately. So TLS, uncommon, but when it happens you can get a CRS-like syndrome that can be serious.

DR LOVE: And CRS you treat the same way as usual?

DR STEIN: Well, I’m not sure exactly. That’s a good question. I’m not sure exactly how to treat it, but a lot of support.

DR LOVE: So here’s a topic, Amir, a lot of people are going to be talking about very, very soon after you present the data looking at this question with venetoclax in younger patients. Obviously, we’re looking forward to your presentation. And when is saw the answers here I was kind of surprised because I briefly read the — your abstract, but then when we chatted here I didn’t realize the patients were all high risk.

DR FATHI: They’re not.

DR LOVE: And you can see the answers there. So can you talk a little bit about today where you see HMA/venetoclax fitting in in a patient eligible for intensive therapy and maybe where you see it heading?

DR FATHI: Yeah. I encourage everybody to come to the plenary session and watch me sweat under the lights. So this is one of those pretest questions where you say well, I’ve got a lot of work to do.

They’re not all adverse risk. In fact, nearly a third of the patients — nearly 30% of the patients are intermediate or favorable risk.

It is important for me when I give this presentation to focus in on the eligibility of this study. We could not include FLT3-mutated patients because we could not randomize to an arm which did not include an approved FLT3 inhibitor. We could not include core-binding factor patients because we could not randomize to an arm that did not include GO, which is also approved in combination with induction. We could not include younger patients with MPN1 because many of those patients subsequently receive HiDAC as consolidation as opposed to transplant.

So what ended up happening is we excluded FLT3, we excluded core-binding factors, we excluded MPN1 in younger patients but not in older patients. So in our older patient population a sizeable proportion of these folks have intermediate and favorable risk. So we have to look at this study, look at the patient population that was studied, and then make the case that among these patients these were the outcomes that we saw. And I think I’ll try and make the case that the outcomes that we saw were pretty impressive in terms of not just EFS, which was the primary endpoint, but also the tolerability, the decreased risk of infection, bleeding complications, quality of life, health care utilization, time in the hospital, ICU, 60-day mortality. I think all of that together would make the case for the appropriate patient, such as the one that would have been enrolled on this study, this would be the right path.

So that’s kind of the answer that I gave, at least based on the experience I had on this study. Now it will take time. Even myself I’m so addicted to induction that it’s hard to step away from it because it’s what we’re accustomed to. I’m happy to talk more about it. There’s so much to talk about when it comes to this data, and I’m sure I’ll have an opportunity at some point in the future to delve into it.

DR LOVE: Yeah. I was thinking we could actually do a 2-hour program just on what’s in this 1 paper. They have a bunch of cases and scenarios. But Tara, first of all, it’s going to be incumbent for people to understand what the entry criteria was. I didn’t even realize it when I saw the abstract to start with. But then the question is what about everybody else. Is it being studied? And what do you think it’s going to show?

DR ERBA: Yeah. I mean, this is very important data. I think for many of us when we saw this data, and it really focused on the adverse-risk patients, many of us were looking for reasons not to give those patients intensive chemo because the only curative pathway for those patients is allotransplant. So if you can get them to a transplant then there’s a survival benefit. And in fact, more patients went to transplant in the PARADIGM trial after aza/ven than after intensive chemotherapy, and hopefully we’ll find out why that is. But a lot of us think we know why that is, because of toxicities.

In terms of how this data’s going to impact us, I want to be careful here because the problem is we don’t get this data back in real time at a lot of institutions. Even at our institution it takes 2-3 days to get this, and in the community sometimes it’s missed during the entire induction. And so if you say that we’re just going to start HMA/ven on everyone now you’re going to miss the opportunity to deliver curative therapy for patients with AML who are younger, and that curative therapy may be given without needing an allotransplant in first remission, and it’s a time-limited therapy.

So that’s my caveat about this data. I think it’s very important that we stress who these patients were. There are other studies going on in myeloMATCH that are looking at the same question for intermediate- and high-risk patients. Dr Lin here is leading the SWOG study of adverse-risk AML, where we randomize patients to 4 — 5 different arms, 7 + 3, 7 + 3 with ven, HMA/ven, aza/ven, CPX-351 and CPX-351 with ven. So we’ll be getting more data if that study actually is able to accrue now that Dr Fathi has stolen our thunder.

DR LOVE: So Tara, can you elaborate a little bit more? And I’m sure patients are going to come in asking about this. 7 + 3 is not necessarily a walk in the park. Again, in the patients who couldn’t have gotten into the study what’s the indirect evidence in terms of whether or not HMA/ven is actually a reasonable alternative for them even now?

DR ERBA: Me?

DR LOVE: Tara.

DR LIN: I think for high-risk patients, for adverse-risk patients, we have data from trials that use 7 + 3 where the efficacy isn’t that good. If you have a remission rate of 20-30% in HMA/ven, in higher-risk patients maybe higher than that, that’s a discussion to have. I think the challenge, like Harry was talking about, with an HMA/ven-based regimen in someone who isn’t going to go to transplant is what is your — what is your endpoint. And so when you have someone with higher-risk disease, where transplant is the goal, then it can be a great way, right? If it’s just as efficacious or more efficacious and less toxic then that’s a great way to get someone to transplant in better shape, because that’s what we think is going to be the curative part of their treatment.

DR LOVE: So a theme that comes up in every program we do no matter almost what the cancer is is allowing the patient potentially to have a role in the decision. I think this is a situation where if the patient is able to understand, and they want to understand, and they want to be involved hopefully you can present the data to them and have them participate in what’s becoming a more and more challenging question.

Selection of Therapy for Younger Patients with AML without a Targetable Mutation; Promising Investigational Strategies — Dr Perl

DR LOVE: Which really leads into our second topic that Sasha’s going to talk about, which is the current approach to the younger patient. So Sasha, take it away.

DR PERL: Thanks, Neil. And thank you for the opportunity to speak today, and thank you everybody for coming out nice and early today, and I hope this is helpful for your treatment approaches.

I don’t have NCCN Guidelines here, but I just have kind of an overview of how AML is treated today, as Tara has introduced in older patients, and this is how we look at patients who are potentially able to get intensive chemotherapy, primarily patients under the age of 60, where we’re still going to use a fit or unfit approach to say what’s the most appropriate way to treat them.

And these are the approved therapies and kind of the general way that we walk through the therapy for these patients, generally leaning towards the top part of the page with intensive chemotherapy, 7+3-based therapy with or without a targeted agent, such as a FLT3 inhibitor if a FLT3 mutation is present or GO if a core-binding factor rearrangement is present, maybe for some MPN1-mutant patients, and then make an assessment of their risk stratification as to whether we should do transplant or, again, time-limited chemotherapy, primarily high-dose or intermediate-dose Ara-C and/or maintenance for the patients who have maintenance options. Transplant is often incorporated into this therapy, and even for patients who are not able to get 7 + 3 you can come back to a more intensive approach if they were too sick to get induction.

I was asked, however, to talk about the patients who make up a large chunk of our practice who don’t have a target for therapy right now where we can say chemotherapy plus a targeted agent is the right choice. So how do we define this? First off, AML, as we’ll know, is a really diverse disease, and just the first target that we care about in our diagnostic workup is APL or not, with everybody else falling into a non-APL. And we don’t think about fitness when we treat APL patients. Just keep that in mind for later discussion. If you have the right therapy the fitness is not the point.

But up until 2017 that was just everybody else just got 7 + 3 if they were able to receive it, but then suddenly we have do you have a core-binding factor rearrangement, do you have a FLT3 mutation or everybody else, as I’ve shown you on the prior slide. Things are a little more complicated now as we have more drugs, more targets and more options coming out from studies. You might want to know if the patient had an IDH mutation. There are targeted inhibitors. You might want to know if they have an MPN1 or a KMT2A rearrangement because there’s now data adding menin inhibitors to front-line chemotherapy. We have no randomized data to say that this is the right approach, but you could easily see how this is going to be a focus and something we’re going to need to integrate into our diagnostic algorithm, which requires quick sequencing. And I’ll talk about how that’s starting to move forward, which Harry’s given an introduction to with the idea of myeloMATCH.

We also want to know about patients who don’t necessarily benefit from intensive approaches, such as TP53, where again really the goal is to get these patients to transplant quickly. And that might be a goal in patients who generally have adverse-risk features so that everyone else is a really different population than what we used to think of before.

So let’s look at the whole genotype of AML here, and unfortunately the type is very small here, so I’ll try to walk you through this. This is the whole genomic landscape of AML. Who would we say is the right patient for 7 + 3 and 7 + 3 alone, not an additional targeted agent? Well, if this is everybody, and we can get rid of the APL patients here in blue, and well the core-binding factor patients, we can get rid of those. I mentioned that half of the patients with MPN1 have a FLT3 mutation, so let’s get rid of those. And yeah, we might give menin inhibitors for the other people who have MPN1 or maybe the patients with KMT2A.

We’re kind of getting to a small number of patients here. Who’s left? Well, it turns out the patients who are left kind of fit into 2 groups here. You’ve got the CEBPA. These are mutated patients, which are quite rare, and that’s why I’ve put the unicorn there. It’s kind of fun to find them. They’re not all that common. And they do generally very well, and nearly no one else on this page does anywhere near as well. And these patients generally we’re going to take to transplant once they are in remission. These include patients with rare gene fusions, some of which might respond to menin inhibitors, and then patients who genetically look a lot like older AML, the patients that Tara was just talking about. They may have TP53 mutations, myelodysplasia-related gene mutations or cytogenetic abnormalities or a whole mishmash of other genetic abnormalities.

Let’s talk a little bit about TP53 because this is a real problem and one that isn’t easily fixed by existing drugs. Patients with TP53 mutations, particularly those who have abnormal karyotype, which is common, often complex karyotype, often monosomal karyotypes, and often deletions of 5q, monosomies of chromosome 17 or deletions of 17p or monosomy 17. And if you see those features, that’s this disease. It’s very hard to cure no matter what you do, with intensive chemotherapy, shown on the left, or with lower-intensity therapy, ven/HMA, shown on the right. And even when you go to transplant outcomes are worse for these patients.

As you see in the top, this is a recent UBMT review from over 1700 patients that were reviewed who had ELN 2022 adverse-risk disease. And you can see the adverse risk does worse than the intermediate risk, which does worse than the favorable risk, but in reality if you actually look at just the ELN 2022 cytogenetic risk here the TP53 patients pulled that adverse risk down even farther. So those differences between adverse and intermediate risk are not all that great when you get rid of TP53. So there’s a big group of patients that can do well if we can just get them to transplant, and we need to identify them.

So how do we identify who’s got the disease to treat with targeted agents versus everybody else? This is what myeloMATCH is trying to answer. We’re doing a rapid screening, 3-day turnaround next-generation sequencing panel, FISH panel and a karyotype so that we can get an ELN 2022 risk category for everybody walking in the door within essentially 72 hours and hopefully have a slot to enroll them to a study at that point.

We have tier 1 treatment trials, which are basically front-line chemotherapy. All of these are randomized Phase II studies that are signal finding, and hopefully we will soon have studies for patients thereafter, either those that don’t respond optimally, those who go to transplant, or those that remain MRD-positive.

There’s been mention of the younger adult high-risk study, a study that Tara is leading, which is looking at 5 arms all looking at the question of whether venetoclax improves outcomes for patients who don’t have an identifiable mutation that a targeted inhibitor would be more appropriate to treat or to answer the question in a randomized setting. And a similar approach is being used in younger patients. And on both of these studies aza/ven is an alternative and one of the study options. And as was mentioned before, that’s going to be presented in terms of some of the new data on randomized studies comparing intensive chemotherapy to HMA/ven in fit patients.

There’s even more intensive approaches with venetoclax that can be looked at. The venetoclax/FLAG-IDA regimen, developed at MD Anderson, is a highly intensive regimen. And I think of this really as AYA therapy in ALL, adolescent/young adult therapy in younger patients. This is for your fittest of the fit youngest patients, and it’s appropriate that we look at these highly intensive regimens really in these patients. I would not look at this as a general fix-all regimen just because of the very high remission rate, which it clearly does have, and a very high MRD-negative rate, which it clearly does have. But this is the kind of therapy that we really should tailor for the right patient, and we have no randomized data to say that it’s better. So while it’s very impressive that 95% of patients can get into a composite remission, with nearly all of these patients going into an MRD-negative remission, we don’t know that this is better than what we can achieve with less intensive approaches.

And it’s kind of interesting when we look at this as a salvage regimen it actually performs for patients who don’t have TP53 and who are in first salvage, very similarly to front-line outcomes. And that, to my mind, makes me wonder whether this should be positioned first of positioned when front-line approaches don’t work or don’t work optimally. And I don’t think we know the answer to that. So it’s a very interesting regimen. It’s one that I think needs some randomized data, but it is quite effective, and it may be the wave of the future. We’re just going to have to see.

What do we know about HMA/ven versus more intensive approaches? This is the first published study that looked at that question from a Chinese group. This is a little bit of a funny study in that they only gave 1 cycle of HMA/ven or intensive chemotherapy, with a second offered if patients didn’t go into remission. And the remission rate on the decitabine/venetoclax study was nearly 90%, so practically nobody needed a second cycle. Everybody on this study got intensive chemotherapy thereafter, high-dose Ara-C consolidation, and some went to transplant.

What we learned from this study was that it was feasible to give low-intensity therapy to younger patients. Here, these were all patients under the age of 60, treated in China in a multicenter study. And indeed the survival was no different between the 2 study arms, 7 + 3 versus HMA/ven. And if you look at the subgroup analysis of the adverse-risk patients they did substantially better in terms of their remission rate to HMA/ven, which was eye opening. More than twice the remission rate and a 50% —percentage absolute increase in the CR and CRi rate, which I think is actually noteworthy recognizing their small numbers.

We will have data from the PARADIGM study, as has been mentioned here. Oh, and just to go back to this. This included all genotypes. This study did not restrict in the way that Amir was mentioning before.

On his study that he’ll present later today, later at this meeting, Amir and colleagues, including myself, randomized newly diagnosed patients, with the restrictions that are shown at the bottom, to either 7 + 3, or CPX-351 if more appropriate, or azacitidine/venetoclax, treated for as many aza/ven cycles as is appropriate. And as was mentioned before, these are patients that primarily would go on to transplant in remission. And the real question was did we see a difference in remission rates, did we see a difference in transplant rates, and the primary end point was event-free survival, recognizing that it was natural that there would be crossover if patients responded or didn’t respond to therapy and had the option of being treated with the other.

These are some of the results, and we may use this just for our discussion here just so you can see what they showed. There was a significant improvement in event-free survival. We did not see a significant improvement in overall survival, but we’ll show that at the presentation. The remission rates for all measurements on the study, all defined responses, were higher in the aza/ven arm. The composite CR rate was significantly higher. The CR rate was numerically higher but not significantly higher for aza/ven. And the toxicity was, in terms of serious infections, was a little bit better in the aza/ven. What was potentially more noteworthy was that there was no 60-day mortality in the HMA/ven arm, and we’ll show the ICU utilization, as well as some other hospital metrics at the talk.

In the few minutes that I have left I do want to mention some novel agents that are being tested in the front-line randomized setting. One that’s called PVEK, or pivekimab, which is an ADC targeting CD123. This is a drug that is under review by the FDA for BPDCN right now. It’s an interesting drug because it’s also being developed in CD123-positive AML, which is a significant subset of patients. And like other ADCs that have been developed for AML, gemtuzumab, vadastuximab, it does have some single-agent activity, but I can’t say in the relapsed/refractory study it was best shown what it can really do. You do see about half the patients here have blast reductions, and about 20% of patients will achieve a composite CR of some sort. But this is probably a drug that’s going to be best combined with other chemotherapy, as I think is generally the case for ADCs in AML, and not in the third-line salvage as was done on this Phase I study.

Data with ven/aza plus pivekimab will be presented by Naval Daver later at this meeting. They do show what looks like a good CR rate, a good CR/CRi rate. And what’s kind of notable about this agent is it’s a drug that doesn’t seem to have much in the way of myelotoxicity, and that seems to be an advantage in adding it to other combination regimens. And there are data on using this with FLAG-IDA that will be presented as a poster at this meeting as well.

Lastly, I think Harry’s going to talk about adding quizartinib to intensive chemotherapy outside of the context of FLT3-ITD-positive patients. That’s a study that is currently enrolling called QuANTUM-Wild. It’s based on a randomized Phase II study done in Spain called QuiWI that showed an appreciable improvement in event-free and overall survival in patients who did not have FLT3-ITD. The mechanism of this we’re still working out, but the observation is rather striking.

And just in conclusion, rapid integrated genomics are now coming to a treatment center near you right now if you can enroll a patient on myeloMATCH, and I think the workflow will change so that this is available in more and more places. We have a 3-day turnaround time in our academic center. With my own patients I’m able to get this even off study, and I think more and more commercial labs will more towards this for a limited panel.

If no molecular targeted therapeutic is found that would improve outcomes the key question is can you get this patient to the benefit of a transplant because the vast majority, with the exception of the CEBPA patients, will benefit from that approach. And if patients don’t clear MRD that may be the right answer or time to add in targeted agents too.

Novel approaches, including adding venetoclax to HMA, will be presented at the plenary, or highly intensive chemotherapy, but that needs randomized study to confirm. And there are a number of novel agents that show promise, but we will hopefully see more randomized data soon, such as PVEK and quizartinib. And thank you for your attention.

DR LOVE: Thanks, Sasha. We’re getting some great questions from the audience. I’ll get to that in a second. So Harry, we’re actually — we’re doing 4 posters next week at San Antonio. One is on the 10 Phase III trials general oncologists have to understand in order to choose second-line endocrine therapy of ER-positive, HER2-negative disease, just pointing out that there’s a lot of data for general oncologists to consider. And looking at your myeloMATCH study, that’s a fairly complicated study to try to digest.

I’m kind of curious, though. How do you get everything — all the assays done in 3 days? Is it central?

DR ERBA: Yeah. It’s central. The NCI is supporting this. When they realized that all of the NCTN cooperative group leukemia committees were going to work together to develop a comprehensive trial for newly diagnosed AML and myelodysplastic syndrome back in 2019 they designated us as a project. We are 1 of the 3 adult precision medicine initiatives that’s being funded by the NCI. And so the NCI is funding this work.

We use a thermal fissure platform for the 3-day turnaround of the myeloid gene panel. We have amazing cytogeneticists in Seattle who do a 3-day turnaround of metaphase chromosomes. Brentwood at UCLA Children's Hospital does flow cytometry. So we get all this data together from samples that are sent in within 72 hours, and then every afternoon on the East Coast at 1 o'clock we get together a group of leukemia docs, laboratory people, IT people, look at the data and make sure that the patient is being assigned to the correct study. I have to be there at 1 o'clock today to oversee some of those patients.

So it’s an amazing platform. The government is supporting this, and it’s caught on very quickly. It’s accrued very quickly to the screening protocol.

DR LOVE: Eytan, I’m wondering if you think someday this will be the case for every patient with AML.

DR STEIN: I think it would be nice if we had a rapid way to identify every patient with AML and then assign them to the right treatment.

DR LOVE: So Tara, some questions from the audience. What’s your strategy for dose reduction of ven with azole use?

DR LIN: So the package insert will give guidance in terms of if you’re using fluconazole or voriconazole or posaconazole. Our preferred azole is posa because of the pattern of fungal infections that we see in the Midwest, which is not insignificant. And with that dose reduction we actually lower the venetoclax down to 50 mg. And very rarely we will see someone come in who has not had an appropriate venetoclax dose reduction, and those patients are at risk for really prolonged cytopenias, so it’s important to really adhere to those package guidelines. And that’s also part of the challenge when you’re starting someone on venetoclax is that you have to get insurance to pay for venetoclax, you have to get insurance to pay for your antifungal, and sometimes you’re not sure how that combination’s going to work out. So for that reason we very often start people in the hospital, where we have all of those drugs available and then can make those adjustments as we need to based on coverage.

DR LOVE: So Amir, another question from the audience. What do you do if a higher-risk younger, fit patient is refractory after 2 cycles of ven/aza?

DR FATHI: Well, I mean, refractoriness, to me, has a specific definition. This is one of the challenges we had on our study. If a patient gets a cycle or 2 of 7 + 3 and goes from 60% blasts to 30% blasts usually people go another way. With HMA/ven you continue, and there is a chance, although diminishingly so after 3 to 4 cycles, that they may get into a remission. I can tell you on our study, on those 86 patients who got HMA/ven on that arm, 2 or 3 patients got their CR after 4 cycles, and 1 of them is still alive, doing well after transplant. So it does depend on the pattern of reduction in blasts that you see over time. If you’re giving cycles of treatment, and there is no significant reduction, and there’s persistence of blasts I would switch course.

DR LOVE: So getting back to our survey. Again, we talked about this before, a 35-year-old, but also who has a high-risk situation, patients who are using HMA/venetoclax. But Eytan, what I was also curious about is responses when we say you have a younger patient eligible for intensive therapy, but they have a secondary AML. We see a lot of CPX, but a number of people are bringing up HMA/ven. How do you think that through and how are you approaching these secondary cancers, Eytan?

DR STEIN: Yeah. So for the secondary it’s based on the randomized study of CPX-351 versus 7 + 3, which showed a survival advantage with CPX-351. We’ve been prioritizing using CPX-351, especially because when you look at the curves after transplant the patients who were transplanted on the CPX-351 arm do much better than the patients who were transplanted on the 7+3 arm. Now I think Amir’s study is going to help us answer this question about which one of — which one of us is going to need to change our answers here or which one of us should be changing our answers.

DR FATHI: This was kind of a tricky question. I’m sorry, Neil, to jump in, but because secondary AML can apply to both MDS — AML with MDS changes, as well as therapy-related AML. There weren’t too many therapy-related AMLs on my study.

DR LOVE: How does the approach, Amir, differ based on the etiology, whether it’s MDS or cancer — cancer treatment?

DR FATHI: Well, I mean, I think there is a decent amount of overlap, but with therapy-related AML oftentimes there’s 2 types, as you know. There’s alkylating and radiation-related myeloid malignancies and then topoisomerase-related malignancies, and they both carry a relatively poor prognosis. On our study we didn’t have too many of those patients. But in patients who have a preceding history of MDS or an MPN and then develop AML generally we call that an AML with myeloid — MDS-related changes. Those patients in general I would posit that our study supports the use of HMA and venetoclax in that setting.

DR LOVE: So Tara, what about the patient who has disease relapse after 7 + 3? Does HMA/venetoclax — it looks like everybody considers HMA/venetoclax. Is that your usual approach and how does it differ from the way you treat people up front?

DR LIN: I think in the relapsed setting, especially in a case like this, we’d always think about a clinical trial, but outside of a clinical trial using an HMA/venetoclax regimen we are able to salvage a lot of people, get them back into remission. There are some older data that shows that maybe an HMA can actually spur some graft versus leukemia effect, which might also be part of the efficacy as well.

DR LOVE: So Harry, I feel like we talk about ADCs no matter what program we’re doing nowadays in oncology. I’m curious what your — we asked the faculty about — I don’t know if there’s a nickname for pivekimab, but —

UNKNOWN: PVEK.

DR LOVE: What is it?

DR ERBA: PVEK.

DR LOVE: Like it, great. So any thoughts, Harry, in terms of the future of this ADC?

DR ERBA: Well, I think when you’re adding a third agent to either 7 + 3 or to HMA/ven you have to always worry about the potential for toxicity and what is really going to be the efficacy data showing a benefit. So outside of a Phase III study I think it’s hard to imagine that the ADC is going to add very much other than toxicity at this point. I mean the data that we have from the Phase II is very intriguing, but until you do the Phase III and actually see the toxicity associated with the addition of that agent that has maybe not so much myelosuppression, but it has some myelosuppression.

DR LOVE: So a final comment from Sasha. It seems like not a huge amount of enthusiasm for CAR T in AML.

DR PERL: There’s a huge amount of hope for CAR T, but until we see it pay off and the right way to move it forward, I think we’re all just kind of cautious about recommending it as a proven methodology in this disease. It’s hard to find the approach that really achieves the kind of benefits we’ve seen in B lymphoid diseases or myeloma in myeloid malignancies, in large part due to the lack of specificity for the target being in the tumor. Everything that’s been developed for CAR T hits normal hematopoiesis, and so it’s been hard to come up with an approach that’s really tolerable in patients and really is leukemia specific.

Role of FLT3 Inhibitors in AML Management — Dr Erba

DR LOVE: Alright, Harry. Let’s talk about FLT3 inhibitors.

DR ERBA: Well thank you for the invitation to be here. Actually when I first got the invitation RTP asked me to be the moderator, and fortunately for all of us Neil Love is actually here today, so thank you, Neil, for all you do.

So I’m going to be talking about FLT3 inhibitors, mostly in previously untreated AML. And we have to start with what is FLT3. It’s a receptor tyrosine kinase that is present on hematopoietic stem cells. It binds the FLT3 ligand, which is shown over there on the right-hand side, caused dimerization of the receptor and then activation of the intracellular kinase. This is important because this is present on AML blasts. FLT3-wild-type is present on AML blasts and can still bind the FLT ligand, and there’s data showing that the amount of FLT3 on the surface of an AML blast may actually be important in prognosis. I’m going to come back to that.

The mutations that we see in FLT3 are the more common ITD mutation found in about 25-30% of patients and the less common tyrosine kinase domain mutations. For either the ITD mutation or the TKD mutation Type I inhibitors are effective. Those are midostaurin and gilteritinib. For the TKD mutations in the tyrosine kinase domain the FLT3 folds into its active confirmation and is not inhibited by Type II inhibitors, such as quizartinib and sorafenib, which only can bind to the inactive conformation. So there’s a bit of a difference between these inhibitors that you need to understand, and it leads to their specificity.

In patients who are fit for intensive chemotherapy up to the age of 75 when the Type II inhibitor quizartinib was added to standard 7+3 induction, up to 4 cycles of consolidation with high-dose cytarabine, and up to 144 weeks of continuation there was an improvement in survival for the entire cohort, with a hazard ratio of 0.78. Now you can argue that this hazard ratio is exactly the same hazard ratio shown in the RATIFY trial, and so then we’re left with arguing that while the RATIFY trial had about 25-30% of their patients, the TKD-mutant patients, and only had patients up to the age of 59. And so if you compare then the younger patients in the QuANTUM-First study who only had the ITD mutation to those in RATIFY the hazard ratio is 0.68 for quizartinib and 0.80, but there we’re left with just comparing between 2 clinical trials.

The data that convinced me, when I saw this data coming out, that quizartinib was something different is that in patients who have a FLT3-ITD mutation who received quizartinib but did not undergo allogeneic transplant in first remission there was still a survival benefit compared to placebo. This wasn’t seen in the RATIFY trial. Only the patients who underwent transplant. Now you might say, as many of my transplanter friends say, everyone gets a transplant. Well, that’s like a plumber saying all of my patients have problems with their plumbing, okay? Yes, all of their patients are going to get a transplant, but only 50% of patients in this trial actually got a transplant. And these were all patients fit for intensive chemotherapy, so you don’t know who is going to get a transplant. And so I think you need to err on the side of giving them a regimen that will benefit patients who will or will not get a transplant. As we know, though, that the benefit wasn’t so pronounced in older patients, and I’m going to come back to.

This is the first — the QuANTUM-First study was the first prospective trial that looked at a FLT3-ITD NGS-based MRD assay prospectively at prespecified times. And in this study what we showed was achieving an MRD-negative status by this FLT3 assay, regardless of treatment, was associated with a better survival than patients who did not. However, patients who received quizartinib had not only a lower median FLT3 variant allelic frequency at the various timepoints shown here, but in this slide more patients were completely negative on that assay for — if they were treated with quizartinib.

The patients who were completely negative did not seem to be benefitting from quizartinib versus placebo, as you can see on the top. But remember, more patients got to that FLT3-negative status, and patients who remained positive at each of those timepoints, those patients were still benefitting from quizartinib.

But as I mentioned, we know that in patients over the age of 60 the hazard ratio for survival with quizartinib was 0.9 and not statistically different. However, I refuse to believe that we should just be ageist about this. There is nothing magical about the age of 60, and it’s really the biology of the disease. This is data that was first presented by our colleague Mark Levis, and the manuscript is in preparation now, looking at whether there’s a genomic signature that might show a benefit — a place where quizartinib will be most beneficial. I’m not showing you the data in the younger patients. In the older patients, those with the triple mutation of FLT3-ITD, MPN1 and DNMT3A, which Elli Papaemmanuil and other have shown is the worst actors of the FLT3-ITD-mutant patients, there was a survival benefit even in the older patients with those mutations with quizartinib. So we shouldn’t just be ageist about this, and I’m going to come back to this triple combination at the end.

Of course, the tendency is many times to start with an HMA and add in a third drug, and we have really great data from MD Anderson showing these triple combinations. I’m going to show you 2, decitabine/venetoclax with quizartinib. One thing to be very careful about in this — when you go to use this in your own practice is that these are studies being done at MD Anderson, and they are constantly learning from their experience and changing the protocol. So along the course of the way they went from 10 days of decitabine down to 5 days of decitabine, they shortened up on the venetoclax, they shortened up on the quizartinib, they did bone marrows at day 14.

So with that caveat in mind, you can see a very high remission rate over on the left. Most of those were MRD-negative.

The median survival had not been reached, but the follow up is only 17 months, with a 1-year overall survival of 72%. That looks really good, but look at the number of censored marks. They’re mostly at the beginning of this curve and very few at the end holding it up. In fact, if you look over on the right the patients who did not get an allotransplant have a remarkably better outcome, mostly because of transplant-related mortality in the patients who did get an allotransplant, in red. Of course, this is a single-center study, and we don’t know what other factors that doctors may have used to decide if you’re going to get a transplant. For example, the MPN1, FLT3 combination may be particularly sensitive to quizartinib, as I just showed you, and maybe those patients weren’t getting transplant and were going to do better anyway.

The other combination from MD Anderson is the aza/ven/gilt combination with very high response rates, and many of these after 4 cycles are MRD-negative. You can see that the relapse-free survival at 12 months was 75%, and at 12 months the overall survival was 83%. This is going to be updated at this year’s congress.

But keep in mind gilteritinib being a Type I inhibitor is also active against the TKD, and they had 8 patients in their study that were TKD mutated. That’s shown in the blue survival curve on the right. Those with the FLT3-ITD are shown in the green curve. And also you have to think about what is your goal here for these patients getting this triple combination. If it’s to bring them to transplant this may be a viable option using this regimen because 13 out of the 30 actually got a transplant, and I would imagine most of those were actually the FLT-ITD-mutated patients.

But let’s come back to quizartinib. Quizartinib actually has single-agent activity against wild-type FLT3. This was shown in the original studies, Phase I studies published by Jorge Cortes that in patients who were FLT3-ITD-negative they saw responses with quizartinib. This was a Phase I study, so they were trying to get patients on quickly. They didn’t necessarily need to have all their patients having an ITD mutation. Now some of these patients actually truly had ITD-negative disease using a very sensitive assay.

And so this led Pau Montesinos and his colleagues in the PATHEMA in Portugal and Spain to do this randomized trial, a Phase II randomized trial of idarubicin, cytarabine and quizartinib at a higher dose than in the QuANTUM-First study, 60 mg daily for 2 weeks, followed by high-dose cytarabine in patients who were FLT3-ITD-negative. And as Sasha showed you, there was an improvement in the event-free survival and overall survival for the entire group up to the age of 70.

The benefit was seen in the younger and the older patients. It was seen in patients who did get a transplant, didn’t get an allotransplant. And there was again a signature that was seen. Patients who had MPN1 mutation, DNMT3A mutation seemed to have the greatest benefit, as did the patients with ELN favorable risk, and there were very few core-binding factor patients. And so most of this are the MPN1, FLT3-ITD-negative patients.

Look at that overall survival. What’s going to beat that? Everyone’s trying to beat that. We have menin inhibitors coming, right? And how is this all going to play out? Do we use a menin inhibitor? Do we keep up with the FLT3 inhibitor?

And there may be a specific FLT3-like expression signature that correlates with this benefit. I’m just going to skip through that very quickly, try to advance.

Now gilteritinib has been compared to midostaurin in the PrECOG study, the HOVON study of 7 + 3 with gilteritinib versus midostaurin, will read out maybe in the next year or so. This study looked at the combination, and the primary endpoint was MRD-negative remissions following induction.

And what they showed was that there was a higher CRC rate with gilteritinib than midostaurin, but by flow cytometry, down at the bottom, there was not an improvement in MRD-negative remissions with gilteritinib versus midostaurin. And this was also true using the FLT3-ITD assay, the patients — or FLT3-TKD assay. Those who were negative still had a lower chance of having FLT3 negativity after — in that remission. It may be that we’re just looking too early.

And I will finish with the BMT-CTN study. This study looked at the addition of gilteritinib versus placebo in patients with FLT3-ITD-mutated AML undergoing allotransplant. There was a trend for an improvement in relapse-free survival that didn’t quite meet statistical significance. No difference in overall survival.

But the patients who were MRD-positive either right before transplant by the FLT3-ITD assay or right after transplant, which were fewer, had a benefit with gilteritinib in terms of relapse-free survival. But keep in mind that there was more hematopoietic toxicity there.

And then finally, just following up on the publication by Sasha Perl and colleagues from the ADMIRAL trial, we have long-term follow up for the use of gilteritinib versus standard chemo for FLT3-mutated AML patients with relapsed/refractory disease. And I’m going to point to the tail on this curve. You can see it’s only around 10%. 26 of these gilteritinib-treated patients were still alive more than 2 years later, and the majority, 18, underwent allotransplant, and 16 of them received gilteritinib post allotransplant. Thank you.

DR LOVE: So let’s take a look at the survey. And Sasha, we presented a situation of a patient with a FLT3-ITD or -TKD mutation eligible for intensive chemo. In general, people go with quizartinib for ITD and midostaurin for TDK. Interestingly, Dr DiNardo brings up the issue of HMA/venetoclax plus either gilteritinib or quizartinib. So Sasha, I’m curious how patients tolerate quizartinib compared to midostaurin. And any thoughts about this MD Anderson strategy of HMA/venetoclax plus FLT3?

DR PERL: So for the tolerability question I think generally the drug is very well tolerated. In terms of monitoring, it’s more important to monitor QT prolongation with quizartinib, and there’s important drug/drug interactions. The likely interactions that Tara had mentioned warrant dose reduction of quizartinib, and azoles are on that list. So you cut the dose of the quizartinib in half if you’re using posaconazole or voriconazole.

In terms of the HMA/ven plus a TKI approach, I’m very much a fan because we’ve been enrolling studies just like MD Anderson has. We’ll be presenting the VICEROY study at this meeting, which shows actually very similar results. Jessica Altman will give that talk I believe on Sunday, actually, just after the plenary. And I think what we have shown is that now on several studies is HMA/ven plus TKI, at least for gilteritinib or quizartinib, leads to very high remission rates that seem to be sustained, and that’s impressive because it’s been hard to do that in this population before. And I think we’re ready for the point where we can do a randomized study, much like Amir has just done, with patients who didn’t get targeted therapy in this setting where they are getting targeted therapy, 7 + 3 plus a TKI versus HMA/ven plus a TKI and figure out what’s the right approach.

Again, to get back to the point I made in my talk, in APL we don’t care if you’re fit or unfit, we just want to give the best therapy. If a lower-intensity therapy can be given to the 80-year-olds we’re giving HMA/ven plus TKI, and they’re 20 years old but it’s better therapy, let’s give the better therapy.

DR LOVE: So Tara, any comments on the management of patients who are older? Here we say 80 with FLT3 mutation, again ITD/TKD. Most people are saying aza/ven, but there’s a sprinkling of people who also bring in the issue of adding in a FLT3 inhibitor. Can you comment?

DR LIN: I think it’s important. I think we have single-institution data looking at triplets, and a lot of people are adding the FLT3 inhibitors to HMA/venetoclax, but we really need more larger trial data to help guide us. I worry about myelosuppression in a triplet, and like Harry mentioned in his talk, the way that MD Anderson has developed some of these regimens they’re learning over time and doing dose reductions. And so if this is something that you’re going to do you want to use the most recent paper and abstract that you can find to choose those dose reductions versus the first one that you might have seen published.

DR LOVE: So Amir, when I heard about these FLT3 signatures it kind of reminded me in a weird way of the HRD assays they do now to look at whether or not to use a PARP inhibitor. Obviously, with PARP you have a BRCA mutation, you know they’re sensitive, but they also came up with this LOH signature, and this kind of reminds me a little bit of that. Any thoughts, Amir, about this strategy, and is it persuasive enough that right now you think it ought to be applied?

DR FATHI: I think Harry just went through a lot of that data. The truth is these are not really FLT3 inhibitors. They’re nonspecific sufficiently promiscuous tyrosine kinase inhibitors that happen to also hit FLT3, and as a result you’re more likely to potentially have activity in other pathways that might be impacted and potentially vulnerable.

Now quizartinib is unique because it’s not really clear why it may be more active. We’ve known for some time it’s been more active. As Harry mentioned, Dr Cortes presented this data almost 10, 12 year ago. We were all at that meeting, and I think we all kind of a little bit forgot about it. And then the Spanish came by and showed us the data from QuiWI, and now we’re doing the larger Phase III study. I wrote too early to tell because I don’t think we all yet add quizartinib to up-front therapy in FLT3-wild-type patients, but I think in the future if this Phase III shows what QuiWI showed then I think we have to. I mean there is certainly then some degree of activity, perhaps even more prominent than what we saw with QuANTUM-First.

DR LOVE: So Eytan, this is a question we added because you suggested it. I wasn’t even exactly clear what you were thinking. Maybe you can explain the idea behind it and now what you think about the faculty answers.

DR STEIN: Yeah. So there’s emerging data that suggests that specifically in the group of patients who have MPN1 mutations cooccurring with FLT3 mutations that if they’re MRD-negative for that MPN1 maybe they don’t need an allogeneic stem cell transplant. And it raises the question whether a patient who is also negative for their FLT3-ITD by this very highly sensitive assay maybe they don’t need an allogeneic stem cell transplant. So I do think there may be — while right now I think we’re transplanting a lot of these patients, I think there may be a time where maybe we can reduce the number of patients that need to go to an allogeneic stem cell transplant even with a FLT3-ITD.

DR LOVE: I mean it looks like at least at this point, it’s getting people not to use maintenance.

DR STEIN: Yeah.

DR LOVE: Most people anyhow.

DR STEIN: Exactly.

DR LOVE: A couple people still say yes.

So Tara, we have a 60-year-old patient who has disease progression after 7 + 3 plus quizartinib. What comes next? You say aza/ven/gilteritinib. Other people say gilteritinib alone. How do you think that one through?

DR LIN: The key parts of this question that kind of, I think, led to my answer and maybe other folks is that this is a fit patient who got intensive therapy, got a transplant. The question for me at the time of this relapse is it — are we still going for cure. Is DLI, a second transplant, something like that on the table? If yes, then I’d want to give a more aggressive regimen like a triplet. If no, then I think that that’s when you may use gilteritinib alone. That would be the — right — the FDA approved regimen would be to use gilteritinib alone.

DR LOVE: And then Sasha, you have an 80-year-old patient, FLT-ITD mutation, disease progression after aza/ven. Our way of doing consensus is if everybody gives the same answer we call it a consensus. This looks kind of near consensus on gilteritinib. What’s your experience with gilteritinib in this situation? Clinically how much benefit does it provide?

DR PERL: Yeah. I mean it’s a good question because the ADMIRAL study, which Harry just showed, was enrolled in the pre-venetoclax era. So we haven’t done a follow up to say that those data hold up in this era or whether to say just add the FLT3 inhibitor to ven/aza, what the right answer here is. I think in most 80-year-olds I’m trying to avoid toxicity. It’s a proven agent, it has activity, and its mechanism of action is sufficiently distinct so that I would think that we would want to try that in the setting where post venetoclax most any other cytotoxic agent is not very effective, but targeted agents can have still activity.

DR LOVE: Harry, any new agents out there targeting FLT3 that you’re excited about? Any new trials out there that you’re excited about?

DR ERBA: Well, tuspetinib is a FLT3 inhibitor that, like Amir Fathi was saying, also targets a number of other kinases, and this actually may be a potential benefit. In the Phase I study of tuspetinib we’ve seen activity outside of the FLT3-ITD-mutated patients. So in terms of FLT3 inhibitors coming along that would be the one I would focus on right now. There are also FLT3 inhibitors that are active against the gatekeeper mutation that can develop.

DR LOVE: So Eytan, going back to the faculty questions. Someone wants to know can you comment on trials evaluating the addition of venetoclax to approved drugs other than HMA or 7 + 3. I’d actually like you to comment on ven plus 7 + 3, but this person wants to know about adding to anthracyclines or GO regimens — anthracyclines and GO.

DR STEIN: Yeah. So I think what — I mean there are trials that are trying to combine a lot of these things together. What we worry about is that there’s going to be a ceiling to the myelosuppression that you can get to the patient. So you may have — and we know this from studies over many years. You can get higher CR rates and clear the blasts more, but you end up killing more of your patients because of the myelosuppression. And I worry if you add GO and ven to an anthracycline you’re going to send that marrow into a toast-like state where it’s unable to recover.

Incorporation of IDH Inhibitors into the Care of Patients with AML — Dr Fathi

DR LOVE: So let’s move on now. Amir, can you talk a little bit about IDH inhibition?

DR FATHI: I guess everybody was standing, so I’m going to stand. Alright. So I’m going to be speaking about an oldie but a goodie, IDH inhibitors.

So this is a slide that Sasha used much more effectively than I’m going to. Just taking the pies off of this thing, I can’t do that. The whole point of this slide is just to show that AML compared to when I was in fellowship and certainly prior to that, 20 years ago, is — it did not become more complex. We just found out how complex it was, so it gets really involved. And we learned — somebody who was wise at some point said the more we learn the more we know what we don’t know, but at the same time we hopefully find targets for therapy and learn how to go about it. And here is the IDH cascade there right in the middle. I’ll get into that a little bit later.

There are now multiple timelines that I’ve stolen from various papers. This is the most recent one that I could find because this keeps updating, and I’ve added ziftomenib there at the end. I think Dr Stein’s going to probably talk about that. But the 3 IDH inhibitors that we currently have available for use are highlighted in the red box here, starting with enasidenib in 2017.

So right after I got done with fellowship, and in many ways I’m lucky, and I think a lot of folks on this panel can also attest to that, we found out about IDH1 mutations based on a paper that emerged from Washington University in St Louis. About 8% of patients, many of them with cytogenetically normal intermediate-risk AML, were found to have this mutation in isocitrate dehydrogenase 1. And then a subsequent study published afterwards found that even a larger proportion of patients had IDH2 mutations.

So IDH (isocitrate dehydrogenase). It’s an enzyme in the KREB cycle, and if you’re like me you’ve learned the KREB cycle about 2 dozen times in your life and then forgotten about it right afterwards. And the only reason I know this is because I explain this to my patients and students and residents every day, so it’s in my brain. But isocitrate dehydrogenase is a key enzyme that converts isocitrate to alpha-ketoglutarate leading to the production of ATP in the case of IDH1 in the cell, and in the case of IDH2 in the mitochondria.

When you have a mutation, however, something bad happens. The altered IDH protein instead makes an alternative catalysis converting alpha-ketoglutarate to 2-hydroxyglutarate, which happens to be an oncometabolite and is key in the inhibition of certain essential enzymes, including TET2 and other hydroxylases, leading to suppression of TET and aberrant hypermethylation of key genes, which then results in a blockage in differentiation of myeloid cells in MDS and AML. That’s how this is thought to work based on very intricate and elegant studies, some of them coming out of MSK.

So the way that IDH inhibitors ultimately were developed to work was to inhibit these altered IDH proteins, suppress 2-HG, and this was shown actually in both preclinical and clinical studies and lead to therapeutic response.

This was one of the earliest studies with an IDH inhibitor, enasidenib. I remember Eytan and I sitting in rooms when the first few patients — one of them I think was treated at MSK. One of the first things that Eytan saw was hair growth, and I was super excited. That didn’t pan out.

So there was Phase I dose escalations, and then there was subsequent dose expansion in a Phase II looking at the promising dose of 100 mg daily of enasidenib, this IDH2 inhibitor. And there was a focus specifically on relapsed or refractory patients. Multiple different cohorts were studied here, and you have the overall response rate there at the top, and then the CR and the CR plus CRi. Suffice it to say overall response rate hovers around 30-40%, and the CR/CRi rate hovers between 20-30%. And this is the range of responses you see oftentimes regardless of the variant allelic fraction proportion in many of the patients we’ve treated in the relapsed/refractory setting.

These are the survival curves. These aren’t knock your socks off exciting survival curves, 9 months, but then you splay out those who have responded, even those who haven’t had a CR response, and you see they do better over time. In fact, a 19.7-month median overall survival for a pill you can take every day when the alternative was intensive chemotherapy or nothing is pretty impressive.

How about ivosidenib? Well, this was the next kid on the block, an IDH1 inhibitor, again made by Agios, currently with Servier. They did a similar study of dose escalation and dose expansion in a Phase II looking at various primary and secondary endpoints.

Similar. This was published by Courtney DiNardo in the New England Journal, where you again see a CR/CRi proportion that hovers around 30% or so. And similar survival curves. Improvement, particularly in patients who’ve had any degree of response and particularly so in those who had a CR.

The part that I didn’t show you with enasidenib but also applies to all these IDH inhibitors is it’s not just that you have these responses with a very relatively well-tolerated IDH inhibitor, you also — at least marrow responses. You also have this transfusional independence emerge, both maintenance of transfusional independence and development of transfusional independence, and this really impacts quality of life. This is the case even in some patients, as you can see here, who had less responses than CR or CRh. In fact, many of these patients, probably because of the differentiating effects of IDH inhibitors, had PRs. So we talk about PRs, which is a reduction in blasts in the marrow to less than — not less than 5% but recovery of blood counts. They don’t see it much with traditional intensive chemotherapy because it knocks everything down, but with differentiating agents we got to see a lot of PRs — a decent amount of PRs.

Alright. So as I mentioned, these drugs work by suppressing 2-HG to oncometabolite, releasing the blockage on key enzymes that suppress maturation and differentiation, allow differentiation to take place, which is good, but then in a subset of patients cause what we would currently describe as a differentiation syndrome, akin to what we saw with APL, for example, and use of ATRA and arsenic. And these generally occur somewhere between 2, 3 weeks after start of treatment all the way out to 4, 6 months.

They can happen at any time, and they manifest like this. Sorry, this is a little bit close to the left edge there, but the most common effects that you see is shortness of breath and fever and infiltrates and rash and a little kidney failure, a little adenopathy. The problem is a lot of these mimic what you would otherwise see also in a lot of AML stuff; AML itself, the treatment of AML, the heart disease and particularly infections. So how do you tell? Well, sometimes you can’t. But if you’re suspicious, if there is concurrent differentiation happening in the peripheral blood, if the white count is rising, if you don’t have another reason, go ahead and treat because guess what? Differentiation syndrome can get out of hand really quickly, and people can get into trouble, particularly with respiratory decompensation. Many end up in the ICU, and we’ve had a few deaths with differentiation syndrome over the years.

What is the best way to manage differentiation syndrome with IDH inhibitors? Steroids. This is not necessarily the case with every agent that differentiates, but in this particular situation steroids is the major course. Now, do you need to stop drug? Usually not with IDH inhibitors. Usually you can treat with steroids, you can use hydroxyurea to bring down the white count or the platelet count if you need to, you can manage TLS and DIC, which can sometimes accompany differentiation syndrome. But typically steroids, if it is IDH DS, should do the trick. The management of other DS, maybe with menin inhibitors and others, may require study drug hold, but I’m sure we can talk about that later.

Alright. Soon after the introduction of IDH inhibitors we found out that they don’t work forever. So in patients who had IDH2 mutations in particular, treated with enasidenib, some after initial response developed therapeutic resistance, and then folks all around the country, including, I think, Eytan, went looking for potential causes, and resistance was caused by various different things. One was second site IDH mutations. The other was development of the other, so sometimes people would have an IDH2-mutated AML, get treated and pick up an IDH1 mutation, versus isoform switching, which is something that was seen.

Then here’s some data for ivosidenib, and this also was seen with enasidenib, where you actually don’t have an effect on IDH per se, but you develop additional RTK mutations over the course of therapy. So you’re going along doing well with your IDH inhibitor and boom! A RAS mutation. Boom! A FLT3 mutation. And then you lose your response. Or, as shown on the right side of this panel, again, you can pick up second site IDH mutations or a separate IDH2 clone in a patient with an initial IDH1 mutation. So these at least have been the discovered mechanisms of resistance.

Alright. The newest kid on the block when it comes to IDH inhibitors is olutasidenib. I had a couple of strokes when I first tried to say this name, but now I’m doing pretty well. Or oluta.

So this data was mainly led by a lot of folks: Jorge Cortes, Stephene de Botton, Justin Watts. A lot of these folks have been involved with the development of this very promising IDH1 inhibitor. Here we have on the left the overall responses, CR/CRi, which I think approximate what you would see earlier. One of the interesting datapoints that’s emerged from the similar Phase I/Phase II studies that were done with this study, as in prior studies, was the median duration of response, which is kind of crazy, 26 months. It’s certainly very impressive. It is hard to know whether it’s the drug itself that caused this impressive duration of response compared to what we had seen with ivosidenib or perhaps a different patient population now several years in, but this was an impressive finding.

Again, similar data on the median overall survival of around 11.6 months, and then the split depending on whether — what type of response you had in terms of the overall survival. Those who responded obviously did better.

This, I believe, was recently incorporated into the slide set that just kind of looks at the 5-year results, an update of the most recent presentation again showing similar data that if you had a CR or CRh you did the best, but other responders less than that also did better than nonresponders. I don’t know what that is. Oh, this is the presentation here at —

Okay. Is there a difference between olutasidenib and ivosidenib? I don’t know, personally, and I’d love to get the panel’s view on this mainly because of the ravages of time make it difficult to look at patient populations from a decade ago and now. But to me this composite CR rate and the overall response rate are relatively similar, but the duration of response that has been reported is obviously markedly different, as you can see in that table.

Watts, Justin, my good friend from Miami, and colleagues published a paper to postulate why this may be, and one of the reasons was potentially that the ivosidenib may bind the allosteric pocket longer in a more constant state, causing interaction with wild-type IDH1, whereas the smaller olutasidenib molecule has weaker affinity for wild-type IDH1, making it a more selective inhibitor of the mutant IDH1. So maybe that’s why. It binds to mutant IDH1 a little bit stronger, or strongly. It also tends to bind some of the resistance IDH1 mutations that have emerged as a response as a result of treatment with ivosidenib. That might be also potentially interesting to look at.

Combinations. Okay. So a lot of us did work on this, where we combined enasidenib and ivosidenib to intensive chemotherapy. All I would say is that these combinations were well tolerated. In general, IDH inhibitors do well in terms of marrow suppression, they’re not particularly marrow suppressive, so that is promising when you try and combine it with up-front therapy. And the 30-day and 60-day mortality was as you would expect with up-front intensive therapy, and the responses were generally promising, except for perhaps secondary AML and ivosidenib. They weren’t as promising.

How about HMA? Now, this is kind of where the interesting data starts to emerge. This is an important point. Enasidenib and HMA were combined and studied initially showing fairly impressive responses, where you actually look at overall response in those who received IDH2 inhibitor enasidenib and azacitidine versus azacitidine only. I mean, look at this data. Overall response 74 versus 36%. Complete remission 54% versus 12%.

But then when you do a non-placebo-controlled study, and then you compare these, you have some problems. So the EFS is there. It’s not significant, but there is a change. The overall survival completely overlapping. And a lot of these patients who I’m sure came off HMA and failed, guess what, went on and got enasidenib. So this was 1 challenge with this particular study.

But nevertheless — but however, ivosidenib and HMA therapy, this is the original Phase I data — of Phase II data looking promising, CR/CRh rate overall.

They did the AGILE Phase III placebo-controlled HMA versus HMA/ivo, and this is a chasm of a difference in terms of overall survival. Some of these patients — I mean, I think it was 29 months in the most latest update, median overall survival in IDH1-mutated patients. So do I think that there is a substantial difference between enasidenib and ivosidenib? No. I think there’s maybe a little bit of a difference, but not this difference, what I just showed — I’m sorry. I guess I’m just going forward. I was trying to go backward, but I’m going forward.

Okay. It’s okay. We’re now moving on to triplets, and given that HMA and ivosidenib in general are so well tolerated, with a good degree of overall survival advantage, that is now the standard of care for this small group of patients that have IDH1-mutated. How about adding something since it’s so well tolerated? Well, maybe we can do HMA/ven and ivosidenib, HMA/ven and enasidenib. And in fact, these studies are being done, mainly at MD Anderson, and take a look at these survival curves. They’re quite promising.

So when folks talk to me and say you’re doing HMA/ven versus induction I think the next path — one of the advantages, perhaps, of doing a study like this is that maybe it’s an entré to other studies, the comparison of triplets versus intensive chemotherapy. Because I think, particularly in younger patients, you have a standing to move forward and be optimistic that it may show an advantage.

This just shows, again from the same paper, that in general the median time to neutrophil recovery and platelet recovery was promising. The range of toxicities was as expected. There was not much. In terms of ivosidenib, I don’t know if I’ve mentioned that the main thing you worry about is QT prolongation. The main thing with enasidenib you worry about is bilirubin elevation, but it’s a benign indirect bilirubin elevation typically. And with olutasidenib it’s liver injury in a sizeable minority of patients.

How about maintenance? Well, we’ve run a few of these post-transplant maintenance studies at Mass General with collaboration also with other sites looking at both ivosidenib and enasidenib following transplant. These were Phase I studies. We’re currently running a Phase II study of ivosidenib. But it’s quite promising in this limited number of about 20 patients. The overall survival, progression-free survival was quite, quite promising among transplanted patients.

This is data that I think was presented relatively recently at EHA by Andrew Wei looking at oluta as a maintenance strategy not after transplant but after any type of therapy that would get patients into a CR or a CRi, if I remember correctly, and it shows promising RFS and OS over time, albeit in a limited number of patients.

So conclusions. IDH inhibitors are now key members of AML therapeutics, thankfully, both in the relapsed/refractory and more impressively in the up-front setting in the case of ivosidenib. Ivosidenib is approved in combination with HMA for up-front AML. Enos, ivo and oluta are approved for relapsed/refractory AML as monotherapy. Triplet therapies are now emerging and showing promise. We’ll see how far they will go. I hope far. And IDH inhibitors may have a role in the maintenance setting, but trials are ongoing. I appreciate your time. Thank you.

DR LOVE: So before we get into the survey, Eytan, at our 11:30 session on CLL, one of the oncologists is going to present a case of a patient who’s progressing on acalabrutinib who has a BTK resistance mutation who’s now responding to pirtobrutinib. Can you talk a little bit about your vision of what — why the resistance develops and also any thoughts you have about the mechanism of action and the translational data of oluta versus ivo, why you see the longer duration of response, and again whether or not it’s effective against resistance mechanisms?

DR STEIN: Yeah. So I think the data’s really interesting. I think it’s interesting to see how that duration of response is so much longer with oluta than with ivo, and it’s hard to ignore that. It may be a different patient population, but maybe not. And I like the hypothesis, although I think it has to be proven experimentally, that Justin Watts has laid out, that perhaps it does have this activity against these resistance mutations, and that’s what prevents the patient from relapsing relatively early. A lot more work to be done because I think it also gets into the question of the sequencing, and of course maybe you want to use your drug that hits a lot of resistance mutations earlier than the drug that has a liability with resistance mutations.

DR LOVE: There are 2 first-line pirtobrutinib studies being presented here, so we’ll see whether that comes in.

Alright. Let’s get into the survey. So Tara, we have an older patient, IDH1 or IDH2 mutation. A lot of questions from the audience about that. Most people are using HMA with IDH1 inhibitor, as well as IDH2. Can you talk a little bit about how you approach these patients?

Also, we were talking earlier about TLS. Do you see more TLS with IDH patients? Any difference in terms of how you monitor them for TLS?

DR LIN: So I think when you’re taking this patient at the time of diagnosis, when you’re considering adding an IDH1 inhibitor and using that instead of HMA/venetoclax, not everybody can get that IDH result back very quickly. So I think in a lot of cases you are starting an HMA/venetoclax up front because you don’t have the data. When you have someone who is very stable, who is outpatient while you’re waiting for all of this diagnostic information to come back, I think that’s when making a choice for aza/ivo becomes practical, it becomes feasible. There are also issues with getting, right, the oral approved through insurance here in the US, and so I think that also plays a part sometimes in your choice of drug up front.

In terms of TLS, I don’t know that I see more TLS in these patients versus in others, but I think you need to always have a high index of suspicion. I think sometimes, especially with our older patients, they will come in and present with spontaneous TLS with an elevated uric acid, even in the setting of variable white counts. And so TLS prophylaxis with allopurinol. You can give them IV fluids when they’re in the clinic, but certainly following those labs closely is important.

DR LOVE: So Sasha, what about the patient who gets 7 + 3 with an IDH mutation, we’re saying IDH1? Any comments in terms of how you manage? You say aza/ven. Other people say ivo/aza. Any comments?

DR PERL: Well, I think anybody who progresses after transplant, this is along the lines that Tara had mentioned before. The big question is are you going to treat them with full curative treatment approach in the post-transplant relapsed setting where long-term survival is unfortunately rare. In a patient where we’re going to be maximally aggressive we would include DLI, we would include whatever’s most likely to get them back into a remission, but also something that's going to be tolerable down the road in case that isn’t going to be satisfactory for long-term disease control. So aza/ven in this setting for this patient is what I recommended, mostly because it’s hard to give intensive chemotherapy after transplant in post-transplant relapses, and this has been pretty tolerable for patients and can be done outpatient. We want to avoid inpatient therapy.

DR LOVE: Harry, anything you want to add to that? And also, can you comment on this other scenario of an 80-year-old patient who gets aza/ven? Most people say oluta in this situation. Also in the prior situation, although most people said aza/ven. Dr Cortes, who I had the pleasure of spending a couple hours with recently, that was really a thrill, says oluta for a patient like that.

DR ERBA: Yeah. So going back to that last patient, and I agree exactly with what Sasha said, these are post-transplant patients, and their ability for their marrow to tolerate intensive chemotherapy is poor. I would say the same is true for venetoclax-based regimens. And so that is why I actually chose ivosidenib with azacitidine. You get maturation, differentiation of those blasts, you get more rapid count recovery in that first cycle, which might be very important. So that’s the other way of thinking about those options for patients.

I would start with ivosidenib there. And based on the AGILE trial data they may have a very long response to that combination by targeting the IDH1 mutation. I’m thinking about the MDS data showing very long responses with single-agent ivosidenib in MDS, and is there something about targeting IDH1 in that maybe the preleukemic clone that keeps the disease at bay?

In this case the patient has a few options; AML and an IDH1 mutation who experience disease progression after aza/ven. I wrote here olutasidenib, and I can’t remember why I wrote olutasidenib, to tell you the truth. Did I do that? Okay. So I was doing this during the Michigan/Ohio State game, so clearly I was —

DR LOVE: Does that mean that’s not what you would do?

DR PERL: Ohio State was up, and that starts with an O.

DR ERBA: Yeah. I’m not sure at what point. It was bad at the end.

So actually, in thinking about this question here, I would use either oluta or ivo. I said oluta because there is a limited amount of data with oluta showing pretty significant responses and durability responses in specifically those patients who failed venetoclax. So that’s why. I had to pick one, and I picked olutasidenib. I picked Michigan, and that was wrong.

DR LOVE: Yeah, well, last year.

So what about the issue of using an IDH inhibitor after a previous one? Amir, a typical situation would be maybe somebody’s who gotten ivo. You say you might use oluta. In what situations have you seen patients benefit?

DR FATHI: A handful of patients. We don’t have a wealth of IDH inhibitors. We have 3 as far as I know. Well, we do also have the one that we studied that — vora. We have vorasidenib. That’s mainly being used in glial malignancies.

So yeah. I think if somebody has progressed beyond ivosidenib the next choice has to be olutasidenib. And just to be clear, I think olutasidenib is a perfectly good agent, IDH1 inhibitor. And if it is true that it causes such a durable response it would seem that it’s one that should be favored, but I personally would feel more comfortable in terms of a head-to-head comparison, I think, as Eytan was saying, to make sure that we are comparing apples and apples. One part that doesn’t make a lot of sense to me is say the fairly impressive durability of response but overall survival just being up by about 2 months compared to ivosidenib. I think we need to flesh out some of that data.

Current and Future Role of Menin Inhibitors in the Treatment of AML — Dr Stein

DR LOVE: Alright. So we’re going to finish up with the newest and one of the most exciting, I think, advances. Interesting, really interesting agents, the menin inhibitors. Eytan.

DR STEIN: Alright. Thank you, Dr Love. Thank you for inviting me to speak. When I stand up it’s not like when Dr Fathi stands up. I’m not that much different sitting down and standing up when it comes to height.

Okay. We’re going to move on. What are menin inhibitors? What is important about menin inhibitors? Menin is an adaptor protein that binds to this MLL complex, and that binding to the MLL complex leads to the upregulation of HOX genes, which freezes those blasts in a dedifferentiated state, which is acute myeloid leukemia. This seems to be important in KMT2A-rearranged acute myeloid leukemia and acute lymphoblastic leukemia, as someone up here said before, I think Amir, that happens primarily in therapy-related acute myeloid leukemia, and in the most common mutation we see in AML, MPN1-mutant acute myeloid leukemia.

So if you can develop a drug, depicted in the red ball, that blocks menin from binding to MLL, the MLL complex, you get a suppression of the HOX gene upregulation, which leads to myeloid differentiation.

Now, there are a variety of different menin inhibitors that are being developed. You can see structurally they fall into really 2 families, with this 1 drug, which is no longer being developed, icovamenib, that falls into a third family.

What does this mean when it comes to treatment? Well, the first drug that has been approved for relapsed and refractory acute myeloid and acute lymphoblastic leukemia with a KMT2A rearrangement is revumenib. Revumenib was studied in a Phase I/II study. The Phase I study was dose escalation. Phase II had 2 arms, KMT2A-rearranged and MPN1-mutant. The primary endpoint of the study was the CR and CRh rate, and the key secondary endpoints you can see on the slide.

Here are the disease characteristics of the patients treated with revumenib for KMT2A relapsed and refractory acute myeloid leukemia. Let me highlight a few things. 104 patients that were treated and are in the FDA label. Most of those patients had acute myeloid leukemia, 83%. 15% of the patients had acute lymphoblastic leukemia. Highly relapsed and refractory population with almost 50% of the patients having previously received an allogeneic stem cell transplant.

The rate of CR and CRh, which is the approvable endpoint in these studies of targeted therapies by the Food and Drug Administration, was 21.2%, with a median duration of CR and CRh of 6.4 months. We’re going to see in one of the subsequent slides the overall response rate is significantly higher, so the blast clearance rate is high, but the rate of CR and CRh just breaks that 20% threshold.

What about for patients with MPN1-mutant relapsed and refractory acute myeloid leukemia? So in I believe October of 2025, or November, it was just a month ago or so, revumenib got an additional approval for relapsed and refractory MPN1-mutant acute myeloid leukemia based on 65 patients treated in that Phase II study. Again, the CR/CRh rate here is around 20%, 23.1%, with a median duration of CR and CRh of less than 6 months of 4.5 months.

There are going to be updates to this data that are going to be presented at this meeting. Please go to that session. It’s going to be a great one. And the FDA then approved a second menin inhibitor for relapsed and refractory MPN1-mutant acute myeloid leukemia based on work done by a number of the investigators that are sitting on this podium, including Dr Fathi and Dr Erba. And this approval, you can see, came through on November 13, 2025.

What was it based on? It was based on 112 patients treated with ziftomenib at 600 mg once a day. You can see the breakdown in the patients here. Most of these patients, as most patients with MPN1 mutations, have de novo acute myeloid leukemia, 85%. 15% of the patients had secondary acute myeloid leukemia. And again, what you’re seeing here is rates of CR and CRh, that’s the top line, that are around 20%, in this case 21.4%, with a median duration of CR and CRh of 5 months, a little bit longer than what we saw with revumenib. Whether that means anything clinically or whether that’s statistically significant, that difference, I think, remains to be seen.

Here is the money slide. Focus on this slide. This shows all of the response rates up to date, except for at this meeting because I didn’t update it for this meeting yet, of all the different menin inhibitors. And you can see a couple of things I want to point out. Very high rates of blast clearance, both for KMT2A and MPN1, in the 60% range over here. But the rate of CR and CRh is much more modest, in the range of 20-30%.

The other thing that it is very important to point out is the rate of MRD negativity. We only have MRD-negative CR/CRh rates for some of the studies. In this study with revumenib for the KMT2A patients the MRD negativity rate was 70%.

In this study of ziftomenib for the MPN1-mutant patients, published in I think the Lancet or Lancet Oncology or maybe JCO, it was 56 — he’s whispering JCO. Thank you, Amir — it was 56%. So that’s really great, isn’t it? We’ve got great rates of MRD-negative complete remission, but yet the duration of remission is quite short. So why, if you’re getting MRD negativity, is your duration of remission so short? It suggests that perhaps MRD negativity is not what we should be looking at in these studies because these patients clearly weren’t MRD-negative if they’re relapsing so quickly after going into a remission.

Why do some patients with menin inhibitors, or specifically with revumenib, relapse? Well, it’s been shown that there are a number of resistance mutations that develop, just like the resistance mutations we see with targeted therapies in all types of cancer. And in fact, with revumenib when this analysis was done on the original patients, a full 40% of the patients developed a resistance mutation at some point when they were on study. We don’t know exactly the rate of resistance mutations with other menin inhibitors. We don’t know yet, getting to Dr Love’s point of acalabrutinib and pirtobrutinib, if you can be treating with 1 menin inhibitor, if you develop a resistance mutation, what’s the next menin inhibitor that we should be using. These are things that we’re going to need to work out in the relapsed and refractory setting.

When you have a good drug for relapsed and refractory leukemia you don’t want to keep it in relapsed and refractory leukemia. You want to move it up front with standard of care agents. This is being done for newly diagnosed acute myeloid leukemia. It’s also being combined with standard of care for relapsed and refractory acute myeloid leukemia. For the relapsed and refractory patients, up here, you can see again very high overall response rates, more modest rates of CR and CRh. For newly diagnosed acute myeloid leukemia we’re getting both very high rates of overall response, in the 90% range, when the menin inhibitors are combined with 7 + 3 or aza/ven, and also very high rates of CR and CRh, in the range of 80%.

Here’s a study that was presented by Josh Zeidner looking at one of those combinations, aza/ven with revumenib. Aza/ven given for 1-3 cycles in combination with revumenib. Patients who achieve a remission can continue on the study.

Very high rates of remission. You can see here CR rates of 70%, overall response rates of 88.4%. This, I would say, is quite impressive data.

Now I’m going to show you the thing that concerns me. These are the swimmers’ plots of the patients, published in the JCO, that were treated on that study. Everyone in blue is an MPN1-mutant patient, okay? These patients do very well with aza/ven. You will see that with those blue bars most of those patients go into remission, that’s the yellow circle, and then you will see that many of the patients have a red star.

Now you might be saying red stars? When I was in fourth grade and got a red star on my spelling test that meant I got a hundred. That’s great, but that is not the case here. Red stars in this swimmers’ plot means the patients have died. So why is a patient who is going into remission so early in their therapy with an MPN1 mutation subsequently dying while on therapy?

So this gets to the point of something important for the community oncologist that you can’t just layer agents on top of agents at their current FDA-approved doses. That’s what happened at the beginning of this study. Menin inhibitor for 28 days, ven for 28 days, aza for 7 days, and patients ended up having complications related to myelosuppression that caused them to die even after going into remission. And the iterative process of this is now this study has been amended so that with the updated data, what we’re hoping with these shorter durations of menin inhibitors, shorter durations of venetoclax, that we will get less deaths of the patients who are on study.

Okay. We’re just going to swing forward. This is also going to be updated, some combination data. This is going to be presented by Ghayas Issa in this meeting.

We have very, very good data looking at the KOMET-007 study, a Phase I study looking at ziftomenib plus 7 + 3 in MPN1 or KMT2A-rearranged newly diagnosed acute myeloid leukemia. Very high rates of complete remission, as I mentioned before. Similarly, we see similar things with aza/ven.

Okay. And ziftomenib is going to be presented also in an oral abstract session, which you can see on this slide — on these 2 slides.

What are the issues that we see with menin inhibitors? So we see a little bit of myelosuppression. We see differentiation syndrome. With revumenib we see QT prolongation. With ziftomenib we see pruritus.

I want to talk a little bit about differentiation syndrome. In the interest of time, I’m actually going to skip through a little bit of this but only to point out that the differentiation syndrome seen with different agents may not be the same. That is the differentiation syndrome with menin inhibitors may be qualitatively different than the differentiation syndrome seen with ATRA and arsenic or with IDH inhibitors or even with FLT3 inhibitors.

We treat differentiation syndrome traditionally by giving steroids, 10 mg of dexamethasone twice a day. But I wanted to present a case here that I think is instructive and that many of us have seen about the kind of differentiation syndrome we see in patients on menin inhibitors. This was a patient of ours with Burkitt’s lymphoma who had a therapy-related acute myeloid leukemia who ended up relapsing after standard therapy, gets revumenib. She got admitted with fevers and hypoxia with what was called menin differentiation syndrome, improves and then subsequently develops steroid-refractory fevers, hypoxemia and hypotension.

And I want to show you what the inflammatory markers are like in this patient. This patient has a ferritin level that goes up to 300,000. They have a CXCL9 that goes up to 100,000. You can see their soluble IL-2 goes up quite high, as does their interferon gamma.

And when we did some — or when our pathologist did some fancy — took some fancy pictures and did some fancy staining what we think we are seeing here is something that is like hemophagocytosis in patients who are getting menin inhibitors and develop this rip-roaring differentiation syndrome that I would argue is uncontrollable when it gets to this point.

This patient received tocilizumab, the patient received anakinra, the patient received emapalumab, the extraordinarily expensive interferon gamma drug, and we could not save this patient.

So I would argue that when you’re giving a menin inhibitor to a patient it is really important to think about differentiation syndrome. I think we as a community need to think about whether there are certain patients who should be getting steroid prophylaxis when they’re on a menin inhibitor, especially those patients who are at high risk of differentiation syndrome.

My concluding thoughts are the same as my prior thoughts, so in the interest of time I’m going to throw it back to Dr Love.

DR LOVE: Thanks so much, Eytan. I really love that slide doing the indirect comparison, although most people when they put those slides up go I’m not supposed to do this, but, but very instructive.

The last part of your talk, where you got into differentiation syndrome, can you elaborate a little bit more in terms of your vision of what differentiation syndrome is? Is it qualitatively different in terms of what you see with IDH versus menin? Is it like the reversal of the malignant process? And what’s the difference in the types of differentiation syndromes pathophysiologically?

DR STEIN: Yeah. So that’s a long answer to a good question. So the differentiation syndrome, what I think of is — I mean when I describe this to patients I describe it as you are trying to tell an immature cell, in this case a toddler, to change. Toddlers don’t like changing.

DR PERL: Impossible.

DR STEIN: They yell and scream, and they throw up their hands and have a fit. So this is what these leukemia cells are doing. They’re releasing cytokines because they don’t want to turn into normal, healthy neutrophils, or maybe they do want to turn into normal, healthy neutrophils. And all that cytokine release is essentially causing what I would say is a spectrum of what we call differentiation syndrome. It’s over here and can be treated with steroids, going to something in the middle, which is a cytokine release syndrome, which I think is very similar to differentiation syndrome, going all the way to the right-hand side in my schema, of something like HOH, which is really uncontrollable. And I think that’s what’s going on pathophysiologically. There are a variety of different things that are going on that we can’t get into right now.

DR LOVE: So Tara, let’s get into some clinical practical questions. A 60-year-old patient with a KMT2A rearrangement, gets disease progression after 7 + 3 plus transplant. You say aza/ven and revu. Other people say just revu. Any comments?

DR LIN: I think the question for all of these post-transplant relapse cases that we’ve done is are you still going for cure or is this with more of a palliative let's get you some quality time. And I think a single agent would be appropriate if you’re thinking about more quality time with minimal toxicity. If the person is young and fit and you’re trying to get them to a second transplant I think there’s enough safety data that I would consider using the combination.

DR LOVE: So Sasha, can you talk a little bit about — now we’re talking an 80-year-old patient with KMT2A. Everybody says revu. Any comments on that? And then on a 60-year-old with an MPN1 mutation?

DR PERL: I think, again, for the 80-year-old your goal is largely quality of life metrics. It’s an approved drug. It’s a very well-tolerated drug as long as you don’t wind up with cardiac issues from QT prolongation, you can see cytopenias from it, but otherwise it’s a pretty easy drug to tolerate. And that’s often my goal in these patients.

In the 60-year-old patient with an MPN1 mutation, again, in the post-transplant setting one of the reasons to give the combination, I think, is just because one point, what Eytan said in terms of durability of remission, which is short with single agent. And the second is actually we tend to see less risk of differentiation syndrome in these combinations if we can cytoreduce, and the aza/ven’s quite effective at doing that without adding a lot of cytotoxicity — or toxicity outside of just myelosuppression.

DR LOVE: What about the issue here of zifto versus Revlimid in a patient with MPN1?

DR PERL: Yeah. Kind of a tossup here. We don’t have comparisons to know which one’s better. I think they’re both well-tolerated drugs in our hands. We do have experience with both of these drugs in the combination setting. I’ve had personally more experience with revumenib than zifto as a single agent.

DR LOVE: So Amir, any comments? Again, we were talking about indirect comparison, but the way you see it in terms of the available menin inhibitors, in terms of efficacy, tolerability. Most people say at this point it looks pretty similar. A couple of people say enzo. Any thoughts?

DR FATHI: Yeah. I think it all depends on who has experience with what drug at their site and comparing that. But unfortunately there is now probably 2 dozen different menin inhibitors so we’re all a little bit biased regarding what we’re kind of comparing to because we may have experience with 1 or 2. Acting as the great prophet Solomon, from a distance I would say they are all of relatively similar activity and tolerability. It's not something that gets in the way of choosing one versus the other. That’s been my experience.

DR LOVE: Any comments, Harry? And also it looks like the people think that the risk for differentiation syndrome seems very similar. Any comments, Harry? And also, have you seen HLH syndrome, Harry?

DR ERBA: I’ll start with this one. I haven’t seen HLH-like syndrome. And the next question is what would I do? I would call Eytan. Actually, that’s what I said before, but knowing the outcome of his patient I’m not sure I’ll call. I’ll call somebody who knows something about HLH. I thought it was really interesting what I saw on the slide that he didn’t say, that the histiocytes were derived from the leukemia cells.

UNKNOWN: Yeah. That’s great.

DR ERBA: Maybe we can talk about that later, more of a nerdy kind of thing to talk about.

In terms of — I’m sorry, go back to the other question that you asked me. If you can go back to the prior. Risk. I have used enzo and zifto and revu, and you have to remember these drugs aren’t all the same. I’m always baffled by how you have these huge molecules and somehow they get through the gut, into the body, into the cell and do what they’re supposed to do. There must be something that is necessary for these molecules. And there are things I don’t understand about medicinal chemistry that tells the chemist how to build something that has the properties to get into a cell.

And when you look at that data the enzo molecule seems to have a lot of the properties that allow it to get into a cell. Okay, but how about on the clinical side? You can’t compare between Phase I studies, you just can’t, but we do. And you asked the question. And when I look at the overall response rate with enzo versus zifto and revu the overall response rate looks higher, and in my experience the tolerability is better. But until we have head-to-head comparisons you won’t really know.

DR LOVE: Final comment from Sasha.

DR PERL: Yeah. The only thing I would throw out there is just we think of these as menin inhibitors for MPN1 or KMT2A rearrangement. The zifto drug development for KMT2A-rearranged patients was halted because of toxicity, so just recognizing the approval is for MPN1 specifically as a single agent. They’re going through further development in combinations but not as single agent, so that’s an important distinction because we don’t know what the right safe dose is. Maybe there is one, but that hasn’t been established by trials.

DR LOVE: Great point. So faculty, thank you so much. Audience, thank you for attending. Come on back at 11:30, we’ll talk about CLL.