Acute Myeloid Leukemia Podcast

Adverse-Risk Acute Myeloid Leukemia

Gail J. Roboz, MD; Alexander Perl, MD


June 16, 2022

This transcript has been edited for clarity.

Gail J. Roboz, MD: Hi, everyone. My name is Dr Gail Roboz, I'm a professor of medicine. I run the leukemia program at Weill Cornell Medicine at the NewYork-Presbyterian Hospital in New York City. It is my great pleasure to work today with my friend and colleague, Dr Alexander Perl, who is associate professor of medicine in the leukemia program from the University of Pennsylvania. Sasha, thank you so much for joining today.

Alexander Perl, MD: My pleasure, Gail. Thanks for inviting me to do this.

Roboz: What we're going to run through today is what to do with adverse risk acute myeloid leukemia (AML). We know that AML is not only one disease; it's many. It's divided by biological subtypes, by mutational profiles, by performance status of the patients; there's lots of different ways to slice and dice AML. But today, the focus is on adverse risk.

What I'll open with is this: Sasha, how does a patient get into that category? What needs to be tested for, and what are the results that define somebody as adverse risk?

Perl: I think there's a lot of different ways to define someone's risk. Obviously, it depends on patient factors. It depends on genetic factors of the leukemia. It sort of depends on how you got to leukemia in the first place. We can first look also at issues totally unrelated to the leukemia, such as is this an older patient; a younger patient; someone with a lot of comorbidity; even if they had a leukemia that itself that you would predict, on the basis of the genetics of the leukemia, that it should respond to therapy. If you have a patient with a favorable genotype who has a lot of comorbidity, you're not going to be able to consider them easily cured with standard therapy because it's going to be very hard to treat them. And whether can they withstand the rigors of therapy in and of itself is a question that relates to the risk of the overall treatment plan, and you're not going to be able to know that early on in therapy.

Then the last thing is, how well does the patient actually respond? Because if you take patients who have a favorable-risk genotype who don't follow the cookbook and their leukemias failed to respond the way you want them to, suddenly their risk changes. And it's an evolving practice. So you can't say on the day that you diagnose somebody, your risk is X and then just look into your crystal ball and say, this is going to go well or this isn't. But let me give you some generalities.

The older patients tend to do worse because it's harder to treat them. They've got more comorbidity. You can't always give them the intensity of therapy that you want. Or if you go to transplant, what might be the best donor or like a sibling in a younger patient is now as old as the patient, and then you have to choose from other options there. The presence of comorbidities is a consideration; you can't give intensive therapy to people with heart disease, lung disease, kidney disease, et cetera. Instead, you have to potentially peel back or adjust your standard approach in this setting, and that could compromise long-term outcomes.

But the thing I think people focus on a lot is, what are the changes in the leukemia itself that impact risk? There's a lot of classifications for these. The oldest ones just dealt with clinical measures, such white blood cell count and karyotype. We've gotten better at looking at genetics at a higher resolution of late, including gene mutations and gene fusions that have a strong impact on outcome.

Then the most recent development is, when you treat patients, how well do they do? Not just in terms of do they go into remission, but do they become minimal residual disease (MRD) negative, as a strong predictor of whether they will be able to stay in deep remission as you treat them effectively with the same therapy again and again using chemotherapy as the backbone? You can look at the European LeukemiaNet (ELN) classification that's well codified and validated in terms of saying, what are the gene mutations in karyotypes that go into the higher-risk diseases? Some of the features there include the presence of adverse karyotypes often associated with TP53 mutation — certain gene mutations, such as RUNX1, ASXL1, put you into this group, and a high allele burden of FLT3 ITD in the absence of an NPM1 mutation puts you in a high-risk group. But there's also gene fusions that can do this in and of themselves. An inversion of chromosome 3, I think of as the top of the pile of really bad-actor leukemias that make us lump somebody into a group that is very chemotherapy-refractory.

So we can look at the genetics of leukemia and say, this gets you into that unfortunate group where options are really hard and we don't have a great standard approach that we can say, if you do this, you'll get a good outcome.

Roboz: I think it's important that you brought that up: What are the things that you're going to look for on day 1? It's not only white blood cell count and chromosomes — that was a long time ago. Now we're doing mutational profiling standardly, but we all struggle with it taking 2 weeks to come back. So you're not knowing that information immediately.

I think it's important to get out there that fluorescence in situ hybridization (FISH) testing and polymerase chain reaction (PCR) can sometimes come back faster and at least give you some of the information that you brought up in your initial comments: Do I have 23 chromosome abnormalities that are coming back? FISH for some places comes back faster than karyotype; I may not know I have a TP53 mutation, but maybe I can see an inversion 3. Or maybe I can see five different abnormalities that are picked up on a FISH test.

These are things that, as you pointed out, are at least giving you a flicker from the beginning that this is falling into an ELN classification of adverse risk. I have multiple complex abnormalities. I have some reason to believe that the biology of this disease is poorly behaving.

Then I think it's worth it for everybody to know that the word "adverse risk" — yes, it's categorizable in a classification scheme, but it also depends on the point in therapy. Is the patient tolerating therapy? How are they doing? I have people all the time who will come in with, for example, a relapse of a core binding factor or an NPM1 mutated AML, and I'm being presented with a patient in third relapse as good risk. Well, unfortunately, no. There is nothing good about that anymore. Even if the initial classification was favorable by ELN, if you're on your third relapse, that's not good anymore. I think in the adverse-risk group, everyone is zeroed in right on your special friend FLT3, because that's something for which everybody goes, Hey, wait a minute, I've got a drug for that.

Talk to me about how it is that when you look at an ELN classification where you've got FLT3 and a couple of different places there, it's not what people think that it's just straight-up adverse risk. Can you walk us through FLT3 and how that classifies you, and what its co-mutations need to be in order to come up with that kind of initial biological risk?

Perl: FLT3, as you mentioned, is a common mutation in AML. It's actually one of the most commonly mutated genes in AML. And it's a kinase, so when it's activated by these mutations, it turns on signaling in the cell that contributes to leukemia growth, and we have drugs that can turn that off, similar to the manner in which we turn off BCR-ABL signaling in chronic myeloid leukemia (CML) or Ph-positive acute lymphoblastic leukemia (ALL). So this gives us an opportunity to treat the disease. And what you hinted at is that if you treat the disease with one of these drugs, you could potentially change its natural history.

It's important to recognize there's not one FLT3 mutation; there's actually two major subtypes. They have different prognostic effects. The FLT3 ITD are internal tandem duplications that historically have been associated with worse outcomes in terms of standard chemotherapy responses and not so much initial response, but the likelihood of relapse and then survival thereafter. So there are patients with FLT3 ITD–positive AML who are cured with frontline chemotherapy, but if they ever have relapse, they've been very hard to get into long-term remissions thereafter. So the risk classifications reflect that the patients who've had the greatest risk for relapse among those with an FLT3 mutation are those with an ITD burden with a high allele burden, however you want to measure that — whether it's by PCR, allele ratio or next-generation sequencing (NGS), varied allele fraction — and the absence of NPM1, which generally is a chemotherapy sensitivity marker, although in some contexts can be associated with resistant disease. We've come to recognize that there are a few things that modulate this risk, such as co-mutation of DNMT3A, which, when co-mutated with FLT3, gets rid of NPM1 as being a favorable marker. So that triple mutation emerges as a very strong predictor of poor chemotherapy outcomes. And I don't even look at the allele burden in those patients.

Roboz: I think that the allele burden issue really gives people a headache because they don't know where to find it on the report. It's expressed differently on different reports. And then there's the VAF — the variance allele frequency — is that the same as the allele burden? I have absolutely heard both in the academic centers and in the communities, people will just say, forget it. If it's an FLT3 ITD, I'm done; it's adverse risk, I move on and I don't even count the allele burden. I get it what you just said that in the context of a triple mutation, we don't do it.

But are they right? Can you just look at an ITD and move on and call it adverse risk? Or do you need to page through that 23-page report and try to figure out where the allele fraction is?

Perl: I can tell you what I do, which is that I fall more into that camp where I say the ITD is bad and I don't want to undertreat these patients. But I am impressed by the emerging data that say when you add frontline FLT3 inhibitors to intensive chemotherapy, the risk actually does go down and there is a less clear role for allo transplant in first remission for patients who are FLT3 ITD–positive, who have a good response to 7+3 and midostaurin as the regimen that's been largely tested and fleshed out. But I don't think we have enough data here to say that FLT3 is good and you should hope to be FLT3 ITD–positive at a low burden. I think there's pitfalls in how this is tested. I think there's pitfalls in the universality of the reports. I agree with you that there's not a harmonized agreement upon how we measure this, how we report this, or how to use this in our therapy.

Because of that, there's difficulty in creating a uniform standard; it's not like the IS [International Scale Real-time Quantitative Polymerase Chain Reaction (IS RQ-PCR)] in CML therapy, where every lab gives you the same report and you can look at it and go from lab to lab and you get the same answer and you know what to do. We don't have that for FLT3 ITD. And so I do largely look at it in a binary way: It's present or it's absent, and I treat based on its presence.

Roboz: Are you serially following FLT3? Because I am part of the ELN guidelines for MRD monitoring, and we have FLT3 as an unstable marker for MRD and not what we should be looking at. And yet I hear this all the time that you have to wait for the FLT3 to come back in remission before you do a transplant, or you have track the FLT3.

Perl: It's not standard to track FLT3 presently, but all of the studies that are looking at the impact of adding an FLT3 inhibitor (and which FLT3 inhibitor) to frontline therapy actually do track it, and they track it in a much more sensitive way than what's standard in therapy. So all of these studies that are doing — randomized comparisons of quizartinib vs placebo will be reported soon (the QuANTUM-First study), and following that, there are comparisons of crenolanib or gilteritinib vs midostaurin that are active control designs, and those are ongoing now being collected. We don't have any data from these trials where they've looked at those changes in burden to say what you should do or what's an optimal response in terms of the cutoff; we have limited data in this area. But what we do know from those data are that there is some role of the persistence of an FLT3 ITD at some cutoff in terms of predicting relapse rate.

A study done in Australia that Andrew Wei led comparing sorafenib vs placebo added to 7+3 showed that patients who were treated with sorafenib were less likely to have detectable FLT3 ITD when they went into remission. And those patients had a lower relapse rate, so it may be an important predictor thereafter. Similarly, when we've looked at the outcomes of transplant, patients who have detectable FLT3 ITD on almost any measure, including very sensitive assays, have a much higher mortality rate. The data are limited, but something that suggests that it couldn't be something that we would want to track yet again that the optimal standard of what the cutoff should be is really not agreed upon. I check it in my own practice, mostly by PCR, just because it's quick and dirty and easy to do. But I don't like to see persistence of leukemia-associated mutations that we think are predictors of relapse persisting after induction or after two intensive cycles of chemotherapy.

Roboz: Do we need to be doing anything right this minute about changing what is added to intensive chemotherapy, and when might we know a next step there?

Perl: It's a great question, and one I don't know the answer to because the studies are ongoing. I think the simple answer to use an FLT3 inhibitor vs don't use an FLT3 inhibitor that's been answered. You should use an FLT3 inhibitor. And the data are clearly positive for midostaurin, for either FLT3-TKD or FLT3-ITD.

Quizartinib is a more potent FLT3 inhibitor and one with greater single-agent activity than midostaurin. It's not yet approved in the United States. It's only approved in Japan. It is active in the relapsed setting, but the data were not convincing enough to allow the US Food and Drug Administration (FDA) to approve it in that setting, ss you're aware. But for FLT3 ITD–positive patients, should we use quizartinib over anything else? We're going to have to see the data from QuANTUM-First to say. And then all the studies with the other drugs are forthcoming. They're currently enrolling. I think the right thing to do is to enroll patients on these studies so that we can answer this.

Roboz: So people are talking doublets, triplets, quadruplets. Let's break down. What does that mean in this context? A triplet therapy, the addition of midostaurin to 7+3 — there are two drugs there, technically three drugs, that are being used as intensive induction for patients with FLT3-mutated disease. And there may or may not be a swap of another agent that's better than midostaurin down the road. But we also know that many patients with AML are not going to be getting intensive chemotherapy, and we know that the addition of venetoclax to hypomethylating agents (HMAs) has really transformed the standard of care for older patients. The remission rates are better; the long-term outcomes are better. Things are better with the addition of venetoclax to HMAs for older patients. But then what about a triplet there?

We don't have any of this approved. This is all based on largely single-center data coming out of MD Anderson. But for older patients, should we be getting on track to add a triplet — something like gilteritinib or an FLT3 inhibitor — to that new backbone?

And what about the fact that there are data suggesting that if you add gilteritinib, for example, to just an HMA alone will that didn't look too convincing? What do we do for the patient who isn't getting intensive chemotherapy but who is older and is going to get an HMA-based upfront induction?

Perl: I am treating my patients with newly diagnosed disease who are not good candidates for intensive chemotherapy, by whatever measure, but have an FLT3 mutation, with frontline venetoclax and azacytidine. The data that that works have been published now. Marina Konopleva's paper just came out in clinical cancer research, and that shows, pretty convincingly, that it's an active regimen in this patient population.

The big concern is particularly for the FLT3-ITD–positive patients who have relapse. It's the same story as you see with chemotherapy. So the question is not what you should use first and what's active there. Rather, it’s how do you keep people in remission for the long run? That's where we're really going to have to figure out the question.

What we don't know is whether we should then use azacytidine and follow that with something that includes an FLT3 inhibitor, or wait till progression or measure MRD or really put all our eggs in one basket and give them all together. You're not going to know the answer to that without a trial.

The other concern that I have is, this is a population whom we've just said are patients who are not good candidates for a more intensive therapy, but you're adding a drug that potentially causes a significant amount of myelosuppression to then an aza going to change that from being a low-intensity regimen, to being much more intensive. The concern is that we don't want to take a patient who is really not a great candidate for an intensive regimen and give them more than they can tolerate. That's why I think it shouldn't be standard that just because it's feasible in single institutions, you should incorporate it into your standard therapy for anybody with this diagnosis. Outside of the realm of what you feel comfortable giving — say, 7+3 or other intensive regimens — we really have to be careful here and develop this as a community, and hopefully through multicenter trials and preferably through randomized trials, figure out the best way forward.

Roboz: It's a very important point that just piling on drugs that, oh, there's an FLT3, I'm going to put on inhibitor and then I'm going to put on this and then I'm going to put on that. I think when you actually look at the pillbox of the patient who is 82, that you're sending home, potentially with an antifungal, an antibiotic, an antiviral, an FLT3 inhibitor, and venetoclax, it's unbelievable to sit down and sort out what the polypharmacy looks like for these patients. I do think that the data are interesting and important, and I think you're right that we should just be careful not to necessarily think that on day 1, we have to pile on every single drug that sounds good.

Let's go backward, though, to the beginning of the adverse risk again. Not everybody has FLT3, and we've got a lot of patients who have a complex karyotype or a monosomal karyotype or some jumble of numbers that includes things like 3s, 5s, and 7s. We just don't love chemotherapy for those patients. Let's just say you have a patient who can tolerate whatever you're going to throw at them, but you're looking at a whole bunch of karyotypic abnormalities; what are you going to do? How are you going to decide whether a patient who has an ugly-looking complex karyotype should get something like a standard regimen vs a CPX-351–based regimen vs a venetoclax-based regimen? How do we unpack that, since we know that those patients don't have great outcomes even when they go on to an allotransplant afterward? What do we do with those patients?

Perl: When all you have is a hammer, the world certainly just looks like a nail and that's what you do. But we have alternatives now, and it's an interesting story that's emerging from this and you bring up a lot of important examples.

In people who evolve to AML from prior myelodysplastic syndromes, and people who have therapy-related AML and may have had prior anthracycline and not be great candidates for retreatment or who've had mediastinal radiation, you worry about toxicity et cetera. We have alternatives to using 7+3-based therapy, or CPX, which is potentially kinder and gentler than 7+3 but is still intensive chemotherapy. What is the benefit of using a lower-intensity regimen that has high activity — say, using venetoclax and azacytidine in people with adverse risk — if you're going to follow that with, say, a transplant? Is there really a difference in terms of how you get them from point A to point B? Maybe there is an advantage to using a lower-intensity regimen because you're more likely to get them to transplant in one piece, and maybe you're more likely to get them to transplant without infections or other complications. So there are interesting emerging data.

Your group has added an interesting dataset presented at ASH this past year, as has our group and largely coming up with the same answer that as we look at intensive chemotherapy and low-intensity chemotherapy with venetoclax and azacytidine, the outcomes in certain patients are not all that different, although that could be important differences in subsets. That's where you really need a randomized trial to answer this. At our site, we’ve just opened a randomized comparison of venetoclax and azacytidine vs intensive chemotherapy. It allows for either 7+3 or CPX, whatever you think is most appropriate. This may not be a definitive study, but hopefully it will give us a lot of information about variations on the same thing that achieve the same goal, or whether there real differences. You really should pick and choose how you give this to patients who may have adverse risk factors.

Roboz: These are incredibly compelling roads, all leading to the same answer, which is that we still need to put our patients with AML onto clinical trials. I am much more excited to be an AML doctor now than I was 20 years ago, when there were very, very limited options. But I agree with you 100% that these patients should be put onto clinical trials.

As my closing up, I want to ask: For adverse-risk patients, is it still the standard of care and the goal that no matter how you get them into response, we are still trying to get those patients into allogeneic stem cell transplantation? Would you say that for patients who are eligible that still, no matter what you're selecting for upfront therapy, is that how we're guiding these patients?

Perl: I'm going to come back to a point that I made earlier, which is that we look at MRD as being a strong predictor overall of how patients are going to do. In adverse-risk disease, there are scant data to say that MRD is helpful to tell you what you should do next. They largely tell you something you already know, which is (A) this patient is more likely to be MRD-positive and (B) they're likely to do poorly with whatever you give them next. So if a patient becomes MRD-negative, great, that's fantastic. But I would not necessarily use that to say that they don't need a transplant. I don't think we have those data, certainly in individuals, to say that should change our therapy if they otherwise have adverse-risk disease. We can debate that for more favorable genotypes. For adverse-risk disease, I still think that allotransplant has the best track record for cure for virtually all ways that you get to a classification of adverse risk from the leukemia genetics. I do think that if you look at the patient definitions — that is, somebody who might not tolerate intensive therapy — that's where it gets a little bit tricky, and we have some data saying that reduced-intensity transplant is similar to allotransplant. But in patients eligible for both myeloablative transplant does better. As we get farther and farther away from a good candidate for myeloablative transplant, I think those data become less compelling.

The bottom line is that I still think allotransplant is a great therapy and is more likely to cure than chemotherapy for the vast majority of patients who have adverse-risk disease and who are candidates for that approach.

Roboz: On that note, I'd like to thank you so much for participating. It's been a great discussion. I'll call you as soon as we're done with this to go over some more cases. And in the meantime, thanks to everybody for listening.


High Prognostic Impact of Flow Cytometric Minimal Residual Disease Detection in Acute Myeloid Leukemia: Data From the HOVON/SAKK AML 42A Study

Diagnosis and Management of AML in Adults: 2017 ELN Recommendations From an International Expert Panel

The Presence of a FLT3 Internal Tandem Duplication in Patients With Acute Myeloid Leukemia (AML) Adds Important Prognostic Information to Cytogenetic Risk Group and Response to the First Cycle of Chemotherapy: Analysis of 854 Patients From the United Kingdom Medical Research Council AML 10 and 12 trials

Diagnostic and Therapeutic Pitfalls in NPM1-Mutated AML: Notes From the Field

Prognostic Impact of DNMT3A Mutation in Acute Myeloid Leukemia With Mutated NPM1

Midostaurin Plus Chemotherapy for Acute Myeloid Leukemia With a FLT3 Mutation

Quizartinib With Standard of Care Chemotherapy and as Continuation Therapy in Patients With Newly Diagnosed FLT3-ITD (+) Acute Myeloid Leukemia (AML) (QuANTUM-First)

A Review of FLT3 Inhibitors in Acute Myeloid Leukemia

Abstract 591: Results of a Phase 2, Randomized, Double-Blind Study of Sorafenib Vs Placebo in Combination With Intensive Chemotherapy in Previously Untreated Patients with FLT3-ITD Acute Myeloid Leukemia (ALLG AMLM16)

AC220 Is a Uniquely Potent and Selective Inhibitor of FLT3 for the Treatment of Acute Myeloid Leukemia (AML)

Daiichi Waits on FDA as AML Drug Quizartinib Approved in Japan

FDA Declines to Approve Quizartinib for FLT3-Mutated Acute Myeloid Leukemia

Safety and Preliminary Efficacy of Venetoclax With Decitabine or Azacitidine in Elderly Patients With Previously Untreated Acute Myeloid Leukaemia: A Non-randomised, Open-Label, Phase 1b Study

Triplet Therapy With Venetoclax, FLT3 Inhibitor and Decitabine for FLT3-Mutated Acute Myeloid Leukemia

Impact of Venetoclax and Azacitidine in Treatment-Naïve Patients with Acute Myeloid Leukemia and IDH1/2 Mutations

Abstract 32: Comparing Outcomes between Liposomal Daunorubicin/Cytarabine (CPX-351) and HMA+Venetoclax As Frontline Therapy in Acute Myeloid Leukemia

Abstract 795: Real World Survival Outcomes of CPX-351 Vs Venetoclax and Azacitadine for Initial Therapy in Adult Acute Myeloid Leukemia

Myeloablative Vs Reduced-Intensity Hematopoietic Cell Transplantation for Acute Myeloid Leukemia and Myelodysplastic Syndromes

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.