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 and director of the leukemia program at Weill Cornell Medicine and the NewYork-Presbyterian Hospital in New York City. It is my great pleasure to welcome my colleague and my friend from the MD Anderson Cancer Center in Houston, also a professor of medicine and from the leukemia program: Dr Gautam Borthakur. We're going to talk about good-risk acute myeloid leukemia (AML). I always struggle with this because AML is a bad disease. I worry sometimes as we're talking to patients and we introduce the words "good" and "favorable"; I think patients sometimes think that means "don't worry about it, we'll live forever," and I don't think that's quite right. But I want to launch the conversation by first defining what are we talking about. I think there's confusion about assigning good or favorable risk just using cytogenetics, which is kind of the old way we used to do things, and then introducing molecular classifications — for example, the European LeukemiaNet (ELN) Classification, which combines cytogenetics and molecular information. Gautam, can you launch into that and talk to us about those differences and teach us how to define good risk?
Gautam Borthakur, MD: Excluding acute promyelocytic leukemia (APL), the group that comes under the good-risk category is what we commonly call the core binding factor AML, mostly centered around two translocation events: inversion 16 and translocation 8;21. The other group that comes under the same category is the NPM1-mutated AML, particularly if it does not have a coexisting FLT3 mutation or a very-low-burden FLT3 mutation. That's the added sophistication from the ELN classification, where originally we pretty much assigned a good risk entirely based on cytogenetics. Now additional nuances are coming into the categorization of this particular group.
Roboz: So if you're looking up a publication or you're looking up something on UpToDate and you're trying to figure out how to manage the patient, it's important to have a quick look at the eligibility of the clinical trial. Because if they've defined good risk only by cytogenetics, or if they've defined intermediate risk only by cytogenetics, that may or may not be exactly what you're looking at — for example, if you have an NPM1-mutated patient. Similar to that, what do we do about patients, for example, who might have an inversion 16 but who also have a FLT3 mutation?
Borthakur: Those are the additional nuances in the core binding factor in AML inversion 16 or translocation 8;21. There are frequent co-mutations, particularly mutations in the KIT pathway or mutations in the RAS pathway. And third, the mutations in the FLT3 pathway, most of the KIT mutations that are considered higher risk is usually within the exon 17; and on the flip side the FLT3 mutations, at least in the setting of core binding factor, it's been predominantly the D835 mutations and less frequent the ITD mutations. But with the traditional 7 + 3 chemotherapy plus high-dose Ara-C consolidation, in that setting repeatedly it's been shown both in inversion 16 and translocation 8;21 that the presence of a KIT mutation does possibly identify a population at higher risk for relapse. And with that one biology, that has always kind of intrigued me — that many of these patients, KIT-positive patients, tend to relapse earlier. At the time of relapse, many of these patients would not have a KIT mutation. So, the KIT is a proliferative mutation defining a bad disease biology rather than KIT per se being the driver event. And it's unlike what we see in, say, FLT3-mutated AML, right? Because most of the time when they relapse, they tend to retain the FLT3 mutation and infrequently lose the FLT3 mutation. So, the KIT story is kind of interesting.
The British group also has published additional nuances that it may not be the presence of KIT mutation or the FLT3 mutation. Maybe it's the allelic burden that matters. It's been arbitrarily defined what the high allelic burden would be in the context of KIT mutation or FLT3 mutation in core binding factor AML. But that's another issue that we need to potentially look at. And there are other nuances also, that not in all settings the KIT becomes a poor prognostic marker; because in a pediatric setting it's been shown that it's not necessarily a poor prognostic marker. So maybe a more effective therapy can potentially overcome that additional prognostic impact that the presence of a KIT mutation has. The FLT3 is, of course, a different challenge because now you have inhibitors that can actually hinder FLT3 mutation, even though there are smaller reports that say maybe there is an increased risk of relapse. But that's not a consistent finding, not as consistent as a KIT mutation. So, whether the FLT3 needs to be targeted along with the induction consolidation chemotherapy, that's still a bit up in the air in the context of core binding factor AML.
Roboz: I think that we should make a couple of really basic but important points about getting the diagnosis right, because within the category of good-risk AML, there are both prognostic and therapeutic implications that are very, very dependent on getting the whole package of diagnostic information correct. So, find out core binding factor upfront cytogenetics and/or FISH; find out if you have an NPM1 mutation. You can send out sequencing that might take 2 weeks to come back, but if you have a rapid PCR, you could get that back. And that package, if you agree with me, will then give you all of the bits and pieces of the initial biology that can help you start building your treatment program.
Borthakur: I completely agree. For example, NPM1-mutated AML with low FLT3 burden. What do you do? Do you add an FLT3 inhibitor? Now, if you look at the FLT3 inhibitor trials irrespective of allelic burden, there is benefit across the board. So, even though somebody has an NPM1 mutation, if you think it's a good-risk disease, you can pretty much do the standard induction consolidation and forget about sending the patient for transplant or additional tyrosine kinase inhibitor (TKI). But in this context, possibly you'd be more inclined to use the TKI because of the added benefit that you get.
Roboz: Okay, let's do that. When you've got your NPM1 you've got a low-allelic-ratio FLT. What am I doing with gemtuzumab for those patients? Is it in? Is it out? Am I now building a cocktail with gemtuzumab and a FLT3 inhibitor? Now I'm starting to get stressed.
Borthakur: No, and that is the biggest challenge with an NPM1-mutated population: Where should we use gemtuzumab or not? In the core binding factor group, I think the story's clearer, that you possibly have the best benefit in that context. And in the NPM1 context, it's a little bit of a problem because in some of the trials, when you add gemtuzumab to the induction regimen, maybe there are a few more induction-related deaths or cytopenia-related deaths. So, it kind of offsets the benefits of gemtuzumab in the setting of NPM1-mutated AML. But for people who can successfully receive the induction consolidation that includes gemtuzumab, their molecular clearance is much better.
Roboz: Let's push hard for a minute on NPM1, and let's make it easy in that we're going to talk about first: an isolated straight-up NPM1 mutation. This is a patient with a cytogenetically normal AML with an isolated NPM1 — a 40-year-old patient who is healthy. What's that patient getting treated with?
Borthakur: At this point? It would be more of a standard induction consolidation.
Roboz: And we're not transplanting those guys if they're an isolated NPM1 mutation, right? We're getting them into remission.
Borthakur: That is correct. You bring up a very important point. The sensitivity of the assay matters, right? Even we are struggling at MD Anderson with the most sensitive assay; we are kind of shifting toward a better-sensitivity assay because right now, the in-house assay has a sensitivity of about one person. But that's not enough. You possibly need about a 10 to the power minus four, minus five level of sensitivity when you can more comfortably make the decision of whether to even consider transplant for this patient or not.
Borthakur: Because even then, just based on when you do it in a panel, the sensitivity may not be as good; where if you do an isolated PCR-based assay, the sensitivity may be much better. So you need to really have the discussion with the lab: What's your level of sensitivity before we make the decision whether to send the patient for transplant or not?
Roboz: For the listener who may not be pondering this type of thing every minute of the day, it's really, really tough because if you have an NPM1 mutation, how are you going to follow to make sure that that person is — you're not going to repeat next-generation sequencing every time. So you're going to be monitoring molecularly defined patients, like an NPM1 patient, by PCR. And here's a plug for the ELN MRD guidelines. That is something where we actually have some pretty basic clinical guidelines for monitoring for molecularly defined patients, like an NPM1-mutated patient, of how to check the PCR and it doesn't actually have to be flat zero. There are sometimes detectable transcripts, but they need to be staying at the same level or in a downward trajectory over time to make sure that that patient is safe. We actually have some monitoring suggestions using peripheral blood because these patients don't necessarily want to have a bone marrow biopsy every five seconds. But I think both in the US and in Europe, there's going to be a tendency not to try, at least not to allotransplant that patient. Would you agree?
Roboz: So let's make life a little bit more complicated. Your healthy 40-year-old has a FLT-TKD. Are you throwing in, first of all, the TKI and induction? Are you throwing in an FLT3 inhibitor for that patient, and is that TKD going to make you transplant a 40-year-old?
Borthakur: The first part may be easier to answer because, yes, we are throwing in a TKI along with that, irrespective of the allelic ratio. Now, if they clear out, if declarative wouldn't be on one end of the FLT3 because the thought is that the FLT3 is possibly a mutation that is acquired later, whereas NPM1 is possibly a little more ancestral than the FLT3. So, if you actually clear out both, then I think you're in a fairly strong, grounded yes, this patient may not benefit from stem cell transplant. And you can just monitor the patient by qPCR at least for a couple of years. So I would not send that particular patient for a transplant.
Roboz: We would actually do the same. And I think that that comes as a surprise, because in the US, there is such a bias for allotransplant, for obvious reasons, for 40-year-olds. And there's such a bias that if you see a FLT, half of the world doesn't even look at the next three letters that come after the F-L-T. So you've got to look and see: Am I dealing with an ITD or a TKD mutation? And I think that's important. So that patient is likely to get a high-dose cytarabine-based consolidation. Just a quick question. Your group has published the eye-popping results from even further intensification of chemotherapy with something like a FLAG-IDA venetoclax or a CLIA-venetoclax. These are super-intensive, both in the salvage and in the induction settings. Do I need a sledgehammer for this patient or can we stick to what we know and not necessarily use that type of intensification, even though venetoclax is so powerful with an NPM1?
Borthakur: Because of delivery of both the venetoclax and the TKI in combination with intensive chemotherapy, that's possibly asking a little too much. More than likely we'll have prolonged myelosuppression and infectious-related complications. So I guess at some point we'll have to make a choice about whether we go with the venetoclax-based regimen or we add a TKI to the induction consolidation in this context. Possibly a TKI would serve us and we can choose something like the gilteritinib, which can hit both the ITD or the TKD mutation. You may not even need to go for the full dosing. Something like 80 mg dosing might be good enough in combination with the induction consolidation. So for this patient with an NPM1-mutated FLT3-TKD mutation, we possibly do not need to throw in the venetoclax component of it.
Roboz: I think that's an important point too — that with the availability in the United States of multiple agents, there is this "Oh, let's throw in a little bit of this, let's throw in a little bit of that." I actually want to discourage that. I feel like if you have a straightforward path with excellent outcomes, then maybe people shouldn't necessarily layer in additional doublets and triplets and so forth, without more data, because I worry about having some unanticipated toxicity there. Now, what are you going to do if the patient that we're talking about, let's say with this isolated NPM1, is now 70? And let's say maybe a little bit less superbly healthy than the 40-year-old. Let's make it a little harder that this is someone who could probably get through intensive induction. You're not going to want to give high-dose cytarabine (HiDAC), real HiDAC — 18 gram kind of HiDAC — to this 70-year-old. So, how do you think about an NPM1 patient in that group? Given that even though the patient is fit enough for intensive therapy, how are you thinking about a hypomethylating agent plus venetoclax-based combination for, again, very, very high results with NPM1, vs trying to get this patient down a more traditional cure path with maybe some less intensive cytarabine as consolidation?
Borthakur: In that, I think our approach generally has been possibly to move to a hypomethylating agent and a venetoclax-based combination, just because the results have been so good. I mean, if you look at the CR rates, CR rates are almost comparable with intensive chemotherapy, particularly in good-risk groups, like NPM1 mutated. And the long-term outcomes also have been very good. After 2 or 3 years of follow-up, in many of these categories we have not reached the median survival, which is excellent. So, our approach in that context, possibly somebody over the age of 70 with an NPM1 isolated mutation, we could possibly go with a hypomethylating agent plus venetoclax-based combination.
Roboz: Before I switch to the discussion over to the core binding factor, I want to bring up very interesting data that are emerging from the QUAZAR trial, the trial looking at oral azacytidine in patients who were intermediate-risk cytogenetics. The way they got into the study was being over 55 with intermediate-risk cytogenetics, and then they got intensive therapy. But something happened. They blew up. They couldn't get a stem cell transplant. I think there are interesting data to know about that are emerging — that actually the NPM1-mutated and also NPM1 plus FLT-mutated patients in that group, again addressing this point that they were intermediate risk by cytogenetics, were favorable risk by the NPM1. They actually had some significant benefits with oral azacytidine maintenance after intensive chemotherapy. So I think it's going to be interesting to watch the maintenance space for these NPM1-mutated patients.
Now taking us back to core binding factor So, we have an 8;21. We have an inversion 16. Is it standard at MD Anderson to use gemtuzumab in combination with chemotherapy for those patients?
Borthakur: Yes, it is. We started around 2007. This program of adding gemtuzumab to the induction consolidation based on some of our historical data… We decided not to go to 2 + 7 backbone but with the more fludarabine, Ara-C and GCSF so-called flag sort of a backbone. The addition of gemtuzumab definitely adds value. But unfortunately, when gemtuzumab was discontinued from the US market, we tried to replace gemtuzumab with another anthracycline-containing agent. So we went on to idarubicin and we have published some of this data. And then we also have some data that's in preparation for a manuscript that we still feel that the idarubicin substitution did not do what gemtuzumab couldn't do. So now we have completely gone back to gemtuzumab. One of the things that we had published was that when we use gemtuzumab, if you use a surrogate marker of early qPCR clearance or more sustained qPCR clearance for the translocation transcripts, the gemtuzumab-based regimens have way outperformed anything that we had to offer with the idarubicin-based regimens, so we have pretty much moved over to use of gemtuzumab in this.
Roboz: I think that's a great point to emphasize, that gemtuzumab should be after you've diagnosed your core binding factor, good-risk, favorable-risk patient. You're going to use gemtuzumab, there is a survival benefit. The dosing you have to be careful about, I think. NCCN guidelines, that type of dosing, whether it's a 6 mg dose or how exactly you dose it, it's not going back to the original dosing of when gemtuzumab was first released. I think the other thing to emphasize is what you just brought up a minute ago: that these patients are going to need ongoing qPCR monitoring. And, actually, commercial labs will do that. You have to remember to send to the patient for monitoring, and I would once again make a plug for the ELN guidelines, just again categorizing how to monitor these patients in remission so that you can make sure that your transcript levels are either stable or going down.
And now in the last 5 seconds, I want you to have the last word. Sometimes those transcripts don't behave, and maybe they're even trending up a little bit without any evidence of relapse. Do you sprinkle in a little bit of a hypomethylating agent at that point to try to clean things up?
Borthakur: There are two contexts where we have added the hypomethylating agent. One is the context where we expect the patient to get at least a four to six induction consolidation regimen total, which we think is optimal treatment now for one reason or the other. Maybe because of an infection or something else, the patient cannot complete the planned number of cycles. Or somebody who had finished the planned number of induction consolidation but now still has some persistent qPCR positivity. Now what do we do? Do we send them for a transplant or do we try another strategy? And that's where we use the hypomethylating agent. The second context: If somebody is molecularly close to negative and had a lesser number of induction consolidations than what you intended. That would be one context that I would possibly advocate for use of a hypomethylating agent or somebody with a very low PCR positivity postinduction consolidation. Now, when we see a higher burden of disease, like less than the so-called optimal response, you may need to start to think about transplant because the hypomethylating agent may not deliver what you want to do in a low persistent disease context. But with a somewhat higher PCR burden, I possibly would still think about transplant.
Roboz: Super-important message to not just look under the microscope. So when you're thinking good-risk AML, you're thinking PCR transcripts and NPM1 transcripts, core binding factor transcripts. Thanks very much for your interest and participation. I look forward to speaking with you again.
Borthakur: Thank you. This was fun.
Refinement of Cytogenetic Classification in Acute Myeloid Leukemia: Determination of Prognostic Significance of Rare Recurring Chromosomal Abnormalities Among 5876 Younger Adult Patients Treated in the United Kingdom Medical Research Council Trials
Venetoclax Plus Intensive Chemotherapy With Cladribine, Idarubicin, and Cytarabine in Patients With Newly Diagnosed Acute Myeloid Leukaemia or High-Risk Myelodysplastic Syndrome: A Cohort From a Single-Centre, Single-Arm, Phase 2 Trial
Medscape © 2022 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Good-Risk Acute Myeloid Leukemia - Medscape - Apr 14, 2022.