Current and Emerging Role of Biomarker-Directed Therapeutic Approaches for Patients with Acute Myeloid Leukemia (AML) — Courtney D DiNardo, MD, MSCE

DR DINARDO: Courtney DiNardo here from MD Anderson in Houston, Texas. My privilege to talk to you today about some of the current and emerging roles of biomarker-directed therapeutic approaches for patients with AML.

And I’ll just say 1 word to kind of overlay all of this, which is that we are becoming increasingly fortunate that we have individualized therapies. We have targeted therapies. Precision medicine is becoming more and more a reality.

So some of the updates, you can see here, is for patients with FLT3-mutated disease, so for our younger patients who are FLT3 mutated, which is about a third of them. So a decent chunk of our newly diagnosed AML patients have a FLT3-ITD mutation. We now have updates from the QUANTUM-FIRST study, which was a randomized study of standard intensive chemotherapy with either placebo or quizartinib, showing that we have an improved overall survival. And this has led to the approval of quizartinib to incorporate for our FLT3-ITD-mutated patients at diagnosis.

And one thing people often are saying is, well, midostaurin is what we typically have used, right, based on the RATIFY study, which you can see on the left. When you add midostaurin the complete remission rate is a little bit higher, a clear improvement, although somewhat incremental, but clear improvement in overall survival with midostaurin compared to placebo. And when I try to superimpose the QUANTUM-FIRST graph it kind of looks the same.

But I think one thing that’s important to remember is even though these studies on first blush are similar there’s 2 main differences. One is that the RATIFY study was only for patients who were 60 years of age and younger, whereas the QUANTUM-FIRST study allowed people up until the age of 75. And then the other main difference is that midostaurin is effective for both the ITD and the FLT3-TKD mutations, so about a quarter of the patients that were treated in the RATIFY study on the left were FLT3-TKD, and those patients don’t have as poor of a prognosis.

So when you think about it that way — so if you’re kind of now looking at the QUANTUM-FIRST study on the top right, only patients who are under 60 years old and who only have, remember, the FLT3-ITD, it actually starts to look a little bit different. And you can potentially see that there’s more of a benefit with the addition of quizartinib.

And then conversely, you look down to the bottom, and you see that patients who are over the age of 60 unfortunately don’t seem to have that same benefit from the addition of quizartinib, and I think some of that just has to do with the challenges with giving older patients intensive chemotherapy, which is why our field is definitely transitioning towards lower-intensity combinations and effective strategies for our older patients.

The next step is looking at why is it, or is it true, that patients that are getting quizartinib are getting better remissions, deeper remissions, and so MRD is becoming more and more important in analyses of AML studies.

And so this was an update from Sasha Perl at ASH showing that at the various different timepoints. Of course if you become MRD negative you’re clearing that clone, not only are you in a morphologic remission, the hematopathologist can’t see any leukemia, but you’ve actually cleared leukemia ever further using a very sensitive assay, like 10^{-5, -6} assays, you do better, right? So not surprising if there’s still residual disease that’s detected with MRD positive your outcomes are not as good. And that’s true at every different timepoint; after induction, after consolidation, after completion of therapy.

So the obvious question is does quizartinib help you obtain deeper remissions, and so that’s what’s shown here, and that is indeed true. So looking at these blue bars, blue is the patients who are getting randomized quizartinib. At each different timepoint, after induction, after consolidation, after completion of consolidation you have more patients getting deeper MRD-negative status. And then along the bottom you can see at the end of the induction/consolidation treatment over half of the patients, 56% of them, are becoming MRD negative on the quizartinib arm.

MRD wasn’t a thing back when we did the RATIFY study, so we can’t really compare it to midostaurin, but we certainly can see that the addition of a FLT inhibitor is obtaining more MRD-negative remissions and hopefully more patients being cured because of that.

So the conclusion of kind of the QUANTUM-FIRST updates over the past year is (1) there is prognostic utility of obtaining MRD measurements. We realize more and more that MRD is important and can help us identify who is going to do best with treatment, as well as in the future help us identify how to treat people differently because of it and that quizartinib is associated with deep responses and more frequent MRD clearance.

DR LOVE: I’m just kind of curious. I’m sure most of your patients go on trials, but in situations where you have a patient who’s eligible, has got an ITD mutation, is eligible to receive both quizartinib as well as midostaurin, do you have any preference?

DR DINARDO: We have increasingly been using quizartinib. I think there’s comfort in what you know, so I feel like in the community they’re still using a lot of midostaurin over quizartinib. I find that the GI adverse events and the rashes and a lot of the challenges with midostaurin are far less frequent with quizartinib. The main side effect with quizartinib is more cytopenias. So you just have to be aware that the myelosuppression is a little bit more prolonged and help manage that. And QTc, of course, you have to monitor for QTc. So it’s a slightly different profile, but we’ve increasingly been using quizartinib and being happy with that.

DR LOVE: In terms of the “GI issues” with midostaurin, I’ve heard that it like smells bad. Is that what it’s about?

DR DINARDO: It does. The pill itself smells bad. One of my patients calls it her skunk pill because it really just has an unpleasant odor. So that causes some of the nausea just as you take it, but the pill itself also does seem to be associated in many people with just more GI complaints.

DR LOVE: So let’s get on to this next part because this is super cool, this wild-type thing.

DR DINARDO: Yes. And so the QUIWI study is a study that everyone is trying to grapple with a little bit and trying to figure out what to do. So the QUIWI study is from Spain. This is a Spanish Cooperative Group study looking at using quizartinib, which is a selective FLT3-ITD inhibitor but does have kinase activity, for patients who are FLT3 wild type, right, which we didn’t really think made sense. But they have this rationale, and they did a clinical trial. It was a 2:1 randomization, so you get 7 + 3 with either quizartinib or placebo, so very similar to the QUANTUM-FIRST study but now we’re looking in a FLT3-ITD wild-type population.

And essentially what they identified and reported on a year or 2 ago was that patients who received quizartinib actually did better. Two-year overall survival was like 63% versus under 50%, 47%, with the placebo. And newly identified, and what they reported this year at ASH was there’s this FLT3-like expression profile that about half of the patients have, and that was what helped predict who was going to respond to quizartinib or not.

And so if you did not have that FLT3-like signature you didn’t need quizartinib. It didn’t benefit you just like you would expect a wild-type patient wouldn’t benefit from quizartinib, but the FLT3-like patients really did benefit, and so that was highlighted.

And the challenge is that you can’t really identify who the FLT3-like people are just with standard mutational panels, which is what we’re using now to decide how to treat patients. It was true that more patients with NPM1 mutations and DNMT3A mutations were FLT3-like, but that wasn’t universal, and there were patients in all of the different prognostic categories that were in both the yes and no FLT3-like.

And so the challenge is trying to figure out can we use this, how do we identify who is going to be FLT3-like to be able to incorporate quizartinib. But it is super interesting, and it makes us of course realize that we think we know how a drug is working, and we don’t always, and there can be other mechanisms.

This story reminds me a lot of the Ph-like story in ALL, where there are patients who don’t have BCR/ABL, don’t have that Philadelphia chromosome, but yet they have an expression profile, a Ph-like profile that shows that they actually do respond better to tyrosine kinase inhibitors. So it’s a very similar story, and so we’re trying to figure out how to actually incorporate this in the clinic.

DR LOVE: I was talking with Naval about this, and I was recollecting, I think it was the sorafenib study that I think showed benefit even in “wild-type” patients, and I guess they never explained that. Can you talk a little bit more about like what FLT3 is and like pathophysiologically, not just clinically but biologically what goes on in these patients?

DR DINARDO: Yeah. So first of all, it’s one of the most common mutations in AML, so FLT3 and NPM1 are 2 of the most common. So about a third of our patients will have a FLT3 mutation, either in ITD, which stands for an internal tandem duplication, or a TKD, which is the tyrosine kinase domain. And so the ITD is exactly that, you get these duplications in the FLT3 which leads to a proliferative phenotype. So FLT3 is involved in the proliferative pathway.

So when you have these ITDs it often leads to kind of a patient that is presenting with AML with a higher white count, with a more kind of aggressive physiology, and that’s why the FLT3-ITD patients have historically been more high risk. They’re the ones presenting with the white count over 50,000, where it’s an emergency, you’ve got to calm things down because they’re having kind of leukocytosis complications.

DR LOVE: I guess it’s too early to say what the magnitude of benefit might be in these people, I guess even if it’s there I guess, right?

DR DINARDO: Yeah. I mean the challenge I have with the incorporation of quizartinib to standard 7 + 3 is that there are a lot of different things you can add to standard 7 + 3 and have patients improve. So for instance there’s also discussion of adding fludarabine or cladribine. There’s the Polish study adding cladribine to standard 7 + 3, and patients do better. There’s a lot of clinical trials here throughout the United States adding venetoclax, right? We use venetoclax with lower-intensity therapy, but adding venetoclax to intensive chemotherapy is also incredibly effective, and I wouldn’t be surprised if some studies are positive in the coming years in terms of that, as well. Our own institution has published Phase II studies of FLAG-IDA/venetoclax and CLIA/venetoclax, so our regimens of these things, which show I mean 80, 85% response rates, MRD negative rates. So I think the challenge is knowing what to add in which patient because adding something to standard 7 + 3 is almost certainly going to benefit a certain percentage.

DR LOVE: Okay. Please continue.

DR DINARDO: So the summary of the QUIWI results is that there is kind of surprising but initial evidence that quizartinib with intensive chemotherapy may benefit patients who are FLT3 wild type. And my second comment here is it’s just hard to obtain these expression patterns. It’s not only hard, it’s essentially impossible. So we can’t really be using these expression patterns yet to identify who should or who should not get quizartinib, but it really is part of this growing body of evidence suggesting a more personalized and targeted approach for AML is increasingly possible as we have these new therapies.

And there is also another FLT3 inhibitor, of course, gilteritinib, which is another kind of second generation. This is more of a kind of pan kinase also that hits both the ITD and the TKD. We are waiting for the results of a large randomized clinical trial of 7 + 3 with gilteritinib.

There is a Phase IB study that Keith Pratz has presented and published in JCO last year I believe, highlighting the fact that this will probably work too. So gilteritinib is approved only in the relapsed setting as a single agent right now, but in combination with intensive chemotherapy in the Phase IB nonrandomized component patients. Both FLT3 wild type and FLT3 mutated received gilteritinib and appeared to do better, both in terms of response and overall survival. And so we’re kind of all eagerly awaiting kind of the results of that larger 7 + 3/gilteritinib clinical trial.

But gilteritinib, it’s the monotherapy relapsed approval right now, but a big story over the past year was the results of the MORPHO trial, which looked at is there a role for using gilteritinib in post-transplant maintenance. So the MORPHO study was a very large study that enrolled patients who were in a remission that had received only 1 or 2 different induction/consolidation cycles, 30 to 90 days after transplant they should be starting gilteritinib or placebo.

An important kind of secondary analysis of this study was looking at the role of MRD. So I talked about MRD already with the QUANTUM-FIRST study, and now we’re talking about that same very sensitive FLT3 MRD assay in this study. And actually about half of the patients who were enrolled in this study were MRD positive, either right before or right after transplant.

And so there was this collective groan when this data was presented because the primary objective was relapse-free survival for the entire population, and the hope was that gilteritinib was going to improve relapse-free survival across the board. And it was just shy of being statistically significant with a p-value of 0.0518, and so the study was negative. And so if you have — you’re a one-liner this study was negative.

But what I think is potentially the right way to use this, again, we’re talking about MRD-directed therapies and the importance of MRD, in those 50% of patients who had low-level FLT3-ITD mutations identified, those were the patients that benefitted from 2 years of gilteritinib. And so along the top in this, in this curve, you can see there’s a clear benefit from adding gilteritinib in those patients who are MRD positive, whereas if you didn’t have any disease, if you were cured, you didn’t need 2 years of chemotherapy. And so I think if you look at it that way it’s, again, highlighting the fact that we’re doing better with personalized medicine.

I love this slide because it shows that MRD is important, but also the level of MRD is important. So when you look at kind of how much residual disease you had, you could detect it at 10^{-4, -5, -6}, you are increasingly likely to do better the less residual disease you have, which of course makes sense. And for all of those patients by adding gilteritinib you’re improving the outcomes.

And then the last thing that I think is really important to recognize is, again, FLT3-mutated patients have that more proliferative phenotype, and so relapses often happen fairly quickly, especially in that post-transplant setting. And so if you are kind of thinking yes, I’m going to add a FLT3 inhibitor as maintenance for my patient with FLT3-mutated disease post-transplant, but you’re waiting 6 months, 9 months, it’s too late. They’ve really already relapsed by then. So it’s important to start that therapy in a patient who has MRD-detectable disease as early as you can, as early as 30 days, once they’ve engrafted, and counts have recovered, and they’re doing well. And so I think that’s just a really important thing to keep in mind because oftentimes we’re waiting in the peritransplant setting to start a maintenance approach.

There is additional evidence about the role of adding gilteritinib in that post-transplant setting. So this was the ADMIRAL study. I had mentioned gilteritinib is approved as a single agent in the relapsed setting, so that’s because of this study, the ADMIRAL study, which highlighted that gilteritinib was not only as effective it was more effective than intensive chemotherapy or other approaches for people in the relapsed setting with FLT3-mutated disease.

And there was a subset of patients who went to transplant in this study, 64 out of the original almost 250 that got gilteritinib, and in those patients that went to transplant 40 of them restarted gilteritinib, and 24 of them did not, and the patients who restarted gilteritinib did better overall. And this is biased analysis, and this of course is just kind of additional evidence to suggest and highlight the fact that patients who are FLT3 mutated that are going to transplant, post-transplant maintenance, especially if they have MRD-positive disease, is definitely of benefit.

And again, just highlighting what I have mentioned already, any level of MRD in patients with FLT3-mutated disease will impact relapse-free survival. You want to clear that as much as you can. If you have detectable MRD that is where there’s clear evidence of improving outcomes in patients receiving gilteritinib post-transplant, and the sooner you can start the better your patient will do because relapses happen fairly quickly post-transplant.

DR LOVE: Just to clarify. Have most of these patients received midostaurin previously?

DR DINARDO: It depends on the part of the world that these patients were treated in. So in the US, yes, many of them had. In Europe it’s some patients had, and then in Asian countries very few had. So that is something that has been looked at, and really it was a benefit in MRD-positive patients across the board.

DR LOVE: That’s interesting. All right. Keep going.

DR DINARDO: Just the long and short of it is that TKDs are not nearly as bad as the ITDs, and so it’s a FLT3 mutation for sure, and you can target it, and you should target it, but it’s really the ITD patients that have the highest risk of relapse and the highest risk of poor outcomes.

And then I have an update. This is the only update of a triplet study that I’ll talk about today. But this is the addition of adding either actually an IDH1 or an IDH2 inhibitor, ivosidenib or enasidenib, to the backbone of HMA and venetoclax. So in this case the hypomethylating agent is oral decitabine.

So one of the neat things about this triplet trial is that it’s all oral, so oral decitabine/venetoclax and either the oral IDH1 or IDH2 inhibitor.

The oral decitabine is given for 5 days per cycle. Venetoclax is only 14 days, and so as many of you know the label of venetoclax is to give continuous venetoclax, which is really not what’s done in the community and not the best for your patients. So when we are designing triplet trials almost always the venetoclax is given for 14 days, and then often in patients in a remission you can shorten that later on in subsequent cycles. But just to highlight that study schema.

We’ve treated now over 50 patients, both newly diagnosed and relapsed/refractory patients, with IDH1 or IDH2 mutations, have been enrolled in this study. And for newly diagnosed patients we are seeing response rates, composite remission rates of over 90%. 85% are becoming flow negative for MRD assays.

And so just as a reminder with the aza/ven study the VIALE-A study, about 40% of responding patients became MRD negative. So it seems like about twice as many patients are becoming MRD-negative, and we’re not seeing early mortality.

I think a triplet trial really depends on what that third agent is. The IDH inhibitors don’t cause a lot of myelosuppression, they’re not really cytotoxic, so we’re not seeing a lot of increased toxicity by doing the addition of the IDH inhibitors.

In the relapsed setting we’re seeing composite remissions overall of about 50%. When you kind of think about how well either aza/ven or an IDH inhibitor is going to do in the relapsed setting it’s about like 40% composite remissions have been described, but what’s kind of important to remember is how heavily pretreated this population is. So when you look at those patients who haven’t received prior venetoclax or who haven’t received multiple lines of therapies, of prior IDH inhibitors and venetoclax, about two thirds of patients are responding. And so I think this really does highlight that this is an effective treatment strategy, and I think you’ll be hearing more and more about some of these HMA/ven triplet trials trying to get deeper remissions and hopefully improved outcomes.

There is another IDH1 inhibitor, olutasidenib. So this is a study that led to the approval of olutasidenib in the relapsed/refractory setting as a monotherapy. There were 8 different cohorts, but the largest enrolled and the one submitted to the FDA was the relapsed/refractory cohort of AML patients. So about 153 patients were treated.

And so one of the questions in the world of IDH1-mutated disease, which is about 10%, so about 10% of AML patients will have an IDH1 mutation, is should you use olutasidenib, or should you use ivosidenib. And there’s no head-to-head trials, of course, and so we kind of just look at the studies that were done and try to make comparisons and identify.

So when you look at the median survival of olutasidenib it was just under 12 months, 11.6. It was 9 months for ivosidenib. When you look at the nonresponders those curves are very similar, and so there is some, I think, appropriateness of comparison here.

What is really interesting is that the response rates overall are incredibly similar. So the composite remission rate is about a third. The overall response rate is in the 40s for both of them, but what’s really different is the duration of remission. And olutasidenib has kind of double the duration of a response rate or a composite remission, and so the question is, is this a better drug or is it just a different patient population. And it’s really hard to know. I’m not entirely sure.

So kind of looking through at the comparison of the IDH1 trials that led to the approvals in the relapsed setting they were both about 150 patients. I’ve kind of highlighted the differences. There were more patients on ivosidenib who had poor-risk cytogenetics, there were double the number of patients with p53 mutations, more patients were kind of multiply relapsed instead of just in their first salvage, and more patients had had a prior transplant. So it may be that the ivosidenib population just was a higher-risk population, and that’s why we’re not seeing as long remissions. We know p53-mutated patients don’t stay in remission very long if they obtain a response in the first place.

But on the flipside there are very few things that work in a post-venetoclax setting. We and others have published outcomes in patients that fail HMA/ven regimens. Their median survival is like on the order of weeks, and really limited survival outcomes with any sort of effective therapy, unfortunately.

But there is a small kind of subcohort analysis of the 17 patients that had received venetoclax that then went on olutasidenib, and it’s really quite impressive actually. The overall response rate in these patients that had failed venetoclax and then went on this study was 40%, and the median response hasn’t been reached when it was last updated, and it had been ongoing for over kind of a year and a half. So there is, to me, some truth that maybe olutasidenib is leading to deeper remissions and may be a quite effective drug. Yes.

DR LOVE: I don’t know, you’ve probably never heard of the STAMPEDE trials out of the UK in prostate cancer. These wild trials with like so many arms and all. I was just jotting down here on my pad here first of all what do you think — I know this trial would never be done, like a 6-arm trial, triplet versus doublet versus single agent, I don’t know what the short name for olu, and ivo.

DR DINARDO: Yeah.

DR LOVE: Any thoughts about what it would show? And also right now I mean more than data what’s your gut about the best long-term strategy? Because you brought up the issue of maybe sequential, you’ve got to look at that. You’d have to look at a lot of things. But globally right now what’s your intuition about where it’s going to land?

DR DINARDO: Yeah. I mean I think you’re right. In a perfect world we would have a STAMPEDE-like trial where we would have arms of aza/ivo versus aza/ven versus aza/oluta versus triplet combinations to figure out which one really is the best approach. The challenge is only 10% of AML patients have an IDH1 mutation, and AML is a rare disease, right? So when you are trying to design these trials that would need hundreds of patients in each arm to show superiority there’s just no way, which is a shame because it would be really nice to have a definitive answer.

I’ve presented — and cooperative groups doing at least 3-way randomization, aza/ivo, aza/ven, aza/ivo/ven, because I do feel like the triplet of aza, ven, and ivo is leading to deeper remissions in a well-tolerated fashion. But it’s hard because patients that get either azacitidine and venetoclax or azacitidine and ivo, those are approved front-line regimens, and they do pretty well, so that incremental benefit to prove in a randomized 3-way clinical trial would be really tricky.

My bias is you have about two thirds of patients responding to either aza/ivo or aza/ven in the front-line setting. When you do the triplet you’re getting like 90% of patients responding, so you’re capturing more responses at the beginning, and it seems like we have more patients obtaining deeper remissions early on.

Just because you start with a 3-drug regimen doesn’t mean they need to stay on that forever and ever. So part of what our field has to do is really help identify maybe 1 or 2 induction-type cycles, and then you move to a more ongoing maintenance strategy where they’re getting maybe 2 or 3 drugs or a shorter regimen. So this is still being designed, but I do believe that adding these therapies together as opposed to sequencing them is going to be the optimal approach.

Whether to use ivosidenib or olutasidenib, that one’s tricky too, because all of the triplet trial data and everything we’ve come up with so far is with ivosidenib. But olutasidenib is as good if not better, and so we really need to be kind of incorporating olutasidenib into our clinical trials, and we actually will be opening a triplet trial with oluta to get some experience with that in the coming months.

DR LOVE: Any biologic potential explanation if oluta does have greater activity, greater duration of response, et cetera? Is the kinome different? Is there any clues there about why it would be?

DR DINARDO: Nothing that I have seen that is compelling to really explain a difference. Olutasidenib is — it may kind of fit into that pocket where it binds because IDH is an enzyme, so it’s kind of blocking. Maybe it fits better. The only main difference in terms of kind of nuances of the compound is olutasidenib is a BID drug, so you take it twice a day. The half-life is shorter than the once a day ivosidenib.

The other thing that’s different, I don’t think I’m going through it in slide detail, so I’ll mention, is the side effect profile is just a little bit different. So they both can cause differentiation syndrome, of course, but ivosidenib has QT prolongation issues, so you have to be monitoring EKGs. Olutasidenib has transaminitis issues, so you have to be monitoring LFTs. And so there is kind of a little bit of a difference there.

DR LOVE: Interesting.

DR DINARDO: Yeah.

DR LOVE: So let’s get into the next great thing that I think oncologists better start learning about real fast.

DR DINARDO: That is exactly right. So this is the, as you said, the next new thing for AML are the menin inhibitors. So there are no menin inhibitors approved right now for use in the clinic, but these are small molecule agents, they’re pills, and they are differentiating agents. So just like the IDH inhibitors, ATRA, arsenic for acute promyelocytic leukemia, this is another class of differentiating agent therapy.

You can see this schema, this is specific, to review a minute, but they’re all the same. These menin inhibitors go in and they kind of bind to the menin protein, and they prevent this kind of leukemogenic complex that is created, that patients who have KMT2A-rearranged leukemias, so that used to be called MLL rearranged, the 11q chromosome translocations. And then NPM1-mutated leukemias also have this upregulation of this complex, which leads to kind of this upregulation of HOXA9 and MEIS1, so again we’re going back to expression signatures where you can identify patients who have this kind of upregulation of this pathway.

There is this expression profile that identifies that patients with KMT2A-rearranged leukemias and NPM1-mutated leukemias are most likely to benefit from menin inhibitors, and that’s where the initial clinical trials initiated. But actually we’re realizing that there are other subsets of AML patients, patients with for instance NUP98 translocations that are often seen in pediatric patients. And so we’re kind of still learning which patients are most likely to potentially benefit from menin inhibitors. But this is a really exciting new foray and new kind of entire class of effective AML strategies.

So you can see there’s 5 of 6 different menin inhibitors that are currently under evaluation in the clinic made by different industry partners. The first 2, revumenib and ziftomenib, are the farthest along and have had the most data presented. Revumenib in particular has already been submitted for FDA filing, and there is a PDUFA date in November, so actually in only potentially a couple of months we may have our first menin inhibitor available in the clinic.

And just to highlight some of the data. This is with ziftomenib, one of the menin inhibitors, and they’re all being evaluated initially as monotherapy in the relapsed setting as that’s how clinical trials work with novel therapies.

So about 40% of patients respond, and the responses are unfortunately not as durable as we would like. So in general the duration of remission with the menin inhibitor as a single agent is about 6 months. Resistance tends to happen pretty quickly.

This is not so different really than FLT3 inhibitors and IDH inhibitors, where you’re looking at a single-agent of a targeted therapy in the relapsed setting, so actually quite similar when you look at response rates and things. And so this is an example with ziftomenib, again highlighting that these are differentiating agents so you don’t see a lot of myelosuppression, you see count recovery, and you can see differentiation syndrome in about 15% to 20% of patients.

Revumenib is the other farthest along menin inhibitor. So there is some data shown here. Pediatric patients, it’s important to highlight, have KMT2A rearrangements. It’s one of the most common AMLs in infants, and so what’s interesting for us in the world of adult leukemia therapy is we’ve been partnering with pediatricians a lot more than we did before to enroll patients on these menin inhibitor studies because there is definitely a need for pediatric patients to have access to this.

Again highlighting that these are patients being enrolled in the relapsed setting, they’ve had multiple lines of therapy, many of them have failed venetoclax like we’ve said before, really highlights a population that doesn’t do well with really anything that we have available, and again we’re seeing an overall response rate here of over 60%, composite remission rate of 23%, and median duration of about 6 months. So again, these single agents work but don’t work that long, so hopefully you can get your patient to a transplant. In the future we’ll be designing clinical trials of combinations where most likely these agents will work in a more durable fashion.

So again just highlighting the fact that these are not particularly cytopenic. We don’t have a lot of challenges with that, but we do see QT prolongation and differentiation syndrome with all of these menin inhibitors.

There have been quite a few patients that have been able to transition effectively to transplant. So you get a responding patient, you’re able to get them to that transplant, especially with a KMT2A-rearranged leukemia that are really high risk you often haven’t had the opportunity to go to a transplant because they are very adverse risk, they don’t respond well to standard therapies, and so this is really exciting.

The KMT2A-rearranged leukemia group is even smaller, it’s about 5% of leukemia patients, but there is a clear need that we need these inhibitors in that population. And then NPM1-mutated disease is one of the most common. Many NPM1 patients do well in the front line but when they relapse they do just as bad as anyone else, so there’s definitely kind of a need for both of these initial AML cohorts.

And then just to highlight the third menin inhibitor that has clinical data that has been presented. This one is still a drug name with lots of letters and numbers, JNJ-75276617. But again just highlighting that patients are heavily pretreated, most of them have had prior venetoclax before, and yet we’re seeing response rates of about 50%, composite remission rates of over 20, 25%, in both the KMT2A and the NPM1-mutated population. We don’t see cytopenias, we do see differentiation syndrome, and responses last for about 6 months.

And the first clinical trial that has been presented thus far, is again a triplet. This is another all-oral triplet where it’s oral decitabine with venetoclax and ASTX. Again the venetoclax is only given for 14 days. We don’t want to kind of compound myelosuppression and toxicity issues.

Only 9 patients have been presented so far. Again, very heavily pretreated. The majority have had venetoclax. The majority here have already failed a transplant also. And now we’re seeing an overall response rate of 100% with a composite remission rate of 44%. If you include the CRPs you’ve got a whole lot of patients with kind of pretty deep remissions.

What we don’t know in this study is the duration of remission. That hasn’t been reported yet, it’s such an early study. So my hope is that once we’re doing these combinations not only are we seeing more responses but we’re seeing them last longer.

And so menin inhibitors are just really exciting, I really do hope that we’ll have the first one approved and available in the clinic by the end of this year. And then just like all of the other targeted therapies that are available and approved as a single agent in the relapsed setting we’ve got to figure out how do we make it work better, what’s the right way to use these combinations and improve the outcomes even further. So I think what’s called the SAVE trial, the all-oral triplet, is a really exciting preview to that.

And so that is the overview. So we have more and more AML therapeutic options, which is great, but it makes it a little more complicated. AML is a rare disease, and you’ve got to figure out how to use the dozen or so new trials that we have — new treatments based on the specifics of your patient. So that’s exciting. Increasing use of MRD, the how deep a remission your patient is in really can help guide treatment strategies. We have more and more trials showing evidence of that. And menin inhibitors are just the next new thing, and we’re really excited about those too.

DR LOVE: So just a couple of final questions. Seeing those trials with the oral decitabine and venetoclax brings that up in my mind because in general when I’ve been asking leukemia investigators about using that outside a trial they like don’t exactly jump on it. I don’t think they’re even doing it. And I was curious whether you are and why the hesitation.

DR DINARDO: I guess oncologists are pretty good rule followers, and so oral decitabine is approved right now only for MDS, so it is increasingly used, I think, in academic settings and in the community for MDS patients. But it just hasn’t really caught on yet in terms of use in AML, and some of that is insurance reimbursement. You can’t get oral decitabine outside of a clinical trial for AML patients in many places. I sometimes try because my patients are of course very interested, right? They would of course prefer an all-oral regimen instead of coming to the clinic for 5 days in row every months. So I think some of it is just awareness and thinking, but I think the other big part is the reimbursement issue.

DR LOVE: No, I mean, I totally get the reimbursement thing for sure, but I still hear people conceptually like questioning it.

I mean, look, I know that it was such a great advance, but it’s still I would say kind of in the palliative world.

DR DINARDO: Yeah.

DR LOVE: I mean, at least pharmacologically it certainly seems like it would be the same. I don’t know. I guess — anyhow. I don’t know.

DR DINARDO: Yes, so pharmacologically it was approved based on the fact that it is 99% equivalent, right? So it’s hard to argue that. It leads to the same kind of effect in a patient.

But it is true, I will say, I guess more anecdotally since there hasn’t been kind of published data about it, but it does seem that patients have more cytopenias. They do seem to kind of drop their counts a little bit lower a little bit longer when they’re getting oral decitabine than IV, and so that is something. It’s nothing that we can’t deal with in the world of MDS and leukemia. We’re very comfortable with low counts and transfusions, but that is something that I have noticed, so that might be something that is making people shy away I suppose.

The other thing I will say is a lot of our patients are quite old and a little bit frail with MDS and AML, and so you want to give them the ease and the comfort of taking a pill. But you also worry a little bit sometimes when people are taking all-oral triplets like are they really taking it the right way or are they taking the right chemotherapy drugs at the right times on the right schedule, and so sometimes having a patient come in for a very short infusion or a subcutaneous injection you know exactly what they’re going to get, and you can lay eyes on your patient and make sure they’re doing okay. So there is some truth to that too.

DR LOVE: And I’ve heard people say too the patients are there all the time getting counts, et cetera, so they have to be there. So 1 final question. With another pan tumor approval in the last couple weeks, T-DXd in HER2 positive, I’m really curious to see how that plays out —

DR DINARDO: Yup.

DR LOVE: — but I’m also curious, and also I didn’t realize another of your particular interests seems like is IDH specifically. I don’t know. You seem like you’re the IDH person there at MD Anderson, or one of the main ones. And I was just curious, again, getting back to that. Of course we’ve been talking about IDH inhibitors in for example biliary tract cancers, and I hear cases along the way where all of a sudden some weird common tumor you would never expect all of a sudden there’s an IDH mutation.

DR DINARDO: Yup.

DR LOVE: So do we know anything about whether these IDH inhibitors work outside of AML and I guess biliary tract cancer?

DR DINARDO: Yeah. So there’s IDH1 and IDH2. IDH2 mutations are almost exclusively found in hematologic malignancies, but you will see them in like certain T-cell lymphomas, and there are some patients with gliomas with IDH2 mutations, but far more frequently are IDH1 in various solid tumors.

So you mentioned cholangiocarcinomas have a pretty high rate of IDH1 mutations, also certain sarcomas like chondrosarcomas, gliomas. And so at least ivosidenib doesn’t have great blood-brain barrier penetration, olutasidenib it sounds like actually does have decent blood-brain barrier penetration. But there is a third IDH inhibitor, it’s a pan IDH inhibitor actually called vorasidenib, which hits the IDH1 and the IDH2, and it is a blood-brain barrier penetrant, and it has been evaluated with positive results in the glioma population. So vorasidenib is something definitely to be paying attention to in the solid-tumor world and gliomas in particular.

Available and Emerging Nontargeted Therapies for AML — Naval Daver, MD

DR DAVER: Hi. My name is Naval Daver. I am a faculty in the Department of Leukemia at the MD Anderson Cancer Center in Houston, Texas. It’s a pleasure to speak to you all today covering the Year in Review with Dr Neil Love. I will be focusing on available and emerging nontargeted therapies for acute myeloid leukemia, which is an area of great interest for us at the MD Anderson Cancer Center in Houston, Texas. I know many people are curious about how the Department of Leukemia works at the MD Anderson Cancer Center in Houston, Texas. We do have probably the largest group of leukemia experts in the world in the Department of Leukemia at MD Anderson and also very, very extensive list of clinical trials. Our focus historically and continues to be the development of novel targeted and immune therapies in combinations especially with the focus on Phase I and II to try to develop new paradigms of treatments that can then move into registration studies. And we do get a lot of patients, almost 70% of our patients, coming from out of the city and 50% from out of state. And so we really do try to develop attractive clinical trials beyond the current standard of care randomized studies so that patients who travel all that distance, spend money, logistical issues will get a treatment that hopefully will be something beyond what they could have got at their local academic or community center. And of course with all the progress happening in leukemia in general not just acute myeloid leukemia but all of the fields it’s a very, very exciting time for myself and many of my colleague researchers who are working in this field.

So I’m going to start with probably the medication that’s discussed the most in most acute myeloid leukemia settings. This is a drug called venetoclax, BCL2 inhibitor. The pivotal study of azacitidine with venetoclax called the VIALE-A study that was led by my colleague and friend, Dr DiNardo, led to the front-line approval of this combination in newly diagnosed older, unfit acute myeloid leukemia. We see here the response rates were all improved. The CR/CRi rate was 66 versus 28%. So more than 2-fold improvement. CR rate was also doubled at 37 versus 18%. And there is an overall survival improvement with median survival of 15 versus 10 months with the combination. So this is very good. Led to the approval. Has been used extensively. Responses were seen across most molecular subgroups.

However, what we have started to see now with long-term follow-up — this was data presented by Keith Pratz at the ASH meeting in 2022 December and then published recently earlier this year with the median of 3.5 year follow-up, 43 months — what we see is that the median survival at 3 years with aza/ven combination is about 24% as compared to a median survival at 3 years of 8% with aza alone. So, yes, this is better, 24% compared to 8% 3-year survival; however, it’s also not optimal. We do not want to be at 24% survival at 3 years in the future and so there are efforts to see how we can improve on this. And also important to realize that aza/ven improved survival but on its own may not be truly curative for a large population of patients and we do continue to see relapses occurring with this treatment out into the third and fourth year of treatment.

So this has led to a lot of efforts to understand who are the patients who seem to benefit the most and have the most durable responses with aza/ven versus those who may have some response benefit but the responses are short-lived and they relapse earlier. And what we found is based on a molecular analysis that was done on the Phase III VIALE-A study there were actually 3 groups of patients who had different degrees of benefit with aza/venetoclax. So the first group is the one that seemed to really have minimum, maybe even no benefit with the aza/venetoclax compared to azacitidine. These were TP53 mutated where the median survival was only 6 months. This is actually similar to median survival of 6 months seen with azacitidine or decitabine alone. So this population really needs new treatments, new combinations with or without venetoclax. The middle group is those who have signaling mutations. So the FLT3 and RAS mutations without a TP53. These patients had some improvement in survival about a 12-month median survival better than aza alone which is 8 to 9 months but definitely needs improvement. And then you have the others who have none of those 3 difficult mutations, no TP53, no FLT3, RAS. This is usually enriched with IDH1, IDH2, NPM1 mutated and here you see the median survival is better at 27 months. And I think this is really important because this will be kind of the ground work on which we can then build combinations with targeted therapies, immunotherapies to know within each molecular risk group how we can further improve on that.

This has now also led to a better understanding of resistance of venetoclax. So a lot of this work was done by Andrew Wynn, the group in Australia, as well as our group here at MD Anderson. We have now understanding of some of the major mechanism of resistance. These include mutational resistance such as TP53 mutations or mutations in activating kinases like the RAS-MAP kinase or FLT3 but also morphological changes such as monocytic differentiation which seems to be resistant to venetoclax or upregulation of other anti-apoptotic proteins such as MCL-1, BCL-XL.

And the nice thing is that we actually now have numerous combinations that could be used to overcome some of these resistance. I will mention one because this paper was just recently published this year looking at the clinical combination of azacitidine/venetoclax with FLT3 inhibitor. This was based on very strong preclinical rationale that resistance to FLT3 inhibitors often occurs through multiple different on and off target pathways including RAS-MAP kinase, BCL2 upregulation, and on-target resistance mutations.

And so we then took this into the clinic in a combination study of aza/venetoclax with a targeted highly specific potent FLT3 inhibitor, gilteritinib. This data was published in the JCO actually earlier this year and what we see is that the response rates with this combination are much better than what we get with aza/venetoclax and FLT3 mutated. Historically the response rate is a CR rate of about 30, 35% and now we’re seeing CR rates of above 90% with the addition of gilteritinib to the backbone. Of course, we had to dose modify the venetoclax down to 14 days to allow for the addition of gilteritinib without significant myelosuppression and we are now seeing good count recovery. And remember these are older, unfit patients so to see a median overall survival here at 2 years of almost 70% is much better than the 25%, 30% we have seen with the VIALE-A.

So this has now led to randomized studies looking at this combination of aza/venetoclax/gilteritinib versus aza/ven and I think this will hopefully establish a new standard of care in the future especially for some of these FLT3 mutated. And this is just one of the efforts. Similar efforts are ongoing with the addition of IDH inhibitors, antibody drug conjugates, immunotherapies to the aza/ven backbone to boost that 24, 25% 3-year survival to something higher, 50%, 60%, 70% in the future.

Now at the same time there’s also an effort to see if we can get away from IV therapy altogether. We all as physicians and patients like oral therapy especially if it can be delivered in a safe and effective manner. So this has led to a combination of oral decitabine with venetoclax in older, unfit AML. So the background here is that there is an oral/oral agent called oral decitabine which was approved by the US FDA for front-line, high-risk MDS patients but seems to have a PK equivalence, almost 99% PK equivalence to IV decitabine. And so it was a logical step to say, could we then use this oral decitabine instead of the IV HMA in combination with venetoclax in the older, unfit AML? So we did this study at MD Anderson looking at 52 patients, 42 of these were front-line. These were definitely high-risk, older, unfit patients, median age close to 80 years, so even older than what we have seen in the VIALE-A study, and we did have a small population of relapsed refractory AML. Importantly, this study also allowed patients with secondary therapy related AML which were excluded from the VIALE-A. So this is a higher risk population, older age, also enriched with secondary therapy related AML than the VIALE-A population. And we gave the oral decitabine per the label dosing that’s approved in MDS which is 5 days of the oral decitabine in combination with venetoclax given for 21 days with a bone marrow done on day 21. If that marrow showed remission or clearance, we would interrupt therapy to allow for count recovery. So overall we see that the CR/CRi rates are quite similar, about 60 to 70%, very similar to what has been seen with IV azacitidine or IV decitabine with venetoclax and of those who had a MRD, measurable residual disease assessment, it was negative in about one-third of the patients. Median overall survival is close around 13 months, a little lower than the 14.7, 15 months with the VIALE-A and this is probably attributable to the older age of this population as well as the allowance of secondary therapy related AML which were not allowed in the VIALE-A.

In general, a well-tolerated combination. It looked like what we see with aza/ven, some cytopenias, some mucositis was seen. We did see infections, febrile neutropenias, bacteremia. These were all generally manageable, but again, just highlighting that because this is oral/oral did not make it safer than what we see with IV HMA/ven, it’s just easier and logistically better but you still have to expect those same febrile neutropenias, bacteremias, sepsis, cellulitis infections and either prophylax or manage those appropriately.

But all in all, I think, suggests that this combination of oral/oral seems quite similar and equivalent to the IV HMA/venetoclax approach.

DR LOVE: So just a couple of follow-up questions. First of all, are you able to access this regimen outside a trial setting right now? Are you using it? And if not, would you like to?

DR DAVER: I don’t think we can access this yet outside of the trial regimen. We are enrolling most of the patients who are getting this on this ongoing clinical trial. There is also a larger multicenter clinical trial that is close to completion looking at this combination and we hope that that trial will support a label expansion or an NCCN guideline at which time then hopefully this could be used in a routine practice. I think there is one important thing to be aware of is that even though the decitabine is oral yes it still is myelosuppressive, the combination. So one should not think that just because it’s oral this has now become much safer. Actually the safety is the same and these patients still need monitoring. And one of the concerns that we have in the academic setting sometimes is that because its oral/oral physicians or patients may get the feeling that we can just take this pill, come back in 2 to 3 weeks. That’s not the case. These patients still need lab work, usually daily for the first 4 to 5 days for tumor lysis, cytopenia and then after that twice a week or so. So oral more convenient logistically easier but still needs to be monitored as we would do with IV therapy at least for that first cycle.

DR LOVE: I’m kind of curious if you sort of take a step back and just think about these patients confronting the diagnosis and all the financial, social, all the issues that come up, putting aside the lab data, what have you observed in terms of quality of life?

Obviously you would think that not coming to clinic would be beneficial but I’m just kind of curious in view of all the other issues they’re dealing with is it really a substantial benefit or maybe not that big a deal?

DR DAVER: Yeah, I think that’s a great question. I think the benefit is more in the subsequent cycles. The first cycle and we at MD Anderson who have probably done this combination the most today still admit these patients because the toxicities that happen are not different with oral and IV. We still see cytopenia. We need transfusions. We need to do hydration, tumor lysis. So for the first cycle effectively it may not matter to the patient. They are going to be in the hospital for 28 days. So it’s oral, IV it’s not going to change things. However, for subsequent cycles I think it’s a nice advantage to them because they can be outpatient. They don’t have to come for those 5 consecutive days of IV decitabine or 7 consecutive days of IV azacitidine and they get much more time away from the clinic and from the infusion center. So I think it helps their quality of life, their expenses, their travel, their logistics much more in cycle 2 and beyond.

DR LOVE: But are you saying that you generally put people in the hospital for 28 days to get aza/ven?

DR DAVER: Yeah, yeah. We’ve actually always put patients in the hospital. Older, unfit patients, remember these are 70 or 75+ patient populations and again this is not something necessarily done across the country but that’s also because of the patient profile. So the majority of our patients come from long distances from far away areas in Texas and Louisiana and Alabama and it’s really hard for them to come drive 3 times a week for blood work from 7, 8 hours away. So because we have such a wide catchment from outside of Houston, we often admit them for the initial hydration, tumor lysis, blood count monitoring, transfusions for cycle 1. After that, we do everything outpatient. Now if a patient were living in Houston, local, very compliant, can come for labs routinely, we may consider not doing that but in general that has been our approach and it’s been quite safe.

DR LOVE: I always assumed that in general in the community setting and other academic settings that people generally don’t get admitted for azo/ven. Am I wrong?

DR DAVER: Actually, I think most academic settings do admit them. They may be for different periods of time. Some admit them maybe for 2 weeks, some admit them for 4 weeks. But most academic settings that I’m familiar with, especially for those older than 75 or 70, actually are admitting the patients initially. I don’t know for sure in the community. I think some of them do admit for the first cycle, some don’t. But this is only for the first cycle and many do admit. In fact, the recommendations for most guidelines are to admit for that first cycle including from the NCCN.

DR LOVE: One other thing I was going to ask you, when you were talking about FLT3 and you were starting to get into the mechanisms of resistance I was sort of triggered a memory. One of the early studies, I can’t remember if it was sorafenib or midostaurin, where it looked like they saw benefit in FLT3 unmutated and is there any reason to think that people who at least start out unmutated might benefit from these drugs?

DR DAVER: Yeah, this is a great question. There is data, actually very recent data, that has come out with quizartinib, in fact, in the wild type FLT3 mutated patients and this is something we had seen initially in the Phase I about 10 years ago whereas a single agent quizartinib had activity about 30%, in fact, overall response rate in relapsed refractory FLT3 wild type. And then recently the Spanish group have looked at quizartinib in combination with intensive chemo in front-line newly diagnosed patients who are FLT3 wild type and showed actually a significant survival advantage with quizartinib added to intensive chemo versus placebo intensive chemo in a wild type population. So this is something that is being explored further with larger planned studies coming, but we do think that this may be due to 2 things. One, there may be off-target other mutations or other pathways that are inhibited by quizartinib. We know, for example, Cbl is one of them, maybe PDGFR or others. And the second is we actually do know that even in the FLT3 wild type without the mutation there is upregulation of FLT3 ligand. This is lower than what you see in the FLT3 mutated. It’s about 10, 20% compared to what you see in a mutated FLT3 but it’s there. And so if you have a very powerful FLT3 inhibitor it may actually be able to suppress that FLT3 ligand expression even in the wild type and get you some activity. So I think these drugs may not be truly just FLT3 inhibitors in the future. They may have a wider scope.

So moving on then, how do we move beyond just the HMA/ven? We are getting survivals of 14 to 15 month median but the 3-year survival as I mentioned is 25% or lower. So one approach could be to add targeted therapies for those who have a targetable mutation. So potentially adding a FLT3 inhibitor like gilteritinib or quizartinib or adding an IDH1 or 2 inhibitor or potentially adding antibody drug conjugate. But what about those patients who do not have a targetable mutation like FLT3, IDH1, IDH2, MLL, NPM1? This is still about 60, 65% of the patients. So for this group, our group led by Dr Kadia has made an effort to look at new backbone combinations that may further enhance both response as well as depth of response, MRD clearance, and overall survival in this older, unfit population. This regimen is called the cladribine low-dose cytarabine or triple-nucleoside regimen. And here, you see the most recent data that has been presented at the ASH meeting last year where this combination has shown very striking response rates, a CR/CRi rate of 92% with a true CR rate of 75%. Remember with aza/venetoclax VIALE-A, the true CR rate was 37% so this is quite a significant higher CR rate and MRD negativity was 81%. But it is a single arm study so that does need to be confirmed. However, given the number of patients we have treated now with close to 120 patients and the survival continuing to look much better than what we have seen with HMA/ven, this has now become our regimen of choice for patients who are 60 to 80 years of age in the front-line setting who do not have a targetable mutation like a FLT3, IDH1, IDH2, MLL, and PM1 and the 2-year survival is about 65% compared to 25, 30% as I just showed you for the HMA/venetoclax and hopefully there will be randomized studies starting in the future to confirm this survival and response benefit.

Just showing that the response seemed to be across all different molecular groups and there was a robust survival in both ELN intermediate favorable but also in the ELN adverse where historically HMA/ven has not done as well with median survival less than 10 months. Here, the median survival was 24, 25 months. So I think this may be a new regimen emerging that could have a higher activity response, depth of response, MRD, and potential survival benefit especially in that intermediate adverse group than what we have seen historically with the HMA/venetoclax regimen.

Now the other question I get a lot from doctors both in community and academics is, how long do we continue aza/ven? Is this a continuous therapy like blood pressure medications that you take it for the rest of your life if you have hypertension? Or is there a way to stop this treatment at some point? And if there is, what molecular markers or MRD markers or response markers can be used to choose patients most suited to stop? So this was something that the French group led by Dr Sylvain Garciaz looked at in a retrospective analysis across the FILO cooperative group centers as well as the US Moffitt Center contributing some patients to this. This is across the last 5 years. These were patients who were adults who received aza/ven and were in a remission, either a CR, CRi, MLFS, and patients should have received at least 1 or more cycles of the treatment and only those who had discontinued treatment more than 3 months before the time of the data cutoff were included.

And the key questions here were, number 1, what is the profile of these patients who stopped? What you see is that these were predominantly front-line patients which makes sense because in salvage setting, we are much more nervous to stop something we know inherently this is a more difficult biological group. Patients had a median of 4 but some of them up to 15, 17 cycles. They all had some form of response. Majority were actually a full CR, 80%, a few of them with MLFS, and a lot of them, almost 78%, had MRD by either a flow or molecular analysis. And the reasons for discontinuation, I think this is quite interesting and something that matches what I have seen in practice, were predominantly due to prolonged cytopenias, what they’re calling hematological toxicities. It’s almost 60% of either the patients or the physicians decided to stop treatment because they were finding that the counts were not recovering putting patients at risk of infection, bleeding, other complications, and a few others this was due to patient preference or due to comorbidities, weakness, and fatigue. And this is similar in the front-line and the relapsed group.

So what was the outcome of stopping? Was there benefit? In fact, yes. Correction of cytopenias was seen in 60% of those patients, so a lot of those patients, 58% stopped due to cytopenia and a large proportion, two thirds of them, seemed to have recovered their counts after stopping aza/ven which biologically makes sense. Once you’re off the venetoclax, we do see a majority of the patients will recover their counts within the next 2 to 3 months. What was really interesting I think is the median overall survival. So here, the median overall survival after stopping in these patients with CR/CRi, many with MRD negative, was 44 months. So, in fact, remembering that these are 75-plus patients, this is quite good, 3- to-4-year survival even after stopping. And again, retrospective analysis, so it’s suggesting, not definitive, but suggesting that there may be a group of these patients with deep responses who have achieved a full CR who may be able to stop treatment at some point in time with a good maintained overall survival as was seen in the study. Now the relapsed is a little bit more tricky. Here, majority of the patients do seem to have relapsed. Median overall survival is lower and so you may want to be more cautious.

So the conclusions here were, is the discontinuation of aza/ven in patients with a CR/CRi, especially those who are MRD negative, may be feasible especially in those who are having prolonged hematological toxicity and there is reversible of the cytopenia after stopping the treatment. A very encouraging OS was seen in the front-line population of 44 months in that specific group of patients who stopped but I think this needs to be looked at in a prospective trial and I do think that this is something that is being looked at by the French group and potentially by others in the future.

So now moving on to other combinations in venetoclax. We talked a lot about the HMA/venetoclax, cladribine/low-dose cytarabine/venetoclax, or HMA/venetoclax triplets adding FLT to the IDH1, IDH2 menin inhibitors. But what about for younger, fit patients? Because all of those are focused on older, 70, 75+, or those with severe comorbidities; however, still, about 50% of our patients are below 65 years of age and fit, can receive intensive chemo. However, intensive chemo alone gives us a 3-year survival for about 35, 40%. Can we improve on that? So to this effect, we started looking at combinations of adding venetoclax to the backbone intensive chemo that we have been using at MD Anderson, either the CLIA, cladribine/idarubicin/cytarabine, or the FLAG-Ida which is fludarabine on a similar backbone of idarubicin/cytarabine with venetoclax added for a 7 day period only to avoid very severe myelosuppression when added to backbone intensive chemo.

So these data were published at the end of 2022 and updated last year. What we see is overall, the response rates are very striking across the board, 86% with the addition of ven to intensive chemo compared to 60% with intensive chemotherapy alone. A large number of these were MRD negative responses, and what we see is that with transplant, there does appear to be a survival benefit with intensive chemo/ven versus intensive chemo. But even in those who did not go to transplant, there appears to be that survival benefit. So this regimen seems to be very potent, the addition of ven to intensive chemotherapy.

And here just looking at the event-free overall survival on a propensity match analysis within our population at MD Anderson when we compared intensive chemo/ven, 64 patients to those who received intensive chemo alone in the past 162 patient, we are seeing a signal for event-free overall survival improvement. So another regimen that is going to be taken forward into randomized Phase III studies. A number of these are actually now being developed and started but suggest that ven may not be restricted to older, unfit population. It may have benefit and implication even in younger, fit patients.

So that brings us to one of these studies that kind of emerged from this data set that we had started to put out from MD Anderson. This is from the SAL group in Germany led by Dr Christoph Rollig. The data was presented by Leo Ruhnke at the ASH meeting a few months ago, December 2023.

The study is called the RELAX study. Here, again, they are using an intensive chemo backbone adding venetoclax to it. The backbone they use is slightly different. This is called the HAM regimen which is looking at cytarabine and mitoxantrone given for 3 to 7 days with venetoclax added for 14 days. So the same concept, adding venetoclax to intensive chemotherapy; however, different chemotherapy backbone here with cytarabine and mitoxantrone.

First the study population. This was in fact just the Phase IB part. The study is now moving on continuing in Phase II and there will hopefully be more updated data coming in the future. But these were younger patients. These were a median of 26 to 74 with a median age of 54 years. So this is not the older unfit 78, 79-year-old population that we talked about with HMA/ven, and these were predominantly relapsed but also some primary refractory AML with a number of them having received previous induction chemo or previous transplant and about half of these would fall in the ELN adverse. So this is kind of like a real world population that we would see based on the molecular ELN and prior therapies.

Overall, the regimen was quite tolerated. I will say, one of the big concerns that people have is how can you tolerate these intensive chemotherapy with ven regimens? But I do think with appropriate reduction of the venetoclax to 14 or even 7 days and with good academic or centers that are used to managing neutropenia infection, count recovery, sepsis, transfusions, this is feasible. This was something that they were able to show in this cooperative group study across 30, 40 centers that, yes, there was febrile neutropenia, infections, sepsis, but these could be managed and the 60 day mortality actually was only 5% which is similar or even lower than has been shown in relapsed setting with intensive chemotherapy. So I do think feasibility was shown.

And then of course the key question is, what is the efficacy of this regimen? And this was actually really quite striking. The CRc rate, remember, these are relapsed refractory patients, was 82% with half of these as a CR, half as CRi, and especially in those with ELN favorable intermediate, there was a very, very high response rate. So all in all, for a relapsed population, this is really quite encouraging. This is very similar to the FLAG-Ida/ven relapsed data that our group had published a couple of years ago showing CRc rates of about 70, 75% and this is kind of a confirmation of that data.

And the overall survival now at 2 years is 60%. Again, this is something we have not seen in relapsed refractory patients except for with intensive chemo/ven based regimens. So I do think this is quite a nice signal, but again, only 38 patients so we do need to see more data and as I understand it, the SAL is expanding this further.

So what other options could there be for adding to intensive chemotherapy? Is it just venetoclax? In fact, we just discussed this briefly. Dr Love brought it up. Could we look at other drugs that could enhance activity in a non-targetable mutation population? And one of those that’s quite interesting is a drug called quizartinib. So here, what I’m showing you is actually the survival curve from a study called the QUIWI study that was presented by Pau Montesinos at the European Hematology Meeting in 2023 where they added quizartinib which we think of traditionally as a FLT3 inhibitor but they added it in non-FLT3 mutated patients to the backbone of 3 + 7 intensive chemo and the comparator was placebo with 3 + 7. And in this non-FLT3 mutated or FLT3 wild-type group, they showed quite a striking overall survival difference at 2 years and ongoing with the addition of quizartinib. So this is kind of a little bit of a surprise. Not something that I think we all expected given we thought that quizartinib is mainly a FLT3 inhibitor, but appears there may be additional impact of quizartinib beyond just hitting the FLT3 mutated population. This is just showing that especially in the intermediate ELN, there was a clear benefit of adding quizartinib to 3 + 7 versus placebo 3 + 7. And this is something that is being evaluated further.

In fact, this is a point I would like to make and I think will be discussed by my colleague, Dr DiNardo, as well is there is now an effort to generate something that I think is quite cool called a FLT3-like signature. We have seen this done in ALL where we have a Philadelphia-like signature which identifies patients who don’t have the BCR-ABL or Philadelphia chromosome but on RNA gene expression have a profile that exactly looked like a BCR-ABL rearranged and in some of those patients, using BCR-ABL inhibitor works. And so the same analogy here is that these patients don’t have FLT3 mutation but they seem to have a gene signature using RNA NanoString that matches exactly FLT3 mutated and were also the same patients who benefitted with an OS with the addition of quizartinib FLT3 inhibitor. And these could make up up to 30% of the non-FLT3. So if you say 30% of patients have a known FLT3 mutation, another 30% could have a FLT3-like signature benefit from quizartinib. That could mean that close to 50, 60% of front-line AML could benefit from the addition of such a kinase inhibitor like quizartinib which could be very meaningful.

DR LOVE: So was this trial a Phase III?

DR DAVER: So it was actually, they called it a randomized Phase II but it was a pretty big study. It was 250 patients, like 120 on each arm. But it was done only in Spain. It was a Spanish PETHEMA group study so we don’t know yet how this will impact the NCCN, the FDA, but there may be other larger similar studies globally that could come in the future is what we think.

DR LOVE: Are you aware of any ongoing Phase III study right now looking at it?

DR DAVER: Not yet. No, not yet. But I think hopefully in the future. But we do think that this data is pretty striking and as you saw, the survival difference is quite large so I definitely think that it would be great to do a confirmatory study and I think a very good chance that study would be positive. There was nothing in the Spanish data they said that raised concerns. The population seemed to be very real world like. So, yeah, I think hopefully it would be nice and that could then expand the use of a drug like quizartinib tyrosine kinase inhibitor from 30% currently to maybe 55, 60% and benefit some of these FLT3-like patients. I think it would be very nice to get this signature used in the future and so along with doing the studies, it will also be nice to see if we can actually develop an RNA signature that could then be CLIA validated and used in the future kind of like what we do for Philadelphia-like ALL.

DR LOVE: Fascinating.

DR DAVER: It is very interesting, yeah. Okay, so now moving away from venetoclax and targeted therapy approaches to immunotherapy, an area that I’m really interested in and an area that is still in its infancy in AML but I think it will continue to grow a lot of interest, a lot of effort, a lot of pharma, a lot of investigators moving there. So the first approach is using antibody drug conjugates. This is something that has been successful across many heme malignancies, ALL with inotuzumab, multiple myeloma with BCMA targeting, lymphoma with CD20 targeting, and AML with CD33 targeting, gemtuzumab already approved. However, we think that there is an avenue to improve beyond gemtuzumab with the use of new antibody drug conjugates using a better payload, using better linkers, and better cellular uptake. One of these is a drug called pivekimab, also previously called IMGN632. This has now been evaluated in combination with aza/ven. It was initially done as a single agent in relapsed setting where it showed about a 25, 30% CR/CRi rate as single agent which was not thought to be sufficient to go for a full registration so we decided to move it up-front in combination with HMA/ven to see if we can improve that depth of response, CR rate, and OS. The nice thing about this agent is the safety profile. It’s actually been very well tolerated and compared to other CD123 antibody drug conjugates that are out there such as tagraxofusp or bispecifics such as the ones we have used in the past, flotetuzumab, we see very limited capillary leak syndrome, severe edema, cytokine release, or VOD. So I think that really separates this agent in our mind from the others. The safety profile compared to all of the other ADCs in AML seems to be favorable. And we are seeing good responses as you see here. 50 patients were presented at the ASH meeting with an MRD negative CCR which is CR + CRi rate of 76% which looks better than the 40% or so CCR MRD negative rate that has been shown historically with HMA/ven and responses were seen in those 3 populations, lower benefit, intermediate benefit, high benefit at a higher rate than what has been shown with HMA/ven. But eventually, this will need to be confirmed in a randomized study.

We’re still enrolling with the triplet in the front-line. We hope to present updated data at the ASH meeting this year with a larger front-line single arm data set and then hopefully eventually move to randomized studies. This drug is also being evaluated by one of my colleagues, Dr Pemmaraju, in front-line BPDCN, blastic plasmacytic neoplasm, and we think that that may be an area where it could actually have a path to quicker registration because it’s very safe, it does not have capillary leak or CRS that has been seen with the other approved agent in BPDCN and may actually get there quite quickly through Dr Pemmaraju and our efforts.

And then lastly about immunotherapy, I’ll talk about another CD123 NK cell engager. This is a Phase I, very early. There has been a lot of efforts to develop bispecific engagers or immune checkpoints that can harness the body’s innate or adaptive immune system to attack the leukemia cells very similar to the successes seen with solid tumor immune checkpoints or bispecifics in multiple myeloma or in lymphoma. And a number of those using T cell targeting bispecifics in AML unfortunately failed either due to lack of efficacy or due to severe toxicity, cytokine release syndrome, and were aborted. However, now we have moved away from T cell adaptive immune harnessing in AML to more of NK cell or innate immune system harnessing. And this is one of those first approaches using an NK cell bispecific that targets CD123 which is a common antigen on leukemia cells along with a NK cell target such as CD16.

And this data was presented at the ASCO meeting in 2023 by Dr Anthony Stein from City of Hope again here showing that this is what we call a TriKE or a triple engager that targets CD123 on the surface of the AML leukemic blast and then CD16 to engage the NK cell which is a component of the innate immune system as well as working on an NK activating receptor at the same time, so not only engaging the NK cell but activating it through the NKp46. So this approach scientifically sounds like it would be very sound, has now gone into a clinical trial in the Phase I and very good to see. We are seeing single agent responses. This is kind of the swimmer plot of responses across dose levels 1 through 8. Overall, the responses were seen mainly in the dose levels 3 and 4 as you can see as well as dose level 8. There were 2, 1, and 2 responses, CRcs, which you can see on that bottom row. And interestingly, those patients were treated at the higher dose levels which were considered to be 1,000 microgram per kilo infusion. We are seeing 5 out of 16 confirmed CR/CRi which is pretty encouraging. Of course, early and we need to continue to see this data. The Phase II is ongoing with expansion. I anticipate it will be presented at one of the large meetings in 2024 either in the summer or ASH later this year and we’re also looking at a number of immune correlates as well as future combinations to further enhance this NK engager whether it’s to be combined with NK cells or combined with venetoclax based approaches. So I think there may be a lot we hear about this hopefully in the near future.

DR LOVE: So do you see like CRS and other CD3-like side effects with this thing?

DR DAVER: Oh, yeah, we do. So to your point, Dr Love, yeah, we do see CRS with this and this was also seen at the doses that were active not surprisingly, so once we got into that 1,000 mcg or higher, we started seeing CRS in patients. It seems to not be a super high rate but it’s there. Now infusion related reactions are seen. These can be seen in 50 to 70% and we have seen some of these that can be Grade 3 like so you do have to monitor these patients whether they’re admitted for a couple of days or close outpatient ambulatory infusion monitoring and if we start seeing any infusion reaction CRS, we usually start a steroid quickly or we’ll start with diphenhydramine/acetaminophen approaches and that seems to have worked in all the patients. I haven’t seen any where we could not control the infusion reaction CRS. But exactly as you’re asking, that is the expected and on target toxicity to be managed which it seems to be manageable.

Now just a couple of other interesting things. We didn’t talk too much about MRD, measurable residual disease or minimal residual disease depending on the publications you look at, but I think this is becoming something more and more in the forefront and interesting in the acute myeloid leukemia space. So I thought this study was really quite interesting. This is now looking at molecular MRD using a high sensitivity PCR for NPM1 which is probably the most established MRD marker among the molecular markers in AML. And these authors actually were able to pull data sets from 2 large UK AML17 and 19 studies and showed that in patients who have an NPM1-positive after induction chemo, induction was daunorubicin cytarabine based, there was a clear benefit for allotransplant. So NPM1 was a molecular MRD marker that could predict those patients who clearly needed allotransplant and benefitted from it whether it was in terms of overall survival, relapsed-free survival, or relapsed. On the other hand, and I think this was the most interesting thing, in patients who did achieve NPM1 MRD-negative after induction consolidation, there did not seem to be any benefit from allotransplant whether it was in terms of overall survival. There was some benefit in relapsed-free survival but I think the point here is that even though you get more relapses without transplant, NPM1 is sensitive and you can salvage them effectively and so the OS in the end seemed to be similar. But basically highlighting that in the future, there may be strategies to select patients who need transplant using NPM1 as an MRD marker.

Also interestingly, one may ask, well, what about if they had a concomitant FLT3? Because this is seen in about 50% of NPM1 mutated. Did this matter? And overall, it did not seem to matter. If you are NPM1 MRD-negative, those patients seemed to have a similar OS whether or not they had a concomitant FLT3 mutation.

I think the key limitations of this study and which is why this is not something that we plan to implement right away is that a lot of this data was generated before the era of FLT3 inhibitors. These studies were 7, 8 years old and also we don’t have a FLT3 MRD assessment on this so we’re not really sure how we would do that because today we are adding FLT3 inhibitors and using FLT3 MRD assessment for most of our patients.