Naval G. Daver, MD, discusses treatment options and the data these options are based off of for the management of acute myeloid leukemia, based on a case scenario of a patient with <em>FLT3 </em>internal tandem duplication –positive acute myeloid leukemia.
Naval G. Daver, MD
During aTargeted Oncologylive case-based peer perspectives program, Naval G. Daver, MD, discussed treatment options and the data these options are based off of for the management of acute myeloid leukemia (AML). Daver, an associate professor in Department of Leukemia at The University of Texas MD Anderson Cancer Center in Houston, explained disease management and the factors that go into treatment decisions during the dinner event based on a case scenario of a patient withFLT3internal tandem duplication (ITD)positive acute myeloid leukemia (AML).
A healthy 48-year-old man visited his primary care physician for flu-like symptoms that had lasted for more than 2 weeks. He was not taking any medications for any medical history.
His laboratory results revealed a white blood cell count of 65,000 x 109/L (90% blasts); platelets, 85 x 109/L; and absolute neutrophil count, 1.5/mm3. His liver and cardiac functions were within normal limits. His bone marrow showed 65% blasts. He was calculated to have a normal cytogenetic risk status by European LeukemiaNet [ELN] classification.
What are your initial impressions of this patient? Is this a typical presentation?
This is a 48-year-old man with some flu-like symptoms, which lasted for a couple of weeks. He had no significant past medical history or medications; high white blood cell count, 65,000 high-blast platelets in the absolute neutrophil count on the lower end; and normal cardiac and liver functions. This is a very typical kind of referral that we will get.
What type of diagnostic testing do you usually do? What type of mutation testing do you order?
The first thing you do when the patient comes in is the bone marrow biopsy and assess for [myelo]blasts. We usually get the results pretty quickly for the differential, within a few hours. Then we are rushing the cytogenetics, both routine cytogenetics and FISH [fluorescence in situ hybridization] panels for core-binding factor. Most centers are doing some variation of molecular testing. Some are doing targeted testing with 5 or 6 important genes; others are doing broader 50-gene, 80-gene, or 100-gene panels.
For this patient it was sent and, let us say, in 3 to 4 days he had some molecular results back. [The results indicated that this patient was]NPM1mutated,FLT3positive, and some of the other mutations that were sent were negative. [It is different] from lab to lab, some of the labs like NeoGenomics or Foundation Medicine, only give youFLT3positive or negative, and some of the academic institution labs are giving you allele burdens. It is kind of questionable at this time what exactly you are going to do with those.
What we do usually when a patient comes in is we do rapid FISH for [chromosomal translocation t(15; 17)], to rule out acute promyelocytic leukemia. [We also test for t(16;21), and the presence of t(8;21) or inv(16), which would indicate core-binding factor AML]. Then we send all of this in parallel because if you do it sequentially, then you are going to lose a lot of time.
Molecular testing revealed a NPM1 mutation and FLT3-ITDpositivity (low allelic ratio). He was wild type for TP53, ASXL1, and RUNX1.
If this patient tested positive for core-binding factor AML, what would be your choice of therapy?
I think if you are going to use gemtuzumab ozogamicin [Mylotarg], the best indication is in patients with core-binding factor AML. I think addition of gemtuzumab in patients with core-binding factor really adds 15% to 20% absolute. So, I probably would do it and so you pick those people out separately.
What induction therapy would you offer this patient?
Basically, you are looking at a patient with anFLT3mutation. That is the critical thing here when you are thinking about targeted therapy, which is really what we are thinking about predominantly in AML therapy. The question is if you can incorporate an FLT3 inhibitor. For that you have to get the test back, and there has been a lot of improvement in testing time. We did a kind of a research analysis with Novartis a few years ago, predominantly in the private setting, and only about 55% to 60% of all practices were sending outFLT3because at that time no drug was approved. Most recently, when we looked at it a few months ago, about 80% to 85% are checkingFLT3. Even the timing used to be about 8.5 days, and now it’s down to 6 days. And, the FLT3 inhibitors are added on day 8. So, even if you get the testing result back by day 5 or 6, which seems to be the average in the community setting, I think you are okay and you can add the midostaurin after starting the 7+3 induction treatment.
I think that the FLT3 inhibitors work through multikinase pathways, and they work in many different situations, [but not in all people]. I think this is where people need to do more trials of Vyxeos [daunorubicin, cytarabine] plus some of the FLT3 inhibitors, not just midostaurin. Probably the better ones that are coming along are quizartinib and gilteritinib [Xospata]. But I think the point is that it is not common to have adverse cytogenetics inFLT3. Usually if you have anFLT3mutation, I think a lot of people would go ahead with FLT3-based therapy. Ninety percent of the time you are not going to have adverse cytogenetics, so this is not something commonly found. Sometimes, you are going to have that problem and then this may be a good person to go for a clinical trial.
The patient was treated with 7+3 (cytarabine plus anthracycline regimen) and midostaurin (Rydapt). Upon count recovery, his peripheral blasts were approximately 40% on day 36 post induction. Bone marrow testing confirmed that the patient had refractory AML.
Do you factor inFLT3allele burden when making your decision?
I think for selecting an FLT3 inhibitor, the allele burden is not showing much difference. People with high allele burden do poorly, but they actually derive benefit from an FLT3 inhibitor and that would lower the burden, as well. I think that the transplant question is probably more driven by the allele burden. I think the ELN data are correct that if a patient has a lowFLT3allelic ratio, especially if they have anNPM1mutation, those people can do almost as good as patients with core-binding factor AML.1In that group, I would use an FLT3 inhibitor, just like how we would treat with gemtuzumab in the patients with core-binding factor AML.
What data contributed to the decision to offer this regimen to this patient?
The phase III RATIFY study took people with newly diagnosed AML with anFLT3mutation, aged 18 to 60 years and randomized them to receive either induction 7+3, the standard before the FLT3 inhibitors, or 7+3 with the addition of midostaurin at 50 milligrams twice a day, given on days 8 through 21.2
This regimen comes based off of phase II data showing that giving continuous midostaurin from day 1 to 21 or 1 to 28 was not tolerated because of gastrointestinal toxicities.3And so, they looked at 6 different schedules and the one that seemed to be most tolerable based on gastrointestinal adverse effects was the 8 through 21 [regimen] where you give 7 days of cytarabine, and add the midostaurin. You added this both in induction phase, during the consolidation phase, and then also during maintenance, which was included as part of this trial for 1 year of maintenance in these patients.
There was not a big difference [between arms] in the RATIFY patient population. One of the things that people do see is that there is a slightly higher female predominance in the placebo group. This was randomized, blinded to both the physician and the patients, so that is just one of those things that occurred. These are all younger patients, <60 years, with a median age of 47 to 49 years of age. And they had a mix of ITD ratios. And then there were some people who had 2 types ofFLT3mutations: the ITD, which is the common one and makes up about two-thirds of allFLT3mutations or more; and then TKD, which is the less common one. Midostaurin actually targets both the ITD and the TKD mutations. We use it in people who have either one.
The overall response rate is slightly improved but has not reached significance. I think based on the response, we would not have been very excited to continue to use this drug. It is more the survival data [that is of interest]. The good thing was that the blood count recovery was not delayed. This has been a big problem in AML with a lot of other therapies. When we added drugs such as gemtuzumab or others, we saw that there was significant cumulative or additive myelosuppression, and blood count recovery was significantly slowed down or prolonged. But, here it is about 35 days to achievement of response with full complete response [CR] count recovery.
The way the trial was done [is that investigators] had to code responses by day 60. When you look at people who got re-induction or had a delayed response, which occurs in about 10% to 15%, then you do see that there is a more significant improvement with the addition of midostaurin for the response rates.
If you look at the phase Ib trial data that were published inLeukemia,where they include CR/CRis [CR with incomplete hematologic recovery] it is about 80% to 82%. It is much higher. But in the phase III trial, it is only the true CRs, and this was an intent-to-treat population. At 60 days, if you are not [determined by investigators to have a] CR, you were not collected for the primary endpoint.
The trial met its primary endpoint [of overall survival (OS)], but clinically, I think this highlights the point that targeted therapies improved outcomes, they are safe, and they can be given. But, I think this is really a starting point. I do not think 5 to 8 years from now we are going to be using midostaurin. There are much better FLT3 inhibitors coming alongquizartinib, gilteritinib, crenolanib. And I think this was the first stepping stone. Just like we start with thalidomide [Thalomid], then you might move to lenalidomide [Revlimid], and so on. At this time, this is what is approved. This was the first drug actually approved out of the 8 drugs that have been approved in the last 2 years in AML, which is quite amazing because we had almost no drugs approved for 40 years. And now, you have 8 new ones.
The other thing that was seen was whether you have ITD high, ITD low, or TKD, there was an equivalent benefit seen in all the populations. [Therefore], you do not want to restrict use of midostaurin in this population to only people who have highFLT3-ITD, or those who have ITD but no TKD. Routinely in the frontline setting we do add it, whether they have ITD or TKD, to the induction treatments that we are using.
What is the safety profile like with this regimen? What toxicities do you commonly see with patients receiving midostaurin?
The tolerability profile is actually pretty good. The only adverse effect that was seen that is at a slightly higher rate was rash. And clinically I somehow have not seen too much rash, but in the study, they did see about 10% to 15% incidence of rash. It usually got better with steroids, it was not a big issue.
The other thing that we have seen which is rare, maybe in <1%, is pulmonary toxicity. We have seen some patients who develop a COP [cryptogenic organizing pneumonia] like a cough or BOOP [bronchiolitis obliterans with organizing pneumonia], and it usually happens after a few weeks or so of midostaurin. Initially, we thought maybe this is a fungal infection, maybe this is some other kind of an idiopathic pulmonary issue. But in fact, even in this trial, they had documented 4 to 5 patients who had this. This is important, especially in these patients who are going to be going for transplant and who are being worked up because you may think that this is related to transplant or post-transplant but actually this could be the midostaurin. It gets better if you give steroids and stop treatment. That is the only rare, but important, adverse effect to know about.
The good thing was the blood count recovery was not significantly inferior, and so no cumulative myelosuppression was seen.
The other thing I have heard is it tastes really bad. A lot of patients have told me it stinks, it is like a rat poison. So, a lot of patients complain about that. But they will take it, they do not really refuse it because of that.
Do you think the chosen therapy should be followed by maintenance midostaurin in this patient?
The difference in Europe is that midostaurin is really approved in the setting of the trial, in patients aged 18 to 60 years, and it is approved with maintenance therapy; whereas in the United States it is actually not approved for maintenance. There has been a lot of debate about that because, at the end of the day, I think it is hard to take a phase III trial and decide which part of the phase III gave us the survival advantage. I mean, for all we know, all the benefit came from maintenance. For all we know, it all came from inductions. But we do not know, so I think personally, at the University of Texas MD Anderson Cancer Center and in our group, we do not really agree with the way the FDA approves everything.
The Europeans actually approved it in a better way because you do a phase III trial and either you take it as it is, or you do not. It is hard to talk 10 to 15 years later and say, “well, we think it was from this.” The data are the data. And now, of course, there are data coming with maintenance in other settings, which are often post-transplant, that are showing benefits. So, I think the maintenance story with tyrosine kinase inhibitors in Philadelphia chromosomepositive ALL [acute lymphocytic leukemia] is evolving to be an important one in any case.
When would you do a bone marrow biopsy for this patient?
We may do it at day 21 to 28, but then if we are not getting a clear idealet us say blast counts are not coming down—we will usually wait and repeat [the bone marrow biopsy] because a lot of these people can have a delayed remission in this population.
What are your dosing concerns with this regimen?
There is clinical data here from the Burnett et al where they used higher doses of daunorubicin (90 mg/m2) and they show that, in fact their rate of OS is equal to patients receiving the standard dose.4You are comparing across trials, but it is equal to what you get with the addition of midostaurin. So, a lot of people, in Britain especially, argue that the midostaurin is basically making up the difference between 60 mg/m2and 90 mg/m2of daunorubicin. When we use idarubicin [Idamycin], and most of the data shows that with idarubicin at 12 mg/m2and daunorubicin at 60 mg/m2or 90 mg/m2, you are not getting a big difference. I personally agree with using daunorubicin and adding midostaurin. I would be anxious about the myelosuppression, cardiotoxicity, other adverse effects. In general, the daunorubicin at 90 mg/m2to me has a higher risk profile.
I think that is the whole point of these new agents is that maybe you do not need to push the bar to the highest where either you are dead or you make it out, right? You will hear about that with venetoclax [Venclexta] and others, that you have a margin that you do not have to start with the maximum kind of dosing.
What options would you consider for this patient at this point?
Okay, so this person started recovering counts but by the time he was recovering, we were waiting to get a bone marrow biopsy around day 30 to 35; and it is showing that, unfortunately, the recovery is with refractory disease.
We could make sure theFLT3does not change. In the RATIFY study, they showed that in almost 45% of people, theirFLT3changes. They become ITD positive, then they become negative. Some people acquired a different allele. So that is very important that unlikeIDH,which is quite constant, it does not change.FLT3is very dynamic. You have to checkFLT3at the time you are going to give the treatment. You cannot assume that the baselineFLT3is true.
But let us assume that this person has a persistentFLT3ITD mutations. What would be the next options be?
There are many options. I do not think there is a right answer really in this situation. I think it is fine to do a second induction, so that is one option. And I think the other is to consider moving to a second-generation FLT3 inhibitor. Gilteritinib has been approved in this setting, quizartinib will probably be approved in a couple of months.
The allele burden did not predict for response for FLT3 inhibitors. We looked at it and people who are >3%: so 3% to 10%, 10% to 50%, and >50%. The response rate is actually the same. Even though it is low, it still works.
I think the big question is whether the second-generation inhibitors should be used as single agents or in combination regimends. And that is the same thing we are seeing in the treatment of ALL with some of the agents such as inotuzumab [Besponsa], blinatumomab [Blincyto]. As a single agent, you know they meet the endpoint.
So gilteritinib had a CR/CRh [CR with partial hematologic recovery] rate of about 23%, based on the study for which it was granted approval.5,6The time to response is about 3.6 months, which is nice. But really, 23% is a pretty low response rate. No one can say “well, would you do that instead of chemotherapy,” which gives you about the same response rate. Probably yes, if I am a patient, I would prefer to have an oral, well-tolerated, outpatient therapy than get MEC [mitoxantrone, cytarabine, and etoposide] or FLAG-ida [fludarabine, high-dose cytarabine, idarubicin, and granulocyte colony-stimulating factor]. What we are seeing is that combining these with HMAs [hypomethylating agents] is probably giving you a much higher response rate. So I think over time you are going to see a lot of evolution combining these with HMAs, which is probably going to give you a more realistic cure rather than a 6 or 10-week survival benefit, which is kind of what you are getting right now.
In any case this was the approval study. This approval raised more questions than answers. Because at this time, we do not know the randomized arm or response rate. We do not know the OS rate. We do not know if the co-primary endpoint of OS has been met. Now, all of these data are there, they just have not been published or released yet. We need to get that information, I think, to make this more robust.
What would you give as a second induction regimen?
Second induction has traditionally been 7+3…and in this study, investigators gave 7+3 and midostaurin again. Now, some of the groups… go to a second induction with something stronger, FLAG-ida, CLIA [cladribine, idarubicin, and high-dose cytarabine]. We have seen a lot of patients where they will receive 7+3 on day 14 or 21. If investigators see residual disease, then they will give FLAG-ida, MEC, CLIA, [or another intensive regimen]. Both of those are considered as re-induction. But the traditional definition includes giving 7+3 again.
Are there any ongoing studies of combinations or sequencing with these agents?
With any diseasenot multiple myeloma or lymphoma—the single-drug therapies are very rarely going to work. But again, the FDA has certain requirements for approving something. I think you will see this more as you combine it with venetoclax/HMA. So those are ongoing; these are not data that are publicly available, but those are all ongoing trials and some of the data are quite exciting. By the next American Society of Hematology Annual Meeting, you will see a lot of venetoclax [combination data], triplets even. It is just like myeloma. You have Revlimid which is good, you added a proteasome inhibitor to Revlimid, and it is better. Then you added daratumumab [Darzalex]. You combine it and you see much, much more improvement.
[As a single agent and as a bridge to transplant, the response rate is] about 40%. I mean, if you look at CR/CRi rates together, it is about 40% now. Mahesh Swaminathan, MD, from our group has looked at the combination of azacitidine and quizartinib and the response rate is about 70% to 75%.7So even if you are using it as a bridge, I will probably do it in combination. I think a couple of years from now, as data emerges, there is an ongoing study of azacitidine with gilteritinib and azacytidine with quizartinib. Most of these will move into that combination. But they are approved right now as single agents and that is good because at least the drug has to be approved so research can keep moving forward. But I do not think the single agent will be the way to go.
There is no public data yet. It is like myeloma 10 years ago. When Revlimid came out, nobody knew you would add it to everything. So those trials are going on but at least historically, we have seen that almost never has a single agent stood out and been better. I think we will see the combinations are better.
This is what is approved today, so this is available for practice. But there is a lot of published data, pre-clinically at least, showing that maybe combination could be good in the future.