Maximizing Response in Acute Leukemia: Pearls From ASH 2017

Jorge E. Cortes, MD; Daniel J. DeAngelo, MD, PhD


January 29, 2018

Jorge E. Cortes, MD: Hello. I am Jorge Cortes, from the Department of Leukemia at MD Anderson Cancer Center in Houston, Texas. Welcome to Medscape Oncology Insights, coming to you from the 2017 American Society of Hematology Annual Meeting in Atlanta, Georgia. Today, we will be discussing some of the highlights of the leukemia sessions at this meeting.

Joining me is my friend and colleague Dan DeAngelo, director of clinical and translational research in the adult leukemia program at the Dana-Farber Cancer Center in Boston, Massachusetts. Welcome, Dan.

You had a fascinating, very prominent plenary [abstract] on a new phase 1 study.[1] Tell us a bit about that.

Daniel J. DeAngelo, MD, PhD: This was a trial of BLU-285, a novel, highly specific KIT inhibitor for patients with advanced systemic mastocytosis (SM). SM is an uncommon disease driven predominantly by KIT mutations. In fact, 90% of patients with SM will have a KIT D816V mutation as the driver. The BLU-285 molecule binds to KIT in the active conformation; in addition, it is more potent at inhibiting the KIT mutant form than the wild type.

BLU-285 lends itself to an optimal toxicity-to-efficacy ratio. In this phase 1 study, we were able to show that almost 100% of the patients achieved disease control; 72% of patients had a response, with 56% complete and partial remissions. Moreover, no patient came off study because of toxicity concerns; [this is] rather unusual in a phase 1 trial, and thus very exciting.

Dr Cortes: Excellent. What is the next step?

Dr DeAngelo: Over the past couple of years, midostaurin has been in clinical trial development, and it was just approved by the US Food and Drug Administration (FDA) in the summer of 2017 for patients with aggressive SM. For the first time, hematologist-oncologists are learning about the disease and how best to handle it. Historically, these patients would come through dermatology or allergy and immunology, and land on the doorstep of the hematologist. Now we actually have a therapy—an even better hammer, if you will—to drive the patient home.

One of the most profound observations with this particular study was that during the short period of time that they were on the study, 60% of patients achieved complete bone marrow remission. We are very excited about the outcome. Moving forward, I believe we should formalize the data in a phase 2 trial, but also should try it in earlier-stage disease. What about [testing it in patients with] indolent SM or smoldering mastocytosis?

This is an unmet need first, but second, there is a lot of education [needed]. We had a separate poster abstract,[2] where we looked retrospectively at patients who had been molecularly profiled and had a KIT mutation. Not surprising, not one of those had been diagnosed with SM, but when we went back and pulled the marrow [samples], every single one of them had SM that the pathologists had missed.

Dr Cortes: Very interesting. You mentioned midostaurin, and of course we recently had the approval of midostaurin [not only] for patients with mastocytosis, but also in acute myeloid leukemia (AML). We have known about midostaurin for a while; it had some activity as a single agent, an FLT3 inhibitor, and was recently approved in combination with chemotherapy for front-line, FLT3-mutated AML.

One issue that has been controversial is that the study design included using it in maintenance treatment, but not too many patients received maintenance, so the label in the United States [does not include using midostaurin as maintenance treatment], although it is approved for that indication in Europe. A very interesting abstract, presented here by Richard Larson,[3] looked at whether there is indeed a benefit or not with the maintenance. What is your take on that study?

Dr DeAngelo: This is one of those regulatory issues. As you mentioned, the RATIFY trial[4] enrolled patients between age 18 and 60 years with newly diagnosed untreated FLT3-positive AML, including both internal tandem duplication (ITD) and tyrosine kinase domain mutant (mostly DA35). Patients received combination chemotherapy (daunorubicin and cytarabine) plus the midostaurin or placebo. Patients who did not receive bone marrow transplant would receive maintenance midostaurin.

The world changed when the study was designed and more and more patients began to undergo transplant. In fact, a large percentage of patients went directly to transplant and did not receive midostaurin. Thus, midostaurin was not tested in a randomized fashion during the maintenance period of the study, so they did not receive an indication for maintenance therapy from the FDA. In my opinion, this was primarily a regulatory issue.

At our center, when a patient goes to transplant, we usually put them post-transplant on a tyrosine kinase FLT3 inhibitor. We have phase 2 data at our institution showing that sorafenib improves survival after transplant. I believe midostaurin, given its indication, is another very reasonable option.

As you already mentioned, the European regulatory agency approved this indication. But we have this conundrum about the way the study was designed, what the different regulators want, and what we as clinicians should do. I believe the data suggest that maintenance therapy is beneficial. This is a difficult disease, and anything we can do to ameliorate relapse makes sense to me.

Dr Cortes: I agree. The data that were presented were interesting. The maintenance phase went for 1 year. During that year, the curves separated a bit at the beginning and then merged back again. But according to the data that were presented, after that year, it appears that a lot of relapses occurred as soon as the maintenance treatment was stopped, and then it plateaued.

Of course, that is an extrapolation, but you wonder whether the patients relapsed because the maintenance was stopped. Were the patients actually experiencing benefits [that were lost when maintenance was discontinued]? With these design issues, we do not have an answer. I get the same feeling as you that there is benefit, at least for a subset of patients.

Dr DeAngelo: The theme has emerged since tyrosine kinase inhibitor therapies were introduced in chronic myeloid leukemia: When do you stop? Rarely, in that disease, and perhaps it is a bad example. In that disease, patients do achieve complete remission, but in the absence of that we continue.

What happens if a patient with Philadelphia chromosome-positive acute lymphoblastic leukemia (ALL) goes to transplant? Patients in the SWOG study,[5] led by your colleague Dr Ravandi, continued treatment after transplant. Why wouldn't it be the same in this disease? AML is as malignant as the others. Anything I think you can do to mitigate relapse is important.

From a biological perspective, unless we know that a patient has achieved complete molecular remission post-transplant or post-therapy—and that was not assessed in the RATIFY study—there may be residual low-volume disease. Continuing the inhibitor should benefit the patient. That is my interpretation from Dick Larson's study.

Dr Cortes: I agree. Speaking about maintenance, another very interesting abstract came from the HOVON study.[6] In this trial, they gave patients with AML the standard chemotherapy and consolidation, and then they were randomly assigned to receive maintenance with azacitidine versus observation, which is the standard approach after chemotherapy. The trial showed a very significant separation of the curves on disease-free survival [in favor of azacitidine]. What's your take? You have experience with a similar trial.

Dr DeAngelo: Maintenance chemotherapy after treatment for AML has been tested for the past decade and a half. These studies are difficult because patients become fatigued. You get the patients through induction, you lose some along the way, then you give them consolidation and at the end of this painful, long, toxic therapy, you offer more chemotherapy. And the patients say, "What?!" Those studies are difficult.

We have completed three maintenance studies within the cooperative groups. They have all shown a trend, but not a statistical benefit. I believe the limitation with those studies was the fall-off [in numbers]. There were so many patients who declined to be randomly assigned, or once assigned decided it's not for them.

The difference in the HOVON study was that many more patients agreed to randomization, [and] many more patients stayed after randomization. Using azacitidine maintenance rather than, perhaps, decitabine, which is a bit more myelosuppressive, was the right choice. I believe the data are provocative.

The devil is always in the details in terms of who benefits. Were the patients who benefited those patients who had evidence of clonal hematopoiesis—for example, abnormal molecular findings? Did they have a molecular profile going into the therapy that was more suggestive of underlying secondary AML?

Our group has looked at the ontogeny of AML, simply because when a person walks in the door with newly diagnosed AML, there sometimes are signs of earlier disease. Perhaps those are the patients who benefited. It would be helpful to know which patients benefited, and if we can learn from that. I believe it provides another intriguing option.

Dr Cortes: I agree. We have conducted a similar randomized study with azacitidine, but ours did not show much. As you mentioned, all of these details [must be analyzed], but these data open new avenues. I do not believe that we should say everyone should receive maintenance with azacitidine right now, but I don't believe it is dead either.

Dr DeAngelo: This opens up the box. In my opinion, the important thing about the HOVON study was that the toxicity was rather minimal.

Dr Cortes: Another drug that was recently approved was the IDH inhibitor enasidenib. A few abstracts presented here provided some follow-up data and some new information.[7,8,9] Let me start with a follow-up study that showed how the responses can be delayed.[8]

It found that initial responses can take four or five or six cycles sometimes. That is important, because we are using these drugs now. At the beginning, when we began the phase 1 study, we may have missed some responders because we did not recognize the differentiation syndrome. What do you think about it?

Dr DeAngelo: I am absolutely intrigued by this drug. The overall response rate of 30%-plus is exciting, but I agree with you. I believe we may have seen more responses had we really known what we were doing. Hindsight is always 20/20, of course, but I believe the eagerness to take patients off study early on was unfortunate.

In my own clinical experience, I have seen patients go into remission after four or five cycles. I remember a patient with an IDH2 mutation who had relapsed after stem cell transplant and went on therapy. It took three cycles before we really began to see a reduction in his blast count. Then he went into complete remission, transfusion-independent, for more than 2 years with taking one pill a day. How great is that?

You can easily see how another physician who had not treated the patient in a clinical trial and did not know that would have been eager to take the patient off the drug after 1 month and no response, after 2 months and no response, after 3 months. But the patient was not progressing, and eventually he responded. That is a very important point. I believe it is a profound point. Recognition of the differentiation syndrome is also somewhat important, and that came out in this abstract as well.

It is important to recognize that when you begin to see this leukocytosis, it is not always disease progression, especially in people who are having cytokine release with some low-grade fevers, shortness of breath, [and] capillary leak. It is reminiscent of all-trans retinoic acid (ATRA) differentiation syndrome in patients with acute promyelocytic leukemia (APL). The agent deletes the IDH2 clone. For patients who have limited options, this agent is really a wonderful tool in our therapeutic belt.

Dr Cortes: The drug was approved, of course, and it was approved for the relapsed/refractory patients. The study included patients who were previously untreated, so they analyzed the subset of patients who were previously untreated and enrolled that study.[9] The response rate in this subset was somewhat modest; the complete response (CR) rate was something like 18%-19%, with an overall response rate of about 30%, similar to the relapsed refractory patients, interesting enough.

A couple of studies looked at [using enasidenib or ivosidenib as] front-line therapy in combination with other agents, one with azacitidine[10] and another with the standard 3+7 type chemotherapy. The results are early, so I don't know that the response rate that we see now is what we will see by the end of the study. What do you think about that concept [of using an IDH inhibitor] as initial therapy, alone or in combination?

Dr DeAngelo: The fantasy that we, or at least I, have is that you take a patient with AML and you molecularly profile them. They have mutation X, so you get inhibitor X. They have mutation Y, you get inhibitor Y. We have more tools in our belt; we have midostaurin for patients with FLT3 mutations. Now we have enasidenib for patients with IDH mutations and others not far down the road. The BLU or midostaurin may be for KIT mutations. If you live a fantasy, that is the way it should go. The problem is proving it, because these are rare diseases.

If you take the RATIFY trial, where you are looking at one of the more common targeted mutations, 25% of AML has an FLT3 mutation, either internal tandem duplications (ITDs) or D835. At most,15% have the IDH2 mutation, and maybe 7% or 8% have IDH1, so it becomes more difficult [to provide clear-cut evidence] from a regulatory agency standpoint. How are they going to approve it? Are we going to need randomized trials for these rare mutations?

That is my take. But it is good to see these trials beginning to accrue patients and starting to show that these drugs are safe. With time, I believe we will learn about some of the efficacy signals. I worry about the regulatory approach for these; I would hate for the perfect to be the enemy of good. I would like to have these options for our patients.

Dr Cortes: I agree. It will be interesting to see as the data mature, because with the FLT3 mutations, those patients have a very poor prognosis, but with IDH1 and IDH2, the prognosis is not that bad. We have these drugs; is chemotherapy needed? Maybe what will help is to debulk, and then we will have to do less consolidation.

Dr DeAngelo: Or maintenance? [We may have a few] different options. When you say that with the IDH1 and IDH2 mutations, the prognosis is not so bad, it depends on whether you are a glass-half-full or glass-half-empty person. Of course, when you look a 35-year-old patient in the eye and say, "It's not a bad prognosis, but you have a 50% chance of survival," to me that still is not so great. The only [good] leukemia is no leukemia. I think anything we can do to improve the outcomes will be a benefit.

Dr Cortes: Let me very briefly touch on one last drug: venetoclax. We know that it has great benefit in chronic lymphocytic leukemia (CLL). Now we have a lot of interest in using it for AML; we have single-agent data, and then some combinations. Let us talk briefly about Andrew Wei's update on his study with low-dose cytarabine (ara-C) in combination with venetoclax.[11] There are also the studies with hypomethylating agents.[12,13] The responses look remarkably good. How do you see it?

Dr DeAngelo: The big problem with leukemia is that our chemotherapy is [just] okay. The problem is that patients are just not eradicating their disease. Anything we can do to improve or increase the apoptotic rate—the programmed cell death—is going to help our patients.

Chemotherapy works through this apoptotic pathway, and venetoclax is the natural drug to try to augment that process. This is an example where this drug as a single agent works, but it has a CR rate of about 20%. With chemotherapy—even low-dose chemotherapy, which by itself has a minuscule response rate—the two together are synergistic, and we are seeing response rates into the 70% area.

The key is to learn how to safely administer it. I don't know about you, but I have not seen responses this high with lower-dose, if you will, chemotherapy, ever. I am fascinated by it. How do we use venetoclax not just with low-dose chemotherapy, such as low-dose cytarabine or the hypomethylating agents, but also with other chemotherapy backbones? You and I, or at least your center, are beginning a study of venetoclax plus chemotherapy for ALL, for example. So we need to take a look at it. I do not believe it is a one-size-fits-all thing, but these two abstracts are very provocative in showing the proof of principle.

Dr Cortes: Absolutely. This has been a great discussion, Dan. Thank you very much for joining us.


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