Lead Researcher Discusses Midostaurin in AML

Richard M. Stone, MD

The recent FDA approval of midostaurin (Rydapt) has excited the field of acute myeloid leukemia (AML). Midostaurin was granted approval for the treatment of adult patients with newly diagnosed FT3-positive AML in combination with standard cytarabine and daunorubicin induction and cytarabine consolidation.

This approval was based on the results of the phase III RATIFY trial (CALGB 10603), which reported that midostaurin, in addition to standard chemotherapy, reduced the risk of death by 23% compared with chemotherapy alone in patients with AML who harbored a FLT3 mutation. After censoring for patients who received stem cell transplants, the overall survival (OS) benefit with midostaurin remained steady at 25%.

The median OS was 74.7 months with midostaurin versus 25.6 months with chemotherapy alone (HR, 0.77; 95% CI, 0.63-0.95; P = .016). The 5-year OS rate for patients in the midostaurin arm was 50.9% versus 43.9% in the chemotherapy-alone cohort.

Although midostaurin has the ability to improve the cure rate of this disease, it will not benefit all patients with AML, says Richard M. Stone, MD, the lead author on the RATIFY trial.

OncLive: What impact do you see midostaurin making in the AML community?

In an interview with OncLive, Stone, director of the Adult Leukemia Program, Dana-Farber Cancer Institute, and professor of medicine, Harvard Medical School, discussed the impact that midostaurin could have for patients with AML, as well as ongoing research with the agent. Stone: AML is a devastating disease that affects about 15,000 people in the United States annually. Unfortunately, about 40% to 50% of the people who get this disease die of it. It is particularly difficult in older adults who get this disease, and the median age is about 68 to 70, so it is a disease of older adults. But, young people do get it as well. Even in young people, a lot of patients die of it.

We have not really done much to change or improve the care of people between the ages of 18 and 60 who have AML, and we have done even less, frankly, for those with AML who are over the age of 60 over the last few years. We are desperately looking for drugs to improve the outcome. There have not been any drug approvals for AML in several decades, so any new drug in AML that is going to improve the cure rate is going to be heralded with some degree of excitement.

What benefit was seen with midostaurin in the phase III RATIFY clinical trial?

With midostaurin, when added to chemotherapy, one has a better outcome in the long run, and in the short run, for that matter, than if one added placebo to chemotherapy. That is why we are excited about this. I wish I could say that we cured everybody who was eligible to get midostaurin by adding it to chemotherapy. That is not the case, but it is an advance compared to chemotherapy alone for people with FLT3-mutant AML. This trial had a simple design with the control group getting standard induction chemotherapy, which was daunorubicin and cytarabine, and if those individuals achieved remission, they got standard high-dose post-remission therapy.

The experimental arm got the same chemotherapy, but instead of placebo being added to it, which was the case in the control arm, they got midostaurin at a dose of 50 mg twice daily orally given on days 8 through 22, which is 14 days after the chemotherapy completed. They got the midostaurin during induction during each of the post-remission courses, and there was also randomization to placebo or midostaurin for 12 months after the chemotherapy was concluded. It was not a re-randomization—the people who got midostaurin got it all along.

With any trial, the important thing to think about is what is the endpoint, so what was the most important finding we were looking for, because that primary endpoint determines the number of patients that had to be enrolled on the trial. The primary endpoint we chose was overall survival—would more people on midostaurin plus chemotherapy survive than people who had chemotherapy plus placebo.

The trial was based on seeing a risk reduction of at least 22% in the risk of dying for those randomized to midostaurin plus chemotherapy versus placebo plus chemotherapy. The trial met its primary endpoint, meaning that indeed the risk of dying was lower by about 24% if you were randomized to midostaurin than if you were randomized to placebo. It took 714 patients to see that difference.

The overall risk of dying was about 24% lower on the midostaurin arm. That does not have to do with the median survival. The median survival depends on where you have the 50% cutoff on the Kaplan-Meier curve. It is true that the median survival for patients on midostaurin was years longer than it was for people who got placebo, but one should not overdraw conclusions based on that seemingly exciting number. It is really the hazard ratio that is important, because the number of people on the trial was large enough. The P value was quite significant, even though the hazard ratio was 24%.

Were there any significant toxicities?

Which patients will benefit from midostaurin?

Another important piece of data that is helpful in understanding the magnitude of the benefit is how many patients were alive at 4 years who were randomized to midostaurin, compared to those who were alive at 4 years who were randomized to placebo plus chemotherapy. Approximately 44% of people on the placebo arm were alive at 4 years, and about 51% on the midostaurin arm were alive at 4 years, so that is a 7% difference. Is that huge? No. But it is highly significant and has clinical benefit. I think anybody in their right mind would want to be on the arm that had a 7% longer chance of being alive, and therefore, probably cured, at 4 years. That’s why median survival is misleading. Overall survival, hazard ratio, and the number of patients alive at 4 years probably has more clinical meaning and is more relevant to the absolute magnitude of the benefit relative to the median survival comparisons. The adverse event rate in both arms was quite similar. The only side effect or life-threatening thing that was worse on midostaurin was skin rash, and that was not very common. Despite the fact that single-agent trials and some of the early trials of midostaurin suggested a fair degree of gastrointestinal toxicity, in terms of nausea and vomiting, we did not actually see that in this placebo-controlled trial. Toxicities were very similar between the 2 arms, suggesting it is a safe drug to give at this dosing schedule. People who would have been eligible for the trial are the appropriate candidates for midostaurin—people between the ages of 18 and 60 who had a FLT3 mutation in the leukemic cells at the time of diagnosis. Those are the ones that would get chemotherapy plus midostaurin—that represents about 30% or 35% of all AML patients.

A couple other important pieces of data are that over half of the patients on the trial got transplanted at some point. About a quarter of the patients got transplanted during the first remission, and about a quarter of the patients got transplanted later after they had relapsed or had not gone into remission. I think the positive total outcome that we saw was due, at least in part, to the fairly high transplant rate. As it turns out, compared to the beginning when the trial was started, we learned that it was a good idea to transplant people that had this type of mutation. That is why the transplant rate was higher than we estimated at the beginning. I think that is an important part of the management.

Is there ongoing research exploring whether it would be better to give midostaurin post-transplant?

For a patient with AML and a FLT3 mutation who was between 18 and 60, I would give them midostaurin. If they were a transplant candidate, I would give them 1 cycle of consolidation, or whenever the transplant was ready I would then transplant them. I think that is going to lead to a pretty good outcome. If you look at the curve of patients who had midostaurin and were transplanted in first remission, the 5-year survival rate was about 60%, so if you were in that category, you did pretty well. The RATIFY trial did not include use of midostaurin either after relapse or transplantation. Subsequently, there had been a lot of trials using medicines like midostaurin, or midostaurin itself, in people that are either at risk for relapse after transplant because they had a FLT3 mutation going in, or people actually had relapse after the transplant, and it seemed like using the tyrosine kinase inhibitors to inhibit FLT3 in the post-transplant setting was a good idea. Many investigators are either studying the use of FLT3 inhibitors after transplant or incorporating it into trials.

Do you think this agent could open the door for other targeted agents in AML?

An important subsequent research issue will be whether to use midostaurin or other FLT3 inhibitors after transplant. There’s actually a big bone marrow transplant Clinical Trials Network trial ongoing now of people that have FLT3 mutations, get a transplant, and then they’re randomized to a different FLT3 inhibitor called gilteritinib versus observation. FLT3 inhibitors, including midostaurin, will be evaluated in the post-transplant setting. A drug like midostaurin is a multikinase inhibitor. It does not just inhibit FLT3. There was some data from my trial that suggests that maybe it did not work by inhibiting FLT3, maybe it worked by inhibiting other enzymes that might be relevant for the leukemia state. As such, there will likely be a trial evaluating chemotherapy plus midostaurin versus chemotherapy plus placebo in FLT3 wild-type patients, so the other 70% of patients could possibly benefit from this, but we have no idea if that’s true or not. We have to do a clinical trial to figure it out. Absolutely. I think midostaurin will help to pave the way for additional targeted therapies. Even now, in the days [following the approval], there is a lot of research with targeted therapies, which is either ongoing or soon to start, specifically of FLT3 inhibitors that are being tested in different situations. There’s also IDH1 and IDH2 inhibitors that are being tested in patients with IDH1 and IDH2 mutations.

The good news is that AML might be dealt with in a better way than we have done in the past by using targeted therapies. The difficulty is that AML is not the most common disease in the world, which is good because nobody wants to have AML. The challenge is that you’re subdividing a relatively rare disease into even more rare subtypes, each of which might require a different targeted therapy.