The Role of IDH Inhibitors for Treatment of AML

, , , ,

Peer Exchange | <b>Acute Myeloid Leukemia: 2021 Perspectives on Testing, Systemic Therapies, and Transplantation</b>

Harry Erba, MD, PhD; Courtney DiNardo, MD; and Dan Pollyea, MD, MS, discuss the role of the IDH inhibitors ivosidenib and enasidenib as single-agent or combination therapy in patients with acute myeloid leukemia (AML).

Harry Erba, MD, PhD: Let's move on to a different target [in acute myeloid leukemia (AML)], and that is IDH. We'll start with talking about ivosidenib for newly-diagnosed patients with an IDH1 mutation, and then move on to talking about patients with an IDH2 mutation and the use of enasidenib. Courtney, I know you have done a lot of work in this area, so why don't you lead us off?

Courtney DiNardo, MD: Regarding the use of the IDH inhibitors in the frontline, older, chemotherapy-ineligible setting—we have an approval for ivosidenib in that setting. We have data that show that there is a CR [complete remission]/CRH [CR with partial hematologic recovery] rate of about 40% and a median survival of about 1 year. That is absolutely a great strategy for our particularly old and frail patients who do not want to receive hypomethylating agents and [who are] going back and forth. The transfusion independence that you see with the IDH inhibitors really can be quite remarkable. The comment I made earlier still stands; we have the approval of these targeted agents as single agents, but trying to figure out how to make them more effective in combination is where much of the field is going, and we will see increased responses.

To get to your second comment—the front-line therapy of azacytidine with enasidenib. This was a frontline, randomized trial looking at azacytidine with or without the IDH2 inhibitor enasidenib. It was not blinded, so patients were randomized, and [they] knew what they were getting. One of the interesting things about this study—which may have similar stories as some of the other, LACEWING-type studies that we will talk about with FLT3 inhibitors, potentially—is that the response rate was impressive with the combination: a true CR rate of over 50% compared with 12% [with use of azacytidine alone], and then an improvement in event-free survival with the combination. The overall survival (OS) was not significantly different, but it was impressive at about 2 years for both arms. It raises an interesting [concept]. As we get more effective therapies and we can sequence effective therapies, OS is, of course, impacted by subsequent therapies. Enasidenib with azacytidine is an effective combination, although very few people are using it now, because, at the same time, azacytidine [plus] venetoclax has become, essentially, the standard of care for our older, unfit patients. There have also been very impressive responses in [patients with IDH mutations], as Dan presented recently.

Harry Erba, MD, PhD: Along those lines—it was a very impressive median survival in both arms, but the event-free survival (EFS) was much worse than the azacytidine-alone arm. I remember the last time we spoke about this, you said that very few patients in the azacytidine-alone arm went on to get enasidenib. Do you know how many went on to get venetoclax yet?

Courtney DiNardo, MD: It took digging to get that information, because it was not part of the original case-report forms, but over one-third of patients on the azacytidine-alone arm did end up getting enasidenib in the salvage setting. We do not have information on venetoclax, but I would presume, in most places where that was available, that was used. Certainly, for some of the patients I am aware of, that was the case.

Harry Erba, MD, PhD: Well, if we are on the subject, there was a very similar trial that was just presented at the EHA (European Hematology Association) conference that is actually not on our agenda today. Courtney, we need to discuss it, because it had a very similar outcome. Can you tell us about the IDENTIFY trial?

Courtney DiNardo, MD: The IDENTIFY trial was the phase 3 study looking at enasidenib alone as monotherapy versus investigator choice for patients with relapsed/refractory (R/R), IDH2-mutated disease. The long and short of it is a similar story to what I just said, which is that the response rates for patients who were receiving enasidenib were significantly improved. The hematologic improvement was 40% versus 10%; the CR rate was 25% versus others. Just the response rates were improved, but the survival was about 6 months in both arms. It was a particularly R/R population; you had to be older, and you had to have had 2 or 3 prior lines of therapy. That is one of those studies [that may be viewed with] different lenses. Having improved response rates, transfusion independence, clinical benefit, and improvement in EFS—things like that are definitely important and would make me comfortable still using enasidenib.

Harry Erba, MD, PhD: Dan, let me get you back in the conversation here. I know you have done a lot of work as well in this area of IDH inhibitors. This is a randomized phase 3 study of the standard of care versus a drug that was approved by the FDA [US Food and Drug Administration] based on a cohort in a phase 1 study of enasidenib. What would you tell [the] ODAC [Oncologic Drugs Advisory Committee] or the FDA if they look at this data and say, “Hey, did we approve a drug that we shouldn't have approved?” What would you say? What do you think is going happen?

Dan Pollyea, MD, MS: Hopefully, nothing, because, like Courtney said—I mean, I completely agree. There is a time and a place for survival as a primary end point. Historically, when there were not really any good salvage therapies, and response was a surrogate for survival—now, things happen between relapse and death that can salvage a patient quite readily. In this population of R/R patients, having a single therapy, a pill, that is very well tolerated—that can get patients into remission and improve their transfusion dependence or free them from the need for transfusion. Those are really worthy end points and really important quality-of-life type of metrics. Even if it cannot show that it makes a patient live longer, I would continue to prescribe it, and I would expect that we would still have it available, because those are really, really important end points. I would not change a thing.

Harry Erba, MD, PhD: It really points out how we, as investigators, need to convince the FDA and industry that it is not just about survival in older patients with AML who may not be eligible for stem-cell transplants. It is about quality of life. We just have not been very good at being able to collect that data and [determine] how it translates into improving the quality of life. [This] boggles my mind, because patients in remission, I know, in clinic feel much better, because they are not coming in for a transfusion or getting admitted for infections. How can I not prove this to somebody? It is really frustrating.

Courtney DiNardo, MD: If we had designed this study differently as a noninferiority study, it would have been an "easy sell" that enasidenib can have equivalent median OS. The 1-year OS, though, is better, in that response rates are better, and transfusion independence is better. If we had incorporated quality of life, it would have been very easy, because, to your point, patients want to be outpatient and taking a pill as opposed to getting intensive chemotherapy or other things. I agree. The onus is on us, in a way, to make these points known.

TRANSCRIPT EDITED FOR CLARITY