The Role of FLT3 Inhibitors for Treatment of AML

Harry Erba, MD, PhD: Let’s move on to talk about some therapies. We’re going to start with FLT3 and our targeted therapy section. There was an intriguing presentation at the ASCO [American Society of Clinical Oncology Annual Meeting] 2021 conference regarding, potentially, a signature that might let us to look at which patients with flickering mutated AML [acute myeloid leukemia] might benefit from midostaurin with chemotherapy. I’m going to start with Courtney, but I’m sure Mark is going to have some comments about this.

Courtney DiNardo, MD: I was going to start with an overview and remind us that FLT3 mutations are some of the most common in our AML patients—about one-third of our patients. We need to be aware of FLT3 mutations and now, in our up-front treatment strategy for intensive chemotherapy, we need to make sure to add a FLT3 inhibitor. Midostaurin is the 1 that’s approved. Phospho-signaling is not something that we’re doing in real time [at The University of Texas MD Anderson Cancer Center]. There’s this interesting abstract you mentioned, by Arran David Dokal, I believe, where using that predictive signature can actually identify who does and doesn’t respond, but I’m not entirely convinced that’s a particularly relevant assay. I look to Mark to tell us whether that’s something we should be looking at, either in a treatment or a research-based setting.

Mark Levis, MD, PhD: I have to confess: I looked at that as very much a work in progress. Proteomics used to be the big thing and I was as enthusiastic as anyone 10 years ago, but it hasn’t panned out. I will wait and see on that.

Dan Pollyea, MD, MS: What I like about this abstract—and Mark is the perfect person to give further commentary on this—is the historical aspect. They talk about patients without a FLT3 mutation who might be benefiting from this therapy because of its off-target effect against protein kinase C, and I like how this drug started as a protein kinase C inhibitor and the off-target effect was FLT3. Now, we’re taking it back around to where this drug began and exploring some of the non-FLT3-targeting properties and seeing whether it might be effective. I love that, because that would explain why some of the patients with TKD mutations did as well as they did in the RATIFY study, and it also could help explain or provide more rationale for some of these FLT3 wild types that are happening with midostaurin. The interesting part is the use of this old generation of a FLT3 inhibitor and trying to figure out all its properties in context with some of these new FLT3 inhibitors that we now have.

Mark Levis, MD, PhD: The feature of those drugs—an older 1 was the dreaded CEP-701 [lestaurtinib]. These were very similar drugs. They are like staurosporine and, in fact, that’s what midostaurin looks like. Staurosporine inhibits all these kinases. I remember doing some of those phospho-kinase assays, and these drugs just blanked out every kinase under the sun. They just turned everything off. It actually wasn’t that good of a PKC inhibitor, either, to be honest. PKC was 1 of the earliest identified kinases in cancer. Everybody went after it, and then it turned out it wasn’t all that key of a thing to go after. Yes, the historical PKC412 [midostaurin] is of interest there, I guess.

Harry Erba, MD, PhD: Dan, this is critically important. If this observation is correct—and I know we debate this—how are we really benefiting this population of patients with midostaurin? It causes a concern about moving the second-generation drugs and FLT3 inhibitors into up-front therapy. There are a number of those trials going on. Do you want to comment about those?

Dan Pollyea, MD, MS: You’re right. The assumption is that the RATIFY trial was positive because all these patients who were FLT3-positive responded to the addition of a FLT3 inhibitor, but what if there were some other off-target impact of this relatively nonspecific kinase inhibitor that was more of a driver in this? Based on this assumption—which was a fair assumption—we have all these next generation FLT3 inhibitors that are superspecific for FLT3. We have crenolanib, quizartinib, and gilteritinib. As we would expect, they’ve rushed into the void and are in different stages of clinical trial and development. Researchers are comparing this with midostaurin plus 7+3 [cytarabine, daunorubicin, idarubicin, mitoxantrone] to see if a more specific FLT3 inhibitor would be better. Everyone doubled down on that idea because if this is FLT3 driven, then we would expect that some of these trials would be positive. If it isn’t, then it’s possible that midostaurin will look better or that there won’t be a difference. It will be interesting to see, over the next couple of years, how these clinical trials read out.

Mark Levis, MD, PhD: This, to me, resembled those genome expression signatures where they would say, “We can now predict who will do well with this treatment,” but going forward it really never worked out. That’s what this reminded me of.

Vinod Pullarkat, MD: The challenge also is to know if it’s FLT3 mutant-like biology. Only a few patients have a high proliferative type of AML, so many of these trials included allelic

ratios. Everyone is looking for an assay that can capture that FLT3-mutant type of biology. It’s an attempt, and the differences in that abstract are pretty significant if that’s real and true, so we don’t have a quick test for that. That’s a challenge, and a lot of people are trying to quote the details of the FLT3-mutation allelic ratios.

TRANSCRIPT EDITED FOR CLARITY