FLT3 Genetic Alterations in AML

Harry Paul Erba, MD, PhD: I am Dr Harry Erba from Duke University School of Medicine in Durham, North Carolina.

Joining me today in this discussion are my colleagues Dr Amir Fathi, from Massachusetts General Hospital in Boston, Massachusetts; Dr Mark Levis, from the Johns Hopkins University School of Medicine in Baltimore, Maryland; Dr Gail Roboz from the Weill Cornell Medical College in New York, New York; and Dr Daniel Pollyea from the University of Colorado School of Medicine in Aurora, Colorado.

Today we are going to discuss a number of topics pertaining to contemporary treatment of acute myeloid leukemia [AML]. We’ll discuss the latest research in the field and the impact of recent clinical trials on making decisions around treatment selection.

Let’s get started on our first topic. I’m going to turn to you, Mark, to discuss the prognostic and predictive implications of FLT3 genetic alternations in AML.

Mark J. Levis, MD, PhD: FLT3 testing is often front and center when a new patient with AML hits the floor, so I think no one in the field who treats these patients doesn’t know that. Unfortunately we still encounter cases, particularly from the community, where FLT3 testing is either not obtained or obtained very late. As most people know, there are 2 broad categories of FLT3 mutations: FLT3 ITD and FT3 TKD mutations. ITD is kind of common and bad; TKD is less common, less bad. The prognostic implications are, if you have an ITD mutation, you’ll do worse in general than a patient lacking an ITD mutation. TKD mutations, less so.

Nowadays, because we have an approved therapy, midostaurin, to be added to induction therapy, at least in a fit patient, the failure to recognize the presence of either of these mutations means you’re not administering a standard of care treatment that’s offering the patient a survival benefit.

You’re also not potentially starting plans for an allogeneic transplant, which may be indicated in the presence particularly of a ITD mutation, and you’re simply not aware of the response that could be expected to chemotherapy. If the patient is unfit, we know that patients, particularly with a FLT3 ITD mutation, don’t respond as well to HMA-VEN [hypomethylating agent and venetoclax] combinations, or if they do respond, response doesn’t last that long. Just starting with that, you’ve got to know about these things.

But you need to know even more. You need to know the amount of mutation present, and you actually need to know the quality of the specimen that you sent for analysis. Did you send peripheral blood that had only a few percentage points’ circulating blast? Or did you send a bone marrow sample that was pure blast? That’s actually going to impact your best guess of how much of a mutation is present. The mutation burden, as we call it, either the allelic ratio or the VAF [variant allelic function], particularly FLT3 ITD mutations, really impacts prognosis in a big way. If you have a high allelic ratio or a high mutation burden of an ITD, particularly in the context of lacking an NPM1 mutation, that’s a disaster. That’s qualified as what we call ELN3, which in the ELN [European LeukemiaNet] classification system is for those patients with very poor prognosis, who do need to know that.

Finally, you need to know what mutations are present if and when the patient is relapsing. The type of FLT3 inhibitor you’re using may be impacted. If you’re going to use a FLT3 inhibitor that is a so-called type 1 inhibitor, it’s not going to work on the TKD mutations.

You’re not going to use sorafenib or quizartinib necessarily if there’s a TKD mutation present. It’s wrapped around your treatment response or prognostic response or plans for the patient, whether it’s a fit or unfit patient.

Harry Paul Erba, MD, PhD: Mark, is it adequate to wait for the next-gen sequencing panel, or are PCR [polymerase chain reaction] assays preferred for any reason?

Mark J. Levis, MD, PhD: Here’s the other big slipup that comes, and it’s even at our institution. Next-generation sequencing assays are typically used to identify the FLT3 TKD mutation because it’s a point mutation.

As you know, a next-generation sequencing assay may be delayed in coming back. Yes, at academic institutions we’re all getting them with a reasonable turnaround of, say, a week, so we can start the newly diagnosed patient who comes back with a TKD mutation on midostaurin. But if your next-generation sequencing assay does not come back for 2 or 3 weeks, which is still unfortunately common around the country, and this is what I’m still seeing, you’ve just missed the opportunity to add midostaurin to the induction regimen for a patient with a TKD mutation. We know those patients did quite well when midostaurin was added.

On the other hand, the FLT3 ITD mutation is still mostly—not always but mostly—done with a conventional PCR assay. The turnaround time is quick, 2 to 3 days. Even your average commercial diagnostic lab will have that available sooner than the next-generation sequencing assay. They’re done differently. The mechanism of estimating the amount of allelic burden will vary according to the technique used. Actually, I would caution anyone to remember that the ITD allelic burden is prognostic, which does not necessarily mean the patient will predict a response to the combination of induction chemotherapy plus midostaurin.

In other words, it doesn’t matter what kind or how much of a FLT3 mutation you have for you to get benefit from the addition of midostaurin to induction chemotherapy. So don’t worry about applying rocket science to interpreting the mutation panel results. FLT3 mutation equals add midostaurin, and be aware of the finer prognostic implications of what kind of mutation and how much.

Transcript Edited for Clarity