Utilization of Second-Generation FLT3 Inhibitors in AML

Harry Paul Erba, MD, PhD: Amir, I want to bring you into this discussion. If you have a young patient, FLT3 mutated, and they’ve relapsed, you know you’ve got to get them to transplant. Doesn’t it just make sense to give them chemotherapy to get them to transplant? Don’t these targeted inhibitors take longer? Do they get you in a deep remission?

Amir Fathi, MD: Eventually. If I may start with some principles that have been alluded to in passing, which are very important in my opinion when it comes to FLT3 inhibitors. One is tolerability. There is a marked difference between tolerability of, say, midostaurin versus gilteritinib. Or even sorafenib versus gilteritinib. Midostaurin does cause a good amount of GI [gastrointestinal] toxicity. It’s uncommon for me to give midostaurin to somebody and not have them at least have mild GI issues, whereas gilteritinib is much better tolerated. That’s an important consideration.

The other consideration is the type of leukemia patient. This is extremely important. Now that we’re doing NGS [next-generation sequencing] panels, we’re doing additional mutational testing. We’re discovering additional FLT3 TKD variance. Some of these FLT3 inhibitors, as Mark just alluded to, don’t really work. Sorafenib and quizartinib don’t have much activity.

What I oftentimes worry about in the community when we see patients coming in with FLT3 mutations or potentially treated with FLT3 inhibitors, for a mutation that would not be expected to respond to that FLT3 patient. Those factors are very important.

Finally, we talk about RATIFY, and this question has come up multiple times in the past. What is a dirty FLT3 inhibitor—let’s say midostaurin or sorafenib—doing to these patients? Is it mainly having a therapeutic impact through suppression of FLT3 mutations? Or is it potentially having other effects on other pathways that it may be inhibiting? It’s probably FLT3 because of the patterns that we have seen, but I’m not sure if that is completely the case. For example, the Soranal study showed that across-the-board patients without FLT3 mutations seemed to have some degree of response.

The quizartinib study showed that patients with FLT3 wild-type disease have a response. These are all questions that are yet to be fully answered. The answer to your question giving to a younger patient who may relapse, I’m not sure if it’s clear-cut that I would necessarily combine it with chemotherapy. That question has not been sufficiently answered, so I could not fault somebody if they want to put their patient on monotherapy with gilteritinib, with remission rates at approximately 50%; you could potentially decide to combine it with an HMA [hypomethylating agent]. The data on that is emerging. It’s not final, and we don’t know if it’s substantially better. I suspect it will be better, but I don’t have any definitive evidence to suggest that. I do not know if I personally would beat up a patient, even a younger one, with intensive chemotherapy, and gilteritinib, in the relaxed refractory setting.

Relapsed/refractory treatment for AML [acute myeloid leukemia], the remission rate, as we all know, hovers around 40% to 50%. Even in younger patients, it is not, depending on the timing of relapse, necessarily more impressive. My own personal preference is either give it as monotherapy or combine it with a hypomethylating therapy.

Gail J. Roboz, MD: I would just jump in with 1 comment that sorafenib gets a little bit of a bad rap. It’s not the easiest drug to take. Mark Levis always says that all his patients’ hands and feet fall off. Mine don’t do that. But that drug has magical powers in the post-transplant setting that are different. I think both the scientific and clinical data suggest that the post-transplant relapses really might be something important to consider for some patients with sorafenib that could actually be different from the others. I just put that plug in that sometimes it’s just dismissed as old and not easy to take. But I think that specific setting is 1 that is actually quite interesting, especially in combination with HMAs.

Transcript Edited for Clarity