Combinations are the Future in FLT3-Mutated AML - Episode 9

Developing Resistance With FLT3 Inhibitors in AML


Jessica K. Altman, MD, and Naval G. Daver, MD, discuss what is known in terms of mechanisms of resistance to FLT3 inhibitors used to treat acute myeloid leukemia.

Jessica K. Altman, MD: We’ve talked a lot about resistance to FLT3 inhibitors. Can we try to walk through our understanding of the mechanisms of resistance to FLT3 inhibitors?

Naval G. Daver, MD: Yes, sure. We recently presented on this at ASH [the American Society of Hematology annual meeting]. One of my fellows and I published on this, and of course others, including Catherine Smith, MD, yourself, and others have also published on this, in terms of with gilteritinib and quizartinib.

To summarize this, there are very different mechanisms of resistance associated with different FLT3 inhibitors. I think this is critical to note, even though it sounds like it’s obvious. This demonstrates the need for multiple FLT3 inhibitors, very much like the CML [chronic myelocytic leukemia] story. And so, you can switch and salvage patients who would not have been salvageable if there were only 1 or 2 TKIs [tyrosine kinase inhibitors].

We have type 1 and type 2 FLT3 inhibitors; these are the 2 big groups. Type 1 drugs include gilteritinib, crenolanib, and midostaurin. These drugs cover both the ITD [internal tandem duplication] and D835 mutations.

Then there are the type 2 inhibitors, which are quizartinib, sorafenib. These predominantly target ITD. In regard to the type 2 FLT3 inhibitors—quizartinib and sorafenib—the most common mechanism of resistance is through acquisition of an on-target FLT3-TKD [tyrosine kinase domain]. Many studies have repeatedly shown that 30% to 40% of patients acquire that mutation, FLT3-TKD. They stop responding to quizartinib or sorafenib. Then you have to go to gilteritinib or midostaurin.

On the other hand, with midostaurin and gilteritinib we very rarely see a true D835 because it’s covered. We usually see a non–FLT3-target parallel pathway activation. Then as you were saying, Jessica, we see what they call the gatekeeper; there’s the F691 of FLT3. And so, we may see that.

Until now, there has been no clear drug. There are efforts and potential drugs that could target that, but this may be the gatekeeper where we will need a very nice drug like ponatinib. I think this is important because we are doing this in our practice, and I know you and others are too. We may give gilteritinib and you may see a RAS/MAP kinase mutation or upregulation. We may switch the patient if we have access to quizartinib in combination with azacitidine, or azacitidine/venetoclax, or others like sorafenib. This way, we can hopefully overcome some of the resistance.

Then one last kind of uncommon but important one is BCR-ABL. This has been shown with both single-agent gilteritinib and in combination with gilteritinib. Although very rare, 5% of people may have an emergent BCR-ABL. Then, of course, you could give them dasatinib, ponatinib sort of drugs going forward.

In terms of resistance, how do you think these would change as we use these drugs in combo—if we give 3+7 [daunorubicin, cytarabine]/gilteritinib, azacitidine/gilteritinib or azacitidine/quizartinib? What do you think we should be looking for?

Jessica K. Altman, MD: Just as you and I discussed before, as we use these agents, the patterns of resistance may change. It may be that we see a completely different mechanism of resistance with the more potent therapy that we give initially. So in addition to the mutations that you alluded to, there also can be an increase and upregulation of the FLT3 ligand. Occasionally we see that based on laboratory work. We may see changes in the bone marrow microenvironment. Hopefully the chemotherapy in combination with the FLT3 inhibitors will be able to minimize that. But this is all the more reason to check for what mutations are occurring at the time of relapse. I don’t know what we’ll see. Maybe we will see more BCR-ABL mutations, and that’s important to mention because not all next-generation sequencing panels have BCR-ABL. That means you have to think about that and remember to ask for that, specifically.

Transcript Edited for Clarity