Harry Erba, MD, PhD: That brings me to 2 questions I want to ask about these drugs. We have midostaurin, and I’ll throw in ceritinib. They’re very nonselective FLT3 inhibitors. Now you have the more selective ones. Do we know at this point what the optimal approach is to these patients?
Alexander E. Perl, MD: I don’t think we know. I have reason to think that it makes sense, logically, that we should use a highly selective, highly potent FLT3 inhibitor. But look at midostaurin, which all the data suggest is pretty broad in its activity, and yet it’s been active. We don’t really know. I think the design of comparing a selective potent inhibitor to a less selective inhibitor, added to the same backbone of therapy, is really what we need. We just need a study like that.
Eunice Wang, MD: Right, exactly. There were data in the FLT3 wild-type patients showing that there was activity. Maybe we need a broader kinase inhibitor. There is some speculation up front where there are many, many clones in high-bulk disease. There are a lot of different subclones in there. Maybe a broader inhibitor would be better while waiting until relapse. A highly potent, dominant FLT3-driven clone might be better, but I don’t think we know.
Harry Erba, MD, PhD: I think that’s going to be a very important question for us to figure out.
Eunice Wang, MD: What do you think, Jorge?
Jorge E. Cortes, MD: I complete my total ignorance. I don’t know what is best. I think that there are elements that are good for both approaches, which you both have mentioned. I think that we at some point are going to just have to do the comparisons. It may be that for subsets, one is better than the other.
Eunice Wang, MD: In CML maybe? If they have different tyrosine kinases for different situations.
Harry Erba, MD, PhD: Let me ask the expert here. I’m just an investigator on these studies, but I have noticed one thing with the single-agent FLT3 inhibitors I’d like for you to explain to me. They have been very effective at clearing peripheral blood blasts and even clearing the marrow. But many patients don’t recover their counts. Why?
Alexander E. Perl, MD: This is a complicated answer, and I know we’re limited on time.
Harry Erba, MD, PhD: No, in 15 seconds.
Alexander E. Perl, MD: These are complex diseases with multiple mutations where you’re using 1 drug to target 1 mutation. You can’t get rid of everything else with that 1 drug. Now, one thing we’ve noticed is that there is differentiation of FLT3 inhibitors, and that pushes toward a myeloid bias. That doesn’t generate a lot of platelets or red cells, potentially. We also do see patients who clear the FLT3-mutated clone, but it may leave behind clonal hematopoiesis with other mutations that aren’t targeted by the drug. So, I think what you really need to use is combinations. Asking 1 drug to do it all is really hard.
Harry Erba, MD, PhD: FLT3 inhibitors, FLT3 mutations, are late events in leukemogenesis.
Alexander E. Perl, MD: Yes.
Harry Erba, MD, PhD: Just as you said, you have this soil—not the soil but the seeds—and those cells can still mutate and still have an underlying hematopoietic stem cell defect.
Alexander E. Perl, MD: Right.
Eunice Wang, MD: Really, with crenolanib in the studies that were done with 7 and 3 chemotherapy, the count recovery was actually not affected. I think c-KIT is a particular kinase that has maybe been associated with hematopoiesis and count recovery, and crenolanib—at least theoretically—doesn’t hit c-KIT.
Jorge E. Cortes, MD: I’m going to throw out a hypothesis about this. I think that one thing we probably have forgotten is, we think about the inhibition of c-KIT in some of this, but FLT3 really has a normal function in hematopoiesis.
We’re talking about the mutated one that is constitutively activated, but FLT3, with its ligands, then has a normal function. If we were inhibiting this—even when the ligand is functioning—in the normal hematopoietic cells, we probably are not letting those cells proliferate. There is this concept that when you give chemotherapy, there is an overexpression of the ligands. There have been assumptions about how best to sequence, but maybe they recover well after chemotherapy plus the inhibitor.
Eunice Wang, MD: Because they need the ligand.
Jorge E. Cortes, MD: Because they have this overexpression of the ligand. It’s only a hypothesis; I have no proof. But could that be the reason why they don’t recover with this single agent?
Transcript Edited for Clarity