Harry Erba, MD, PhD: FLT3 mutations may be found in 30% to 35% of adult patients with acute myeloid leukemia. About 20% to 30% will be the internal tandem duplication or ITD mutations. The other 5% to 10% will be the tyrosine kinase domain substitutions and point mutations, so-called TKD mutations. The prognostic significance of the FLT3 TKD is uncertain; however, there are abundant data about the prognostic significance of the internal tandem duplication. Most studies show that the presence of an FLT3 ITD does not affect the rate of complete remission with intensive chemotherapy; however, patients with an AML FLT3 ITD mutation often have a higher chance of relapse and therefore, a worse survival rate. In fact, even following allogeneic stem cell transplant, patients with an FLT3 ITD mutation are more likely to relapse than those who do not have one. The effect of the FLT3 mutation can be affected or modulated by things like the allelic burden and other mutations as well.

Naval G. Daver, MD: The FLT3 ITD and the IDH (isocitrate dehydrogenase) are 2 major therapeutic mutations that, in the last 5 years, have become important in acute myeloid leukemia. These are quite different in the way that they impact the disease, as well as their prognostic impact. The FLT3 ITD is usually associated with a highly proliferative, more aggressive disease, with a high risk of relapse. These patients usually need an allogeneic stem cell transplant, which, now in most guidelines—including the NCCN [National Comprehensive Cancer Network] and ELN [European LeukemiaNet] guidelines—is considered a standard approach for FLT3 ITD-mutated patients. And, of course, there are a number of FLT3 inhibitors that have been developed that show good response rates, either when combined with traditional chemotherapy or as single agents in the relapse.

The IDH mutation, on the other hand, is not an adverse mutation by itself. There have been a number of studies, some showing that there is neutral impact with IDH mutation, some showing that there is a favorable impact, and a few showing negative impact.; But, in general, most experts believe that IDH is a neutral mutation. So, traditionally, when we have IDH-mutated patients, we would treat them with chemotherapy and possibly, but not necessarily, take them to stem cell transplant, unlike FLT3 ITD mutations.

IDH inhibitors, although they’re called IDH inhibitors—like how we say FLT3 inhibitors—don’t actually inhibit the IDH receptor. They, in effect, inhibit a metabolic pathway that is driven by the IDH gene, which is a conversion of 2-hydroxy-beta glutarate to a metabolic enzyme that results in differentiation. So the IDH inhibitors, when we use these, exhibit a differentiation of the leukemia cells, and it is this differentiation that allows the leukemia cells to be terminated. Furthermore, it’s not a direct inhibition that we see with FLT3 ITD inhibitors that work directly on the receptors.

The FLT3 ITD mutations in general are associated with more proliferative disease. Preclinical studies have shown that patients who have the FLT3 ITD mutation on blast tend to have a higher incidence of proliferation. It also seems to be difficult to eradicate the FLT3 ITD clone completely. There has been some testing done with digital deep sequencing, which is one of the most sensitive ways to look at any residual disease. It has been shown that in many FLT3 ITD-mutated patients who achieve a traditional remission with chemotherapy, there is a low level of disease burden that can be identified with high-sensitivity sequencing methods.

We think one of the main reasons that these patients relapse is because they do have minimal residual disease, which can be picked up on when you use very sensitive techniques looking at FLT3. The other thing is that FLT3 itself is a driver mutation. If you do have any cells, even a few cells out of the billions of cells that remain, that bear the FLT3 ITD clone, this clone will, over time—6 months, a year, 2 years—result in relapse of their disease. It is really important to try to eradicate every single FLT3-mutated cell, and that is not often possible with chemotherapy alone. That’s where we think the addition of FLT3 inhibitors, allogeneic stem cell transplant, or ideally both may help us get rid of 100% of the disease, preventing the relapses.

Harry Erba, MD, PhD: In general, there are 2 basic types of mutations that we distinguish in the pathogenesis of acute myeloid leukemia: those that will block differentiation, the so-called type 2 mutations; and those that give a proliferative advantage to the cells, the type 1 mutations. FLT3 mutations are examples of a type 1 mutation and proliferative, and DNA [deoxyribonucleic acid] methyltransferase 3A are examples of mutations that will block differentiation. Therefore, it appears that you need mutations in both classes to induce acute myeloid leukemia. Now, these mutations also interact in terms of prognosis. For example, the prognostic significance of the nucleophosmin mutation, FLT3 mutation, is most dramatic in patients with intermediate risk or normal karyotype AML. Additionally, in acute promyelocytic leukemia, high-risk patients have a higher incidence of FLT3 ITD mutation. However, this does not appear to affect the prognosis of these patients treated with all-trans retinoic acid plus arsenic trioxide, either alone or in combination with chemotherapy.

There are interactions of these mutations with each other in terms of prognosis. There was a large analysis of over 1500 patient samples from the German study group. This was published by Elli Papaemmanuil about 2 years ago in the New England Journal of Medicine. Now, the 3 most common mutations that were found were FLT3 mutations, nucleophosmin, and DNA methyltransferase 3A. She was able to look for interactions between these 3 mutations, and what she found was quite dramatic: Regarding the survival of patients with a FLT3 ITD mutation, the effect on survival was most profound in patients who had both mutated nucleophosmin and mutated DNA methyltransferase 3A. In the presence of wild-type alleles of both mutated nucleophosmin and mutated DNA methyltransferase 3A, or either one alone, the effect of FLT3 ITD was not as dramatic.

Transcript Edited for Clarity