ASH 2015

First targeted therapy for genetically defined subset with acute myeloid leukaemia significantly improves survival

Dr Richard Stone - Dana Farber Cancer Institute, Boston, USA

You’re looking at a multi-kinase inhibitor, a new approach to patients with acute myeloid leukaemia. Can you tell me what was the big issue here? What were you trying to investigate?

Well, we know that about 30% of patients with AML have a mutation in a gene called FLT3 and most of those patients who have the mutation that confers a poor prognosis, they have a high relapse rate and a poor overall survival. So it was an unmet need and we were looking to improve their lot, as it were. Since they have an activating mutation of this enzyme called FLT3, we wanted to inhibit the activity of this enzyme along with chemotherapy hoping that this would effect a better outcome than with chemotherapy alone.

So you use a TKI but why does it have to be a multi-kinase inhibitor?

It does not have to be a multi-kinase inhibitor, in fact it was originally thought of as a FLT3 inhibitor, that’s why it was used in this subtype. But we know factually it inhibits more enzymes than just FLT3 so factually it is a multi-kinase inhibitor but it was used in just FLT3 mutant patients because of its activity in that particular enzyme.

Tell me about the study, what did you do?

We took a large number of AML patients who were between age 18 and 60 and we screened them to see if they would have this mutation and we actually screened 3,300 patients and came up with about 714 that we could randomise, based on the fact that they did have a mutation in the FLT3 enzyme, to either midostaurin or placebo in addition to standard chemotherapy.

And what happened?

The patients who received the midostaurin as opposed to those who received the placebo plus the chemotherapy did better in terms of overall survival and event free survival. So it was indeed a positive study.

What were the numbers in fact?

The hazard ratio in favour of midostaurin was 0.77 which means that there was a 23% less likelihood of dying if you had been randomised to the midostaurin arm as opposed to the placebo arm.

And you had big numbers so this is a very meaningful result, it would seem?

Yes, I think so. It actually shows a result that did not happen by chance as best we can tell by the statistics. It’s a meaningful improvement in the outcome with a significantly lower risk of dying so it well could become the standard of care to treat this patient subpopulation.

Is there a downside from using midostaurin?

There may be but at least in terms of side effects there were no excess of side effects except for a slight increase of severe rashes in those who were randomised to midostaurin as opposed to placebo. There were no increased deaths in the midostaurin arm so, at least from a side effect profile, there was no real downside.

How big an impact do you think this could have on therapy for AML?

Well, again, we have to note, at least for the moment, that the impact is going to be limited to those with this subtype of AML. It’s important to point out that to go on the trial you had the patients’ AML cells had to be rapidly assessed for the presence of the mutation. This was done at one of nine academic labs throughout the world – a patient would walk in the door with leukaemia, their blood or marrow would get sent to one of these nine labs depending on the country that this was taking place, and then the answer about whether there was a mutation or not would come back in 48 hours and then the patient could go on the trial. So, again, to emphasise the fact that you need rapid diagnosis of the mutation in addition to standard diagnosis of AML and then you get chemotherapy plus midostaurin.

Has this sort of opened the doors towards more molecular tinkering with acute myeloid leukaemia, do you think?

I really think it has. Number one it shows that the trial is feasible, that’s one of the most important findings as well, that you can actually tell somebody’s molecular subtype at diagnosis before you even give them a drop of significant chemotherapy and you can base therapy upon that. So AML is a very heterogeneous disease which means that one patient’s AML is not like another’s and we have to take advantage of that heterogeneity intelligently.

Now FLT3 is fairly common, is this just the low hanging fruit or would other things be harder to win?

It’s probably the lowest hanging fruit, it’s one of the two most common mutations but it’s from the other two most common ones, that would be FLT3 and NPM1, it’s the most druggable one because the mutation gives an increased function of the enzyme and it’s easier to inhibit an enzyme that’s overactive than fix one that’s underactive.

How do you think this might impact therapy in the near future then?

The drug company that has the rights to midostaurin, Novartis, will likely submit this for approval in this disease and one other relatively rare disease called systemic mastocytosis where it also works pretty well. They’ll submit it for approval to the FDA and if it gets approved I think it well may become the standard of care for this subtype of AML patients.

What should doctors take home from this right now, then, do you think?

They should take home, number one, that it is possible to target a molecular lesion along with standard chemotherapy and that targeting, at least in this case, was beneficial for patient survival. Number two, they should stay closely tuned to FDA approvals to see when they can give this to their patients. Although Novartis will probably have a compassionate use programme to make it available to patients before the actual approval.