IDH Inhibitors in AML Therapy

Transcript: Daniel A. Pollyea, MD, MS: Enasidenib, the IDH2 inhibitor, and ivosidenib, the IDH1 inhibitor, are both now FDA-approved therapies for relapsed and refractory IDH2- and IDH1-mutant AML [acute myeloid leukemia] patients. And these are really beneficial therapies, and they’re in widespread clinical use. The response rates with both of these therapies tends to be in the 20% to 30% range, and responses on average can last a year or more. The tolerability for most patients is quite good, but there are several adverse events of special interest that need to be carefully considered.

The first would probably be differentiation syndrome, and perhaps as many as 10% or more of patients who are exposed to IDH inhibitors will experience differentiation syndrome.

In the clinical trial this was managed effectively in most cases, and with conservative measures like adding steroids or sometimes holding drug. And there were some patients who, if they could get through the differentiation syndrome, went on to have nice responses. But it is a very difficult and potentially deadly complication adverse effect of this medication that has to be really carefully looked for. And that’s a difficult challenge because differentiation syndrome can take many different forms. Fevers can be differentiation syndrome, fluid accumulation, rash. Those are also things that we frequently see in AML patients outside treatment with an IDH inhibitor. So it’s a real challenge to try to pick that up.

Patients also with enasidenib, the IDH2 inhibitor, can have an increased indirect bilirubinemia, which has no real clinical significance. So the biggest thing to be cautious about there is not to inappropriately stop or hold it there when you see that. Because that’s a largely benign finding.

Ivosidenib can cause some QT prolongation in the clinical trials. This wasn’t terribly significant, but it’s something that should be looked for. And then finally episodes of leukocytosis, which can be associated with differentiation syndrome but not always, are something that needs to be watched out for very carefully because that can also be very dangerous.

Given that enasidenib and ivosidenib are now FDA-approved therapies for relapse and refractory AML patients, I think it’s incumbent on us to know and understand if that mutation is present because this can be an effective strategy for some patients in that situation. Whether it’s necessary to know this at the time of diagnosis is less clear because there are not very clear prognostic implications from IDH mutations at the time of diagnosis, and to date there’s no strategy to treat IDH-mutant patients with inhibitors in the up-front setting. Now that could change, and we heard some data at this meeting about how that could change. But at the moment, that’s not the case.

Differentiation syndrome is a potentially deadly complication of the use, with the use of IDH inhibitors in AML. In the clinical trials it looked to be around 10%, but we have some recent data to suggest that maybe it’s a bit higher—at least now that it’s in the clinical practice, we’re seeing incidences in the community and perhaps even a little bit higher than was reported in the clinical trials.

Regardless, it’s something that is a very difficult complication to detect sometimes and then to manage. With respect to detection, there are many different manifestations of differentiation syndrome that can overlap with magnifications of leukemia or other complications—rash, food accumulation, fevers—so a clinician has to really be on the lookout for this. When detected, steroids have to be initiated very quickly. In some cases, it’s even appropriate to hold the drug for a bit. It’s important, though not to abandon treatment necessarily within differentiation syndrome, because we have the patients who experience this and get supported through it can get through it and get to a remission or have some degree of response.

Using IDH inhibitors in the frontline setting is a very logical thing to consider. This is the path that FLT3 inhibitors went down. So another targeted therapy had some activity in the relapsed setting, got moved into the up-front setting, and had a clear survival benefit. So I think that’s of great interest. We now have some data showing in pilot studies induction chemotherapy with enasidenib and ivosidenib. We have some data showing enasidenib and ivosidenib as single agents in the up-front setting for unfit patients. And there are emerging data on enasidenib and ivosidenib with azacitidine in that setting as well that looks pretty promising. So I think all those strategies are percolating through, and in the next couple of years we should know more about that.

There’s really no preclinical biological rationale for adding an IDH inhibitor with chemotherapy. I think this is what we do in drug development in which we see an effective drug in the relapsed setting and then just move into the upper setting with induction chemotherapy. But I’m not aware that there’s any suggestion that these would be additive or synergistic.

We have now heard the results of several dozen patients who receive this combination, and it doesn’t appear to be antagonistic. But thinking to do this combination is simply following sort of the way clinical drug development usually happens for AML.

Transcript Edited for Clarity