Toxicities Associated With IDH Inhibitors

Harry Erba, MD, PhD: Now, let’s turn our attention to the other side of the coin of using the IDH inhibitors. There are a number of toxicities: hyperbilirubinemia, unconjugated, with enasidenib—that is not something I usually worry about—and QT prolongation with ivosidenib. You need to take that into account and watch your CYP3A4 inhibitors, and your electrolytes, and all those things we learned how to do when we started having second-generation ABL kinases in our portfolios. What about differentiation syndrome? Can you walk us through your approach to identifying that and treating that?

Vinod Pullarkat, MD: One thing I would say up front is, sometimes, regarding some of these oral agents or so-called low-intensity therapies, there is a perception that these are less toxic, or the patient needs less monitoring. This is one example where you have an oral drug, which is well-tolerated generally, but you have this dangerous toxicity of differentiation syndrome that can happen in about 20% of patients. Patients need to be closely followed, if you see [high] white [blood] cell counts, they need to be started on steroids, and sometimes you need additional therapy like hydroxyurea, or in some situations, they even need…. I think the key point is that, for these patients, it is a slow-acting drug, but they do need to be followed very closely. I often see in practice patients being brought back in 2 or 3 weeks after starting the pill. That is an important message, that differentiation syndrome can be lethal and patients need close follow-up.

Harry Erba, MD, PhD: The challenge, I think, for the practicing physician who does not see a lot of these patients is that there is no diagnostic test for differentiation syndrome, which about 20% of patients can experience with ivosidenib or enasidenib. Think about how these patients come back. They come back sometimes with a leukocytosis, fevers, dyspnea, hypoxia, pulmonary infiltrates. How do you distinguish progressive AML [acute myeloid leukemia] with pneumonia or leukostasis from differentiation syndrome? That could be a challenge; I would be shocked if any of you say, “This is a simple thing to do.”

Courtney DiNardo, MD: I was going to say you can only identify it in retrospect.

Harry Erba, MD, PhD: Exactly.

Mark Levis, MD, PhD: You only identify it if you’re looking for it.

Harry Erba, MD, PhD: That is right. You must consider it.

Vinod Pullarkat, MD: Unlike in the real world, patients are inpatient, and if you are checking laboratory [tests] every day, these patients often come back sick. That is the challenge. I do not know if there’s a standard recommendation of how closely patients should be followed, but I like to see them twice a week for some treatment to make sure it doesn’t happen.

Harry Erba, MD, PhD: You must see them. I tell them to check their weight at home, to check for fevers, and shortness of breath, I am not just giving them another red [blood] cell transfusion. It might be differentiation syndrome. Once you bring them into the hospital, you do everything. Of course, you treat pneumonia with antibiotics, and of course, you treat the high white [blood cell] count with hydroxyurea. Most importantly, you treat with dexamethasone 10 mg twice daily. The least important thing to do, in my opinion, is stopping the ivosidenib or enasidenib. I usually end up doing it, especially if they are going to the ICU [intensive care unit], because if I don’t do it, the ICU attending [physician] will do it. They will not care what I say, but it has a long half-life. Courtney, I think your point is well taken. In retrospect, you know because if they get better quickly, it is probably differentiation syndrome, or if you see differentiation on the peripheral blood or the platelet count is going up. I’ll tell you, it is unusual or less common to see responses with IDH inhibitors in patients who have mutations in receptor tyrosine kinases like FLT3 or the Ras/MAPK kinase, so Ras mutations, multiple different mutations. I use that mutational profile to tell me, is it likely to be proliferation and refractoriness? After all, the minority of patients are the ones who are having a response. We have to keep that in mind.

TRANSCRIPT EDITED FOR CLARITY