Decision-Making for Patients With IDH1-Mutated Relapsed or Refractory Acute Myeloid Leukemia

Eytan M. Stein, MD, is a hematologic oncologist and director of the Program for Drug Development in Leukemia at Memorial Sloan Kettering Cancer Center, New York. His practice focuses on the treatment of acute and chronic leukemias, myelodysplastic syndromes, and myeloproliferative neoplasms. His research interests include developing novel early-phase clinical trials of compounds that target the genetic and epigenetic basis of myeloid malignancies.

How do IDH1 mutations change the clinical course of AML and the treatment response in patients?

Mutations in isocitrate dehydrogenase come in 2 flavors: IDH1 and IDH2. The IDH1 mutation occurs in approximately 10% of patients newly diagnosed with AML.¹ Among those 10%, the mutation tends to occur in patients who are a little older — above the age of 60 — and who might be unable to tolerate intensive chemotherapy. What’s exciting about targeted therapeutics for IDH1-mutated AML — both ivosidenib, which is approved for clinical use, and others being developed — is that they offer a way to get rid of leukemia by causing differentiation of cancer cells rather than cancer-cell death. Clinicians might be familiar with differentiation therapy from treating acute promyelocytic leukemia. The basic mechanism is that, instead of causing cancer cells to die, it causes them to mature into healthy, normal cells.

What are the treatment options for patients with IDH1-mutated relapsed or refractory AML

In a patient who has an IDH1 mutation, has had treatment for AML, and experiences relapse, the IDH1 inhibitor is the only option. Still, treatment of relapsed and refractory AML breaks down into different buckets.
 
In younger patients who are eligible for intensive chemotherapy a second time, an option is to give only intensive chemotherapy again. Another option is to add targeted treatment.
 
Older patients start out with nonintensive therapy, so that when they relapse, the only option is giving them a targeted IDH1 inhibitor. But these inhibitors are also newly approved as a first-line treatment for patients with newly diagnosed AML who aren’t eligible for intensive chemotherapy.² The single-agent IDH1 inhibitor is actually a good potential treatment option because it’s oral, so it can be taken at home without intravenous administration or clinic visits. Also, because this agent is targeted, patients don’t develop the low blood count seen with other treatments, so they are less susceptible to infection and don’t have to come in twice a week for infusion, which means that they experience fewer challenges to their quality of life.

Is ivosidenib more effective for certain patients?

There might be some correlation between the IDH1 mutation and other mutations that are part of that patient’s AML, because their malignant cells have multiple mutations. Sometimes, even if treatment targets only the IDH1 mutation, everything else goes away; if there are concurrent mutations in RAS, however, the patient is less likely to respond to targeted IDH1 mutation therapy.³ Therefore, if a patient has both mutations, they can be given an inhibitor if that’s the only option and they still might respond.

Does ivosidenib work better in combination therapies versus as a monotherapy?

That’s an open question. We conducted a study that combined ivosidenib with induction chemotherapy; the outcome was that the combination is safe and you can administer it without increasing adverse effects. The second thing we saw, in terms of efficacy, was that the remission rate was high at approximately 70%, and the survival rate was approximately 60% at 1 year.⁴ We think that, if you administer chemotherapy and an IDH1 inhibitor, the combination will certainly be effective, but will it be more effective? That is the subject of a randomized study sponsored by HOVON [from the Netherlands; ClinicalTrials.gov Identifier: NCT03839771]. Still, that’s for younger patients who can tolerate intensive chemotherapy. For older patients, there is an ongoing randomized study combining ivosidenib with azacitidine [ClinicalTrials.gov Identifier: NCT02677922] to answer the question of whether combining therapies for older patients is safe and efficacious.⁵

When might ivosidenib be ineffective?

Cancer cells can mutate to get around any targeted therapy. A number of mutations have been described that make ivosidenib ineffective by blocking its ability to bind to the mutant IDH protein. There is an ongoing trial of Ly3410738, an IDH1 inhibitor that binds differently to the mutant IDH protein [ClinicalTrials.gov Identifier: NCT04603001]. The drug might be more effective than ivosidenib or at least able to overcome resistance mutations while a patient is taking ivosidenib.⁶

What adverse effects should clinicians monitor for during ivosidenib treatment?

Differentiation syndrome happens in 10% to 15% of patients because, as malignant cells differentiate, they release cytokines that can cause pleural effusion, pericardial effusion, or pulmonary edema.⁸ If these problems aren’t treated immediately, respiratory failure and death can occur. It’s important to be aware of this possibility; typically, it would occur once improvement in the patient’s blood count begins.
 
If a patient comes in after taking ivosidenib for about 1 month and complains of shortness of breath, swollen ankles, or weight gain, the physician should order a chest radiograph and rule out other possible causes of infection. If the problem is differentiation syndrome — and even when the level of suspicion is low — dexamethasone 10 mg twice daily should be started immediately. When addressed promptly, differentiation syndrome is treatable; otherwise, it can be fatal.

Are there enough data to indicate that ivosidenib improves outcomes in the longer term?

Whether ivosidenib improves long-term outcomes is what we’ll learn from the Agile and HOVON150AML [ClinicalTrials.gov Identifiers: NCT03173248 and NCT03839771, respectively]; we don’t really know yet how much this drug improves overall survival. We’re eagerly awaiting the data. Agile outcomes will be available in 1 or 2 years; HOVON150AML outcomes will take longer.