Patient Profile 2: Patient with de novo AML and IDH2 Mutation

Richard Stone, MD: I’ll present our second case. This is a 67-year-old man who presented with bruising and fatigue on his daily 3-mile walk. He had open-angle glaucoma being treated with timolol drops but was otherwise quite robust and healthy. For this bruising and fatigue, he went to see his doctor and was found to have a white blood cell count of 2800 per mm3, a hemoglobin count of 9.1 g/dL, and platelet count of 40,000 per mm3, making him leukopenic, anemic, and thrombocytopenic, respectively. He had pancytopenia. As implied, his ECOG performance status was good. His bone marrow blasts were 55%. They were myeloid, so he had AML [acute myeloid leukemia]. He had normal male karyotype and an IDH2 mutation with a 45% variant allele frequency, a RUNX1 mutation with a 25% variant allele frequency, and a DNMT3A mutation with a 55% variant allele frequency.

This is a patient who has de novo AML. There’s no history of prior therapy for another cancer. There’s no history of prior MDS [myelodysplastic syndromes]. In fact, the mutations seen here are more typical of pan-AML mutations, particularly IDH2. DNMT3A is something we see in CHIP [clonal hematopoiesis of indeterminate potential], along with TET2 and ASXL1 if the variant allele frequency is quite high. RUNX1 is often considered a progression mutation and is present in a smaller variant allele frequency, suggesting but certainly not proving that it’s a subclone that grew out IDH2 and could be a founder mutation seen in MDS and in primordial stem cells, or it could sometimes be a secondary mutation occurring as a progression mutation in people who go from MDS to AML.

IDH2 is a very interesting enzyme. It’s found in the mitochondria, and it catalyzes a reaction in the oxidative decarboxylation cycle or Krebs cycle. It catalyzes isocitrate to alpha-ketoglutarate. When it’s mutated, it catalyzes a metamorphic reaction product called 2-hydroxyglutarate. That phenocopies the epigenetic changes the current pace of the TET2 mutations. It’s definitely a leukemogenic enzyme, and if we could decrease the 2-hydroxyglutarate concentration within the cell by inhibiting IDH2, we can affect some biological and clinical efficacy with an IDH2 inhibitor. We’re faced with deciding how to treat this older gentleman who is robust and has an IDH2 mutation and a couple of other mutations. Eunice, how do you view this case? What would you consider for induction?

Eunice Wang, MD: I’m going back to individualizing treatment. This gentleman is also relatively healthy. He’s robust, as you said. He has an intermediate karyotype and doesn’t have any unfavorable cytogenetic or necessarily mutational abnormalities. Similar to our first case, I’d offer him intensive chemotherapy. As far as the options, standard 7+3 works very well, particularly in IDH2-mutated disease. Standard 7+3 chemotherapies certainly would be an option for the patient.

A more interesting question comes up. We know that IDH2-mutated leukemia is said to also overexpress BCL2, and we know that we also have the availability of small-molecule inhibitors of IDH2, specifically enasidenib, which have been developed and highly effective in bringing down the IDH2 variant allele frequency. Other intriguing options that we could discuss would be whether a combination of venetoclax–7+3, which is being explored on clinical trial, would be of interest, or maybe adding enasidenib to 7+3 would be an option. Both have been explored in clinical trial. I’ll let you speak to the venetoclax–7+3 data.

What we have known from adding IDH1 and IDH2 inhibitors to 7+3 is that they’re well tolerated. They don’t appear to negatively impact the overall survival or short-term survival. But how much they’re adding to the standard 7+3 isn’t known. That’s being explored in a phase 3 setting where those IDH1 and IDH2 inhibitors are being added to half the patients with IDH-mutated AML and omitted in the other half. It isn’t certain whether those inhibitors are going to add much in a cytotoxic fashion, but they seem to be well tolerated. There are some early data suggesting that that type of approach works better for de novo IDH-mutated disease than secondary IDH-mutated disease, but we’re awaiting the results of those phase 3 trials. Maybe you could address the possibility of adding venetoclax to 7+3.

Transcript edited for clarity.