Harry Erba, MD, PhD: Let’s move on and turn our attention to isocitrate dehydrogenase: the IDH inhibitors. Dan, you had a leading role in the development of those drugs.

Dan Pollyea, MD, MS: Yeah. This has been a fascinating story. At this point, most people are aware of what’s happened in the last several years. IDH1 was first identified in the first cancer genome, full genome sequencing, and was not on the radar of anyone in the AML [acute myeloid leukemia] world. Now we know that it’s a recurrent mutation. It and its isoform, IDH2, make up upward of 20% of AML cases. Very quickly, the field has developed inhibitors for these therapies. These inhibitors have very quickly been approved by the FDA for general use. We all have quite a bit of familiarity with these, and we’re excited. At this meeting [American Society of Hematology Annual Meeting & Exposition], some of the new data that we’re hearing about are perhaps combining them with hypomethylator backbones and moving them into the up-front setting. Like Mark was commenting on his enthusiasm for that strategy with very specific FLT3 inhibitors, the same principle may apply with IDH. That’s where a lot of colleagues and people at this table are moving toward.

Rami Komrokji, MD: There are some challenges because the patients with IDH mutation do well with other combinations as well. The last common question you get to ask now: If you have a patient with IDH mutation up front, not fit for intensive chemotherapy, are you going to do azacitidine-venetoclax or are you doing to do an IDH inhibitor? I don’t think we even have that answer. Both are good options. One of the areas I find helpful, because I also do a lot of MDS [myelodysplastic syndrome], is patients in MDS after HMA [hypomethylating agents] failure, when they progress to AML, they don’t have options. That HMA-venetoclax combination is not as viable for them, and many of them are not candidates for intensive chemotherapy. If they have the mutation, I think single agent in that setting is reasonable because they don’t necessarily even do well with intensive chemotherapy.

Mark Levis, MD, PhD: And it can be dramatically effective.

Rami Komrokji, MD: Right.

Harry Erba, MD, PhD: Didn’t Courtney DiNardo present some data about IDH mutations in those secondary AML patients?

Naval Daver, MD: Yeah, so we at [The University of Texas] MD Anderson [Cancer Center] have a study. It’s a multicenter study through the MDS cooperative group looking at exactly what Rami is saying: patients with MDS who had AZA [azacitidine] and either at MDS relapse had IDH or transformation to AML had IDH. If I recall, the response rate was similar to what you see with IDH in relapsed AML: about 35% to 40%, which is good because those people, if they’ve had at least 4 to 6 HMAs, it is a good trial. They don’t have much durability with HMA—VEN [venetoclax] and they may have a short response, but it’s very short-lived. Getting to that point, I think the same for IDH. I don’t feel bad if I started HMA—VEN [venetoclax] and I find a new AML patient has IDH and is not fit, that’s OK. The data are so neck and neck. For FLT3, there were data showing that, and these are data that we’ve been analyzing internally, with HMA—VEN [venetoclax]. The response rates are good: about 60% to 65%, but the survival is lower than the general. It’s about 11 to 12 months as shown by the AbbVie, combining everything even after the FLT3, and that’s also been shown in the lab. This is where doublets, adding maybe a FLT3 inhibitor may help. There were some data with a combination of venetoclax-gilteritinib showing very high responses in relapse. More than IDH, I think FLT3, we may feel that HMA—VEN [venetoclax] is not optimal at this time.

Harry Erba, MD, PhD: Courtney DiNardo, at this meeting [American Society of Hematology Annual Meeting & Exposition], presented data for enasidenib in IDH2-mutated patients in that randomized phase II study of enasidenib—AZA [azacitidine] versus AZA [azacitidine] alone. Of course, compared with AZA [azacitidine] alone, there were much higher response rates. It does become an option for your previously untreated patient. Before that, she presented data for ivosidenib.

Mark Levis, MD, PhD: Last year.

Harry Erba, MD, PhD: Last year, ivosidenib with azacitidine, overall response rate of 80%, 67% CR [complete response], CRI [constant rate infusion], 80% of patients were alive at 1 year.

Mark Levis, MD, PhD: I know you don’t like MRD [minimal residual disease], but in fact, the molecular response went down. In that New England Journal of Medicine article on ivosidenib, the molecular response is nicely correlated with survival, although it was a small data set.

Harry Erba, MD, PhD: OK, just to be devil’s advocate. In the paper by Eytan Stein, in which they looked at a larger subset of enasidenib, I was impressed that, when they looked at their CR, patients with enasidenib, MRD status made no difference in long-term survival. I thought that was…

Mark Levis, MD, PhD: I agree. I think IDH1 and IDH2 are different. I think IDH1 is a cleaner mutation.

Naval Daver, MD: In the New England Journal of Medicine paper, the visual impression is very good.

Mark Levis, MD, PhD: Oh, I know.

Naval Daver, MD: It’s good to see that we’re all happy, but we don’t know.

Harry Erba, MD, PhD: Let me talk about 1 thing I don’t think Courtney DiNardo presented for the enasidenib combination, but she did present it for the ivosidenib combination. One thing that impressed me in the few times I’ve used it is that the blood counts don’t go down. They go up immediately, and that might be the 1 distinguishing feature between the IDH inhibitor with an HMA versus venetoclax with an HMA. Of course, we don’t have any comparative data. At the end of the day, I try to make it easier for the people who do prior authorizations. VEN [venetoclax] is approved up front. Although ivosidenib is approved in the first-line setting, it’s also approved in relapsed/refractory. It just makes it easier for my staff to sequence it that way.

Dan Pollyea, MD, MS: The other logistics issue that’s a little easier is with respect to venetoclax: do you have to wait for molecular mutations to come back or can you feel comfortable moving forward with that?

Mark Levis, MD, PhD: You don’t have to wait for the IDH, it’s true. On the other hand, if you’ve got an indolent, older patient, AZA [azacitidine], HMA—VEN [venetoclax] will occasionally be a rough ride. Yet you want something to give you a little more oomph than just the single agents, so the combination has a lot of appeal.

Transcript Edited for Clarity