Daniel Pollyea, MD, MS: IDH is a really exciting new target. Somewhat unexpectedly and fairly recently, we became aware that IDH was a recurrent mutation in upward of 15% of our patients. And very quickly, companies, industry and clinical trials have been coalesced around this to deliver 2 FDA approved therapies, ivosidenib for patients with an IDH1 mutation and enasidenib for patients with an IDH2 mutation. Both are approved in the relapsed/refractory setting for patients with an IDH mutation as a single agent. IDH1, as we discussed before, ivosidenib also approved for a newly diagnosed patient as a single agent. But both of these can be very effective. About 30% of patients experience remissions. Other patients who fall short of that have improvement with respect to transfusion requirements, fewer or independence [from] transfusions. And median response duration, somewhere between 9 and 12 months. So these are really good options for patients and fairly well tolerated.

The big toxicity we all think about is differentiation syndrome. About 10% of the time we see this. It’s not necessarily the immediate differentiation syndrome that we see in the treatment of APL [acute promyleocytic leukemia]. It can be delayed. Recognition of this, supportive care measures, attempts to treat through, because these patients can go on to have very good outcomes, employing these steroids—all these are very important principles of using these drugs.

Jorge E. Cortes, MD: But one thing I would like to emphasize, not all of the differentiation syndromes come with an increase in the white cell count and the blast count, but some do. And I think it’s very important to recognize, 1) because they have to be managed for the safety of the patients and all that but also because, and probably in the early days of the development, we missed some of these and we thought the patients were progressing and we changed to another therapy. And that may be a nervous kind of response. So patients need to be taken care of and recognize the differentiation syndrome because that patient may still benefit from therapy. Once you manage these, you can go back to the therapy and that patient may still respond. We need to be careful not to take them off prematurely.

Harry Erba, MD, PhD: How about combinations of HMAs [hypomethylating agents] with these drugs or 3 and 7 [cytarabine/daunorubicin]?

Daniel Pollyea, MD, MS: The hypomethylator combinations, very early data. We’ve seen some glimpses of these at meetings and presentations. It looks to suggest that responses are even higher. Particularly in the upfront treatment setting, these combinations really look very promising. And there’s been a program going on for the last several years combining these with intensive induction chemotherapy for eligible patients. Very early days of this, but like Sasha [Alexander] was saying before, getting these targeted therapies into the upfront or earlier in the treatment course, I think we all have a sense that that would be better. It was better for midostaurin. It think it might apply to other targeted therapies, too.

Harry Erba, MD, PhD: I’ll finish with gemtuzumab, and I’ll come back to basically chemotherapy but targeted to the leukemic blasts. We all know that in September of 2017, the FDA reapproved gemtuzumab ozogamicin for the treatment of patients with AML as a single agent or in combination with chemotherapy, so quite broad label, reflecting the trials that led to the approval. But I think the trial that I focus on the most is the ALFA-0701 study, where patients between the ages of 50 and 70, who were fit for chemotherapy, were given standard 3 and 7, 60 [mg] of daunorubicin, followed by an intermediate dose of Ara-C and a dose of daunorubicin with or without [gemtuzumab ozogamicin] GO, 3 mg/m2 capped at 4-and-a-half mg on days 1, 4, and 7. In the 2 arms there was no difference in response rate, but the event-free survival was significantly different and better with gemtuzumab.

At 5 years there was still a trend of an overall survival benefit. In the subset analyses, the benefit was seen in the older patients, 60 to 70. It was seen mostly in the intermediate-risk group, and there was no benefit as we’ve seen for addition of GO, to unfavorable [subset], so again a great reason to get karyotypes and FISH [fluorescence in situ hybridization] at baseline. No to GO for unfavorable. Maybe liposomal daunorubicin cytarabine though in that group. So we use that information. Interestingly, they had 18 percent of patients with a FLT3 ITD [internal tandem duplication], and the event-free survival benefit with GO was quite impressive there.

And those patients didn’t get midostaurin, which of course leads me to where the future is, combinations. Now we have 7 and 3 in GO and we have midostaurin and other FLT3 inhibitors. We’ve really got to get out of the studio and back to work in developing these clinical trials, so we have safety data and hopefully efficacy data. I know people are concerned about the toxicity of GO though. There was a little bit more myelosuppression, but there was no difference in early deaths, no difference in transfers to the unit, and actually no difference in hepatotoxicity even though there were 3 VOD [veno-occlusive disease] cases on the GO arm. Now only 6 of those patients who got GO went to transplant. None of them who went to transplant after GO actually developed VOD after the transplant. There were other patients off study afterward who got GO as a salvage who then went to transplant and did have veno-occlusive disease. So clearly a benefit in the core binding factor leukemias and in intermediate risk. I think we have to figure out how to combine these agents now safely.

Transcript Edited for Clarity