Harry Erba, MD, PhD: Just to finish up this discussion, since we’re running out of time, you know we’ve been using antibody-drug conjugates [ADCs] and identifying markers that are more specific for AML [acute myeloid leukemia] to target with ADCs, with BiTEs [bispecific T-cell engagers], with DARTS [dual-affinity retargeting antibodies], and all of these new things. And, of course, we have now gemtuzumab ozogamicin [G/O] back with us based primarily on the results of the ALFA-0701 study showing a benefit of the fractionated gemtuzumab—3 mg per meter squared on days 1, 4, and 7 in patients between the ages of 50 and 70 with previously untreated AML. There were similar CR rates compared with chemotherapy alone when combined—G/O with 7 and 3 [regimen]. But there was a better event-free survival and a trend for a better overall survival.

At this meeting we heard about veno-occlusive disease. There were only 5 cases that were identified. Two were fatal, and actually 2 of the cases occurred after a patient got G/O in the relapsed setting. They had been on the chemotherapy arm. So it’s something to be concerned about, but in a study of about 300 patients, 150 on the G/O arm, it’s not a very high number that, in my mind, dissuades us from using a drug associated with an event-free survival benefit. The challenge we have is that it doesn’t work very well. There’s no benefit and poor risk, and what do we do with the intermediate risk with FLT3-mutated AML?

So I think all of those are issues out there. It’s also been looked at by the AML-19 study, showing a survival benefit versus best supportive care. But no one’s going to get best supportive care any more. I hope not. I think maybe that should be our one message. We have a lot of options for patients who aren’t so frail that they can’t get anything. We have many things we can give them.

But I want to turn to you, Naval, at this point. What’s coming, in terms of immunotherapies here and targets?

Naval G. Daver, MD: I think there’s a lot of immunotherapy activity, and I think the paradigm is falling very much to solid tumor, melanoma, lung where the targeted therapy, the inhibitors, etcetera came first, and then you had this autoimmune therapy activity coming in. So I think there are 2 big areas. You’ve got the ADCs. You talked about Mylotarg, the antibody drug conjugates and the targets we traditionally use, which we continue to use, CD33, CD123. There’s a newer one, CLL-1 or CLEC12A, that seems to be maybe more specific for AML.

So at this year’s [2018] ASH [American Society of Hematology meeting] there were a lot of immunotherapy presentations in different arenas. Some of the ones that we saw of interest included ImmunoGen, 2 drugs that they have. One is a CD33-targeting ADC. One is a CD123-targeting ADC. These have chemical toxins which we have always felt or had the perception are better tolerated, and maybe have less capillary leak and other adverse effects with the bacterial toxin. And indeed, both of these agents seem to not have at least a capillary leak problem.

As a single agent in the relapse, the CD33 had modest activity in the way of getting true CR, CRis [complete responses with incomplete hematologic recoveries], but they did see blast reductions in a broad variety of patients. So, again, hinting that we need to do combinations, but overall to say there were no major cytopenia or capillary leak issues. So amenable combinations, and that’s what the company is looking at.

And then in the CD123, which is one of the studies that we were involved with, we are seeing responses of CR, CRi rate of about 28% to 30% when you take all those levels. One of the interesting things with that, which we still don’t clearly know why and from the presentation it wasn’t clear to us as well is, why are you seeing responses at all the dose levels? So we were seeing responses at 1, 2, 3, 4, 5—and that’s good, because we just had the discussion and we’re going to expand multiple dose levels because maybe you don’t need to go to a very high dose level.

Harry Erba, MD, PhD: That was seen with the 123 ADC from ImmunoGen as well. At the lowest dose levels.

Naval G. Daver, MD: Yes, which will be nice if that’s so. You know we kind of lean on the gilteritinib design because I like that very much. You can put 8 or 10 more patients at each dose and get more data, and we’re doing PD/PK [pharmacodynamics/pharmacokinetics]. And if it is true and we can use a low dose that had very little-known myelosuppression, that will be wonderful for combinations. You can add it to AZA [azacitidine], AZA/VEN [venetoclax]. You know, other things.

Harry Erba, MD, PhD: That’s the very important point. There were 4 oral abstracts that dealt with targeting CD123, even with ADCs or BiTEs, or things like that. And every speaker made the same point. They weren’t seeing myelosuppression. So maybe combinations will be easier in that setting.

Naval G. Daver, MD: Which was very interesting. I was happy to hear that because I was a little surprised when we actually looked, trended these people, and their neutrophils, over time, were going up, which is like after 2, 3 cycles, and platelets were a modest drop. This is very different from what anybody expected, but this may be good.

And then, as you mentioned, the bispecifics. I think with the bispecifics, we mentioned this a little bit. We’re not getting a stop-the-disease-in-your-track situation. Maybe we shouldn’t have expected it. We know between BLINA [blinatumomab] and INO [inotuzumab] in ALL [acute lymphoblastic leukemia], INO is a very powerful agent in the big bulky disease. It is better than BLINA. But in the low burden disease, MRD [minimal residual disease], BLINA was very good. I think in AML, of course, with less experience we’re seeing the same thing. If you have MRD, low volume diseases, primary refractory, your inflammatory immune signature is still favorable. You haven’t killed and wiped out every T-cell with fludarabine, clofarabine, MEK and CLAG [cladribine/cytarabine/granulocyte-colony stimulating factor], and you can actually get some activity. But once you go to multiple relapse, once you go to people who have had a lot of HMA [hypomethylating agent] therapies, have 20%, 30%, 40%, 50% blasts, these are not working.

So I agree. I think these should be developed like BLINA, you know, for MRD eradication.

Harry Erba, MD, PhD: So, Naval, I think a lot of us look at you as Dr Checkpoint Inhibitor in the myeloid malignancies. Are those ready for prime time in AML?

Naval G. Daver, MD: No. I think that’s the answer. And I actually don’t ever know if they’re going to be prime time in the commercial setting in that way. When we’re looking at immunotherapies, we’ve had a lot of experience in immune checkpoints and bispecifics and CAR [chimeric antigen receptor] T-cells. And these are all a continuum gradient for toxicities.

So CAR T-cells, by far, are the most potent. Nobody ever thinks about commercializing, doing CARs in community. Bispecifics are going the same way. And I think with checkpoints it’s kind of similar. They’re less intense in the toxicities, but these do need to be done in big academic centers or really good community practices.

So what we are seeing is in the relapsed population, we presented this data, especially in the early salvage. And that theme is again coming up—early salvage, T-cell population well preserved, functional T-cell markers. We actually are getting very good response rates of almost up to 55%, 60% in people who were prior HMA-naïve, and the survival is 11 months, which is actually what we would get with a single-agent AZA up front.

So I think these agents can be developed, but the key is going to be that they should be done in big centers that have a lot of experience with bispecifics and others going forward.

Harry Erba, MD, PhD: Well, let me say that in a slightly different way. AML is still a rare disease; 3 out of 100,000 Americans are diagnosed every year. Only 20,000 cases per year are diagnosed. This is a very rare disease. And basically you can summarize from this whole hour and a half or longer discussion, we have 8 new agents and we, who see all these patients on a daily basis, aren’t being sidetracked by patients with solid tumors. We don’t know how to incorporate them. And so, what I would really argue is, some patients won’t want to go to major centers and get treatment there, but at least get a consultation.

Naval G. Daver, MD: Yes.

Harry Erba, MD, PhD: This is difficult stuff. We don’t know what to do. You have the patient in front of you. Partner with somebody close to you that might be able to help you through this decision-making.

Transcript edited for clarity.