My Treatment Approach: Acute Myeloid Leukemia (AML) - Episode 8

Importance of Molecular Testing in Acute Myeloid Leukemia to Guide Treatment


Richard Stone, MD, and Eunice Wang, MD, explain how molecular biomarker testing influences their treatment decision-making in patients with acute myeloid leukemia (AML).

Richard Stone, MD: There are a lot possibilities for this case. In addition to being chemotherapy responsive, the patient would also be responsive in some way to IDH2 inhibitor enasidenib and to venetoclax along with chemotherapy of some sort. There are a lot of potential options, even though we’d all agree that for this patient right now, 7+3 would be the standard of care. As you pointed out, the Europeans are going to do a trial of 7+3 plus enasidenib vs 7+3 in this type of patient. I don’t know whether they’ll allow older patients to go on that trial, but that will be interesting. It’s a very difficult trial to do because only 10% of patients have an IDH2 mutation, so it’s going to take a lot of patients in the top of the funnel to get enough patients with IDH2 mutations to answer the question.

You mentioned venetoclax and 7+3. We’re doing a trial with that, but we can’t give it to people who are over age 60, even if they’re healthy, because we think it’s too toxic. The same is true for the FLAG-IDA [fludarabine, cytarabine, idarubicin, granulocyte colony-stimulating factor] and venetoclax regimen recently published by Courtney DiNardo that has a very high complete remission rate and a very high MRD [minimal residual disease]–negative rate in those who achieve remission. But it’s pretty toxic. I wouldn’t give it to even a healthy 67-year-old patient, so we’re left with 7+3 or possibly azacitidine plus enasidenib based on work that Dr DiNardo recently published of azacitidine vs azacitidine plus enasidenib in people who are unfit for chemotherapy. It wouldn’t apply to this gentleman who’s fit for chemotherapy.

That trial showed an improved event-free survival for those who got the doublet compared with azacitidine, like the LACEWING trial, with a much bigger response rate for FLT3 inhibitor plus azacitidine compared with azacitidine all along. But they didn’t pan out in terms of overall survival, so enasidenib-azacitidine can’t be considered a standard—even in unfit patients—although it certainly has a high response rate. The big question as we talked about earlier is whether venetoclax and azacitidine might be a reasonable option as opposed to 7+3 in this space because of the very high response rate and tolerability. We don’t know the answer to that, so 7+3 is still the way to go with this patient, although we expect that to change in the future.

Eunice Wang, MD: Given that the recommendation for this patient is to give 7+3 in the absence of a third agent, do you think it’s important for these patients eligible for intensive therapy to have their IDH1 and IDH2 resulted? Because in the current era, you aren’t going to do anything differently. We do something differently for FLT3. But for your intensive patients, what type of mutational testing are you doing prior to starting therapy? In our case, we test for FLT3 and IDH1 and IGHV, but for this type of patient, we wouldn’t necessarily alter anything we’re doing based on it.

Richard Stone, MD: You’re right. But we feel strongly that we should get an NGS [next-generation sequencing] panel plus cytogenics on every patient before deciding on therapy, because if this guy had MDS [myelodysplastic syndromes]–related cytogenics, we might have used CPX-351, as we did in your case. We wanted to get that back rapidly, and you can’t do that if you pursue cytogenetics. We’d also like to get the NGS panel if we can. We’re lucky because we get it back in 4 days, but quickly so we can decide if they have a secondary-type mutational pattern in which we might be more willing to use azacitidine-venetoclax as opposed to 7+3, which seems to be more chemotherapy responsive.

Even in the community, if at all possible, we should be getting cytogenetics and an NGS panel back quickly before deciding, particularly in an older patient like this who’s got a fairly low white blood cell count. There’s no rush because the white blood cell count isn’t high, and data from Germany suggested you can wait. I like to get as much information as I can from cytogenetics and the genetic panel before making a decision, because there are so many options. You’re right: right now, if your patient is going to get 7+3 or CPX-351 [cytarabine, daunorubicin], that requires cytogenetics. If the patient is unfit for chemotherapy, you’re probably going to give only azacitidine-venetoclax. Although with an IDH1 mutation, you might consider the doublet. Do you want tell us about [what you’d do] if this person had an IDH1 mutation and was unfit for chemotherapy? We’re changing the case a little.

Eunice Wang, MD: I agree with you. I always get a next-generation sequencing panel. We have an in-house rapid heme test that we can get back in 3 to 5 days. But we have patients like this: we think 7+3 is fine, we get the panel back, and they have secondary mutations. All of a sudden you’re switching to CPX-351 [cytarabine, daunorubicin]. Or God forbid they have a TP53 [mutation], which would be an unusual in normal karyotype. But if they have a TP53 mutation, I’m thinking venetoclax-azacitidine. Or with something really bad, I definitely agree.

In terms of IDH1-mutated patients, we have more options because the results of the phase 3 AGILE trial were just published in the New England Journal of Medicine. They demonstrated clear superiority in patients with newly diagnosed IDH1-mutated AML [acute myeloid leukemia] who were treated with the combination of ivosidenib and azacitidine vs placebo and azacitidine. Unlike the enasidenib-azacitidine data or the gilteritinib-azacitidine data, this trial was properly done in terms of a placebo control, and patients didn’t get taken off study and moved onto the inhibitor in the follow-up setting, which may have confounded the overall survival results.

The AGILE trial showed a clear advantage with a median overall survival of over 24 months and markedly enhanced outcomes on every parameter looked at, favoring the ivosidenib and azacitidine. However, if we’re going to be using this in individuals who are unfit for chemotherapy, it will require you to know that the patient is IDH1-mutated prior to starting induction therapy, because IDH1 mutations are uncommon, occurring in 1 in 20 patients, even though in older patients we can wait for that result. When you do a cross-trial comparison—which one should never do, but we do anyway—looking at the results of VIALE-A with venetoclax-azacitidine in IDH-mutated patients vs the AGILE trial results with ivosidenib-azacitidine, both of those trials seem to be resulting in a median overall survival of about 24 months or more. About two-thirds or more of patients in both of those trials are achieving overall responses.

As you mentioned, we have a lot of options. For older unfit patients, I always do a next-generation sequencing panel. Using an IDH inhibitor in the up-front setting—whether it’s IDH1 or IDH2—requires me to not only perform that testing but wait for that result before starting therapy. That’s clinically able to be done, but I don’t know how that translates into the community and whether community practitioners are going to wait for that IDH1 result, which occurs in 1 in 20 patients, or whether they’re going to go with venetoclax-azacitidine, which in the best-case scenario has very similar outcomes. What are your thoughts on that?

Richard Stone, MD: I agree with everything you said. The results of the AGILE trial in the combination arm were so good that they rival anything you’d expect with 7+3 alone in a patient like this. It brings up the question of whether you should treat these people with enasidenib and azacitidine. Of course, there are no data to support that.

Eunice Wang, MD: Then just get rid of the 7+3? Is that what you’re saying?

Richard Stone, MD: That’s what I’m saying, potentially. Of course, you said you can’t compare trial A with trial B, but you’ve got unfit patients who do so well with that combination. Don’t you think fit patients would do just as well?

Eunice Wang, MD: Yes. There’s no reason to say, “Just because you’re fit, we’re going to give you a more intensive regimen just because we can.” Just because we can do something doesn’t mean we should.

Richard Stone, MD: It’s like punishing someone for staying in shape all their life and not getting coronary artery disease.

Transcript edited for clarity.