Evolving Treatments Address Challenging AML Subtypes

Richard M. Stone, MD

Professor of Medicine

Harvard Medical School

Director of the Adult Acute Leukemia Program

Dana-Farber Cancer Institute

Boston, MA

TARGETED ONCOLOGY: Can you explain the evolution of the acute myeloid leukemia (AML) treatment landscape and what progress has been made over the last 5 years?

Stone: Historically, AML was [considered] a single entity, and patients of almost all ages were treated with the same regimen: 3 days of daunorubicin [Cerubidine] or idarubicin [Idamycin] and 7 days of cytarabine [Cytosar-U]. It was recognized about 25 years ago that patients with acute promyelocytic leukemia [APL] should be treated differently, especially after the advent of 2 nontraditional chemotherapeutic, antineoplastic drugs for APL, namely retinoic acid and then arsenic trioxide. [I’ll also] mention that 10% of all AML subtype APML can be seemingly cured with over a 90% likelihood using a combination of retinoic acid and arsenic trioxide. But up until recently, the therapy for the other 90% of AML patients hadn’t changed. It was 3 plus 7 [days’ regimen, as mentioned earlier], regardless of the specific biologic subtype and regardless of their age. About 10 years ago, for older adults who weren’t deemed to be fit for chemotherapy, we started using hypomethylating agents based on a few trials that suggested that [agents such as] decitabine [Dacogen] or azacitidine [Vidaza] might be reasonably effective and not too toxic for older adults, but the increased understanding of the molecular biology of AML led to parsing treatment into different subcategories of AML and treating [AML types] differently. This coincided with the approval by the FDA of approximately 8 new drugs in AML over the last 4 years.

So it’s been a changing landscape recently. The 3 plus 7 that I described earlier is still the backbone of therapy for most patients, [although] it has undergone significant changes. For example, if a patient with core binding factor AML comes in the door, we’ll often consider using the 3 plus 7 plus the antibody-drug conjugate called gemtuzumab ozogamicin [Mylotarg], which is approved for all types of AML but…used mainly in this risk core binding factor subtype. That’s one [option] for a patient with FLT3, a mutated AML, and they’re fit for chemotherapy. They’re going to [be prescribed] 3 plus 7 plus a FLT3 inhibitor. For now, that FLT3 inhibitor is midostaurin [Rydapt]. Another category of people that have therapy-related AML, or AML that’s arisen after prior MDS [myelodysplastic syndromes] or even AML with cytogenetic features that suggests they may have an occult MDS prodrome, may receive a different type of 3 plus 7, namely the liposomal encapsulated molecule CPX-351. Then there’s an increased appreciation that patients with TP53-mutant AML––often associated with multiple karyotypic abnormalities and a very poor prognosis––who might not be served well by 3 plus 7 might be better getting the same type of AML therapy that we use for older adults with AML. Because of the results of the Viale-a trial [NCT02993523],1 patients over age 75 and/or those who have significant comorbid diseases [that] make them seemingly unfit for 3 and 7 should receive a hypomethylating agent such as azacitidine plus venetoclax [Venclexta], the BCL-2 inhibitor. That [has] revolutionized the cure [for] patients who are older and who can’t take standard chemotherapy. Of course, this leads to a lot of debates, [such as] what’s the best way to treat a patient who’s not 75 but still might be a reasonable candidate for 3 plus 7 chemotherapy or not.

[Primarily], it’s the addition of new targeted therapy to the 3 and 7 in younger adults and the use of venetoclax and azacitidine in older adults. That is what has changed.

Based on the treatments you just discussed, could you explore some of the newer strategies that are undergoing evaluation now?

Well, there are a lot of new strategies being explored…right now. [A highlight is] induction therapy. We’re trying to figure out whether patients with so-called targeted mutations should be treated with azacitidine and venetoclax or whether they should receive the targeted agent plus venetoclax, the targeted agent plus azacitidine, or all 3: venetoclax, [the] targeted agent, and azacitidine. Those [types of treatment] are so-called triplets, [which] are being promulgated with increasing interest but [are] not necessarily ready for prime time.

I’m talking about a patient with IDH1 and IDH2 mutations, which account for about 8% and 12%, respectively, of patients with AML. These IDH mutations can be so-called targeted, or the proteins that they elaborate can be targeted with the IDH1 inhibitor ivosidenib [Tibsovo] and the IDH2 inhibitor enasidenib [Idhifa]. Those drugs are approved for relapsed AML simulators. Ivosidenib was approved for upfront IDH1-mutant AML, but the question is whether they might be used in the early cases with azacitidine, or azacitidine plus venetoclax, or azacitidine plus venetoclax plus the targeted agent. [It’s the] same [question] in patients with a FLT3 mutation. The thing about patients with IDH mutation is they do very well with venetoclax plus azacitidine, along the same [evidence] that was [found] in the Viale-a trial. In the patients who are FLT3 mutant it’s less clear. It seems like it might be a good idea to give them a FLT3 inhibitor plus venetoclax or plus azacitidine or all 3, but we don’t know yet. And in younger patients, we’re also trying to weave in IDH inhibitors to those 20% of patients who have mutant IDH1 or IDH2. We’re coming up with new drugs that may target patients with mutant p53, which I already told you is a very bad group in terms of outcome. [Patients] don’t do well.

There’s a series of drugs, …supposed to be so-called p53 refolding agents, where they actually take your mutant p53 and change the conformation of the mutant protein to a better protein, thereby potentially making it easier to kill the cells with chemotherapy. That’s still [under] investigation. There’s a drug called magrolimab and similar anti-CD47 drugs which actually bind the CD47 on the surface of leukemia cells. If you bind to CD47 on the surface of leukemia cells, you cover up a… “Don’t eat me signal,” which by covering…up you may allow the eating of the AML cell by a macrophage by immune cells. So that’s a very important therapeutic strategy. There are a host of other ones, but those are the key [options]. I would say p53-specific drugs, anti-CD47 [agents and] IDH1 and IDH2 inhibitors [are] the big ones.

There are some other ones out there like a drug called I-131-apamistamab [BC8], …a radionuclide [antibody-] conjugate, which delivers focused radiation to the target that the antibody binds to. There’s a trial called SIERRA [NCT02665065] that randomized patients with relapsed AML to either this anti-CD45 iodinated radioactive molecule, which binds to all white cells, [or] salvage chemotherapy.2 The primary end point [results are] not available yet [as] the trial just stopped. So that’s a pretty exciting agent. And there are other agents involving inhibition of IRAK4, an inhibition of MCL1, and other important moieties that make a leukemic cell a leukemic cell.

What are some of the new targeted therapies that scientists are looking into for AML?

You have targets that are thought to be the optimal ones. Those are mutated and are based on mutations that are only in the leukemic cells and aren’t in the normal cells. We’ve talked about FLT3, ITD, and TKD mutations, which are present in leukemia cells, not normal cells. Those are function mutations. We target those. There are 2 approved agents in that sphere: midostaurin [Rydapt]andgilteritinib[Xospata]. There are the IDH1- and IDH2-targeting agents, which I mentioned, or the p53-targeted agent, [which] we’re not so [sure] on yet. As far as other mutations such as KIT, which has mutated in about 25% of patients with CBF [core binding factor], there are some drugs that may target KIT. Those [targeted] mutations, I think, are most prominent in AML.

Then there’s the surface targeting of AML. These [targets] aren’t taking advantage of mutations, but rather, molecules on the surface of AML cells that may be expressed more potently on AML cells than on normal stem cells. This would be CD33, which is the target for gemtuzumab ozogamicin, which I mentioned earlier in reference to using it with core binding factor AML. There’s CD123, the most interesting target right now, [for] which [there] is one drug approved…called tagraxofusp [Elzonris]. That’s approved not in AML but in blastic plasmacytoid dendritic cell neoplasm [BPDCN].

There’s a host of what I would call proleukemic events, like IRAK4 expression, bromodomain proteins, and menin inhibitors. Menin inhibitors represent one of the important mutationally targeted aspects of drug development. About 15% of patients have either MLL [mixed-lineage leukemia] rearrangement, or about 25% have an NPM1 mutation. Both of these types of mutations cause increased expression of the HOX-MEIS gene cluster. To turn on that gene cluster, you need to have a [great number of] proteins sitting on the surface of the DNA that allows the RNA polymerase to transcribe it. One protein that sits on the surface in the complex on the DNA, that might be important in transcribing genes that are important to leukemogenesis, is a moiety called menin. And menin is critically important in leukemogenesis in patients with mixed-lineage leukemia (MLL), rearranged leukemia, or NPM1-mutant leukemia, and there is a host of menin inhibitors out there. So that’s another mutationally targeted [inhibitor]. It’s not inhibiting the mutation directly, but it is inhibiting an important piece of physiology that is relevant to [patients] who have those type of mutations. So if you add up IDH1, IDH1, FLT3, TP53, C-KIT, and menin, you have a [large] percentage of AML patients who can be treated [with] this specific inhibitor plus chemotherapy, or perhaps the specific inhibitor plus low-dose chemotherapy.

What are the key unmet needs right now?

The key unmet needs are in TP53-mutant AML. It’s a very devastating disease; fortunately, it accounts for about 15% of all AML, but it’s very hard to cure if not impossible to cure if it’s a patient with a transplant. Another unmet need is our [patients] who are older [who] we still probably can’t [cure yet]. They relapsed after getting these azacitidine-plus-venetoclax [treatments]. There are many different subtypes of AML that are [labeled] “good” subtypes but are [still difficult to treat]. We need to address that problem as well.

References

1. DiNardo CD, Jonas BA, Pullarkat V, et al. Azacitidine and venetoclax in previously untreated acute myeloid leukemia. N Engl J Med. 2020;383(7):617-629. doi:10.1056/NEJMoa2012971

2. Thol F, Ganser A. Treatment of relapsed acute myeloid leukemia. Curr Treat Options Oncol. 2020;21(8):66. doi:10.1007/s118