AML Research Focus Adapting to Address Heterogeneity

Gary J. Schiller, MD, irector of hematologic malignancies in the Stem Cell Transplant Unit in the David Geffen School of Medicine at University of California, Los Angeles

Gary J. Schiller, MD

Within the past few years, research in acute myeloid leukemia (AML) has shifted to showcase the heterogeneity of the disease. Now, the availability of both targeted and broad-spectrum drugs demonstrates the focus to approach AML management in a number of different ways, explained Gary J. Schiller, MD.

"AML affects 20,000 to 25,000 people in the United States every year and is associated with a very high mortality. It's a disease that occurs over the entire age spectrum, from pediatrics to geriatrics," said Schiller, the director of hematologic malignancies in the Stem Cell Transplant Unit in the David Geffen School of Medicine at University of California, Los Angeles.

"There is a big unmet medical need, especially in older people who might not tolerate conventional chemotherapy. Fortunately, there are a number of drugs that have been approved for management of AML that take different approaches," added Schiller.

In an interview during the 2019 OncLive^® State of the Science Summit^TM on Hematologic Malignancies, Schiller discussed these therapeutic advances, the importance of genomic testing, and other agents in the pipeline.

OncLive: In the newly diagnosed and relapsed/refractory settings, what are the biggest therapeutic advances over the last couple of years?

Schiller: With venetoclax (Venclexta) and glasdegib (Daurismo), these drugs are paired with hypomethylating agents or with low-dose cytarabine. Glasdegib has approval with low-dose cytarabine, and venetoclax [has approval in combination] with hypomethylating agents. These drugs induce a very high likelihood of a morphologic leukemia free state and of complete remission. Both of them are affected by low blood counts and both induce severe cytopenias. They're difficult to use and are advertised as outpatient management; however, supportive care with transfusions and antibiotics is absolutely critical to getting patients through these drugs because they are very myelosuppressive. They offer an option for eradicating much of the leukemia clone without cytotoxic chemotherapy.

In the same realm of broad-spectrum therapy, is chemotherapy packaged in a different way? The liposomal product [CPX-351; Vyxeos] that was FDA approved is a combination of doxorubicin and cytarabine in a fixed molar ratio and is administered in the outpatient setting. It induces severe myelosuppression and requires extensive supportive care. Most patients require hospitalization when their blood counts get low, but the remission rates are significantly higher in patients with therapy-related AML, secondary-AML, or AML that can look like it has risen out of a previous myelodysplastic clone. These are the non-targeted therapeutic approaches.

The targeted agents, which have gained approval for the most part in the relapsed/refractory setting—but not exclusively—are targeted against identifiable molecular events that characterize AML in certain populations. There are several drugs that have been developed in the FLT3 area and 2 that are FDA approved. With IDH1/2 mutations, which are less commonly seen in AML, you can use midostaurin (Rydapt), which is FDA approved and couples well with chemotherapy.

Finally, there is the immunotherapeutic approach of transplant, but there are other potentials such as the antibody-drug conjugate gemtuzumab ozogamicin (Mylotarg). [Transplant] does have a role, especially in core-binding factor AML. These three approaches have changed the therapeutic paradigm for management.

With these available targeted agents, what do you recommend regarding genomic testing?

AML is not one disease and [we have known that it] hasn't been one disease for 25 years. For a long time, I would go out on the “campaign trail” and try to convince people to do cytogenetic [testing], such as fluorescence in situ hybridization. Since then, it's become routine. However, molecular testing, next-generation sequencing, and testing with probes has not become common and the turnaround time is slow. Yet, the implications of those results are great, not only for initial therapy but for therapy relapse and the use of transplant.

Do molecular testing. Develop molecular panels at your institution or in commercial laboratories and assess AML, both upon diagnosis and relapse, with extensive molecular testing.

How do you feel about the FDA’s decision to not approve quizartinib?

I was surprised; I predicted it would be approved just like gilteritinib (Xospata). [Quizartinib] has a lot of single-agent activity and its profile is not significantly different from gilteritinib. I don't know what happened. I do not think the drug is “dead in the water,” but it has been dealt a setback.

How are you best using gilteritinib in your clinical practice?

On the label, it is used as a single agent in the setting of relapsed/refractory FLT3-mutated disease—in those with FLT3-ITD or FLT3-TKD mutations. It's a very active drug as a single agent, although it is myelosuppressive. In a clinical trial, there is a combination [of gilteritinib] in the upfront setting and the relapsed/refractory setting. The addition of the drug has a lot of activity and perhaps is better tolerated than midostaurin.

What other exciting agents are in the pipeline?

There is a lot of talk about expanding the immunotherapeutic portfolio. That has not yielded yet to bifunctional monoclonal antibodies or CAR T cells. It is a fertile ground for testing, but it is a very tough “nut to crack” because AML bears similarities to normal hematopoiesis. Almost any target you can come up with is shared with hematopoiesis and shared widely. Thus far, these agents have been very toxic, but there continues to be attempts to develop an antibody bifunctional or CAR T-cell approach against more discreet antigens.

What are your thoughts on the early data with immunotherapy in AML?

It is not impressive. Neither pembrolizumab (Keytruda) nor nivolumab (Opdivo) produced impressive responses. Maybe the threshold is a bit too high, but I don’t know if that kind of broad immunotherapeutic approach is going to work here. We need something that is very discrete, like the molecular inhibitors.

You mentioned the heterogeneity of AML. Has the biology of this disease evolved over time or were panels not available to test for abnormalities?

A little of both. The panels did not test [many modern mutations], but we could have seen chromosome 3 mutations because we've been doing conventional cytogenetics for a long time. I also think that people live longer, relapse later, and there is more upfront management with allogeneic transplant—even in an older population. That delays time to relapse and the character of the relapse is different than the original disease. We have affected the biology. We sometimes see very late relapse, which is worrisome. I used to tell patients you will never see a relapse after 3 to 5 years in remission. That is not true [today]. The disease is also changing as well as our ability to make detailed diagnosis.