Novel Agents Advance Field of AML

Gabriel Mannis, MD

Gabriel Mannis, MD

After a 40-year treatment paradigm with little to no change, 2017 saw 4 regulatory approvals in the field of acute myeloid leukemia (AML).

The FDA approvals of gemtuzumab ozogamicin (Mylotarg), CPX-351 (Vyxeos), midostaurin (Rydapt), and enasidenib (Idhifa) are among the exciting therapeutic advancements for the treatment of patients with AML.

Additionally, venetoclax (Venclexta) is showing promise in combination with cytarabine, based on results from a phase I/II trial. In this study, the combination achieved complete responses (CRs) in two-thirds of the patients. The combination led to CR or CR with incomplete blood count recovery (CR) in 38 of 61 patients (62%) treated with the recommended dose of venetoclax. One additional patient had a partial response. In the subgroup of patients with intermediate-risk disease status, the combination resulted in a CR of 76%.

OncLive: Please provide an overview of your presentation on novel therapies in AML.

In an interview during the 2017 OncLive^® State of the Science Summit^TM on Hematologic Malignancies, Gabriel Mannis, MD, an assistant professor of hematology and blood and marrow transplantation, University of California, San Francisco (UCSF) Helen Diller Family Comprehensive Cancer Center, discussed novel therapies shifting the treatment paradigm for patients with AML.Mannis: The treatment of AML has essentially been the same for the last 40 years since the development of 7+3 chemotherapy. Over the past 30 years, the main improvements that have been attempted have been various additions or modifications to 7+3. Over 4 decades, none have been shown to improve outcomes for patients with AML.

In the last 40 years, there has only been one drug that has been FDA approved for the treatment of patients with AML and it was taken off the market shortly after it was approved. There is a lot of excitement in the field and we expect that, within the next year, there may be 1 or 2 more drugs approved. The treatment paradigm for AML has really changed dramatically over the last 6 to 9 months.

Of those 4 FDA-approved drugs, 1 is the FLT3 inhibitor midostaurin, which was approved in combination with 7+3 for patients with FLT3 mutations. CPX-351 was approved, which combines cytarabine and daunorubicin. In patients with secondary AML or de novo AML with myelodysplasia-related changes, CPX-351 was shown to improve overall survival, which was almost doubled for patients receiving CPX-351 compared with 7+3 chemotherapy.

For patients with secondary AML or therapy-related AML, CPX-351 is the new standard of care for fit patients who are able to undergo intensive chemotherapy. Patients like it because it has the added benefit that they do not lose their hair and it is much less toxic than standard 7+3 chemotherapy. The main downside is it concentrates in the bone marrow; therefore, the time to recovery is longer—causing patients to stay in the hospital longer than needed otherwise.

There is also an IDH inhibitor that is showing promise. One of the steps forward regarding drug development for AML is that we are realizing that not all AML cases are the same. Every patient has a unique form of AML and by understanding the molecular heterogeneity, we are able to target their AML in a personalized fashion.

IDH mutations are present in somewhere between 6% to 13% of patients. There are IDH1 and IDH2 mutations. The IDH2 inhibitor enasidenib was approved earlier this year in the setting of relapsed/refractory AML. It showed a 20% CR rate and a 40% overall response rate. It is well tolerated so that even older and frailer patients can receive it.

We participated in the phase I study at UCSF investigating the IDH1 inhibitor ivosidenib. It seems that the regimen will be FDA approved in early 2018.

Reflecting on the 2017 ASH Annual Meeting, what were some exciting abstracts in AML?

Finally, of the 4 FDA-approved drugs, gemtuzumab ozogamicin (Mylotarg) was initially approved over a decade ago and was taken off the market. It has now been reapproved by the FDA after further studies show that a lower dose improves OS when added to 7+3 chemotherapy. The groups who benefitted were those with favorable- and intermediate-risk AML. While it benefitted patients of all ages, it is likely to be used in favorable-risk patients with AML in combination with 7+3 chemotherapy. One of the exciting things from the meeting was the emerging data about venetoclax (Venclexta) in AML. Venetoclax was FDA approved for the treatment of relapsed/refractory CLL, but it is being studied in combination with both hypomethylating agents and cytarabine for the treatment of elderly patients who are not fit for intensive chemotherapy. Those patients now are likely to get single-agent hypomethylating treatments or low-dose cytarabine with response rates of 20% to 30%.

In the early studies, combining venetoclax with azacitidine or decitabine showed response rates of more than 60% and, in some cases, 90%. There was a significant increase in the response rates by adding venetoclax. It tends to be very well tolerated and unlike in CLL, we have not seen some of the tumor lysis that goes along with the treatment.

There were several abstracts looking at various combinations of drugs with venetoclax in AML. There was an ongoing study looking at venetoclax in combination with either an MDM2 inhibitor or a MEK inhibitor for patients with relapsed/refractory AML. The early data look encouraging but those studies are ongoing.

One of my interests is in immunotherapy. AML has been left behind in the immunotherapy field. It started in the solid tumor world with cancers such as melanoma or kidney cancer, and has then shown promise in lymphoma and myeloma. Early responses to those same drugs have not been as great in AML. It is harder to find immunotherapy targets for myeloid malignancies, partly because the antigens are expressed in all sorts of healthy cells. To find an appropriate immunotherapy target, you need to have an antigen that is expressed on the disease cells and not on the healthy cells. This has been a tough challenge in AML.

There were 2 or 3 abstracts presented at the 2017 ASH Annual Meeting that highlighted the progress that has been made for immunotherapy in AML. One of those abstracts discussed a novel bispecific antibody—much like blinatumomab (Blincyto)—called flotetuzumab that targets CD3 on T cells and CD123 on AML blasts. CD123 is overexpressed on the majority of blasts and has relatively low-level expression on other normal cells. In the early study, a heavily pretreated group of patients with refractory AML and adverse prognostic features saw a 57% antileukemic response and several CRs. This is a promising study that we are opening at UCSF soon.

What are some other mutations that have been identified, aside from IDH1/2 and FLT3?

The other big area of interest is with CAR T cells for AML. Targeting CD123 data were presented at ASH. There were 6 patients who were treated with a CD123 CAR T cell. All the patients had received a prior allogeneic stem cell transplant and had relapsed, so these were the worst in terms of hard-to-treat AML. Of those 6 patients, 4 had responses and 2 went into CR. Even on the immunotherapy front, which has been a challenge in AML, we are seeing substantial steps forward. We are learning a lot about the prognostic significance of mutations in AML. In the last 5 to 10 years, the standard has been to look at FLT3, NPM1, and CEBP-alpha as the main 3 mutations that can tell us about prognosis. We now know that IDH1/2 is important because there are targeted agents for that mutation.

As we continue to look at an unexpanded mutational profile in AML, we may identify more targetable mutations. At this point, it is important to continue investigating these mutations because it does help us determine prognostic significance.

In the new European leukemiaNET classification, they have added a category for RUNX1-mutated AML, realizing it has an adverse prognostic impact. Those patients who might otherwise have been considered as intermediate-risk patients now fall into an adverse-risk category and therefore should be considered for allogeneic transplant.

ASXL1 has an adverse prognostic impact of MLL partial tandem duplication. We are learning a lot about the molecular hierarchy of some of these mutations. By investigating them, we will be able to better treat patients by ultimately de-escalating the therapy that we are giving.

What does the future of AML look like in 5 years?

A trend that has changed across all hematologic malignancies is the idea of giving as much therapy as possible to beat the disease into submission. With advanced technology and genetic testing, we are learning that we can be more elegant in how we choose to treat patients and potentially minimize the toxicities of the therapies.The treatment paradigm has the potential to change dramatically. Around 10 years ago, we lumped AML into one big category and later into 3 categories of low-, intermediate-, and high-risk AML. As we learn more about the genetics of AML, we are dividing the pie into many more categories. In some ways, it is going to be harder to do research because clinical trials will have to focus specifically on patients with certain genetic mutations and therapies will become more personalized.

However, with immunotherapy approaches, one could argue that we do not need to pay attention to any of those molecular determinants if we can find an immunotherapy approach that captures all AML. Perhaps combining an immunotherapy approach with a targeted agent could be what ultimately ends up replacing allogeneic transplant.

Weis A, Strickland SA, Roboz GJ, et al. Phase 1/2 study of venetoclax with low-dose cytarabine in treatment-naive, elderly patients with acute myeloid leukemia unfit for intensive chemotherapy: 1-year outcomes. In: Proceedings from the 59th ASH Annual Meeting and Exposition; December 9-12, 2017; Atlanta, Georgia. Abstract 890.