Genomic Breakthroughs Offer Hope for New Strategies in AML

Hetty Carraway, MD

Hetty Carraway, MD

Results of a deep sequencing analysis demonstrated that mutations in 8 high-risk genes are associated with an acute myeloid leukemia (AML) diagnosis. These findings, explained Hetty Carraway, MD, lay the ground work for early detection studies and novel treatment strategies.

“It is pretty provocative work, pretty exciting, and it really allows us to think more deeply about how we can impact and change that trajectory for patients,” said Carraway, who is vice chair of Strategy and Enterprise Development at Taussig Cancer Institute, Cleveland Clinic.

In the study, researchers identified 212 women from the Women’s Health Initiative who eventually developed AML at a median follow-up of 9.6 years. Deep sequencing was performed on peripheral blood DNA and compared with age-matched controls that did not develop disease. Results showed that mutations in IDH1, IDH2, TP53, DNMT3A, TET2, SRSF2, SF3B1 and U2AF1 were significantly associated with an increased risk of developing AML.

Moreover, all patients with TP53 as well as IDH1 and IDH2 mutations eventually developed AML. The presence of these identified mutations years before an AML diagnosis suggests that there is an inactive period that precedes AML, the authors concluded.

OncLive: What is the prevalence of actionable mutations in AML?

In an interview during the 2018 OncLive^® State of the Science Summit^TM on Hematologic Malignancies, Carraway highlighted ongoing genomic research in AML, as well as recent therapeutic advances in the field.Carraway: Things that are really important to remember in the treatment of patients with AML is to think about doing mutation testing by next-generation sequencing for patients at the time of diagnosis, as well as at the time of relapse. These types of mutations can inform how patients will do in terms of prognosis, and it can also inform choice and selection of therapies for them.

In your presentation, you highlighted research about mutations that can predict AML before it is actually diagnosed. Could you discuss that?

Another thing to remember is that as we get novel therapies, know that clinical trials are going to be very important in order to better understand how to combine these [treatments], how to optimize response, and to be mindful of the toxicities that may happen as a result of combination therapy.There has been some fascinating work on the presence of mutations in patients who do not have a hematologic malignancy—to understand the presence of mutations and clonal hematopoiesis of indeterminate potential. [They looked at the] patients that evolve from having these particular mutations that then go on to have dysplasia, that then go onto have clonal events, that then lead to myelodysplastic syndrome (MDS).

Then, there are specific mutations that those groups went on to identify to be more associated with a higher risk of progression to disease, such as hematologic malignancies. They also found that there was an increased risk of cardiovascular disease, early myocardial infarction, and stroke by twofold. That work was really seminal in other groups who then tried to ask, “Well, let’s look at larger cohorts of patients. Are there patients who had AML as a result [of these mutations]?”

That brings up an interesting point. Could this knowledge help prevent the disease from occurring in the first place?

Work by Pinkal Desai, MD, et al, and another group went on to look at larger cohorts of patients and identify those who were diagnosed with leukemia. Then, they went back 10 years before that diagnosis and looked at serial samples of blood to say, “What happened with the detection of mutations?” They were able to identify differences in patients over that time span in the presence of mutations and the variant allele frequency, the types of mutations, and the double mutations in specific genes that then predispose, or what could be used as identifiers, in saying that those patients were likely to progress to AML compared with those patients who did not have AML and did not have those mutations.It really is a whole different way to think about things that we have not really had the luxury of thinking about previously. The other side of that, which many excellent researchers are working on, is after you give patients chemotherapy, can you really evaluate minimal residual disease in this population? Can we do things earlier, so that they can be cured—or change the outcome for them in terms of relapse?

What therapies have become available in this landscape in the last 2 years?

There is also very important work that is led by groups not only in the United States, but also in Europe, really focusing on the presence of these mutations at the time of complete remission and what the implications are with the presence of those mutations at those times.What has really been the most exciting has been in the upfront setting. We had changes for patients with FLT3-mutated AML with the addition of midostaurin (Rydapt) to induction 7+3 along with consolidation chemotherapy for those patients. That has been a game changer and kicked off some of the news that we had in terms of approvals last year.

That agent is very well tolerated with limited toxicities, and so making sure to add midostaurin (Rydapt) to those patients with FLT3-positive AML is important for us to make sure it’s on our radar. For those patients who have AML with MDS-related changes, CPX-351 (Vyxeos) was approved. It is a liposomal reformulation of doxorubicin and cytarabine, and that is also a pretty exciting drug in patients over the age of 60 and up to age 75. Those 2 agents in the upfront setting for patients with newly diagnosed AML are at the forefront.

Other agents that are important for us to highlight are those in the relapsed/refractory setting and thinking about IDH1/2-mutated patients with relapsed/refractory AML. Drugs that are IDH1/2 inhibitors, like enasidenib (Idhifa) and ivosidenib (Tibsovo) are pretty novel agents. Taking a pill once a day as the therapy for your relapsed AML is also a novel [approach]—not like a drug we have ever had before. They have unique toxicity profiles, so that is the most important thing to teach patients about—so they know to say, “Hey, something is going on.” That is pretty exciting, although it is a very small subset.

What specific combinations are under investigation now?

What does the treatment for patients with AML look like now and going forward?

Interestingly, combining those agents with induction chemotherapy as well as hypomethylating therapy are things we are excited to see come in the next year or so.People are excited about the outcomes of some of the studies looking at IDH1/2 inhibitors in combination with 7+3 as upfront therapy. We haven’t thought too much about the sequencing of these drugs, so that is another question—we don’t know if that matters. There are other agents, such as venetoclax (Venclexta) plus low-dose cytarabine or venetoclax plus hypomethylating therapy, that we are also quite excited to see the results of. There is interest of what will happen for these populations. Venetoclax plus hypomethylating therapy seems to be pretty well tolerated and is led by Courtney DiNardo, MD, of The University of Texas MD Anderson Cancer Center. Some people are already testing out venetoclax plus low-dose cytarabine, if they can get access to venetoclax.A lot of it is, if you’re identifying a newly diagnosed patient, what mutations and cytogenetics exist? How am I going to profile this patient? What agents am I going to choose in terms of how old or how young they are? How fit or unfit are they? What mutations are present? Depending on that information, we can choose a FLT3 inhibitor—an induction therapy regimen that may be more in tune with MDS that has transformed to AML—or a regimen in the relapsed/refractory setting where we have selected the IDH1/2 inhibitor.

We discussed the upfront setting for core-binding factor leukemia and intermediate cytogenetics; there are patients who benefit from the addition of gemtuzumab ozogamicin (Mylotarg). That is also something that’s important to consider. The toxicities of that are also important, and the concerns around veno-occlusive disease, so we still have some institutional biases one way or the other. We also have some considerations if that patient is headed to transplant.

In my dreams, AML is cured. Whether that is in 5 years or 10 years—I don’t know when it is, but I want that to happen.

Desai P, Mencia-Trinchant N, Savenkov O, et al. Somatic mutations precede acute myeloid leukemia years before diagnosis. Nature. 2018;24(7):1015-1023. doi: 10.1038/s41591-018-0081-z.