Molecular Profiling May Hold Key to More Effective Therapy for AML

Daniel De Angelo, MD

Molecular profiling of every patient with acute myelogenous leukemia (AML) may hold the key to better treatment in the future, Daniel De Angelo, MD, said in a review of the therapeutic landscape presented at the 2015 Society of Hematologic Oncology annual meeting.

After reviewing data from a negative trial, examination of molecular profile findings showed that some patients had myelodysplastic syndrome¬—like disease associated with “secondary-type” mutations. Others had specific mutations associated with de novo AML, said De Angelo, who is director of Clinical and Translational Research in Adult Leukemia at Dana-Farber Cancer Institute.

As a result of the observations, “We’ve been molecularly profiling every patient, because I would argue that for the vast majority of our patients, we have no idea what they have,” De Angelo said. “[AML] is a multitude of different diseases.”

Further proof of the point came from a review of data for 105 consecutive, unselected patients with AML aged ≥65 years. On the basis of prior findings, De Angelo and colleagues would have expected 45% to 50% of the patients to respond to standard treatment. The hypothesis held true for some patients but not others. Some of the patients did much better than expected, not requiring secondary induction therapy, for example.

“There is a group of older patients who do pretty well,” he said. “If we can identify these patients upfront, we might be more aggressive and push therapy, define a strategy for cure, which we tend not to do with older patients.”

“I would argue that this is an approach we could use with all of our patients as we move into the molecular era, especially with use of targeted therapies,” De Angelo added.

The widespread availability of next-generation sequencing assays has given most oncologists in the United States the potential to request molecular profiling for all patients with AML. With that capability, clinicians can begin to apply new molecularly targeted therapies as soon as the therapies become available.

Molecular profiling has pointed the direction toward new therapies for AML. One class that has attracted considerable attention focuses on the isocitrate dehydrogenase (IDH) gene. Most of the attention has centered on IDH-2, but therapies that target IDH-1 also have emerged, said De Angelo.

AML is associated with IDH mutations in about 10% to 15% of cases but does not appear to be a harbinger of poor prognosis. The extent to which alterations represent founder mutations that account for a majority of patients or late events that manifest in subgroups of patients with AML remains unclear, said De Angelo. Therapeutic options will differ according to the nature of the mutations. To date, therapeutic development has focused on mutations in IDH-2. Clinical investigation of agents that target IDH-2 has revealed inconsistent patterns of differentiation. Some patients begin with a 90% blast count, which declines to 10% within treatment. Three months later, the blasts have disappeared completely, said De Angelo. In other cases, the blast response is less dramatic or predictable. “As we molecularly profile patients initially and throughout therapy, it will be interesting to see which clones are propagating and which are disappearing,” he said.

“It will be important to know which drugs experience this, in terms of how to develop this class of drugs in the future.”

Clinical experience to date suggests some patients have durable responses with agents that target IDH-2 and IDH-1. Not uncommonly, responses evolve gradually over time, but in some cases, the responses prove to be durable. Most studies thus far have involved patients with relapsed/refractory AML, and the nature of IDH mutations associated with the disease—founder or late event—has not been determined.

Beyond IDH-targeted drugs, other therapies that have shown some promise in AML include venetoclax (ABT-199), which has been evaluated primarily in chronic lymphocytic leukemia. The drug targets apoptosis, and investigators have been interested in evaluating the drug’s apoptotic potential in AML.

De Angelo said that, within his own institution, studies have shown that some patients have greater apoptotic potential than others. Those who are “primed” for apoptosis respond better to apoptosis-inducing therapy than do patients who have less evidence of priming.

A 32-patient phase I study showed that 6 (19%) patients responded to treatment with venetoclax, including two complete responses. The drug is problematic to use, as tumor lysis syndrome can occur quickly and without warning, said De Angelo. Limited evidence suggests patients with IDH mutations are more likely to respond to the drug.

The findings regarding activity in patients with IDH mutations suggest venetoclax “lends itself to combination therapy. I don’t think any of the drugs being developed should be used by themselves. They need to be developed for use in combination with more cytotoxic therapies.”

In closing, De Angelo said AML researchers have interest in evaluating immunotherapies and a variety of other classes of targeted agents, including inhibitors of HIT, FLT3, DOT1L, and HDAC.

“We need to better characterize patients with AML in order to define treatment recommendations,” he said. “Not everyone who is old has old-person leukemia, and some of those patients do very well. Some patients who are younger and have AML may have a stem-cell disorder, based on their molecular profile.”