Molecular Testing Mainstays for Acute Myeloid Leukemia

Navel D. Daver, MD

Before treatment is decided for a patient with acute myeloid leukemia (AML), oncologists should perform molecular testing, according to the National Comprehensive Cancer Network Clinical (NCCN) Practice Guidelines. However, the questions remains: Which forms of testing are most important for the beginning of treatment?

“These are usually the 2 major sets of information, molecular next-generation sequencing [NGS] and chromosome analysis that we want on top of the diagnosis of AML before selecting the frontline therapy,” said Navel D. Daver, MD, in an interview with Targeted Oncology™.

According to the NCCN, familial genetic alterations of interest in AML include RUNX1, ANKRD26, CEBPA, DDX41, ETV6, GATA2, MBD4, MECOM/EVI1 complex, SAMD9, SAMD9L, TERC.TERT, ATG2B, and GSKIP.

Currently, several new markers have emerged in AML, and some older markers have newly available therapies that can target these mutations. Among the targets, FLT3 appears to be the most exciting in research right now, according to Daver, associate professor in the Department of Leukemia at The University of Texas MD Anderson Cancer Center (MD Anderson).

In the interview, Daver, discussed the molecular characteristics of AML, agents available for the treatment of mutated AML, and how to potentially improve treatment in the future.

TARGETED ONCOLOGY™: What are some of the testing considerations for AML?

In AML, the main testing considerations currently are to do baseline next-generation sequencing looking for different mutations in patients. We usually do what we call an 81-gene panel here at MD Anderson, and that can look at all the possible mutations that would be both important prognostically because there are certain mutations like tP53, RUNX1, ASXL1, and FLT3 that are known to be adverse and would make us more inclined to move towards allogeneic stem cell transplant at first remission. Then, there are certain mutations, such as CEBPA and NPM1 that could be considered favorable and would make us think less likely to go to stem cell transplant at first remission.

In addition to the prognostic value of these mutations, they are now being used for directing or selecting therapy. For FLT3 mutations, for example, we use FLT3 inhibitors for the initial therapy. Similarly, for tp53 mutations, we're using different tp53-directed therapies. The same approach is being used for IDH1 and IDH2 mutations. So, these mutations may really impact the treatment selection, which has, of course, a major impact on the response rates and overall survival.

Chromosome analysis is also quite important, and there are certain chromosome subsets for which we select particular therapies. For example, for core-binding factor chromosomes, or APL chromosomes, there are completely different treatments. These are usually the 2 major sets of information, molecular NGS and chromosome analysis that we want on top of the diagnosis of AML before selecting the frontline therapy.

Can you discuss the prevalence of FLT3-mutant AML?

For FLT3, we kind of divide it based on the specificity and potency of the different agents towards targeting and blocking FLT3, both preclinical and clinically. The first-generation drugs are the ones we started evaluating about 15 years ago in clinic, and these include sorafenib [Nexavar], midostaurin, lestaurtinib, and others.

The second-generation ones are more recent within the last decade or so, and these are much more potent and were specifically designed to target FLT3 after was identified as an important high-risk mutation in AML. The second-generation FLT3 inhibitors include drugs gilteritinib [Xospata], quizartinib [Vanflyta], and crenolanib. What we're seeing is that as single agents, the second-generation FLT3 inhibitors have much higher potency. Single-agent response rates in the relapsed/refractory FLT3-positive AML are about 45 percent to 50 percent with the second-generation FLT3 inhibitors as compared to only 5 percent to 10 percent single-agent remission rates that could be achieved with the first-generation inhibitors. More importantly, though, combinations of the second-generation FLT3 inhibitors with drugs like venetoclax [Venclexta] or hypomethylating agents, like azacitidine [Vidaza], or even all 3 together are showing very encouraging response rates.

What is important for oncologists to know about resistance to these agents?

For most targeted therapies, whether it's an AML, chronic lymphocytic leukemia, or chronic myeloid leukemia, we're now starting to see patterns of resistance that cross diseases and are emerging as similar resistance mechanisms. So, in FLT3, what we see is that there are 2 major mechanisms of resistance, and 1 of them is on-target resistance. So, if for using a FLT3 inhibitor that targets ITD, sometimes we can see that the patient will have the emergent or expansion of PKD mutations that that drug may not cover. This happens sometimes with sorafenib and quizartinib. So, there's an on-target, acquired FLT3 resistant mutation that drives the relapse.

On the other hand, with some of the other type-one FLT3 inhibitors that can target both ITD and PKD, we're seeing that the resistance is not on-target, but actually in mutations in parallel pro-survival pathways that involve other molecules, such as RAS, MAP kinase, and BCR Abl. It's really important when we see a patient is at risk of losing response or being refracted with an inhibitor to repeat the FLT3 molecular sequencing, but also to repeat the general NGS, because we now have published research showing that there could be very different mutations in 40 percent to 50 percent of FLT3-mutated AML. When they relapse and have new emerging mutations, a number of these could be targetable like IDH1/IDH2 and tP53. So, it has both academic and clinical implications.

To continue to induce optimal responses to therapy and extend survival in the future, what do you think is needed?

I think in general, in AML, the emergence of venetoclax-based therapies has been a major shift and improvement in treatment. And what we've learned now in the last 4 or 5 years is that azacitidine/venetoclax is a very powerful combination, giving us readmission rates close to 70%. But when you look at the 3-year survival, it's only about 35%, which is still better than the 10% 3-year survival we had historically in older unfit AML.

But of course, we want to do better and so I think the next stage of development in clinical trials setting at MD Anderson is a big focus on developing other combinations of venetoclax both in the older unfit patients as well as in the younger patients.

Similarly, in FLT3, tP53, and IDH-mutated AML, we're adding the specific targeted agent to the HMA/venetoclax backbone. So HMA/venetoclax with a FLT3 inhibitor like gilteritinib or quizartinib, HMA/venetoclax with an IDH inhibitor like ivosidenib [Tibsovo] and axitinib [Inlyta], or HMA/venetoclax with drugs like magrolimab, which a CD47 antibody that shown activity in tP53. Also, in phase 1 and 2 studies, we’re seeing that the addition of targeted therapy to HMA/venetoclax is pretty much doubling the CR rates and MRD clearance, and the early survival looks good.

The hope for AML is that eventually we can get rid of or avoid high-dose chemotherapy for the majority of patients similar to what's happened in multiple myeloma over the last 2 decades.

_Reference:

_{NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines: Acute Myeloid Leukemia. NCCN.org, Accessed December 3, 2021. https://bit.ly/31oOR5L}