Current guidelines recommend testing for genomic alterations as part of the diagnostic workup of acute myeloid leukemia (AML). Broad genetic testing is warranted, as established treatment algorithms integrate genomics into classification and risk stratification systems.

A standard genomic workup is required for all patients with AML. Some mutations are more common in children and others are more common in adults, but clinicians need to run the whole panel to adequately classify their patients. The multigene panels are fairly standard now and include all the important genes, in terms of prognosis and identifying potential targeted therapy. So, the challenge is not getting the testing done; rather, the challenge is getting the testing done in a reasonable amount of time so that you can make decisions based on the results. Although guidelines recommend that you have results within approximately 5 days, I would say that most testing in the community still takes closer to 2 weeks.

Cancer genomics can be complicated. Even the cases that we think are simple at first sometimes are not. For example, a small percentage of patients will have normal cytogenetics and few mutations, so you wonder what genetic abnormality could be underlying their leukemia, and sometimes it can be “cryptic” translocations such as KMT2A fusions that are not captured by routine cytogenetics. The physician would need to order a special test to identify the KMT2A rearrangement.

There is another level of complexity in the genomics of our relapse patients because the mutations can be different at relapse than they were at diagnosis. For example, if a patient with an FLT3 mutation is treated with an FLT3 inhibitor, there is a reasonable chance that they will no longer have an FLT3 mutation if they relapse (approximately 40%). Treatment could have successfully eradicated the FLT3 clone but not the full underlying leukemia, or their leukemia could have mutated under the selective pressure of the therapy. Examples like this help us realize how important genetic testing is.

We cannot test patients once at diagnosis and expect that to carry us through all time points. To determine the best use of the available targeted therapies, we need to retest at each new time point to assess the underlying leukemia characteristics to understand how best to treat it. Testing for measurable residual disease (MRD) by mutational testing is still challenging because it is often not covered by insurance. However, ongoing MRD testing by flow cytometry or polymerase chain reaction can be an option for some patients, including the approximately 30% of those who have NPM1 mutations.

As Dr DiNardo mentioned, we need the results of genomic testing to characterize the best possible treatment because we are getting better at developing therapy that is tailored to patients with specific genetics and molecular markers. However, genomic testing can also depend on financial feasibility. This is one reason I think that most—if not all—newly diagnosed patients should be treated at academic centers. A lot of the tests that we might order are not commercially available, although they are generally covered by insurance providers at the time of diagnosis.

Finally, we do MRD testing for the majority of patients at several time points during treatment and then after treatment. MRD testing is already relevant to treating patients, and I believe that it is going to become increasingly important. However, I do want to note that next-generation sequencing panels are probably not sensitive enough for MRD testing.

The importance of genetic testing is the same in pediatric patients as it is in adult patients. Our main challenge, as Dr DiNardo noted, is how quickly we can get results. For example, with FLT3 disease, it is very important that we have the results as soon as possible.

We have seen progress by reducing treatment intensity in pediatric patients with secondary or treatment-related AML. In the past, the standard of care was to give 2 courses of induction therapy, 2 courses of consolidation therapy, and then transplant, and the results were dismal. Reducing chemotherapy to 2 courses of induction followed by transplant improved survival. Further, patients with secondary AML can attain remission with 1 cycle of CPX-351. Then you can treat with venetoclax, for example, through transplant and achieve excellent results.

It is possible that choosing the appropriate intensity of therapy depends on the genotype of the AML. For example, a retrospective study investigating adults with t(8;21) or inv(16) AML who received standard dose or intermediate dose induction, with all patients receiving consolidation therapy after achieving first complete remission, found that those with t(8;21) AML who received the intermediate dose had a better complete response rate than those who were treated with standard therapy.

Regarding selecting treatment for a patient with a previous FLT3 mutation at the time of relapse, some patients with FLT3-mutated disease relapse without an FLT3-mutated clone, even before the introduction of FLT3 inhibitors. However, this can also depend on the sensitivity of the MRD assay (ie, if the assay is not sensitive enough, you might not see the clone). And with effective targeted therapies, you might lose your MRD target.