New Approach to Assessing MRD in Acute Leukemia

Alexander M. Castellino, PhD

March 28, 2018

In hematologic malignancies, minimal residual disease (MRD) has been a benchmark for identifying patients who are at risk for relapse after having achieved complete remission with initial therapy.

In acute myeloid leukemia (AML), MRD is typically assessed by using flow cytometry and, at some centers, by quantifying NPM1 transcripts — an independent and powerful prognosticator of AML.

But a new approach is now being proposed. A study from two European cooperative groups found that next-generation sequencing NGS can be used to monitor for additional somatic mutations from 54 genes that drive AML. The risk for relapse or relapse-free survival after 4-year follow-up increased almost twofold when these mutations were present.

In teasing out which mutations in the NGS panel were relevant, the researchers showed that this risk was not seen with the persistence of three mutations associated with epigenetic regulation, which are collectively called the DTA mutations (DNMT3A, TET2, and ASXL1). These DTA mutations are normally acquired in blood and bone marrow during aging but not in the context of disease diagnosis — a process known as clonal hematopoiesis of indeterminate origin.

When NGS was compared with flow cytometry, the method typically used to detect MRD in AML, the researchers showed that the 4-year relapse rate was high (about 75%) when MRD was detected with both methods and was low (about 25%) when it was not detected with both methods.

"In this study, gene sequencing and multiparameter flow cytometry each had an independent and additive prognostic value with respect to rates of relapse and survival in patients with AML," the researchers write.

The study, published March 28 in the New England Journal of Medicine, was a collaboration between the Dutch-Belgian Cooperative Trial Group for Hematology-Oncology (HOVON) and the Swiss Group for Clinical Cancer Research (SAKK).

David P. Steensma, MD, and Benjamin L. Ebert, MD, PhD, from the Dana-Farber Cancer Institute and Harvard Medical School in Boston, Massachusetts, provided comments in an invited editorial.

"Although the concept of persistent minimal residual disease strikes fear in the hearts of oncologists because of its implications in acute lymphoblastic leukemia and other diseases, assessment for minimal residual disease in AML is more nuanced — one must take into account not only whether a mutation is present after initial therapy but what that mutation is," they write.

And with more than a touch of optimism, they add: "In some cases, as Dante pointed out, the devil is not so black as he is painted."

Non-DTA Mutations Identify Relapse Risk

For their study, the researchers analyzed bone marrow or peripheral blood samples, which came from 482 patients who had newly diagnosed AML and who were in remission following induction therapy; some of the patients had also received consolidation chemotherapy or stem-cell transplants. Mutations most common at diagnosis were NPM1, DNMT3A, FLT3, and NRAS.

Of 482 patients, 430 had at least one mutation, which could be a potential marker for MRD, at diagnosis (89.2%). These 430 patients were randomly assigned to a training (n = 283) or validation (n = 147 patients) cohort.

After induction therapy, DTA mutations were most persistent, occurring at rates of 78.7% for DNMT3A, 54.2% for TET2, and 51.6% for ASXL1. Mutations in the RAS pathway were cleared and occurred at low frequencies for patients achieving remission.

The allele frequency of mutations persisting after complete remission ranged from 0.02% to 47%.

While DTA mutations often persisted at allele frequencies of greater than 2.5%, mutations in IDH1, IDH2, STAG2, and TP53 rarely persisted at allele frequencies of greater than 2.5% after induction and were considered to be typically consistent with the state of morphologic complete remission — defined as less than 5% blast cells in the bone marrow.

"This finding suggests that residual mutation-bearing cells could constitute a minor population of the cells or perhaps even a majority of the cells," the researchers write.

The persistent DTA mutations might have represented nonleukemic clones that repopulated the bone marrow after induction therapy, the researchers suggest. Indeed, in patients who had both DTA and non-DTA mutations at diagnosis, the latter were generally cleared after induction.

"These observations are consistent with the notion that residual cells bearing DTA mutations after induction therapy represent nonleukemic clones rather than persistent malignant disease," the researchers write.

Non-DTA Mutations Dictate Relapse and Survival

In the training cohort, the researchers showed that persistent DTA mutations were not associated with risk for relapse or survival compared with patients with no detected mutations (P = .29).

However, in patients with persistent DTA mutations, non-DTA mutations were associated with significant relapse risk: Four-year relapse rate was 66.7% for patients with mutations vs 39.4% for patients without the mutations (P = .002).

"Thus, in patients with persistent DTA mutations, the presence of residual disease that specifically included coexisting non-DTA mutations represented a predictor of impending relapse," the researchers write.

Persistent non-DTA mutations were reported to be associated with reduced 4-year rate of relapse-free survival (56.7% vs 36.6%; P = .006) and reduced overall survival (65.3% vs 43.7%; P = .01).

The results in the validation cohort confirmed the observations seen in the training cohort.

When the researchers combined the two cohorts, persistent non-DTA mutations were associated with a twofold increased risk for relapse, relapse or death, and death at 4 years:

  • Relapse rate: 55.4% vs 31.9% for no detection; hazard ratio (HR), 2.14; P < 0.001;

  • Relapse-free survival: 36.6% vs 58.1% for no detection; HR for relapse or death, 1.92; P < .001; and

  • Death: 41.9% vs 66.1% for no detection; HR for death, 2.06; P < .001.

In a multivariate analysis, sequencing-based detection of non-DTA mutations remained a significant prognosticator for relapse, relapse-free survival, and survival.

NGS vs Flow Cytometry

The researchers had samples from 340 patients to determine MRD for both flow cytometry and NGS analysis. Concordant results were seen in 69.1% of patients (30 patients with detection and 205 patients with no detection).

The 4-year relapse rates were 73.3% in patients who tested positive with both sequencing and flow cytometry, 52.3% for those with MRD on sequencing but not flow cytometry, 49.8% for patients with MRD on flow cytometry but not sequencing, and 26.7% for patients with MRD negativity on both tests.

"[T]he combined use of sequencing and flow cytometry during complete remission warrants further development and evaluation in clinical practice," the researchers write.

For the researchers, this represents a way to increase the sensitivity of sequencing-based approaches to identify risk for relapse.  

Sequencing-based detection of MRD is imperfect, they point out. They note that not all patients with residual mutations in their cells have a relapse. On the flip side, some patients with no MRD experience relapse.

They indicate that relapse estimation can be improved with new technologies that also incorporate sequencing-based approaches and that may be more sensitive.

"In this respect, it is of particular interest that the use of multiparameter flow cytometry…can increase the yield of identification of residual leukemia during complete remission," they write.

In their editorial, Steensma and Ebert note that the most intensive therapy — allogeneic stem cell transplantation — has not consistently improved outcomes for patients with MRD.

"Therapeutic targeting of specific mutations that are present during remission could delay or prevent relapse," they write.

"In gaining a further understanding of the genetics of minimal residual disease in patients with AML, we are given the opportunity to refine postremission therapy," Steensma and Ebert note.

The study authors have disclosed no relevant financial relationships. Steensma reported receiving consulting fees from Celgene and has been an independent data monitoring committee member for Onconova, Takeda, and Janssen. He has also received grants and/or consulting fees from H3 Biosciences, Acceleron, Tesaro, Kura, and Otsuka. Ebert reported receiving a research grant from Celgene and personal fees from Genoptix.

N Engl J Med. Published  March 28, 2018. Abstract, Editorial

Follow Medscape Oncology on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: