Rise of the Clones: Assessing Molecular Response to Induction Chemotherapy in AML

Patients with acute myeloid leukemia (AML) who have achieved morphologic remission after induction chemotherapy often continue to exhibit clonal hematopoiesis – of a clonal population often closely related to the initial leukemic clone, which confers an increased risk of relapse. However, it remains unknown how induction chemotherapy influences the clonal evolution of a patient’s non-leukemic hematopoietic cell population.

The authors of an analysis recently published in Blood, used next-generation sequencing (NGS) approaches to assess molecular responses to induction chemotherapy in patients with AML in a subset of 15 patients who had achieved complete morphologic remission with one or two rounds of intensive induction chemotherapy (including both cytarabine and an anthracycline).

Researchers, led by Terrence N. Wong, MD, PhD, from the Department of Oncology at Washington University School of Medicine in St. Louis, Missouri, observed three patterns of hematopoiesis:

  • In four cases, they detected no evidence of clonal hematopoiesis after induction therapy with no somatic mutations at a variant allele fraction (VAF) ≥2.5 percent.
  • In six cases, one or more mutations identified in the diagnostic AML sample reappeared in the long-term follow-up sample, representing relapse from the AML founding clone or one of its subclones.
  • In five cases, they observed a novel pattern of hematopoiesis.

Among the latter group, somatic variants not detected in the AML diagnostic sample were identified and validated in one or more remission samples with no evidence of any AML-associated mutations, “suggesting that a hematopoietic population (termed a ‘rising’ clone) unrelated to the initial AML expanded following chemotherapy,” the researchers explained.

In each case, the rising clone remained detectable at stable or expanding levels throughout the observation period (161-544 days). Unlike patients with clonal hematopoiesis, whose clonal population often remains stable or expands slowly over multiple years, the patients with a rising clone following induction therapy were relatively younger and exhibited a rapid expansion of their clones (30 to 150-fold within 1-2 months of chemotherapy initiation).

Mutations associated with these rising clones included: TP53 (two cases), TET2, DNMT3A, and ASXL1. Patients with evidence of a rising clone after induction therapy, however, had similar clinical and hematologic characteristics to patients lacking identifiable clones.

Two of the five patients with rising clones (harboring mutations in ZNF318 R1195* and TET2 Q1828*) subsequently relapsed. NGS of both patients’ bone marrow samples revealed that the majority of the original leukemia-associated mutations were re-identified, with three subclonal mutations no longer detectable. After salvage therapy, and the achievement of morphologic remission, with one clone further expanded, eventually comprising the majority (76%) of hematopoietic cells.

“These results suggest that non-leukemic hematopoietic stem and progenitor cells (HSPCs) harboring specific aging-acquired mutations may have a competitive fitness advantage after induction chemotherapy, expand, and persist long after the completion of chemotherapy,” Dr. Wong and colleagues concluded.

However, since both relapses evolved from the initial AML clone – not from rising clones – the clinical consequences of clonal hematopoiesis arising from non-leukemic HSPCs following induction chemotherapy are unclear. In the future, it will be important to distinguish non-leukemic from leukemic populations when assessing for molecular responses to induction chemotherapy.

“Given the sensitivity limitations of NGS and the difficulty in distinguishing rising clones from co-existing persistent leukemia-related populations, this study likely underestimates the genetic complexity of patients with AML after induction therapy,” the authors wrote, noting some limitations of the current analysis. “Additional patients and longer follow-up are needed to define the extent and prognostic implications of non-leukemic clonal expansion following AML therapy.”


Wong TN, Miller CA, Klco JM, et al. Rapid expansion of pre-existing non-leukemic hematopoietic clones frequently follows induction therapy for de novo AML. Blood. 2015 December 2. [Epub ahead of print]