Clinical Pearls & Morning Reports
There is a growing understanding of the clonal heterogeneity of acute myeloid leukemia (AML) and its evolution over time with therapy. Read the NEJM Case Records of the Massachusetts General Hospital here.
Q: What is our current understanding of the pathogenesis of AML?
A: Our current understanding of the pathogenesis of AML is that cancerous cells arise from an early hematopoietic progenitor through the acquisition of somatic mutations, which leads to clonal expansion of immature myeloid blasts and a malignant phenotype. The mutations leading to this presentation vary; some are more directly leukemogenic, such as the t(15;17) PML-RARA translocation, whereas others require additional cooperating mutations to progress, such as cases in which mutations in DNMT3A, NPM1, and FLT3 are present. These mutations may all arise within the same originating clone, or they may be acquired in various subclones with differing leukemogenic potential until a dominant clone emerges, resulting in the leukemic phenotype.
Q: What are some of the challenges associated with monitoring disease activity in patients with AML?
A: Somatic mutations that are specific to leukemic cells provide a unique opportunity, not only for risk stratification and targeted therapeutics but also for distinguishing these cells from healthy hematopoietic progenitors. If such mutations are truly specific to the leukemia and can be measured with sufficient sensitivity, they can serve as a measure of depth of remission and can be tracked and monitored during therapy. However, there have been challenges with this approach in monitoring AML. First, the mutational heterogeneity lies not only in the genes involved but also in the specific alteration present (e.g., missense, nonsense, frameshift, and insertion or deletion lesions, which are distinct from variants of undetermined significance). Second, cooperating mutations can vary over time, or subclones that were not readily detected at diagnosis may expand.
A: Complicating the assessment of measurable residual disease is the increasing frequency of detectable clonal hematopoiesis with advancing age, a finding seen in a number of otherwise healthy persons. We often face the clinical uncertainty of a patient who is in remission but still has some of the mutations that were present at diagnosis. Although some mutations are clearly part of the leukemic clone, we may not know whether some of the other persistent mutations merely indicate underlying clonal hematopoiesis, particularly among older adults with AML in whom clonal hematopoiesis is otherwise common. Some studies suggest that residual mutations that frequently characterize clonal hematopoiesis — mutations in DNMT3A, TET2, or ASXL1, so-called DTA mutations — may not be predictive of AML relapse after chemotherapy and therefore are unlikely to be associated with the leukemogenic clone.
A: A mutation that has been studied extensively with respect to measurable residual disease assessment is the mutation in NPM1. There are several factors that make disease monitoring based on NPM1 alterations appealing. For example, this mutation seems to be relatively specific to the leukemic clone. Although the NPM1 mutation is detectable at relapse in most patients with NPM1-mutant AML who have a relapse, a small percentage of patients (with estimates ranging from 1 to 14%) will no longer have a detectable NPM1 mutation. One series showed that the majority of such patients had concurrent DNMT3A mutations. This observation raises the question of whether tracking NPM1 mutation levels alone is sufficient in all patients with NPM1-mutated AML.