Clonal hematopoiesis and MRD assessment in acute myeloid leukemia

Clonal hematopoiesis (CH) is an expansion of a clonal population of blood cells with one or more somatic mutations. The process is associated with natural aging but also hematologic malignancy. The presence of CH of indeterminate potential (CHIP) in cytopenic patients with a high variant allele fraction (VAF > 10%) and certain mutation patterns is associated with a high risk of progression to a hematologic neoplasm. It is  also linked to an increased risk of morbidity and mortality from non-neoplastic causes.^1,2

The most common CHIP-associated mutations prevalent in acute myeloid leukemia (AML) and other myeloid neoplasms have been described. However, our understanding of the significance of CH during treatment for myeloid neoplasms is not complete. For example, in patients with AML in complete remission (CR), the persistence of mutations found at the time of diagnosis could indicate an imminent relapse, reversion to residual CHIP that carries a small risk for leukemia recurrence, or a clonal event unrelated to prior AML.

In recent years it became clear that minimal/measurable residual disease (MRD) negativity is an important therapeutic endpoint that can supplement CR in monitoring response to therapy. The persistence of certain mutations after treatment can be difficult to interpret with regard to the likelihood of disease recurrence, as some mutations occur as either a somatic mutation in an AML clone or as a germline mutation. While successful treatment would eliminate an AML clone with a somatic mutation, germline mutations would persist.³ Exploring the clinical implications of detectable mutations in treated AML is further impeded by the lack of consistency of terminology used.

In a perspective recently published in Blood, Robert Hasserjian and colleagues discuss the types of CH detected post-treatment in patients with AML, including the molecular markers associated with true residual AML disease.⁴ The authors also proposed a standardized terminology for distinct types of CH.

Current criteria for complete remission and relapsed disease in AML

Currently, the International Working Group (IWG) and European LeukemiaNet (ELN) define

• Complete remission (CR) following therapy for AML as
  • achievement of a morphologic leukemia-free state, characterized by < 5% blasts in the bone marrow (BM), no Auer rods, and no evidence of extramedullary disease
  • peripheral blood count recovery, characterized by absolute neutrophil count > 1 × 10⁹/L and platelets > 100 × 10⁹/L, in the absence of growth factor treatment
• Recurrent disease as
  • the reappearance of leukemic blasts in the blood or ≥ 5% leukemic blasts in the BM
  • the development of cytologically proven extramedullary disease
  • the appearance of new dysplastic

These criteria are used in determining treatment response. However, morphologic CR does not provide a complete reflection of response as patients in CR may be MRD-positive by karyotype, molecular genetics, or flow cytometry. The presence of MRD in morphologic CR is associated with an increased risk of relapse compared with undetectable MRD, and therefore, may require additional or more intensive therapy.

Molecular genetic interrogation of AML post-therapy

A wide variety of genetic abnormalities, such as mutations and gene rearrangements, have been associated with AML, with several simultaneous abnormalities often present (Table 1). Post-treatment genetic abnormalities occurring in CHIP or premalignant AML diseases, such as myelodysplastic syndrome (MDS), are not necessarily linked to the increased risk of relapse. In contrast, aberrations occurring late in AML ontogeny characterizing certain genetically defined AML subtypes are closely associated with disease burden and can be used as markers of MRD. To complicate the matter further, certain mutations can occur either early or late in AML ontogeny and cannot be used to assess treatment response due to uncertain implications on clinical outcomes. The significance of the post-therapy persistence of genetic abnormalities commonly seen in AML is presented in Table 1, and the distinguishing aspects of AML-related and CH-type mutations are shown in Table 2.

Genetic aberrations in AML can be detected using a range of techniques that vary in their applicability and sensitivity:

• Conventional karyotype
• Fluorescent in situ hybridization (FISH)
• PCR-based, including quantitative (qPCR) and droplet digital (ddPCR)
• Next-generation sequencing (NGS)-based approaches

The minimal recommended sensitivity for the detection of MRD is 1 × 10^-3, which can be achieved using the methods listed above.

Table 1. The significance of post-therapy persistence of genetic abnormalities commonly seen in AML

Table 2. AML-related versus CH-type genetic abnormalities⁴

Proposed terminology to express post-AML CH states

The interpretation of genetic abnormalities after therapy depends on whether

• The post-treatment abnormality was detectable before the start of treatment
• The abnormality is known to be AML-related (Table 1)
• There is morphologic evidence of a background MDS, myeloproliferative neoplasm (MPN), or MDS/MPN
• There is a difference in VAF of the mutation(s) between that at the time of diagnosis and post-treatment

The authors propose specific terms to express post-AML CH states, which are summarized and defined in Table 3.

Table 3. Proposed terminology for the post-AML CH states

Additional relevant definitions include

• Complete molecular remission: morphologic CR with two successive genetic measurable residual disease (gMRD)-absent samples obtained at least 4 weeks apart
• Molecular genetic relapse: an increase of the MRD level of at least 10-fold (1 log) between two positive samples in a patient who previously had undetectable MRD, even if criteria for morphologic recurrence are not met

Limitations

Several factors limit the interpretation of results of the post-AML genetic profile, including

• Unclear relationship of some common mutations, such as IDH1/2 and RUNX1, with disease burden
• Uncertain significance of less common genetic aberrations that persist after AML therapy
• A limited number of AML-related genes with sufficient data to allow treatment decisions based on the presence or absence of gMRD
• Difficulty to differentiate between CH and a residual myeloid neoplasm
• Lack of access to sensitive gMRD-detection technologies
  • Sensitive gMRD-detection technologies are available only for certain genes and hotspot mutations
  • NGS sensitivity can vary between genetic loci and can result in a false MRD-negative status
• Responses of AML-related and CH-type mutations may vary according to the type of therapy and could differ between hypomethylating agents, selective therapies, and intensive induction chemotherapy

Conclusion

The interpretation of a genetic landscape following therapy is complex. Some genetic aberrations have been demonstrated to truly reflect residual AML and can be used as markers of MRD, helping to inform on the risk of relapse and guide therapeutic decisions; while mutations in other genes reflect CH but do not correspond with a residual AML disease. However, understanding of the altered hematopoietic microenvironment post-AML therapy is incomplete, and implications of post-AML CH states compared to CHIP require further study.

Hasserjian et al. recommend the use of uniform nomenclature in pathology reports and clinical records of patients with AML who undergo NGS and other genetic testing post-treatment. They hope that this will facilitate further study of therapeutic intervention and optimal patient management. Although they recognize that the proposed nomenclature may currently not always be applicable and may require refinement as additional evidence accumulates, they urge others to consider the proposal as a starting point for discussion.

Future directions include use of more sensitive NGS technologies, analysis of changes in mutation profiles after hypomethylating agents or targeted therapies, and tailoring of post-treatment mutation analysis to the specific type of therapy.