Evaluating measurable residual disease in acute myeloid leukemia

Farhad Ravandi, Roland B. Walter and Sylvie D. Freeman

Article Figures & Data



  • Table 1.

    Risk status stratification by genetic abnormality per ELN 2017 and NCCN 2017 guidelines

    Risk category*ELN criteria10NCCN criteria6
    Favorablet(8;21)(q22;q22.1); RUNX1-RUNX1T1Core binding factor: inv(16), or t(16;16), or t(8;21), or t(15;17)
    inv(16)(p13.1;q22) or t(16;16)(p13.1;q22); CBFB-MYH11Normal cytogenetics: NPM1 mutation in absence of FLT3-ITD or isolated biallelic (double) CEBPA mutation
    Mutated NPM1 without FLT3-ITD or with FLT3-ITDlow§
    Biallelic mutated CEBPA
    IntermediateMutated NPM1 and FLT3-ITDhigh§Normal cytogenetics
    Wild-type NPM1 without FLT3-ITD or with FLT3-ITDlow§ (without adverse-risk genetic lesions)+8 alone
    t(9;11)(p21.3;q23.3); MLLT3-KMT2A||t(9;11)
    Cytogenetic abnormalities not classified as favorable or adverseOther nondefined
    Core binding factor with KIT mutation
    Poor/adverset(6;9)(p23;q34.1); DEK-NUP214Complex (≥3 clonal chromosomal abnormalities)
    t(v;11q23.3); KMT2A rearrangedMonosomal karyotype
    t(9;22)(q34.1;q11.2); BCR-ABL1−5, 5q–, –7, 7q–
    inv(3)(q21.3q26.2) or t(3;3)(q21.3;q26.2); GATA2,MECOM(EVI1)11q23 – non t(9;11)
    −5 or del(5q); –7; –17/abn(17p)inv(3), t(3;3)
    Complex karyotype, monosomal karyotype#t(6;9)
    Wild-type NPM1 and FLT3-ITDhight(9;22)
    Mutated RUNX1**Normal cytogenetics: with FLT3-ITD mutation††
    Mutated ASXL1**TP53 mutation
    Mutated TP53‡‡
    • * The prognostic value of a marker is treatment-dependent and may change with new therapies.

    • Presence of KIT mutations in patients with t(8:21) and, to a lesser extent, inv(16), confers a high risk of relapse; these patients should be considered intermediate risk and considered for HSCT if available.

    • Other cytogenetic findings in addition to these do not alter risk status.

    • § Low, low allelic ratio (<0.5); high, high allelic ratio (≥0.5); recent studies indicate that AML with NPM1 mutation and FLT3-ITD low allelic ratio may also have a more favorable prognosis and patients should not routinely be assigned to allogeneic HCT.75,76

    • || Presence of t(9;11)(p21.3;q23.3) takes precedence over rare, concurrent adverse-risk gene mutations.

    • 3 or more chromosomal abnormalities in the absence of 1 of the World Health Organization–designated recurring translocations or inversions: t(8;21), inv(16) or t(16;16), t(9;11), t(v;11)(v;q23.3), t(6;9), inv(3) or t(3;3); AML with BCR-ABL1.

    • # Defined by the presence of 1 single monosomy (excluding loss of X or Y) with at least 1 additional monosomy or structural chromosome abnormality (excluding core-binding factor AML).

    • ** These should not be used as adverse prognostic markers if they occur with favorable-risk AML subtypes.

    • †† FLT3-ITD mutations are considered to confer a significantly poorer outcome in patients with normal karyotype; there is controversy about whether FLT3-TKD mutations carry equally poor prognosis.

    • ‡‡ TP53 mutations are significantly associated with AML with complex and monosomal karyotype.

    • NCCN, National Comprehensive Cancer Network.

  • Table 2.

    ELN recommendations for MRD testing

    Flow cytometry
     1Use the following markers in a MRD panel:
    CD7, CD11b, CD13, CD15, CD19, CD33, CD34, CD45, CD56, CD117, HLA-DR (backbone: CD45, CD34, CD117, CD13, CD33, FSC/SSC).
    If necessary, add a “monocytic tube” containing:
     2Integrate the classic LAIP approach with the DfN approach. To trace all aberrancies (at and beyond diagnosis, including newly formed postdiagnosis aberrancies), apply a full panel both at diagnosis and follow-up.
     3Aspirate 5-10 mL BM and use the first pull for MRD assessment. At present PB, with its lower MRD content, should not be used for MRD assessment.
    Pull as low as desirable BM volume because contamination with PB increases with BM volume.
     4Estimate the contamination with PB, especially when a first pool of BM was impossible.
     5Use 500 000 to 1 000 000 white blood cells, use the best aberrancy available and relate it to CD45+ white blood cells.
     6To define “MRD-negative” and “MRD-positive” patient groups, a cutoff of 0.1% is recommended.
     7If true MRD <0.1% is found, report this as “MRD-positive <0.1%, may be consistent with residual leukemia.” If applicable, the comment “this level has not been clinically validated” should be added.
     8In a multicenter setting, transport and storage of full BM at room temperature for a period of 3 d is acceptable.
     9Single-center studies with no extensive experience on MFC MRD are strongly discouraged.
    Molecular biology
     1Molecular MRD analysis is indifferent to the anticoagulant used during cell sampling; thus, heparin or EDTA can be used as anticoagulant.
     2Aspirate 5-10 mL BM and use the first pull for molecular MRD assessment.
     3WT1 expression should not be used as an MRD marker unless no other MRD marker is available in the patient.
     4Do not use mutations in FLT3-ITD, FLT3-TKD, NRAS, KRAS, DNMT3A, ASXL1, IDH1, IDH2, or MLL-PTD and expression levels of EVI1 as single MRD markers. However, these markers may be useful when used in combination with a second MRD marker.
     5We define molecular progression in patients with molecular persistence as an increase of MRD copy numbers ≥1 log10 between any 2 positive samples. Absolute copy numbers should be reported in addition to the fold increase to enable the clinician to make his or her own judgments.
     6We define molecular relapse as an increase of the MRD level ≥1 log10 between 2 positive samples in a patient who previously tested negative. The conversion of negative to positive MRD in PB or BM should be confirmed 4 wk after the initial sample collection in a second sample from both BM and PB. If MRD increases in the follow-up samples ≥1 log10, molecular relapse should be diagnosed.
     1Refine morphology-based CR by assessment of MRD, because CRMRD is a new response criterion according to the AML ELN recommendation 2017.
    Use MRD to refine risk assessment before consolidation treatment, the postinduction time point closest to consolidation treatment is recommended.
     2MRD monitoring should be considered part of the standard of care for AML patients.
    Monitoring beyond 2 years of follow-up should be based on the relapse risk of the patient and decided individually.
    Patients with mutant NPM1, RUNX1-RUNX1T1, CBFB-MYH11, or PML-RARA should have molecular assessment of residual disease at informative clinical time points.
     3Not to assess molecular MRD in subtypes other than APL, CBF AML, and NPM1-mutated AML.
     4For AML patients not included in the molecularly defined subgroups here, MRD should be assessed using MFC.
    During the treatment phase, we recommend molecular MRD assessment at minimum at diagnosis, after 2 cycles of standard induction/consolidation chemotherapy, and after the end of treatment in PB and BM.
    During follow-up of patients with PML-RARA, RUNX1-RUNX1T1, CBFB- MYH11, mutated NPM1, and other molecular markers we recommend molecular MRD assessment every 3 mo for 24 mo after the end of treatment in BM and in PB. Alternatively, PB may be assessed every 4-6 wk.
     5Failure to achieve an MRD-negative CR or rising MRD levels during or after therapy are associated with disease relapse and inferior outcomes and should prompt consideration of changes in therapy.
     6In APL, the most important MRD end point is achievement of PCR-negativity for PML-RARA at the end of consolidation treatment.
    For patients with PML-RARA fusion and low-/intermediate-risk Sanz score who are treated with ATO and ATRA, MRD analysis should be continued until the patient is in CRMRD in BM and then should be terminated.
     7Detectable levels of PML-RARA by PCR during active treatment of APL should not change the treatment plan for an individual patient.
     8A change in status of PML-RARA by PCR from undetectable to detectable, and confirmed by a repeat sample, should be regarded as an imminent disease relapse in APL.
     9Patients with CBF AML should have an initial assessment of MRD after 2 cycles of chemotherapy, followed by serial measurements every 3 mo for at least the first 2 y after the end of treatment.
     10MRD should be assessed pretransplant.
     11MRD should be performed posttransplant.
     12All clinical trials should require molecular and/or MFC assessment of MRD at all times of evaluation of response.
    • Reprinted from Schuurhuis et al.24

    • ATO, arsenic trioxide; ATRA, all-trans retinoic acid; BM, bone marrow; HLA-DR, HLA–antigen D related; PB, peripheral blood.

  • Table 3.

    Pros and cons of methods used to detect MRD in AML

    MFC∼10−4 to 10−5Wide applicability (>90%)Challenging and somewhat subjective interpretation requires experienced pathologist
    Relatively quick (results ≤1 d)Sensitivity dependent on antibody panel used
    Single result interpretableLimited harmonization and standardization across laboratories
    High specificity when using defined LAIPLeukemic phenotype not necessarily stable over time (eg, initial LAIP may not identify subclones leading to relapse)
    Can detect cells with leukemia-stem cell phenotype
    Can distinguish between live and dead cells
    Ease of data storage
    Provides information about whole sample cellularity
    NGS∼10−3 to 10−5Relatively easy to performLimited standardization
    SensitiveError rate leads to low sensitivity of mutated sequences
    Applicable to specific subgroupsMutated genes can be detected in healthy people without hematologic abnormalities
    Persistence of some genetic abnormalities in patients in long-term remission
    Risk of contamination
    RT-qPCR∼10−3 to 10−5Wide applicabilityResults may take multiple days
    May be run by any certified laboratory with RT-qPCR capacityExpensive (computationally demanding and time-consuming)
    High sensitivity (≥MFC)Requires high-level expertise
    Well standardizedRequires setting of threshold limits
    Quality assurance routinely incorporatedInterpretation often requires trend of results
    Different mutations have different biological consequences in AML
    Molecular targets applicable to only ∼50% of all AML cases and <35% in older patents
    FISH∼1 to 10−2Superior to PCR-based assays for detection of numeric cytogenetic abnormalities (gains and losses of whole chromosomes or deletions/duplications)Considerably less sensitive than PCR or MFC
    Not useful for patients with normal karyotype
    Quality-assured probes are expensive; technique is labor intensive
    Chromosome banding analysisNAMore common in routine clinical practiceLabor-intensive, comparatively costly, low-throughput technique reliant on highly trained technical staff
    Evaluates the dividing proportion of cells∼10× less sensitive than FISH
    Not useful for patients with normal karyotype
    • FISH, fluorescence in situ hybridization; NA, not available. QA, quality assurance.

  • Table 4.

    Testing technologies and potential markers for monitoring MRD in AML

    Testing technologiesPotential markers
    t(10;11) – KMT2A-MLLT10
    t(11;19) – KMT2A-ELL or KMT2A-MLLT1
    t(6;11) – KMT2A-MLLT4
    t(9;11) – KMT2A-MLLT3
    • MLL/MLLT3, mixed lineage leukemia.