AML World Awareness Day 2020 | The value of measurable residual disease in acute myeloid leukemia

AML World Awareness Day, April 21, 2020, provides an opportunity for the acute myeloid leukemia (AML) community to unite. The occasion encourages the exchange of knowledge on the most recent advances in the prevention, management, and treatment of AML. To find out about how we are raising awareness, head to www.know-aml.com.

Introduction

Although up to 80% of patients with AML achieve morphologic complete remission (mCR) following induction therapy, disease recurrence is common—with 50–60% of patients relapsing. Residual leukemic cells that have survived intensive chemotherapy often go undetected by conventional morphological analysis and can reside in the bone marrow (BM) at levels of up to 10¹⁰–10¹². Morphologic assessment alone is, therefore, an insufficient biomarker for response in AML remission, and it has become of utmost importance to develop effective prognostic assessments to determine the presence and estimate the extent of residual disease.

A successful prognostic tool would not only be able to detect residual disease and elucidate treatment effectiveness, but also provide crucial information on the disease biology, mechanisms of resistance, and drug–patient interaction. The European LeukemiaNet (ELN) recommendations for the diagnosis and treatment of AML were revised in 2017 to incorporate measurable residual disease (MRD) status (Table 1). A recent review by Giovangiacinto Paterno and colleagues evaluated the techniques being explored and implemented to detect MRD (Table 3) and was recently published in Current Opinion in Hematology, there follows a summary and discussion of how these techniques could aid MRD assessment in the AML setting.¹

The AML Hub previously reported on the latest data set regarding the use of MRD in AML as presented at the 2019 annual meeting of the Society of Hematologic Oncology (SOHO). Read a summary here.

Table 1. Response criteria for AML categorization

AML, acute myeloid leukemia; ANC, absolute neutrophil count; BM, bone marrow; CR, complete response; CRi, CR with incomplete hematologic recovery; CR MRD-, complete remission without minimal residual disease; MFC, multiparameter flow cytometry; MLFS, morphologic leukemia-free state; MRD, minimal residual disease; PD, progressive disease; PR, partial remission; RT-qPCR, reverse transcription-quantitative polymerase chain reaction ; WBC, white blood cell
Category	Definitions
CR MRD-	CR with negativity for a genetic marker by RT-qPCR, or CR with negativity by MFC
CR	BM blasts < 5%; absence of circulating blasts; absence of extramedullary disease; ANC ≥ 1000/ml; platelet count ≥ 100 000/ml; MRD+ or unknown
CRi	All CR criteria except for residual neutropenia or thrombocytopenia
MLFS	BM blasts < 5%; absence of blasts with Auer rods; absence of extramedullary disease; no hematologic recovery required
PR	All hematologic criteria of CR; BM blasts 5–25%; decrease of pretreatment BM blast percentage by ≥ 50%
Primary refractory disease	No CR or CRi after two courses of intensive induction treatment
SD	Absence of CR MRD-, CR, CRi, PR, MLFS; and criteria for PD not met
PD	Evidence for an increase in BM blast percentage and/or increase of absolute blast counts in the blood
Hematologic relapse	BM blasts > 5%; or reappearance of blasts in the blood; or development of extramedullary disease
Molecular relapse (after CR MRD)	Reoccurrence of MRD as assessed by RT-qPCR or by MFC

Multiparametric flow cytometry

Multiparametric flow cytometry (MFC) employs fluorochrome-labeled monoclonal antibodies (mAbs), multiple lasers, and advanced analysis software to create an extensive antigen expression profile of a patients’ hematopoietic cells. The mAbs identify aberrant antigen expression on leukemic cells, not otherwise present on wildtype BM cells, to determine leukemia-associated immunophenotypes (LAIPs). LAIPs possess a number of phenotypical characteristics which can be recognized using MFC, such as cross-lineage expression of lymphoid-affiliated antigens and reduced, absent or over-expression of myeloid markers.

Clinical relevance

Until recently, BM was the preferred source for MFC analysis. However, recent studies have revealed a correlation between BM and peripheral blood (PB) readouts. In fact, the absence of normal myeloid progenitor cells in PB results in highly specific phenotypic profiles. A threshold of 0.1% residual leukemic cells of ≥ 0.5—1.0 × 10⁶ wildtype cells has been suggested by the ELN to determine MRD-positive/negative patients, as those patients with levels < 0.1% have been shown to demonstrate favorable survival outcomes. The question of when to carry out MFC-MRD analysis remains unanswered. Nevertheless, numerous studies have highlighted MRD determined by MFC as an independent prognostic factor for relapse, disease-free survival (DFS) and overall survival (OS), when performed following both induction and consolidation therapy.

Reverse transcription-quantitative polymerase chain reaction

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) effectively amplifies genetic abnormalities associated with leukemic cells to identify MRD; the principal targets are illustrated in Table 2. The applicability of RT-qPCR depends on the molecular marker of interest (Table 3). An attempt to implement the technique for clinical application was initiated by the Europe Against Cancer (EAC) program, which used a systemic evaluation approach across international expert laboratories to determine optimal RT-qPCR assays.

Table 2. Molecular targets for detection using RT-qPCR¹

RT-qPCR, reverse transcription-quantitative polymerase chain reaction
RT-qPCR target	Example
Fusion genes	PML-RARA/t(15;17) RUNX1- RUNX1T1/t(8;21) CBFB-MYH11/(inv(16)/t(16;16) DEK-CAN (NUP214)/ t(6;9)
Gene insertions/duplications	NPM1 FLT3-ITD MLL-PTD
Point mutations	CEBPA IDH1/2 KIT RAS RUNX1 TP53
Gene overexpression	WT1 EVI1 ERG

Clinical relevance

The ELN MRD Working Party have identified genetic aberrations, detectable by RT-qPCR, of significant prognostic value. In particular, continued expression of NPM1, and fusion proteins RUNX1-RUNX1T1, CBFB-MYH11, and PML-RARA, following therapy is strongly associated with disease relapse. It has therefore been recommended that patients harboring these mutations should undergo MRD assessment at the point of diagnosis, after two cycles of induction/consolidation therapy, and every three months for 24 months following treatment termination. Additionally, detection of NPM1 by RT-qPCR has proven to be a critical leukemia-specific target, which may be able to decipher subgroups of patients with differential prognosis, even where FLT3 mutations are also present. The National Cancer Research Institute (NCRI) AML17 trial uncovered that RT-qPCR was successful at detecting NPM1 expression in 346 patients carrying as many as 27 mutations, while maintaining high sensitivity. In the MRC AML15 trial, quantification of the CBFB-MYH11 fusion gene was implemented as a means of MRD assessment. At a threshold of ≥ 50% in the BM and ≥ 10% in the PB, the technique predicted relapse in 100% of cases. Other well-established AML markers are under investigation for their use in MRD monitoring and, although results have been variable, it may be of benefit to combine their evaluation with reliable markers to establish a robust approach to disease characterization.

Digital polymerase chain reaction

Although relatively novel, digital PCR (dPCR) stands as a highly promising approach to MRD valuation; possessing the additional capability to identify unknown mutations within patient samples. dPCR has demonstrated superiority over qPCR when evaluating samples with low levels of nucleic acids, as well as those that may be contaminated. Therefore, dPCR stands as a more precise and reproducible technique.

Clinical relevance

The technique has been particularly effective at detecting MRD as mutations in NPM1, DNMT3A, and IDH1/2, even in patients with unknown or rare mutations. Digital PCR is comparable with qPCR when it comes to fusion protein identification, and the two methods may be exploited mutually to evaluate MRD.

Next-generation sequencing

Next-generation sequencing (NGS) is especially useful in the AML setting due to its ability to identify numerous genetic variants in a single assay. When considering the clonal heterogeneity of AML, NGS may be the only available tool that is applicable to all patients.

Clinical relevance

The capability of NGS to accurately detect mutation combinations makes it a promising approach to refining patient MRD status. Hypothetically, NGS holds the potential to personalize MRD monitoring throughout an individual’s treatment schedule, allowing the fine-tuning of patient treatment regimens through the additional information it bears. A recent study set out to investigate whether the degree of mutation clearance following intensive chemotherapy, as determined by NGS, could be predictive of survival and relapse in patients with AML. Significantly superior survival and relapse rates were observed in patients with variant allele frequencies of < 1%, and therefore NGS-based MRD analysis could be deemed a valuable prognostic tool in the AML setting.

Table 3. Comparison of methods used to detect MRD in patients with AML^1,2

AML, acute myeloid leukemia; CHIP, clonal hematopoiesis of  indeterminate potential; dPCR, digital polymerase chain reaction; MFC, multiparametric flow cytometry; MRD, measurable residual disease; NGS, next-generation sequencing; RT-qPCR, reverse transcription-quantitative polymerase chain reaction
Method	Molecular target	Applicability, %	Sensitivity threshold	Advantages	Limitations
MFC	Cross-lineage expression of lymphoid-affiliated antigens   -CD7   -CD19   -CD56 -Overexpression, reduced or absent expression, and asynchronous expression of myeloid markers   -CD33	90	10-3 – 10-5	-Quick -Cost effective	- Requires experience -~ 30% of patients deemed MRD-negative by MFC analysis -Sensitivity can be impaired by hemodilution of PB samples -Often fails to detect the entirety of a heterogeneous leukemic population
RT-qPCR	NPM1 RUNX1-RUNX1T1 CBF-MYH11 PML-RARA	30 1–5 5 1–2	10−4 – 10−6	-High reproducibility -High specificity and sensitivity for leukemic cells -Target stability throughout therapy -Low risk of contamination	-Limited applicability
dPCR	NPM1 DNMT3A IDH	30–35	10−4–10−5	-High sensitivity -The possibility to monitor several mutations simultaneously	-A specific assay needs to be developed for each mutation
NGS	Potentially all leukemia-specific genetic aberrations	≥ 90	1–5%	-Broad applicability -The ability to study the entire leukemic genome - Unaffected by novel mutations -Ability to detect low-frequency variants -Low limit of detection -High sequencing capacity	-Consideration of CHIP at interpretation of results -Requirement of highly specialized bioinformatic approaches and experience -Low specificity -Costly -Time consuming

 Conclusion and future directions

The presence of MRD, as detected using MFC, RT-qPCR or NGS, prior to allogeneic stem cell transplantation (allo-SCT) has been linked to higher instances of relapse and unfavorable survival outcomes in patients with AML. Despite this, allo-SCT remains superior over the alternatives, proving more effective than further chemotherapy or autologous SCT. For optimal application to the AML setting, MRD assessment still requires standardization. The prognostic significance of MRD needs further investigation in prospective clinical trials to consolidate its value to different treatment regimens. Despite this, MRD evaluation has been incorporated in to ELN guidelines for AML classification, and the status of CR MRD-negative is now considered a significant prognostic biomarker for patient response.

Although MFC, RT-qPCR and NGS have generated promising outcomes in the clinical setting, the search for innovative, more robust technologies is ongoing. Next-generation flow (NGF), for example, can process up to 10⁷ cells and is as sensitive as RT-qPCR. As an advance on MFC, mass-cytometry time-of-flight (CyTOF) combines single cell FC with mass spectrometry, using transition-element-labelled mAbs to detect up to 130 parameters of one cell. These advanced technologies threaten to overcome the complexities of AML characterization.

When considering novel targeted therapies, such as FLT3, IDH, and BCL2 inhibitors, defining a patient’s MRD profile may contribute towards the goal of optimized, personalized dosing regimens.