Development of a 29-Color Single-Tube Full Spectrum Flow Cytometry Assay for the Detection of Measurable Residual Disease and Leukemic Stem Cells in AML

Introduction: Multiparametric flow cytometry is widely utilized to evaluate measurable residual disease (MRD) in acute myeloid leukemia (AML). However, the limited number of available channels in conventional flow cytometry necessitates dividing samples into multiple tubes, thereby restricting both the number of analyzable cells and the complexity of immunophenotyping. Conventional assays also require an additional tube to supplement leukemia stem cell (LSC) analysis and lack the evaluation of hemodilution. Full spectrum flow cytometry (FSFC), capable of simultaneously employing up to 40 fluorescence channels, overcomes the limitation of flow channels and offers the potential for multifunctional design using a single tube.

Aim: To develop a single-tube assay based on FSFC which adheres to the recommendations of the European Leukemia Net Flow-MRD Working Party (Heuser et al, Blood doi:10.1182/blood.2021013626) and incorporates the simultaneous evaluation of leukemia associated immunophenotypes (LAIP), leukemia stem cells (LSC), monocytic markers and sample hemodilution at the same time.

Methods: Measurements were performed on a Cytek®Aurora (Cytek) and a FACS CantoTM (BD) using the Infinicyt^TM 2.0 software (BD) for analysis. KG-1α AML cell line cells with 4 leukemia-associated immunophenotypes (LAIP) were diluted into healthy bone marrow (BM) for limit dilution experiments. We established the limit of blank (LOB), limit of detection (LOD), and limit of quantification (LOQ) for 4 LAIPs in the 29-color single-tube assay. Assay reproducibility was evaluated by comparing intra- and inter-assay variations. The assay's capability to detect the background signal was tested by measuring the percentages of 24 LAIPs and 10 LSC immunophenotypes in three different healthy BM samples. We compared the consistency of the conventional 8-color assay with the new 29-color assay in quantifying major leukocyte subpopulations, LAIP and LSC on primary AML samples as well as during follow-up, acquiring ≥ 4 million CD45⁺ cells per sample. Various established parameters, including immature granulocytes (IGRA), neutrophils (N), mast cells, and plasma cells, were quantified to assess the degree of hemodilution in the tested samples.

Results: The Complexity Index of the single-tube 29-color assay was 9.08. Limit dilution experiments further demonstrated that the 29-color assay easily achieved the minimum detection sensitivity requirement of 0.1% for MRD. The LOB, LOD, and LOQ for CD34⁺CD13⁺CD2⁺ were 0.00862%, 0.00882%, and 0.00927%, respectively. For CD34⁺CD13⁺CD56⁺, the LOB, LOD, and LOQ were 0.00446%, 0.00544%, and 0.00697%, respectively. For CD34⁺CD13⁺CD7⁺, the LOB, LOD, and LOQ were 0.00413%, 0.00421%, and 0.0123%, respectively. For CD34⁺CD13⁺CD22⁺, the LOB, LOD, and LOQ were 0.00522%, 0.00542%, and 0.00620%, respectively. The coefficient of variation for intra- and inter-assay measurements ranged from 0% to 10.93%, demonstrating good reproducibility. We were able to verify the background range for 24 LAIPs in three different healthy bone marrow samples, which spanned from 0.000022% to 0.1% (Mean = 0.00862%). The background for 10 LSC immunophenotypes ranged from 0% to 0.00021% (Mean = 0.0000307%). Our results indicated a high concordance between the established 8-color assay and the new 29-color assay in quantifying lymphocytes (R² = 0.9912), granulocytes (R² = 0.9955), monocytes (R² = 0.9994), immature blasts (R² = 0.9956), LAIP (R² = 0.9786), and LSC (R² = 1.000). Additionally, our new assay was able to quantify IGRA/N ratio and the percentages of mast and plasma cells. These parameters can help evaluate the degree of hemodilution in each tested sample, thereby determining the reliability of MRD status.

Conclusion: In summary, we provide a versatile FSFC-based MRD assay that reduces cell requirements, includes LSC evaluation, and can simultaneously estimate hemodilution in the tested samples. This assay has the potential to improve the reliability and simplicity of MRD detection.