Technology Spotlight
Faster, Cheaper, Better: Spectral Flow Cytometry in the Clinical Laboratory Free

Flow cytometric immunophenotyping of hematopoietic neoplasms has been the standard of care for well over two decades. In that time, the clinical laboratory has followed the state of the art in flow cytometry, albeit with a 10- to 20-year delay. From two-color flow cytometry for T-cell subsets in the early 1990s to more recent reports of 10- and 12-color cytometry for diagnostics in leukemia and minimal residual disease (MRD) assessment, each succeeding generation of machines and reagents has yielded faster turnaround times, at lower cost, with better resolution. However, while the past decade has seen an explosion of interest in spectral flow cytometry in the research arena, clinical applications have thus far been limited. In brief, conventional flow cytometry follows a one fluorophore–one detector relationship with “compensation” for fluorescence spillover into other detectors, while spectral flow cytometry follows a one fluorophore–multiple detector relationship.¹  In rewriting the relationship between fluorophore (and its conjugated antibody marking the cells of interest) and detector, superior resolution can be achieved. Even more importantly, many more markers can be interrogated simultaneously.

In the research realm, 40 markers or more can be used in a single panel, obviating the need for multiple panels with redundant markers²  (e.g., a panel for various B-cell and T-cell subsets), which allows for sparse samples (e.g., cerebrospinal fluid) to be fully interrogated or for deep phenotyping of sparse subsets.³  Within our own clinical laboratory, a panel with 37 markers covering more than 95% of hematologic malignancies was recently developed, incorporating markers for B cells, T cells, natural killer cells, and myeloid neoplasms. Through careful design and fluorophore assignments, we have been able to demonstrate at least equivalent performance on all markers compared with our conventional 10-color immunophenotyping panels, with superior performance seen for many markers. For the practicing hematopathologist, this means better resolution of dimly expressed markers and less struggling with samples showing atypical phenotypes or degeneration. But even more importantly, rare markers that did not make sense to interrogate upfront in a 10-color paradigm now become very useful, as the utility of these markers spans many lineages and malignancies. For example, CD25 is a useful marker in several rare diseases such as systemic mastocytosis, hairy cell leukemia, T-prolymphocytic leukemia, and acute myeloid leukemia (AML). To interrogate for these malignancies in lower color flow, this marker would need to be incorporated into a B-cell, T-cell, and myeloid panel to be usable in the context of gating and aberrancy markers for each of these diseases. Likewise, other markers such as CD200 are seen in chronic lymphocytic leukemia, T follicular helper cell lymphomas, B-cell lymphoblastic leukemia, and AML, while CD30 is seen in many T-cell lymphomas, AML, and a subset of large B-cell lymphomas. While the utility of any marker is dependent on the prevalence of the disease being investigated, the utility of these once marginal markers increases significantly when interrogating for many diseases at once.

Another advantage to incorporating many markers into a single panel is the ability to evaluate for maturational abnormalities in many lineages simultaneously. For example, we are able to evaluate for granulocytic maturational abnormalities seen in myelodysplastic syndromes (Figure 1) as well as evaluate hematogone maturation down to the so-called stage 0 hematogone⁴  (Figure 2) in the same setup. Likewise, T-cell malignancies can be fully interrogated (Figure 3) without the need for add-on studies. Indeed, this is one “killer app” for this technology, enabling operational efficiency within the lab by obviating the need for add-on studies in most cases.

The second practical application involves MRD testing, where many cells are evaluated for many markers. Conventionally, several tubes are used in most evaluations, with significant redundant markers needed such that a three-tube, 10-color AML MRD panel may only use 20 unique antibodies.⁵  Not only are these redundant antibodies not reimbursed and therefore considered a waste, but the sample is split into three parts to be run on each tube, so the true number of events evaluable is decreased by that factor — a major issue on short or minimal bone marrow aspirates. Additionally, the benefits of flexible panel design allow for the addition of many novel markers that have been published in the flow MRD space for both AML and B-cell lymphoblastic lymphoma MRD. While markers such as CD371 (CLL-1), CD366, CD47, and CD224 have been proposed, the number of novel potential markers is such that it is challenging to include them in a panel without significant redundant tubes.

As can be seen in the scenarios described above, conventional analysis is still possible and indeed desirable, and the increased dimensionality of the data does not pose computational issues on modern computers. While interpretation may seem complex from the view of a practicing hematopathologist, our typical diagnoses are easily handled without need to analyze the plethora of new marker combinations possible. For example, the presence of CD13 or CD33 on an otherwise typical CLL population does not dissuade me from that diagnosis. In conclusion, the introduction of spectral flow cytometry in the clinical laboratory represents a major leap in assay sensitivity, but perhaps even more excitingly a major leap in operational efficiency by decreasing reagent usage and labor requirements.