Rodak's Hematology: Clinical Principles and Applications, 5th Ed.

CHAPTER 32. Flow cytometric analysis in hematologic disorders

Magdalena Czader

OUTLINE

Specimen Processing

Flow Cytometry: Principle and Instrumentation

Pattern Recognition Approach to Analysis of Flow Cytometric Data

Concept of Gating

Analysis of Flow Cytometric Data

Cell Populations Identified by Flow Cytometry

Granulocytic Lineage

Monocytic Lineage

Erythroid Lineage

Megakaryocytic Lineage

Lymphoid Lineage

Flow Cytometric Analysis of Myeloid Disorders (Acute Myeloid Leukemias and Chronic Myeloid Neoplasms)

Acute Myeloid Leukemias with Recurrent Cytogenetic Abnormalities

Acute Myeloid Leukemias Not Otherwise Specified

Myeloproliferative Neoplasms and Myelodysplastic Syndromes

Flow Cytometric Analysis of Lymphoid Neoplasms (Lymphoblastic Leukemia/Lymphoma and Mature Lymphoid Neoplasms)

B Lymphoblastic Leukemia/Lymphoma

T Lymphoblastic Leukemia/Lymphoma

Mature Lymphoid Neoplasms

Other Applications of Flow Cytometry Beyond Immunophenotyping of Hematologic Malignancies

Objectives

After completion of this chapter, the reader will be able to:

1. Describe the technique of flow cytometry, including specimen selection and preparation, instrumentation, data collection, and a design of an antibody panel.

2. Discuss the pattern recognition approach to analysis of flow cytometric data for diagnosis and follow-up of hematologic malignancies.

3. Identify basic cell populations defined by flow cytometric parameters.

4. Recognize the key immunophenotypic features of normal bone marrow, peripheral blood, and lymph node tissue, and specimens from patients with acute leukemia or lymphoma.

5. Discuss novel applications of flow cytometry beyond the immunophenotyping of hematologic malignancies.

CASE STUDIES

After studying the material in this chapter, the reader should be able to respond to the following case studies:

Case 1

A 58-year-old man had a 5-month history of extensive right cervical lymphadenopathy and night sweats. His complete blood count (CBC) results were within normal limits. Physical examination showed additional bilateral axillary lymphadenopathy. The cervical lymph node was excised. Histologic examination revealed nodular architecture with predominantly medium-sized lymphoid cells with irregular nuclear outlines. Flow cytometric data are presented in Figure 32-1.

1. What cell subpopulation predominates on the forward scatter (FS)/side scatter (SS) scattergram?

2. List antigens positive in this population.

3. Does the pattern of light chain expression support the diagnosis of lymphoma?

Case 2

A 3-year-old girl was brought to the physician because of fatigue and fevers. The CBC revealed a WBC count of 3 × 109/L, HGB level of 8.3 g/dL, and platelet count of 32 × 109/L. Review of the peripheral blood film showed rare undifferentiated blasts with occasional cytoplasmic blebs. No granules or Auer rods were identified. Bone marrow examination showed a marked increase in blasts (79%) and decreased trilineage hematopoiesis. Flow cytometric analysis was performed. In addition to the markers shown in Figure 32-2, the population of interest was positive for CD34, CD33, CD41, and HLA-DR.

1. What abnormal features are observed on the CD45/SS scattergram?

2. What is the most likely diagnosis considering the constellation of markers expressed by the predominant population?

Image 

FIGURE 32-1 Scattergrams showing immunophenotypic features of lymphoid cells from the patient in Case 1. FITC, fluorescein isothiocyanate; FS, Forward scatter; RD, rhodamine; SS, side scatter.

Image 

FIGURE 32-2 Predominant bone marrow population for the patient in Case 2. ECD, phycoerythrin-Texas Red; PE, phycoerythrin; SS, Side scatter.

Flow cytometry was originally designed to measure physical properties of cells based on their ability to deflect light. Over the years, it has evolved to include detection of fluorescent signals emitted by dyes bound directly to specific molecules or attached to proteins through monoclonal antibodies. The development of monoclonal antibodies is the most significant factor contributing to today’s broad application of flow cytometry. Although the term flow cytometry implies the measurement of a cell, this technique is applied successfully to study other particles, including chromosomes, microorganisms, and proteins. The main advantage of flow cytometry over other techniques is its ability to rapidly and simultaneously analyze multiple parameters in a large number of cells. When one adds the capability of identifying and quantifying rare-event cells in a heterogeneous cell population, the value of flow cytometry to clinical hematology becomes obvious. Currently, this technique not only is applied to analysis of cell lineage in acute leukemia or a detection of clonality in lymphoid populations but also makes it possible to discern abnormal populations in chronic myeloid neoplasms, quantitate minimal residual disease, and monitor immunodeficiency states. Immunophenotypes that originally were used to supplement morphologic classification frequently correlate with specific cytogenetic or molecular abnormalities. According to the classification of hematopoietic neoplasms recommended by the World Health Organization,1 one no longer can rely solely on morphology for a diagnosis of hematologic malignancies. Current diagnostic algorithms integrate morphologic, immunophenotypic, and genotypic information. This approach emphasizes the central role that flow cytometry plays in the hematopathology laboratory.

The focus of this chapter is on the use of flow cytometry in a routine hematopathology laboratory. The chapter follows a “life” of a flow cytometric specimen that starts with specimen processing and ends with a final diagnosis. The discussion is divided into preanalytical (specimen processing), analytical (flow cytometric instrumentation and analysis), and postanalytical (immunophenotypic features of hematopoietic disorders) sections.

Specimen processing

Flow cytometric analysis is particularly useful in diagnosing hematologic malignancies. The specimens most commonly analyzed are bone marrow, peripheral blood, and lymphoid tissues. In addition, immunophenotyping is often performed on body cavity fluids and solid tissues when they are suspected to harbor a hematologic malignancy.2

Prolonged transport or transport under inappropriate conditions may render a specimen unsuitable for analysis. Peripheral blood and bone marrow specimens should be processed within 24 to 48 hours from the time of collection. Certain specimens, such as body cavity fluids or samples from neoplasms with a high proliferative activity, may require even more rapid processing.

When cells are suspended in a fluid, as in peripheral blood and bone marrow, minimal sample preparation is required. These specimens are collected into a tube or container with an anticoagulant, preferably heparin, and are transported to a flow cytometry laboratory at room temperature. Bone marrow biopsy specimens and solid tissue specimens, including core biopsy samples, are submitted in culture media to maintain viability or on saline-moistened gauze. Tissue fragments are mechanically dissociated to yield a cell suspension, usually by mincing with a scalpel.

To obtain a pure population of nucleated cells, red blood cells (RBCs) are lysed. The analytical process depends on cellularity and viability of a specimen; both are routinely assessed before a sample is stained. Cell count can be obtained using automated cell counters or flow cytometry. A specimen is stained with propidium iodide or 7-amino actinomycin to test viability. A cytocentrifuge slide (Chapter 18) is prepared for a morphologic inspection of a cell suspension.

As soon as these steps are completed, a sample is stained with a cocktail of fluorochrome-conjugated monoclonal antibodies. The analysis of intracytoplasmic markers requires an additional fixation and permeabilization step to allow antibodies to pass through a cell membrane. A predetermined panel of antibodies may be used to detect membrane-bound and intracellular markers. Simultaneous analysis of multiple markers, known as multicolor or multiparameter flow cytometry, has numerous advantages. It facilitates visualization of antigen expression and maturation patterns, which are often disturbed in hematopoietic malignancies. In addition, regardless of a complexity of a specimen, analysis can be accomplished using few tubes and with a lower total number of cells, which saves reagents, time, and data storage. There is no consensus on the standardized panel of antibodies to be used in routine flow cytometric evaluation. The U.S.-Canadian Consensus Project in Leukemia/Lymphoma Immunophenotyping recommends the comprehensive approach with multiple markers for myeloid and lymphoid lineage.3 Selected markers commonly analyzed by flow cytometry are presented in Table 32-1.

TABLE 32-1

Lineage-Associated Markers Commonly Analyzed in Routine Flow Cytometry

Lineage

Markers

Immature

CD34 

CD117 

Terminal deoxynucleotidyl transferase

Granulocytic/monocytic

CD33 

CD13 

CD15 

CD14

Erythroid

CD71 

Glycophorin A

Megakaryocytic

CD41 

CD42 

CD61

B lymphocytes

CD19 

CD20 

CD22 

κ Light chain 

λ Light chain

T lymphocytes

CD2 

CD3 

CD4 

CD5 

CD7 

CD8

Flow cytometry: Principle and instrumentation

The most significant discovery that led to the advancement of flow cytometry and its subsequent widespread application in clinical practice was the development of monoclonal antibodies.4 In the originalhybridoma experiments, lymphocytes with predetermined antibody specificity were co-cultured with a myeloma cell line to form immortalized hybrid cells producing specific monoclonal antibodies. For this discovery, which not only fueled the development of flow cytometry but also had innumerable research and, more recently, clinical applications, Köhler and Milstein received a Nobel Prize in 1984. Over the years, numerous antibodies were produced and tested for their lineage specificity. Categorization of these antibodies and associated antigens is accomplished through workshops on human leukocyte differentiation antigens that have been held regularly since 1982. These workshops provide a forum for reporting new antigens and antibodies and define a cluster of antibodies recognizing the same antigen, called cluster of differentiation (CD) (Table 32-2; see also Table 32-1). Consecutive numbers are assigned to each new reported antigen. The Ninth International Conference on Human Leukocyte Differentiation Antigens brought to over 350 the total number of antigens characterized.5

TABLE 32-2

Hematolymphoid Antigens Commonly Used in Clinical Flow Cytometry

Cluster of Differentiation

Function

Cellular Expression

CD1a

T cell development

Precursor T cells

CD2

T cell activation

Precursor and mature T cells, NK cells

CD3

Antigen recognition

Precursor and mature T cells

CD4

Co-receptor for HLA class II

Precursor T cells, helper T cells, monocytes

CD5

T cell signaling

Precursor and mature T cells, subset of B cells

CD7

T cell activation

Precursor and mature T cells, NK cells

CD8

Coreceptor for HLA class I

Precursor T cells, suppressor/cytotoxic T cells, subset of NK cells

CD10

B cell regulation

Precursor B cells, germinal center B cells, granulocytes

CD11b

Cell adhesion

Granulocytic and monocytic lineage, NK cells

CD13

Unknown

Granulocytic and monocytic lineage

CD14

Monocyte activation

Mature monocytes

CD15

Ligand for selectins

Granulocytic and monocytic lineage

CD16

Low-affinity IgG Fc receptor

Granulocytic and monocytic lineage, NK cells

CD18

Cell adhesion and signaling

Granulocytic and monocytic lineage

CD19

B cell activation

Precursor and mature B cells

CD20

B cell activation

Precursor and mature B cells

CD22

B cell activation and adhesion

Precursor and mature B cells

CD31

Cell adhesion

Megakaryocytes, platelets, leukocytes

CD33

Cell proliferation and survival

Granulocytic and monocytic lineage

CD34

Cell adhesion

Hematopoietic stem cells

CD36

Cell adhesion

Megakaryocytes, platelets, erythroid precursors, monocytes

CD38

Cell activation and proliferation

Hematopoietic cells, including activated lymphocytes and plasma cells

CD41

Cell adhesion

Megakaryocytes, platelets

CD42b

Receptor for von Willebrand factor

Megakaryocytes, platelets

CD45

T and B cell receptor activation

Hematopoietic cells

CD56

Cell adhesion

NK cells, subset of T cells

CD61

Cell adhesion

Megakaryocytes, platelets

CD62P

Homing

Platelets

CD63

Cell development, activation, growth, and motility

Platelets

CD64

High-affinity IgG Fc receptor

Granulocytic and monocytic lineage

CD71

Iron uptake

High density on erythroid precursors, low to intermediate density on other proliferating cells

CD79a

B cell receptor signal transduction

Precursor and mature B cells

CD117

Stem cell factor receptor

Hematopoietic stem cells, mast cells

Ig, Immunoglobulin; NK, natural killer.

Monoclonal antibodies have various applications, including immunohistochemistry, immunofluorescence, and Western blot. These methods study cellular proteins in fixed tissues or in cellular extracts; however, they do not provide the ability to examine antigens in their native state and cannot decipher composite cell populations with a complex antigen makeup. In contrast, flow cytometry can define antigen expression on numerous viable cells. Currently, 17 antigens can be detected simultaneously on an individual cell.6 This is accomplished by the conjugation of monoclonal antibodies to a variety of fluorochromes that can be detected directly by a flow cytometer. In a flow cytometer, particles are suspended in fluid and pass one by one in front of a light source. As particles are illuminated, they emit fluorescent signals registered by detectors. These results are later converted to digital output and analyzed using flow cytometry software. The flow cytometer consists of fluidics, a light source (laser), a detection system, and a computer. A brief discussion of these basic components is presented.

To be analyzed individually, cells must pass separately, one by one, through the illumination and detection system of a flow cytometer. This passage is accomplished by injecting a cell suspension into a stream of sheath fluid. This technique, called hydrodynamic focusing, creates a central core of individually aligned cells surrounded by a sheath fluid (Figure 32-3). The central alignment is essential for consistent illumination of cells as they pass before a laser light source.

Image 

FIGURE 32-3 Diagram of a flow cytometer. As cells are injected into pressurized sheath fluid, they are positioned in the center of the stream and one by one exposed to the laser light. Forward scatter (FS) and side scatter (SS) are collected by separate detectors.

A laser is composed of a tube filled with gas, most commonly argon or helium-neon, and a power supply. Current is applied to the gas to raise electrons of a gas to an excited state. When electrons return to a ground state, they emit photons of light. Through an amplification system, a strong beam of light with light waves of identical direction, polarization plane, and wavelength is produced. This narrow coherent beam of light is used to illuminate individual cells, each stained with antibodies conjugated to specific fluorochromes.

After absorption of laser light, the electrons of fluorochromes are raised from a ground state to a higher energy state (). The return to the original ground level is accompanied by a loss of energy, emitted as light of a specific wavelength. Flow cytometers are equipped with several photodetectors, each specific for light of a unique color (wavelength). The fluorescence from an individual cell is partitioned into its different wavelengths through a series of filters (dichroic mirrors) and directed to the corresponding photodetector. Fluorescent signals derived from different fluorochromes attached to particular antibodies are registered separately. Figure 32-4

Image 

FIGURE 32-4 Jablonski diagram showing a principle of fluorescence. When electrons absorb energy, they are raised to the excited state. Subsequently, on their return to ground state, the absorbed energy is emitted in a form of fluorescence.

In addition to fluorescence, scatter signals are recorded. The detector situated directly in line with the illuminating laser beam measures forward scatter (FS or FSC), which is proportional to particle volume or size. A photodetector located to the side measures side scatter (SS or SSC), which reflects surface complexity and internal structures such as granules and vacuoles. FS, SS, and fluorescence are displayed simultaneously on the instrument screen and registered by the computer system.

Pattern recognition approach to analysis of flow cytometric data

Concept of gating

Cell populations with similar physical properties such as size, cytoplasmic complexity, and expression of a specific antigen form clusters on data displays generated by flow cytometers. A gate is an electronic boundary an operator uses to delineate cell clusters. Thus, gating is a process of selecting, with a cursor or computer mouse, a population of interest as defined by one or more flow cytometric parameters.

Gating can be applied at the time of data acquisition (live gate) or at the time of analysis. For diagnostic purposes, data are collected ungated; that is, all events detected by the flow cytometer are recorded. This allows comprehensive testing and retention of positive and negative internal controls. In addition, unexpected abnormal populations are detected. Gating is most commonly applied after a specimen is run through a flow cytometer, when a target population is already known. In contrast, live gating focuses on the acquisition of data for a specific cell population as defined by flow cytometric parameters. For example, one can collect data only on CD19+ B cells to facilitate detection of a small population of monoclonal B cells.

Analysis of flow cytometric data

As with microscopic examination, an evaluation of flow cytometric data is based on the inspection of visual patterns. First, the data are scanned to detect abnormal populations. Subsequent analysis focuses on the antigenic properties of these abnormal cells.

Analysis begins with inspection of dot plots presenting cell size, cytoplasmic complexity, and expression of pan-hematopoietic antigen CD45. As in microscopic examination at low magnification, an operator detects specific cell populations based on their physical properties (). The identification of particular populations can be confirmed and further resolved on the scattergram of CD45 antigen density and SS (Figure 32-5Figure 32-6). This display also provides information on the relative proportion of specific cell populations in the flow cytometric sample. Lymphocytes show the highest density of CD45, with approximately 10% of the cell membrane occupied by this antigen. Granulocytic series show intermediate CD45 density; late erythroid precursors and megakaryocytes are negative for CD45. The CD45/SS display is particularly useful for detection of blasts, which overlap with lymphocytes, monocytes, or both on the FS/SS display.78

Image 

FIGURE 32-5 Main cell subpopulations of normal bone marrow. A, Bone marrow is composed of a heterogeneous population of cells of different sizes and variable complexity of cytoplasm (Wright-Giemsa stain, ×1000). B, Dot plot of forward scatter (FS, cell size) versus side scatter (SS, internal complexity) reflects the heterogeneity of bone marrow subpopulations. Lymphocytes are smallest with negligible amount of agranular cytoplasm and are located closest to the origins of the axes (aqua). Monocytes are slightly larger with occasional granules and vacuoles (green). Granulocytic series shows prominent granularity (navy).

Image 

FIGURE 32-6 Scattergram showing differential densities of pan-hematopoietic marker CD45 on marrow leukocytes. Lymphocytes (aqua) and monocytes (green) show highest density of CD45 antigen. Intermediate expression of CD45 is seen in the granulocytic population (navy) and blasts (black). Erythroid precursors (red) are CD45−. ECD, phycoerythrin-Texas Red; SS,Side scatter.

The FS/SS and CD45/SS displays allow the initial identification of the target population. Further analysis focuses on patterns of antigen expression, including qualitative data (antigen presence or absence) and fluorescence intensity as a relative measure of an antigen density.

Cell populations identified by flow cytometry

The surface and cytoplasmic markers expressed in hematologic malignancies resemble those of normal hematopoietic cell differentiation. Frequently, neoplastic cells are arrested at a particular stage of development and display aberrant antigenic patterns. Diagnosis and classification of hematologic neoplasms is based on the knowledge of normal hematopoietic maturation pathways.

In the past, the differentiation of hematopoietic cells was defined by morphologic criteria. Over time, it became clear that specific morphologic stages of development are accompanied by distinct changes in immunophenotype. Approximate morphologic-immunophenotypic correlates exist; however, because hematopoiesis is a continuous process, transitions between various developmental phases are not discrete.

All hematopoietic progeny are derived from pluripotent stem cells. These cells are morphologically unrecognizable and are defined by their functional and antigenic characteristics. They usually express a combination of CD34, CD117 (c-kit), CD38, and HLA-DR antigens.9 As hematopoietic cells mature, they lose stem cell markers and acquire lineage-specific antigens. A brief discussion of the maturation sequence of major hematopoietic cell lineages is presented in the following sections.

Granulocytic lineage

The stages of granulocytic lineage development, as defined by the expression of specific antigens, correspond closely to the morphologic sequence.10 The first morphologically recognizable cell committed to the granulocytic lineage is a myeloblast (Chapter 12). A myeloblast is characterized by an expression of immature cell markers CD34, CD38, HLA-DR, and stem cell factor receptor CD117. Pan-myeloid markers CD13 and CD33, present on all myeloid progeny, are first expressed at this stage. As a myeloblast matures to a promyelocyte, it loses CD34 and HLA-DR and acquires the CD15 antigen. Further maturation to a myelocyte stage leads to an expression of CD11b, a temporary loss of CD13, and a gradual decrease in the density of CD33. Finally, as granulocytic cells near a band stage, CD16 is acquired, and the density of CD13 increases.

Monocytic lineage

The earliest immunophenotype stage of monocytic development is defined by a gradual increase in the density of CD13, CD33, and CD11b antigens. Subsequent acquisition of CD15 and CD14 marks the transition to a promonocyte and mature monocyte. In contrast to the granulocytic series, strong expression of CD64 and HLA-DR antigens persists throughout monocytic maturation.

Erythroid lineage

The majority of erythroid precursors do not express pan-hematopoietic marker CD45. The earliest marker of erythroid differentiation is the transferrin receptor, CD71. The density of this antigen increases starting in the pronormoblast stage and is rapidly downregulated in reticulocytes.11 In contrast, glycophorin A, although present on reticulocytes and erythrocytes, first appears at the basophilic normoblast stage.

Megakaryocytic lineage

The maturation sequence of megakaryocytes is less well defined. CD41 and CD61, referred to as glycoprotein IIb/IIIa complex, appear as the first markers of megakaryocytic differentiation. These antigens are present on a small subset of CD34+ cells believed to represent early megakaryoblasts.9 CD31 and CD36, although not entirely specific for megakaryocytic lineage, also are present on megakaryoblasts. Subsequent maturation to megakaryocytes and platelets is characterized by the appearance of additional glycoproteins, CD42, CD62P, and CD63.

Lymphoid lineage

The B and T lymphocytes are derived from lymphoid progenitors that express CD34, terminal deoxynucleotidyl transferase (TdT), and HLA-DR. Lymphoid differentiation is characterized by a continuum of changes in the expression of surface and intracellular antigens. The earliest B cell markers include CD19, cytoplasmic CD22, and cytoplasmic CD79.12 As B cell precursors mature, they acquire the CD10 antigen. The appearance of the mature B cell marker CD20 coincides with the decrease in CD10 antigen expression. Another specific immature B cell marker is the cytoplasmic μ chain that eventually is transported to the surface and forms the B cell receptor. At this stage, the immunoglobulin chains in so-called naive B cells have become rearranged. The normal mature B cell population shows a mix of κ and λ light chain–expressing cells. The exclusive expression of only κ or λ molecules is a marker of monoclonality, seen frequently in mature B cell neoplasms. The differentiation of mature naive B cells, often recapitulated by B cell malignancies, is discussed in detail in Chapter 36.

Similar to B cell precursors, immature T cells express CD34 and TdT.13 The first markers associated with T cell lineage include CD2, CD7, and cytoplasmic CD3. CD2 and CD7 are also present in natural killer (NK) cells and, along with the CD56 molecule, are used to detect NK cell–derived neoplasms. In T cells, the expression of CD2, CD7, and cytoplasmic CD3 is followed by the appearance of CD1a and CD5 and coexpression of CD4 and CD8 antigens. Finally, the CD3 antigen appears on the cell surface, and CD4 or CD8 is lost. The sequential transition from double-negative (CD4CD8) through double-positive (CD4+CD8+) stages generates a population of mature helper (CD4+) and suppressor (CD8+) T cells. T cell differentiation occurs in the thymus.

Flow cytometric analysis of myeloid neoplasms (acute myeloid leukemias and chronic myeloid neoplasms)

In myeloid malignancies, flow cytometry is used for initial diagnosis, follow-up, and prognostication. Specific immunophenotypes are associated with select cytogenetic abnormalities. Because most myeloid malignancies are stem cell disorders, the evaluation of blast population and maturing myeloid component is considered mandatory. Almost invariably, blasts are characterized by a low-density expression of CD45 antigen. In normal bone marrow, a blast gate includes a relatively low number of cells showing the immature myeloid immunophenotype (Figure 32-6). In acute myeloid and lymphoblastic leukemias, this region becomes densely populated by immature cells, which reflects the increased number of blasts seen in the bone marrow (Figure 32-7). The exact position of the immature population on the CD45/SS displays depends on the subtype of acute myeloid leukemia (AML). In this chapter, the immunophenotypic features of AML and chronic myeloid neoplasms are discussed in the context of the World Health Organization classification, which introduced categories defined by recurrent cytogenetic abnormalities.1 These leukemias often show specific immunophenotypes and are discussed separately in the following sections.

Image 

FIGURE 32-7 Bone marrow specimen showing acute leukemia. Note uniform cytologic and flow cytometric characteristics. A, Bone marrow aspirate from a patient with acute lymphoblastic anemia (Wright-Giemsa stain, ×500). B, CD45 versus side scatter (SS) plot shows a homogeneous population of blasts with a marked decrease in normal hematopoietic elements. Compare with the heterogeneous pattern of normal bone marrow in Figure 33-6. ECD, Phycoerythrin-Texas Red.

Acute myeloid leukemias with recurrent cytogenetic abnormalities

In most cases, AML with t(8; 21)(q22; q22); RUNX1/RUNX1T1 shows an immature myeloid immunophenotype with high-density CD34 and coexpression of CD19 (Figure 32-8).14 In addition, numerous myeloid antigens, including CD33, CD13 and myeloperoxidase, are expressed. Frequently, there is asynchronous coexpression of CD34 and CD15. TdT is commonly present.

Image 

FIGURE 32-8 Acute myeloid leukemia with t(8; 21)(q22; q22); RUNX1-RUNX1T1A, CD45 versus side scatter (SS) showing increase in blasts (red) with residual lymphocytes (aqua). B and C,Blasts are positive for CD33 and CD34 with characteristic coexpression of CD19 antigen. ECD, Phycoerythrin-Texas Red; FITC, fluorescein isothiocyanate.

AML with inv(16)(p13.1q22) or t(16; 16)(p13.1; q22); CBFB/MYH11 is characterized by the presence of immature cells with expression of CD34, CD117, and TdT and a subpopulation of maturing cells showing monocytic (CD14, CD11b, CD4) and granulocytic (CD15) markers.15 The aberrant coexpression of CD2 on the monocytic population is common.

Acute promyelocytic leukemia, AML with t(15; 17)(q22; q12); PML/RARA, shows a specific immunophenotype. In contrast to most less-differentiated myeloid leukemias, acute promyelocytic leukemia manifests with high SS, which reflects the granular cytoplasm of leukemic cells (Figure 32-9). The constellation of immunophenotypic features used to diagnose acute promyelocytic leukemia includes lack of CD34 and HLA-DR antigens, presence of homogeneous strong CD33 along with myeloperoxidase, and variable CD13 and CD15.16

Image 

FIGURE 32-9 Acute promyelocytic leukemia. A, Typical side scatter (SS) pattern in acute promyelocytic leukemia corresponding to prominent granularity of leukemic cells (red). Residual lymphocytes are shown in aqua. B, Numerous leukemic promyelocytes with distinct granules and occasional Auer rods (Wright-Giemsa stain, ×1000). C, Leukemic cells show high-density expression of CD33 antigen and lack HLA-DR. D, Similarly, CD34 antigen is absent or present in only a few leukemic cells. ECD, phycoerythrin-Texas Red; PC5, phycoerythrin-cyanine 5; PE, phycoerythrin; SS, Side scatter.

AMLs with t(9; 11)(p22; q23); MLLT3/MLL in adults most commonly present with monocytic differentiation. The immunophenotypic features are nonspecific and can be seen in any acute myelomonocytic or monocytic leukemia (negative for CD34 and positive for CD33, CD13, CD14, CD4, CD11b, and CD64).

Acute myeloid leukemias not otherwise specified

In the least-differentiated AMLs—AML with minimal differentiation and AML without maturation—blasts are present in the region of low-density CD45 antigen and display low SS reflecting their relatively agranular cytoplasm. Even the least differentiated AML with minimal differentiation is usually positive for myeloid markers. The expression of CD13, CD33, and CD117 is common. Primitive hematopoietic antigens such as CD34 and HLA-DR are often seen. Myeloperoxidase is absent or is expressed in only a few cells. The immunophenotypic profile of AML with maturation is similar, but more mature myeloid markers such as CD15 and myeloperoxidase are often expressed.

Occasionally, there is aberrant coexpression of antigens. Simultaneous expression of early and late markers of myeloid differentiation on the leukemic blasts is not uncommon (asynchronous antigen expression). Similarly, markers specific for other lineages, such as lymphoid lineage may be seen on myeloid blasts. The most common example is CD7 antigen, which is usually present in the T/NK cell population (). Figure 32-10

Image 

FIGURE 32-10 Myeloblasts of acute myeloid leukemia show aberrant coexpression of CD7 antigen. PE, Phycoerythrin; PC5, phycoerythrin-cyanine 5.

Acute myelomonocytic leukemia and acute monoblastic leukemia usually show higher expression of CD45, similar to normal monocytic precursors. In addition, in acute myelomonocytic leukemia, a population of primitive myeloid blasts is often seen (). The expression of myeloid markers and antigens associated with monocytic lineage, such as CD14, CD4, CD11b, and CD64, is commonly seen. Although CD14 is present on all mature monocytes, it may be absent in monocytic leukemias.Figure 32-1117 More immature monocytic markers, such as CD64, are more consistently expressed.

Image 

FIGURE 32-11 Peripheral blood immunophenotyping in acute myelomonocytic leukemia. A, CD45 versus side scatter (SS) display shows myeloid blasts (red) and a monocytic population (green). B through D, Primitive leukemic blasts are positive for CD34 and negative for CD14. In contrast, monocytic population does not express CD34 and shows positivity for mature monocyte marker CD14 and characteristic monocytic pattern of CD11b and CD15 expression. ECD, Phycoerythrin-Texas Red; FITC,fluorescein isothiocyanate; PC5, phycoerythrin-cyanine 5.

Acute erythroid leukemias are categorized into two subtypes: pure erythroid leukemia and erythroleukemia (erythroid/myeloid leukemia). In the latter, primitive myeloid blasts and erythroid precursors are present. Leukemic cells are positive for erythroid markers such as CD71, glycophorin A, and hemoglobin (HGB). In more immature erythroid leukemias, glycophorin A and hemoglobin may be absent. In these cases, the diagnosis is based on the absence of myeloid markers, high expression of CD71, and scatter characteristics.

Acute megakaryoblastic leukemia usually shows low SS and low to absent CD45. Early megakaryocytic markers, CD41 and CD61, are frequently expressed.18 Occasionally, the late megakaryocytic marker CD42 is present. The expression of stem cell markers CD34 and HLA-DR on the population of leukemic megakaryoblasts varies.

Myeloproliferative neoplasms and myelodysplastic syndromes

The knowledge of antigen expression in the normal differentiation of myeloid lineages allows us to define the aberrant expression patterns frequently seen in chronic myeloid disorders. The abnormalities detected by flow cytometry reflect morphologic features (e.g., hypogranulation of neutrophils in myelodysplastic syndrome detected by low SS) and show changes in antigen expression. Qualitative (presence or absence of a particular antigen) and quantitative abnormalities (differences in the number of antigen molecules) can be used for diagnostic purposes. The interested reader is referred to review articles discussing the details of immunophenotyping in myelodysplastic syndromes and myeloproliferative neoplasms.19-21 A few examples are highlighted to illustrate the role of flow cytometry in diagnosing these diseases.

SS abnormalities related to hypogranulated neutrophils are seen in approximately 70% of myelodysplastic syndromes (). In high-grade myelodysplastic syndrome and myeloproliferative neoplasms undergoing transformation, the increase in immature cells is detected easily. Blasts have a variety of aberrant immunophenotypic features, most commonly coexpression of CD7 and CD56 antigens. Blasts and maturing granulocytic precursors may show asynchronous expression of myeloid markers, including retention of CD34 and HLA-DR in late stages of maturation or late myeloid markers presenting early in differentiation, such as CD15 on myeloblasts. Asynchronous coexpression of markers can also be seen in monocytic and erythroid lineages. Aberrant immunophenotypes are seen in 98% of cases of myelodysplastic syndrome. More importantly, immunophenotypic abnormalities can be seen in cases with minimal or no morphologic dysplasia.Figure 32-1219 Other studies underscore the significance of immunophenotypic abnormalities in predicting the outcome after stem cell transplantation.2223

Image 

FIGURE 32-12 A, Low side scatter (SS) of hypogranular neutrophils seen in most cases of myelodysplastic syndrome (navy). B, Corresponding photomicrograph of markedly dysplastic, hypogranulated neutrophils in myelodysplastic syndrome (Wright-Giemsa stain, ×1000). ECD, Phycoerythrin-Texas Red.

The utility of flow cytometry in myeloproliferative neoplasms is less well established. Specifically, the application of flow cytometry as a diagnostic tool in chronic myelogenous leukemia is limited to the accelerated or blast phase, in which a lineage of an expanding blast population needs to be determined. In the chronic phase, the presence of BCR/ABL1 rearrangement (Philadelphia chromosome) demonstrated by conventional karyotyping or molecular studies remains the defining feature of the disease. Other myeloproliferative neoplasms are not well studied. In general, flow cytometric abnormalities are seen in most cases with abnormal karyotype.20 No consistent set of immunophenotypic features that can be routinely used in the workup of myeloproliferative states has been described.

Flow cytometric analysis of lymphoid neoplasms (lymphoblastic leukemia/lymphoma and mature lymphoid neoplasms)

Similar to myeloid neoplasms, a diagnosis of lymphoid malignancies relies on the expression of lineage-associated markers corresponding to specific stages of lymphoid development. No single marker can be used for lineage assignment, and a diagnosis is typically based on the presence of several B cell or T cell antigens. The sentinel feature of mature B and T cells is the presence of surface receptor complexes. The immune system responds to a wide array of antigens; in healthy individuals, B and T cells express a great diversity of surface immunoglobulin and T cell receptor complexes (polyclonal populations). A neoplastic lymphoid population is characterized by the monoclonal expression of a single B or T cell receptor. In most cases, clonality confirms the malignant nature of lymphoid proliferation. In contrast, lymphoid precursors are generally negative for surface immunoglobulin and T cell receptors and instead carry immature markers. In lymphoblastic (precursor-derived) neoplasms, an expansion of a population with homogeneous marker expression, rather than clonality, is diagnostic of malignancy. The following section presents the key immunophenotypic features of lymphoblastic leukemias and lymphomas. Selected examples of the association between the immunophenotype and the genotype are discussed.

B lymphoblastic leukemia/lymphoma

B lymphoblastic leukemia/lymphoma (B-LL) is also referred to as B acute lymphoblastic leukemia or B lymphoblastic lymphoma. B lymphoblasts are positive for CD19, CD22, CD79a, HLA-DR, and TdT (). The expression of CD34 and CD10 is frequently seen. Surface immunoglobulin light chains are not present. Cytoplasmic μ chain or surface immunoglobulin M may be detected, however. Because B-LL can arise at any stage of B cell differentiation, the presence of several specific markers usually defines early precursor, intermediate, and pre-B stages. Frequently, immunophenotypes correlate with specific cytogenetic and clinical features. In routine practice, confirmation of cytogenetic abnormality using conventional karyotyping or molecular techniques is necessary. Figure 32-13

Image 

FIGURE 32-13 B lymphoblastic leukemia/lymphoma. A, Low-density CD45 antigen characteristic of the blast population. B, Uniform expression of CD34 and CD19 on leukemic blasts. C,High-density CD10 on CD19+ blasts. D, Lack of surface κ and λ light chains signifies immature B cell population. ECD, phycoerythrin-Texas Red; FITC, fluorescein isothiocyanate; PC5, phycoerythrin-cyanine 5; PE, phycoerythrin; SS, Side scatter.

B lymphoblastic leukemia/lymphoma with t(v; 11q23); mll rearranged

MLL gene rearrangements occur most frequently in infant B-LL. Unlike in most B-LLs, blasts in this leukemia are negative for CD10 antigen.24 CD19, CD34, TdT, and occasional myeloid markers are present. The more mature B cell marker CD20 is absent.

B lymphoblastic leukemia/lymphoma with t(9; 22)(q34; q11.2); bcr/abl1

Philadelphia chromosome, t(9; 22); BCR/ABL1, is a hallmark of chronic myelogenous leukemia but also can occur in pediatric and adult B-LL. These cases benefit from an addition of tyrosine kinase inhibitor to the chemotherapy regimen, so it is important to identify them promptly. Most BCR/ABL1–positive cases have a classic intermediate or common B-LL immunophenotype with the expression of CD19, CD10, CD34, and TdT. The expression of myeloid markers CD13 and CD33 and the lack or decreased expression of CD38 antigen on leukemic blasts are common. The density of antigens and their homogeneous or heterogeneous expression within leukemic populations correlates closely with the presence of BCR-ABL1.25

T lymphoblastic leukemia/lymphoma

T lymphoblastic leukemia/lymphoma (T-LL) is derived from immature cells committed to T cell lineage. Designation of leukemia or lymphoma depends on the primary site of involvement: bone marrow or lymph node. T-LL expresses a combination of markers reflecting the stage of T cell differentiation. CD3 is the most specific T cell marker. As with normal T cells, this antigen is seen initially in the cytoplasm before appearing on the cell surface. Other T cell antigens include CD2, CD7, CD5, CD1a, CD4, and CD8. Usually a series of these antigens is detected, recapitulating the T cell differentiation (). CD34 and CD10 may be present. As in other lymphoid neoplasms, the panel of markers determines the lineage. Figure 32-14

Image 

FIGURE 32-14 T lymphoblastic leukemia/lymphoma. A, Predominant population in the blast gate. B and C, Although CD3 antigen is absent from the surface of leukemic cells, it is present in blast cytoplasm, confirming the precursor T cell origin of the leukemia (C). Note residual normal T cells (aqua) positive for surface CD3 and CD5 antigens. D, Simultaneous expression of CD4 and CD8 antigens. ECD, phycoerythrin-Texas Red; FITC, fluorescein isothiocyanate; PC5, phycoerythrin-cyanine 5; PE, phycoerythrin; SS, Side scatter.

Mature lymphoid neoplasms

B and T cell lymphomas display immunophenotypes resembling their normal counterparts. The immunophenotypic features of lymphomas are discussed in detail in Chapter 36 and are summarized in Table 36-2. The flow cytometric workup of lymphomas is facilitated by the clonal origin of mature lymphoid neoplasms, which implies that the malignant population is derived from a single cell. Therefore, all neoplastic cells typically show similar genetic and immunophenotypic features. This stands in strong contrast to variable immunophenotypes of normal lymphoid populations, reflecting a process of antigen-driven selection.

Mature B cell neoplasms

Normal precursor B cells randomly rearrange immunoglobulin heavy and light chain genes. As a result, a mature B cell population expresses a mix of heavy and light chains (, Figure 32-15 A). In contrast, a monoclonal surface light chain expression, exclusively κ or λ, is seen in most B cell lymphomas (Figure 32-15B). Light chain monoclonality along with the expression of pan–B cell markers is diagnostic of B cell lymphoma. Rarely, lymphomas may lose the expression of surface light chains, a feature not seen in normal mature B cells.26 In most cases of plasma cell myeloma, neoplastic plasma cells lack surface immunoglobulin light chains and express only cytoplasmic κ or λ.

Image 

FIGURE 32-15 Comparison of surface light chain expression in reactive and malignant B cells. A, Reactive B cells show heterogeneous expression of κ and λ. B, B cell lymphomas are monoclonal, with the entire lymphoma population expressing only one type of light chain. FITC, Fluorescein isothiocyanate; PE, phycoerythrin.

Mature T cell neoplasms

In T cells, similar to B cells, clonality in most cases indicates malignancy. In the past, the clonality of T cells could only be confirmed by using a molecular analysis of T cell receptor genes. Recently, a flow cytometric assay has been shown to detect clonality in most cases of T cell lymphoma.27 This technique uses a broad array of antibodies against variable regions of T cell receptors. Because this methodology is not widely available, often a diagnosis of T cell lymphoma is based on aberrant immunophenotype. In most cases, a loss or atypical expression of a lymphoid marker can be shown using flow cytometry. For example, mycosis fungoides/Sézary syndrome is characterized by a mature T cell immunophenotype with expression of CD2, surface CD3, CD5, and CD4 and with a loss of the CD7 antigen (Figure 32-16). Over the years it has been shown that the aberrant immunophenotype is a reliable diagnostic feature when the neoplastic population is sizeable. However, small numbers of T cells with unusual antigen makeup can appear in inflammatory conditions;28 thus the aberrant immunophenotype alone cannot be considered pathognomonic of T cell malignancy.

Image 

FIGURE 32-16 Mycosis fungoides. A, T cell population is positive for CD4 antigen (red). B, Neoplastic T cells show loss of CD7. Red represents CD4 and the expression of CD7 (green) is very low. FITC, Fluorescein isothiocyanate; PE, phycoerythrin;PC5, phycoerythrin-cyanine 5.

Other applications of flow cytometry beyond immunophenotyping of hematologic malignancies

The immunophenotyping of hematolymphoid neoplasms is one of many applications of flow cytometry. Other common applications include a diagnosis and monitoring of immunodeficiency states, diagnosis of paroxysmal nocturnal hemoglobinuria (PNH), stem cell enumeration, cell cycle analysis, detection of fetal hemoglobin, and monitoring of sepsis.

Select primary (inherited) and secondary (acquired) immunodeficiencies can be diagnosed using flow cytometry. Both a loss of specific antigens (e.g., CD11/CD18 in leukocyte adhesion deficiency) and functional defects (e.g., oxidative burst evaluation in chronic granulomatous disease) can be assayed by flow cytometry.

Human immunodeficiency virus infection causes a progressive decrease in the number of CD4+ helper T cells. The absolute number of helper T cells in peripheral blood correlates with the stage of the disease and with patient prognosis. The enumeration of T cells and their subsets is easily accomplished by flow cytometry using antibodies against CD4 and CD8 antigens. The absolute numbers are derived by performing a routine white blood cell (WBC) count on the concurrent peripheral blood specimen or by running calibrating beads simultaneously with the patient sample. The CD4:CD8 ratio in healthy individuals is typically greater than 1. There is a significant decrease in numbers of CD4 positive T cells in HIV-positive patients resulting in a reversed CD4:CD8 ratio. Since the CD4 lymphocyte depletion is associated with various infections, the absolute number of CD4 positive lymphocytes serves also as a guide for antibiotic prophylaxis in HIV positive patients.

The diagnostic approach to PNH is a prime example of how an application of flow cytometry increases understanding of hematologic disorders and directly contributes to clinical decision making (Chapter 24).29Before the development of the flow cytometric assay, PNH diagnosis was based on detection of increased susceptibility of RBCs to lysis by the Ham or sucrose hemolysis tests, both of which showed inconsistent sensitivity. Flow cytometry significantly improved the sensitivity and specificity of PNH testing. The absence or decreased expression of glycosylphosphatidylinositol-anchored proteins on RBCs, granulocytes, and monocytes as measured by flow cytometry is diagnostic of PNH. In addition, the levels of CD59 expression correlate with clinical symptoms (Figure 32-17).

Image 

FIGURE 32-17 Diagnosis of paroxysmal nocturnal hemoglobinuria (PNH) is based on the decreased expression of glycosylphosphatidylinositol (GPI)–linked molecules. Different levels of GPI-anchored proteins are best visualized in red blood cells (RBCs) using an antibody against CD59 antigen. A, RBCs from a healthy volunteer show a high number of CD59 molecules and correspond with type I cells. B, Varying percentages of type I cells (normal level of CD59 antigen) and a population with slight decrease in CD59 expression (type II cells) can be seen in PNH patients. C, Granulocytes with a complete loss of CD59 (type III cells) in a patient with PNH. This patient received numerous red blood cell transfusions; the loss of CD59 is best shown in granulocytic and monocytic populations. PE, Phycoerythrin.

Another important application of flow cytometry is cell sorting. During sorting, a heterogeneous cell population is physically divided into subsets according to their physical or immunophenotypic properties. High-speed sorting is achieved by charging droplets containing individual cells of interest. As the charged droplet passes through the electrostatic field, it is isolated from the remainder of the sample and collected into a separate container. The primary clinical application of cell sorting is in stem cell transplantation.

For years, flow cytometry remained confined to the hematopathology and research laboratories. Currently, this methodology is used in bone marrow transplantation, transfusion medicine, coagulation, microbiology, molecular pathology, and drug development. Specific examples of novel applications include tissue typing, molecular testing for neoplasia-associated translocations, and follow-up of drug response, such as by monitoring platelet activation after antiplatelet therapy.

Flow cytometry is a mature field that in recent years experienced a revival with a focus on high-throughput testing for simultaneous analysis of multiple biologic constituents. New approaches to a single-cell analysis such as spectral flow cytometry and an integration of mass spectrometry with single-cell fluidics provide a superior resolution and expand the number of parameters that can be measured in any given cell. These methodologies are in development and open new avenues to diagnostic immunophenotyping in hematopathology.3031

Summary

• Flow cytometry measures physical, antigenic, and functional properties of particles suspended in a fluid.

• Multiparameter flow cytometry is a technique routinely used for a diagnosis and follow-up of hematologic disorders.

• The characterization of complex specimens is achieved through the analysis of individual cells for multiple parameters and the simultaneous display of data for thousands of cells. The cell size, cytoplasmic complexity, and immunophenotypic features detected by monoclonal antibodies directly conjugated to various fluorochromes are analyzed in clinical specimens.

• A key starting point in flow cytometric analysis is a high-quality fresh specimen.

• A flow cytometer consists of fluidics, a light source (laser), multiple detectors, and a computer.

• As with microscopic examination, an evaluation of flow cytometric data is based on the inspection of visual patterns. Initially, the entire sample is scanned for the presence of abnormal populations. Subsequently, detailed immunophenotypic features of cell subsets are studied.

• The immunophenotyping of hematologic specimens is based on knowledge of the maturation patterns of hematopoietic cells. In comparison, myeloid and lymphoid malignant cells and cell populations in nonneoplastic hematologic disorders show significant qualitative and quantitative differences in antigen expression.

• Flow cytometric analysis of acute leukemia determines a lineage of leukemic cells. In select entities, immunophenotype corresponds to the underlying genetic lesion.

• Immunophenotyping of myelodysplastic syndromes and chronic myeloproliferative neoplasms is an emerging application of clinical flow cytometry.

• The clonality of mature B cell and T cell neoplasms can be detected by flow cytometry.

• Flow cytometric analysis is used for diagnosis and monitoring of immunodeficiencies, stem cell enumeration, detection of fetal hemoglobin, tissue typing, molecular analysis, and drug testing.

Now that you have completed this chapter, go back and read again the case studies at the beginning and respond to the questions presented.

Review questions

Answers can be found in the Appendix.

1. What is the most common clinical application of flow cytometry?

a. Diagnosis of platelet disorders

b. Detection of fetomaternal hemorrhage

c. Diagnosis of leukemias and lymphomas

d. Differentiation of anemias

2. Which of the following is true of CD45 antigen?

a. It is present on every cell subpopulation in the bone marrow.

b. It is expressed on all hematopoietic cells, with the exception of megakaryocytes and late erythroid precursors.

c. It is not measured routinely in flow cytometry.

d. It may be present on nonhematopoietic cells.

3. Erythroid precursors are characterized by the expression of:

a. CD71

b. CD20

c. CD61

d. CD3

4. In Figure 32-2A, the cell population colored in aqua represents:

a. Monocytes

b. Nonhematopoietic cells

c. Granulocytes

d. Lymphocytes

5. Antigens expressed by B-LL include:

a. CD3, CD4, and CD8

b. CD19, CD34, and CD10

c. There are no antigens specific for B-LL.

d. Myeloperoxidase

6. Which of the following is true of flow cytometric gating?

a. It is best defined as selection of a target population for flow cytometric analysis.

b. It can be done only at the time of data acquisition.

c. It can be done only at the time of final analysis and interpretation of flow cytometric data.

d. It is accomplished by adjusting flow rate.

7. Collection of ungated events:

a. Facilitates comprehensive analysis of all cells

b. Does not help in detection of unexpected abnormal populations

c. Allows the collection of data on a large number of rare cells

d. Is used for leukemia diagnosis only

8. Mycosis fungoides is characterized by:

a. Loss of certain antigens compared with the normal T cell population

b. Polyclonal T cell receptor

c. Immunophenotype indistinguishable from that of normal T cells

d. Expression of CD3 and CD8 antigens

9. Mature granulocytes show the expression of:

a. CD15, CD33, and CD34

b. CD15, CD33, and CD41

c. CD15, CD33, and CD13

d. CD15, CD33, and CD7

10. During the initial evaluation of flow cytometric data, cell size, cytoplasmic complexity, and expression of CD45 antigen are used to define cell subpopulations. Which of the following parameters defines cytoplasmic complexity/granularity?

a. SS

b. FS

c. CD45

d. HLA-DR

11. The most important feature of the mature neoplastic B cell population is:

a. The presence of a specific immunophenotype with expression of CD19 antigen

b. A clonal light chain expression (i.e., exclusively κ- or λ-positive population)

c. A clonal T cell receptor expression

d. Aberrant expression of CD5 antigen on CD19+ cells

References

1.  In: Swerdlow S.H, Campo E, Harris N.L, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France : IARC Press 2008.

2.  Czader M, Ali S.Z. Flow cytometry as an adjunct to cytomorphologic analysis of serous effusionsDiagn Cytopathol; 2003; 29:74-78 2003.

3.  Wood B.L, Arroz M, Barnett D, et al. 2006 Bethesda International Consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry optimal reagents and reporting for the flow cytometric diagnosis of hematopoietic neoplasia. Cytometry B Clin Cytom; 2007; 72:S14-S22.

4.  Köhler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificityNature; 1975; 256:495-497.

5.  Human Cell Differentiation Molecules. Retrieved from Available at: http://hcdm.org Accessed 06.11.14.

6.  Perfetto S.P, Chattopadhyay P.K, Roederer M. Seventeen-colour flow cytometry unravelling the immune system. Nat Rev Immunol; 2004; 4:648-655.

7.  Stelzer G.T, Shults K.E, Loken M.R. CD45 gating for routine flow cytometric analysis of human bone marrow specimensAnn N Y Acad Sci; 1993; 677:265-281.

8.  Borowitz M.J, Guenther K.L, Shults K.E, et al. Immunophenotyping of acute leukemia by flow cytometric analysis use of CD45 and right-angle light scattered to gate on leukemic blasts in three-color analysis. Am J Clin Pathol; 1993; 100:534-540.

9.  Macedo A, Orfao A, Ciudad J, et al. Phenotypic analysis of CD34 subpopulations in normal human bone marrow and its application for the detection of minimal residual diseaseLeukemia; 1995; 9:1896-1901.

10.  Terstappen L.W, Safford M, Loken M.R. Flow cytometric analysis of human bone marrow III. Neutrophil maturation. Leukemia; 1990; 4:657-663.

11.  Loken M.R, Shah V.O, Dattilio K.L, et al. Flow cytometric analysis of human bone marrow I. Normal erythroid development. Blood; 1987; 69:255-263.

12.  Ciudad J, Orfao A, Vidriales B, et al. Immunophenotypic analysis of CD191 precursors in normal human adult bone marrow implications for minimal residual disease detection. Haematologica; 1998; 83:1069-1075.

13.  Terstappen L.W, Huang S, Picker L.J. Flow cytometric assessment of human T-cell differentiation in thymus and bone marrowBlood; 1992; 79:666-677.

14.  Andrieu V, Radford-Weiss I, Troussard X, et al. Molecular detection of t(8; 21)/AML1-ETO in AML M1/M2 correlation with cytogenetics, morphology and immunophenotype. Br J Haematol; 1996; 92:855-865.

15.  Adriaansen H, Boekhorst P.A.W, Hagemeijer A.M, et al. Acute myeloid leukemia M4 with bone marrow eosinophilia (M4Eo) and inv(16)(p13q22) exhibits a specific immunophenotype with CD2 expressionBlood; 1993; 81:3043-3051.

16.  Lo Coco F, Avvisati G, Diverio D, et al. Rearrangements of the RAR-alpha gene in acute promyelocytic leukaemia correlations with morphology and immunophenotype. Br J Haematol; 1991; 78:494-499.

17.  Krasinskas AM, Wasik MA, Kamoun M, et al. The usefulness of CD64, other monocyte-associated antigens, and CD45 gating in the subclassification of acute myeloid leukemias with monocytic differentiationAm J Clin Pathol; 1998; 110:797-805.

18.  Helleberg C, Knudsen H, Hansen P.B, et al. CD34+ megakaryoblastic leukaemic cells are CD382, but CD61+ and glycophorin A1 improved criteria for diagnosis of AML-M7. Leukemia,; 1994; 11:830-834.

19.  Stetler-Stevenson M, Arthur D.C, Jabbour N, et al. Diagnostic utility of flow cytometric immunophenotyping in myelodysplastic syndromeBlood; 2001; 98:979-987.

20.  Kussick S.J, Wood B.L. Four-color flow cytometry identifies virtually all cytogenetically abnormal bone marrow samples in the workup of non-CML myeloproliferative disordersAm J Clin Pathol; 2003; 120:854-865.

21.  Zhang S, Zhou J, Nassiri M, Czader M. Diagnostic approach to myelodysplastic syndrome. In: Sayar H. Myelodysplastic Syndrome. Hauppauge NY : Nova Science Publishers 2013; 25-60 chapter 2.

22.  Ogata K, Nakamura K, Yokose N, et al. Clinical significance of phenotypic features of blasts in patients with myelodysplastic syndromeBlood; 2002; 100:3887-3896.

23.  Wells, D. A, Benesch, M, Loken, M. R, et al. Myeloid and monocytic dyspoiesis as determined by flow cytometric scoring in myelodysplastic syndrome correlates with the IPSS and with outcome after hematopoietic stem cell transplantation. Blood, 102, 394–403.

24.  Harbott J, Mancini M, Verellen-Dumoulin C, et al. Hematological malignancies with a deletion of 11q23 cytogenetic and clinical aspects. Leukemia; 1998; 12:823-827.

25.  Tabernero M.D, Bortoluci A.M, Alaejos I, et al. Adult precursor B-ALL with BCR/ABL gene rearrangements displays a unique immunophenotype based on the pattern of CD10, CD34, CD13 and CD38 expressionLeukemia; 2001; 15:406-414.

26.  Li S, Eshleman J.R, Borowitz M.J. Lack of surface immunoglobulin light chain expression by flow cytometric immunophenotyping can help diagnose peripheral B-cell lymphomaAm J Clin Pathol; 2002; 118:229-234.

27.  Beck R.C, Stahl S, O’Keefe C.L, et al. Detection of mature T-cell leukemias by flow cytometry using anti T-cell receptor V beta antibodiesAm J Clin Pathol; 2003; 120:785-794.

28.  Alaibac M, Pigozzi B, Belloni-Fortina A, et al. CD7 expression in reactive and malignant human skin T-lymphocytesAnticancer Res; 2003; 23:2707-2710.

29.  Richards S.J, Rawstron A.C, Hillmen P. Application of flow cytometry to the diagnosis of paroxysmal nocturnal hemoglobinuriaCytometry; 2004; 42:223-233.

30.  Sander C.K, Mourant J.R. Advantages of full spectrum flow cytometryJ Biomed Opt; 2013; 18:1-8 037004.

31.  Bendall S.C, Simonds E.F, Qiu P, Amir el-A.D, et al. Single-cell mass cytometry of differential immune and drug responses across a human hematopoietic continuumScience; 2011; 332:687-696.