Revised guideline on immunophenotyping in acute leukaemias and chronic lymphoproliferative disorders

Abstract

In 1994, the General Haematology Task Force (GHTF) of the British Committee for Standards in Haematology (BCSH) published two guidelines on immunophenotyping in acute leukaemias (General Haematology Task Force of the BCSH, 1994a) and chronic lymphoproliferative disorders (General Haematology Task Force of the BCSH, 1994b). Over the past five years there have been major technical advances in this field (Table 1). In addition, some new monoclonal antibodies (McAb) have been shown to be highly specific for the lymphoid or myeloid lineages, respectively, and their use is therefore recommended in the diagnosis of acute leukaemias. This revised guideline will focus on new techniques and reagents and define new panels of McAb recommended for lineage assignment in acute leukaemias and for the characterization of chronic lymphoproliferative disorders. This guideline does not consider in any detail the detection of minimal residual disease, the diagnosis of biphenotypic acute leukaemias or the evaluation of stem cell harvest. Nor does it address the selection of antibodies for therapeutic purposes. Furthermore, the antibody panels outlined here are not absolute and other McAb may be equally appropriate. This guideline was written by a working party comprised of five members selected by the BCSH because of their expertise in this field. In addition the draft guideline was reviewed by eight UK laboratories in the forefront of the field of immunophenotyping. Their comments were discussed and incorporated into the draft when appropriate. The final manuscript was approved by all members of the BCSH and by the Committee of the British Society of Haematology (BSH). The guideline represents the opinion of the BCSH Task Force and has been approved by the BSH. Peripheral blood and/or bone marrow samples should be collected and placed in an anticoagulant, either 0.34 M di- or tri-potassium ethylenediamine tetra-acetic acid (K2 or K3 EDTA) or preservative-free heparin. The choice of one or other anticoagulant will depend on the additional tests to be performed on the same specimen. For instance, heparin will be the anticoagulant of choice if cytogenetic analysis is to be performed while EDTA gives better preservation of the cell morphology. There is no need for anticoagulation of specimens such as pleural, ascitic or cerebrospinal fluid. For samples referred to central laboratories, fresh blood or marrow smears should be made at the time of collection. If the sample is not tested in the local laboratory but referred to a central laboratory it should arrive within 24 h of collection and should be kept at room temperature (between 10 and 30 °C) during transportation. Suspected cases of Burkitt's lymphoma/leukaemia and cerebrospinal fluid samples require rapid transport and processing of the sample. It is recommended that a viability test, e.g. trypan blue exclusion test (Sigma Chemical Co., St. Louis, USA), be performed on samples received for immunophenotyping beyond 24 h (e.g. 24–72 h) to ensure that there is a good viability, e.g. at least 80% viable cells (Braylan et al., 1997). In this situation results should be assessed with caution as antigen loss, particularly of markers which are weakly expressed, may occur. Addition of tissue culture medium is recommended if the specimens are likely to be stored beyond 24 h. • Results need to be interpreted in the context of the clinical features and cell morphology, particularly in specimens containing a mixture of normal and neoplastic cells. When reporting on the marker results, the whole composite phenotype should be taken into account as most immunological markers are not strictly lineage specific; for instance cluster designation (CD)7 is a T-cell marker but it is also expressed in a proportion of cases of acute myeloid leukaemia (AML). • Controls. Positive and negative controls for each McAb should be carried out only if a small number of tests are performed infrequently or if a specific antigen is used infrequently, to monitor the validity of the immunostaining and the lysing procedures. This is essential when a new McAb and/or batch of reagents is introduced, after an instrument service or calibration and/or when a new technique is applied. Positive and negative controls should be either leukaemic or normal cells by which the antigen is known to be expressed or not expressed. Normal residual cells within the sample may also act as an internal control. Laboratories with high workloads and well characterized McAb will acquire evidence of positive and negative controls in the range of clinical samples tested. • Directly conjugated McAb are recommended for flow cytometry when available. Optimal dilutions of McAb and second layer reagents should be defined in each laboratory. Titration should be carried out using known positive and negative controls. Whenever possible, the manufacturer's guidance on the use of McAb should be followed. If dilution is undertaken, relevant documentation of experimental evidence supporting the use of antibodies at the relevant dilution should be available. • Before setting up the techniques for immunophenotyping, the laboratory worker should spend a period of training in a department with experience of immunophenotyping. Competency should be documented. • There is no consensus on the cut-off point for considering a sample to be positive with a marker but commonly used criteria are: positivity in greater than 20% of leukaemic cells in acute leukaemias and positivity greater than 30% of leukaemic cells in chronic lymphoproliferative disorders. Although these cut-off points are arbitrary, they have been used by many groups. In certain circumstances, expression of a marker on the neoplastic or leukaemic cells is of interest rather than the expression on the total mononuclear cell fraction. This is the case with regard to the expression of CD5 in B cells from chronic lymphoid disorders and the expression of kappa and lambda on a minor B-cell population. • It is essential that laboratories participate in a external quality assurance (EQA) program. Flow cytometry can be applied to blood or bone marrow samples without the need to isolate mononuclear cells, thus simplifying laboratory procedures and making immunophenotyping of high risk, e.g. HIV-positive, samples safer. The samples should be treated with a hypotonic erythrocyte lysing solution with NH4Cl-based reagents which causes minimal selective loss of cell populations. Care has to be taken in the lysing procedure as prolonged exposure to lysing agents may cause changes in the forward and side light scatter (SSC) patterns resulting in selective loss of certain populations, whilst inadequate exposure to lysing agents leaves some intact red cells, an excess of debris and makes results inaccurate. Cytoplasmic and nuclear antigens can be detected by flow cytometry using a variety of commercially available permeabilization/fixation solutions. Not all commercially available reagents are equally reliable for detecting intracellular and, in particular, cytoplasmic antigens. Careful attention is needed to ensure that these solutions do not affect the light scatter patterns, that they are suitable for routine use and that they can be combined to permit simultaneous detection of membrane and cytoplasmic/nuclear antigens. It should be noted that there are some solutions suitable for testing whole blood or bone marrow specimens for nuclear antigens, e.g. terminal deoxynucleotidyl transferase (TdT), that are not reliable for cytoplasmic antigen detection. A study evaluating four commercially available solutions: FACS Permeabilization solution (BD Biosciences, San Jose, CA, USA), Fix and Perm cell permeabilization kit (Ander Grub, Vienna), Optilyse B-lysing solution (Immunotech, Marseille, France) and Permeafix (Ortho Diagnostic Systems, Raritan, NJ, USA) has shown that the four reagents have very minor effects on the light scatter pattern (Groeneveld et al. 1996) (Table 2). Fix and Perm and Permeafix gave similar results when detecting the nuclear enzyme TdT, cytoplasmic CD3 (cytCD3) and cytoplasmic immunoglobulin (cytIg) with results being comparable to those obtained by immunofluorescence microscopy (Pizzolo et al., 1995; Groeneveld et al., 1996). FACS Permeabilization solution and Optilyse were reliable for the detection of TdT but gave equivocal results for cytCD3 and cytIg. Fix and Perm seems to be superior to Permeafix and other reagents for the detection of myeloperoxidase (MPO) by flow cytometry (Groeneveld et al., 1996) (Table 2). The technique for detection of intracellular antigens by single immunostaining involves: • incubation of the specimens with the permeabilization/fixation solution for a variable period of time depending on the reagent (ranging from 10 min for Optilyse B and FACS Permeabilization solution to 40 min for Permeafix), • washing with a buffer composed of phosphate buffered saline (PBS) and bovine serum albumin (BSA) (PBS/BSA; pH: 7.3), • incubation for 10–15 min with the fluorochrome-conjugated McAb against the antigen to be investigated, washing again in PBS/BSA and reading on the flow cytometer. The time required to perform the test ranges from 40 min (FACS Permeabilization solution) to 80 min (Permeafix). For double immunostaining, e.g. detection of membrane and intracellular antigens, samples should first be processed using the standard technique to assess the expression of the surface marker. Ideally the membrane marker should be investigated using a fluorochrome-conjugated McAb involving only a single step incubation. Subsequently the samples are washed and processed further, as described earlier, to detect the intracellular marker. The use of phycoerythrin (PE)- instead of fluorescein (FITC)-conjugated McAb is recommended for the detection of antigens which are weakly expressed as PE gives a brighter signal and hence increased sensitivity. Control preparations using isotype-matched fluorochrome-conjugated mouse immunoglobulins for membrane and cytoplasmic markers should be run in parallel. Control samples should be treated in the same way as the test sample. As for the standard inmunophenotyping procedures, blocking of the Fc receptors should be performed when analysing isolated mononuclear cells but is not required when using whole blood or bone marrow specimens (General Haematology Task Force of the BCSH, 1994a). The detection of intracellular antigens can be crucial for the diagnosis of acute leukaemias and rare cases of lymphoid disorders such as those arising from plasma cells. This is because the most specific markers for the myeloid and lymphoid lineages are only detectable in the cytoplasm and/or are present earlier in the cytoplasm than on the cell surface, e.g. CD3 for the T-lymphoid lineage (Janossy, Coustan-Smith & Campana, 1989), CD79a for the B-lymphoid lineage (Buccheri et al., 1993) and antimyeloperoxidase (anti-MPO) for the myeloid lineage (Buccheri et al., 1992). In the past, intracellular staining could only be performed on fixed mononuclear cells spread on cytocentrifuge slides or on blood and bone marrow smears either using direct or indirect immunofluorescence and reading on a fluorescence microscope or applying an immunocytochemical technique (General Haematology Task Force of the BCSH, 1994a). Advances in sample preparation have permitted the estimation of the expression of intracellular and nuclear antigens by flow cytometry and thus have minimized subjective interpretation of the results and permitted a more accurate This technique incubation of the samples with a McAb and with the McAb for which cell to be with an In the step of the is up on a (SSC) This procedure the of and the exclusion of and is performed which the cells positive with the McAb et al., A number of be collected in to reliable results, particularly when the of cells such as cells et al., It should also be noted that there is a laboratories as to the number of collected from 10 to 10 If with is required it is to into account the of positivity of the antigen being tested in to a number of on cells is of for the estimation of stem cells in blood and bone marrow used for and et al., for analysis of the plasma cell in or other plasma cell for detection of rare and for the detection of of red cell is not required for routine Over the or four years new McAb that are for the diagnosis of acute leukaemias or chronic lymphoid disorders have available (Table Their in either or antibody panels is recommended for the routine diagnosis of acute leukaemias or chronic lymphoproliferative disorders. these antibodies are: • CD79a McAb (e.g. which an intracellular of the of the B-cell This is a in B which of the B-cell antigen and is highly specific for the B-lymphoid In addition to when tested by flow CD79a has a high as it is expressed from the of B-cell all the B-cell up to the plasma cell plasma cells from a proportion of cases of are The of acute leukaemias are while are negative in acute myeloid leukaemias (Buccheri et al., Although there have been two CD79a expression in lineage et al., et al., the of these is of the two et al., cases cells expressed CD79a two of these cases were also as biphenotypic acute leukaemia et al., et al., et al., The other a of CD79a expression in a of cases of using et al., of these positive cases were tested only with CD3 and only a were tested with analysis for the of the immunoglobulin was not carried in these two and the that some of the cases to leukaemias of lymphoid cells not to the B or lymphoid lineages could not be The of CD79a expression in the to • a McAb that a the for The is expressed in a of cell some of which are to the myeloid and in a minor proportion of which are in the of the and are to in into cells of myeloid or T-lymphoid lineage et al., study to samples from cases of acute leukaemia to be expressed in two of cases of of the while than of cases of were et al., of these cases either expressed other myeloid antigens, e.g. or and/or to The of for the myeloid lineage is to that of and greater than that of or & The for the of has also the of in the of acute leukaemias which are likely to acute myeloid leukaemias et al., is also an myeloid marker for the of biphenotypic acute et al., It is recommended that be in the of McAb for the routine analysis of cases of acute leukaemias. • a McAb that an of the B-cell and is specific to In the B-cell it is expressed than with a of cases of B lineage being et al., 1996). cases to and is of in the diagnosis of chronic lymphoproliferative disorders because of chronic leukaemia in which cells are negative or membrane weakly when a McAb is and other B-cell in which the cells or expression et al., et al., et al., • that detect which is not expressed in cells from cell and other B-cell disorders which the such as cells from cell leukaemia and from some B-cell and may weakly et al., 1996). of can be detected by flow as described by et al. with and when with for and and with the of by fluorescence in in detecting by or flow cytometry are not when cell is and immunological results by either technique are other e.g. for the detection of are in acute leukaemias. In the light of the new McAb and the use of a revised for the characterization of cases of acute leukaemia is recommended (Table and The is on the by the et al., with some a two step with the first of markers being to all cases of acute leukaemia a myeloid can be by and If with are into a clinical immunophenotyping is required for purposes. In other circumstances, immunophenotyping is if it is not to a diagnosis of on the of a and standard is essential in all cases of myeloid leukaemia leukaemias and in some cases of leukaemias and those with cells as the leukaemic A second is when to with specific 1). may also be performed in cases of acute leukaemia with evidence of myeloid if required by and/or for purposes. • and/or stem cell terminal deoxynucleotidyl transferase • or CD79a cytoplasm CD3 the myeloid was instead of as the a proportion of cases et al., et al., This will be used to the results of the first The first the diagnosis of the of cases of acute leukaemias and their into the major B and and, the diagnosis of the of biphenotypic acute leukaemias for whilst the second is at the of of such as those with or and at the exclusion or of a diagnosis of a The second • antibodies to light if cells have a B-cell phenotype and TdT is • and A if the lineage is not • and markers for if results with the first and second panels are negative and, in if the neoplastic cells only • this is a McAb that in the cytoplasm and is not specific for it is expressed in this and may be more than • against and cells. • which is positive in the of cytoplasmic with and therefore expression in may be of such a • which is positive in acute leukaemias. Although it is not specific for this as it may be positive in when the of the cells is negative and expression a diagnosis of • which is expressed a of cases of and most but is negative in and leukaemias and up to of leukaemias. in chronic lymphoproliferative disorders. are in the This is the same as the one recommended (General Haematology Task Force of the BCSH, with minor As for the diagnosis of acute it a first of markers applied to all cases and a second to be applied if by the results with the first • cell is recommended instead of CD3 because a proportion of T-cell positive or not with are CD3 • B-cell lineage markers which a marker (e.g. two B-cell markers surface immunoglobulin expression with estimation of the fluorescence and light using and reagents and of the fluorescence of membrane or • T-cell and B-cell The first will permit of B-cell from T-cell and, within the B-cell will or not the of is present (Table et al., This will be applied depending on the cell and the results with the first • and in cases with or lymphoid cells and/or when a diagnosis of cell leukaemia is on clinical Although of these markers are specific for cell when assessed they permit the of cell leukaemia from cases of and with et al., 1994b). • McAb if cell or B-cell leukaemia is • Cytoplasmic expression of immunoglobulin and light cytoplasmic CD79a and membrane for cases in which neoplastic cells were negative with the first of McAb and a and/or plasma cell is on clinical and/or • and in cases in which the first a T-cell • e.g. and may be investigated in cases with a diagnosis of or not cells T-cell specific markers such as CD3 or T-cell of these McAb is recommended in those cases in which cells are positive but expression of specific and B cell • a McAb that an in and cells may be in cases in which an of cells is as it may to from the of cases of other T-cell leukaemias that are e.g. T-cell leukaemia and are negative et al., 1996). In addition, this marker is expressed in a well defined of T-cell the T-cell which from cells the • of expression of the nuclear enzyme TdT is but is recommended in those cases shown to have a T-cell phenotype and or in to or a diagnosis of • with such as may be assessed in T-cell leukaemias and in which the cells expression of most T-cell and markers and/or when a diagnosis of positive leukaemia is McAb may be used if a diagnosis of T-cell is The expression of the antigens detected by the McAb in the membrane or can be assessed by flow cytometry and direct or indirect immunofluorescence techniques the guidelines outlined In cases in which is not available for flow an such as the can be carried out on bone marrow smears or thus making use of The and the of in the diagnosis of acute leukaemias and chronic lymphoproliferative disorders is in have the clinical and of antigen such an it has been to define antigen on normal cells, monitor changes in antigen expression that during & and et al., and also to leukaemia diagnosis and minimal residual et al., et al., et al., et al., & 1997). for antigen have been indirect immunofluorescence & et al., calibration of the flow with of fluorescence and the use of antibody with known et al., et al., and, more immunofluorescence et al., et al., et al., have flow cytometry to normal B-cell from their leukaemic by the high expression of and in In addition, et al. have that the combined of CD3 and expression is of in normal from cells. these that such is of in the neoplastic cell from normal cell and may additional with regard to the in lineage the of antigen techniques there are a number of technical and quality that have to be For two have for expression in acute leukaemia et al., et al., In of the interest in antigen UK for the technical with one It was that in incubation lysing and antibody can all have effects on the resulting antibody et al., For the for on normal was shown to be at and greater than 40 being whilst for the obtained was on the used and on single or analysis was Furthermore, when antigens are (e.g. CD3 and no are in their when using either or are for antigens that do not have a normal such as to the of when are et al. have also shown that monoclonal antibodies to a antigen (e.g. can when the same cells are by et al. the need for a standard to and for the at single staining using conjugated with all reagents at and incubation and lysing at should be as the using this a high of can be for and et al., In whilst antigen to have some clinical the techniques require greater they should be into the routine of with new antibodies that need or for which a has not been defined the (Table • the McAb and that the in cases of the and of The pattern of is specific and of for these et al., & either nuclear or cytoplasmic staining in with a number of nuclear in normal or the of other It has been that is more specific than & of the pattern of to be carried out by fluorescence microscopy or by et al., as it is not by flow • The McAb as being expressed in a of acute and characterized by of the et al., et al., et al., As this of acute leukaemias a it be to be to detect by a technique such as flow a study that expression of this McAb in is highly with with of and this McAb has been to a of cases with and expression of but without evidence of et al., Before routine use of this antibody can be a number of cases need to be investigated to • McAb that the of the and which are very specific for the T-lymphoid Although such antibodies have been available for they have not been applied in the diagnosis of acute leukaemias. have that cases with expression of than those which In addition, expression of evidence of T-lymphoid in and biphenotypic acute leukaemias. • expressed in a variety of cells from Although there is as to is a marker for with et al., et al., or a marker of et al., et al., et al., an could be made for in the of markers for chronic lymphoproliferative disorders. of biphenotypic acute leukaemias is relevant because of the of these and with cytogenetic markers et al., et al., et al., of biphenotypic acute leukaemia by using criteria is in to this of acute leukaemia of from and with expression of a marker more of The has recommended a et al., with some on the one et al., 1997). This is on the number and of of the antigens expressed by the leukaemic cells (Table acute leukaemia is defined when for the myeloid and one of the lymphoid lineages are two rare cases may or a B cell phenotype et al., 1997). The criteria for the of biphenotypic acute leukaemia here are those in the of et al.,