Quality assessment for immunophenotyping of leukaemias and lymphomas: the Dutch experience. The SIHON Study Group

Optimal number of reagents required to evaluate hematolymphoid neoplasias: results of an international consensus meeting

At the ISAC 2000 Congress, the Clinical Cytometry Society organized a meeting of international experts to reach consensus on the minimum number of antibodies required for a full evaluation of hematologic and lymphoid neoplasias. A questionnaire was distributed prior to the meeting to numerous experts from US and European institutions and 13 responses were received. At the meeting, 25 individuals, including most of those who returned responses, participated in the discussions and voted on the issues presented. In chronic lymphoproliferative disorders (CLD), 9 antibodies (anti-CD5, CD19, kappa, lambda, CD3, CD20, CD23, CD10, and CD45) were deemed essential for initial evaluation by 75% of the participants. There was near unanimity that additional markers (selected from CD22, FMC7, CD11c, CD103, CD38, CD25, CD79b and heavy chains for B-cell disorders, and CD4, CD7, CD8, CD2, CD56, CD16, TCRa/b, and TCRg/d for T-cell disorders) would be needed to fully characterize CLD, although not every marker would be useful in all cases. Tissue lymphomas were believed to be similar to CLD, needing a minimum of 12--16 markers. However, for some cases, CD30, bcl-2, TdT, CD71, CD1a, and CD34 were cited as useful by the participants. Markers mentioned for plasma cell disorders included kappa, lambda, CD38, CD45, CD56, CD19, CD20, CD138, and heavy chains. Of 17 voting participants, 16 agreed that between 5 to 8 markers would be essential reagents for plasma cell disorders. For acute leukemia (AL), 10 markers (CD10, CD19, CD13, CD33, CD34, CD45, CD7, CD14, CD3, and HLADR) were considered essential by 75% of participants for initial characterization of the leukemia lineage. Most (>75%) agreed that at least one more B (CD20, CD22, CD79a, IgM), T (CD1a, CD2, CD4, CD5, CD8), myeloid (CD11b, CD15, CD64, CD117, myeloperoxidase), erythroid (CD36, CD71, glycophorin A), and megakaryocytic (CD41, CD61) reagents should be included in the essential panel. However, there was no agreement as to which was optimal. Thus, approximately 13--15 of those reagents would be considered essential in all cases of AL, whereas others (CD16, CD56, CDw65, TdT, and cytoplasmic CD3) were mentioned as useful in some cases. Almost all voting participants believed that the appropriate number of markers for complete characterization of AL would average 20--24. The majority of the responders (11 of 13) indicated that fewer reagents could be used in monitoring or staging patients with previously characterized disease, but not all ventured a specific number of reagents. From the above results, we conclude that the phenotypic analysis of hematologic and lymphoid neoplasia requires a rather extensive panel of reagents. Supplementary reagents might even be necessary if they prove to become relevant for diagnostic purposes. Reducing the number of antibodies could significantly compromise the diagnostic accuracy, appropriate monitoring, or therapy of these disorders.

Laboratory practices in reporting flow cytometry phenotyping results for leukemia/lymphoma specimens: results of a survey

Laboratory directors who routinely utilize flow cytometry for at least part of their diagnostic evaluations in leukemias or lymphomas were surveyed by mail. The survey consisted of 12 questions about the flow cytometry procedures used by the laboratory in evaluating leukemias and lymphomas and on the format and content of their official report. It also requested an example of a typical leukemia/lymphoma report and solicited write-in comments about additional important aspects of using flow cytometry to evaluate leukemia and lymphomas not covered by the questionnaire. The goal of the survey, which was sponsored by the Clinical Cytometry Society (CCS), was to document what directors of flow cytometry laboratories currently consider to be the appropriate contents of a clinical leukemia/lymphoma phenotyping analysis and in what manner and detail they report such flow cytometry results to clinicians. The survey indicated that a large number of markers are routinely evaluated to phenotype leukemias (mean = 19) and lymphomas (mean = 16). Light scatter gating, using CD45/14 to monitor the gate selected, is currently employed by a 2:1 ratio over the next most population gating strategy (CD45 vs. 90 degrees LS). Peripheral blood, bone marrow, and lymphoid tissue constitute the majority of clinical specimens evaluated for leukemia and/or lymphoma. Two-color analysis, primarily for surface markers, is currently the standard method for flow cytometry measurements in routine diagnostic studies of leukemia and lymphoma. The official flow cytometry laboratory report is most commonly an individual-lab-generated, paper report form. A discussion of the potential benefits that might result from the development of improved computerized reporting software and from the increased use of antibody-defined, lineage gating is offered. A composite report format is presented that demonstrates the flow measurements and quality control data included in the best of the example clinical reports submitted as part of the survey and considered important by a majority of our survey respondents. The example report is intended to be a basis for further discussion within the flow cytometry community on whether minimum reporting standards for leukemia and/or lymphoma flow cytometry results can and should be developed.

Photomultiplier voltage setting: possible important source of variability in molecular equivalents of soluble fluorochrome (MESF) calculation?

We evaluated the effect of flow cytometer photomultiplier (PMT) voltage setting on molecular equivalents of soluble fluorochrome (MESF) calculation by using two different quantitative microbead calibrating kits as calibrating standards and a third one as an unknown testing sample of stable intensity of fluorescence. Based on the analysis of residual values derived from linear regression and on the determinations of the resolution index, we demonstrated that, on a FACScan instrument, the window of photomultiplier setting for fluorescein isothiocyanate (FITC) fluorescence determinations is 590-670 V. However, the best region was in the range of 610-650 V. Within the 590-670 PMT voltage region, the evaluation of the same testing sample at different PMT voltages, at 590 V, gave MESF percentage differences of 15% and 13% on the two lower unknown standards of intensity compared to the values obtained at the nearby 630 V setting. In the analyzed PMT region, logarithmic amplifier response was globally maintained, whereas evaluation of the same unknown testing sample in both linear (Lin) and logarithmic (Log) amplification demonstrated (at the higher intensities of fluorescence) Log-Lin minimal differences within the 610-650 PMT voltage range. According to our data, it can be stated that PMT voltages between 610 and 650 provide the best instrument performances and that PMT setting may be proposed as a source of variability in MESF calculation that deserves a careful evaluation in quality-control trials.

CD34+ leukemic cells assessed by different CD34 monoclonal antibodies

CD34 monoclonal antibodies (McAbs) are widely used to identify and isolate hemopoietic progenitors and to classify acute and chronic leukemias. We assessed the reactivity of 17 CD34 McAbs from the 5th International Workshop on Leukocyte Differentiation Antigens with a variety of cells types: normal bone marrow hemopoietic progenitors, 10 AML, 6 ALL, 11 CML. The reactivity for these McAbs was compared with that of reference CD34 McAbs (Q-Bend 10 and 8G12). For each cell population the % of McAb binding cells, the peak channel and the mean fluorescence intensity (MFI) of the positive cells was evaluated. The peak channel, the MFI and the number of positive cells varied significantly from case to case, depending on the McAb and the type of leukemia. According to the spectrum of reactivity three groups of McAbs were defined; however, 7 McAbs do not belong to any of these subgroups. These groups were not entirely in accordance with McAb classification based on enzyme cleavage that identified three epitopes of the CD34 molecule. Some reagents were found to be more specific for AML, other for ALL, CML or normal CD34+ cells. Normal bone marrow light density cells showed a significantly higher percentage of positive cells for 43A1 and MD34.2 McAbs compared to that documented for the remaining McAbs. AML cells showed the most variable pattern of expression for the CD34 McAbs. In leukemic samples, MESF (molecular equivalents of soluble fluorochrome) values ranged from 18,200 to 322,000 and the number of binding sites per cells was 5,000-81,000.(ABSTRACT TRUNCATED AT 250 WORDS)