Report
Section 1B: Rh flow cytometryCoordinatorˈs report.Rhesus index and antigen density: an analysis of the reproducibility of flow cytometric determination

https://doi.org/10.1016/S1246-7820(01)00213-0Get rights and content

Abstract

Fifty-seven IgG monoclonal anti-D antibodies were evaluated in the Rh flow cytometry section, in which 12 laboratories participated. Staining protocols and a fluorescein (FITC)-conjugated Fab fragment goat anti-human IgG (H+ L) as a secondary antibody were recommended but not mandatory. A CcDEe red blood cell (RBC) sample that was shown to be homozygous for RHD by molecular methods was supplied and used as internal ‘standard RBC’ throughout all experiments. An RBC panel comprising two partial D and four weak D types was supplied as well. The use of standard RBC reduced the variability of the data among the laboratories and allowed the conversion of fluorescence data into epitope densities, which were compounded in an antigen density (antigen D per RBC). The highest antigen density was determined for DVI type III, followed by DVII and weak D type 3; the lowest antigen density were determined for weak D type 1 and type 2. Nine of the 12 participating laboratories discriminated three groups of aberrant RhD that had similar Rhesus indices (RI): D category VI with RI = 0; weak D type 2 and type 3 with an high RI; and D category VII and weak D type 1 with an intermediate RI. The antigen densities and the Rhesus indices obtained correlated well among the laboratories of this Workshop section despite different staining protocols, secondary antibodies and instrumentation.

Introduction

This is the second Rh flow cytometry section to be an integral part of this series of International Workshops and a follow-up to the previous Rh flow cytometry section of the 3rd Workshop in Nantes 1996 1, 2. The aim of the current section was to evaluate several technical features and objectives for their utility in analyzing aberrant RhD phenotypes (table I).

The variability of flow cytometry results among laboratories was well recognized 1, 4 and an obstacle for the broader use of flow cytometry in the characterization of the molecular polymorphism of both Rhesus proteins 5, 6, 7. Part of this variation may be attributed to differences in staining protocols and reagents 〚1〛. In the previous Workshop section 1996 a standard protocol, fluorescent beads and a secondary antibody were mandatory and evaluated to overcome these problems. The supplied secondary antibody was an Fab fragment anti-human IgG (H+ L) and was anticipated to reduce the amount of agglutination produced when staining sensitized red blood cells (RBC). Fluorescent beads were supplied, too, and used for calibration in 1996.

In the current Workshop section, controls and data evaluation differed from the previous section (table II). Less emphasis was put on protocol and reagent standardization. A secondary antibody was recommended but not supplied; no fluorescent beads were used. A staining protocol as published previously 8, 9 was recommended; however, the participants could apply any staining procedure including their ‘in-house’ or the 3rd Workshop’s staining protocol. Specific instructions for setting up analysis regions were given; however, as a major difference to the previous Workshop section, the analysis regions should include all events. No attempts were made to restrict the analysis region to ‘positive events’. To evaluate the utility of a different layer of standardization, ‘standard RBC’ were provided and to be used in all tests 〚10〛. The calculation of epitope densities was based on the relative fluorescence detected with ‘test RBC’ compared to the ‘standard RBC’ by applying a published algorithm 〚3〛 for which a spreadsheet calculation program was provided.

IgM monoclonal antibodies were generally spared from analysis to avoid problems caused by direct agglutination 〚9, (see figure 2d)〛. There were 66 IgG anti-Rh monoclonal antibodies available in at least one laboratory, of which 60 were anti-D representing 56 different anti-D clones. The testers in the 12 participating laboratories of this Workshop section elected to evaluate between 22 to 55 different anti-D, including two IgM anti-D (BS223 and BS225) tested in one lab. The aggregate analysis of their results is presented in this coordinator’s report. The complete set of original data is available online (see Methods). In addition, seven groups of testers detailed their individual results, which were presented as separate reports 11, 12, 13, 14, 15, 16, 17. Kumpel and Jackson 〚14〛 provided a detailed study with discussion of the literature on the influence of the sensitization procedure on the quantification of IgG bound to RBC.

Section snippets

Testers and antibodies

As in the Nantes 1996 Workshop section, 12 laboratories participated in the flow cytometry evaluation (table III). The monoclonal antibodies were supplied as culture supernatant, most of which contained 0.1% sodium azide as preservative. Although it might have been feasible to use some of the monoclonal antibodies in diluted form (particularly when evaluating RBC with strong antigens), results were to be provided for the undiluted antibody to ensure that near saturating conditions were used.

Utility of using ‘standard RBC’ as an internal standard

Twelve antibodies were used by all laboratories. These antibodies were tested with the six RBC samples of the cell panel. Whilst the results are expected to differ depending on anti-D and RBC sample, for each pair of anti-D/RBC the result among the 12 laboratories should be identical under ideal conditions. Then, the coefficient of variation (CV) of the results among the 12 laboratories would be 0. Larger CV implies inter-laboratory variability, which of course always occurs and is of practical

Conclusion

The testers in the 12 participating laboratories of this Workshop section evaluated between 22 to 55 different anti-D in flow cytometry with a RBC panel comprising two partial D and four weak D types. The use of standard RBC reduced the variability of the data among the laboratories and allowed the conversion of fluorescence data into epitope densities represented as antigen density (antigen D per RBC).

The panels of anti-D used had little effect on the determination of antigen density. However,

Acknowledgements

This study was supported by the DRK-Blutspendedienst Baden-Württemberg, Stuttgart; the Universitätsklinikum Ulm (Forschungsförderungsprojekte P.531); and the Deutsche Gesellschaft für Transfusionsmedizin und Immunhämatologie (DGTI/fle/00-01).

Several parts of this Workshop section were based on the work of the previous Rh flow cytometry section of the 3rd Workshop in Nantes 1996, which was coordinated by Dr. Anne Fletcher. The coordinator acknowledges the expert technical assistance of Marianne

References (19)

There are more references available in the full text version of this article.

Cited by (54)

  • Correlation among automated scores of agglutination, antigen density by flow cytometry and genetics of D variants

    2019, Transfusion and Apheresis Science
    Citation Excerpt :

    Several studies have already quantified and characterized D variants antigen density. [3,6,15,18–21] Although flow cytometry is the most recommended technique for antigen density determination and widely applied a variability in its results is still observed depending on the antisera, protocol and standard RBC used [15,22,23]. In our study, we performed the flow cytometry test with a single monoclonal anti-D due to the limited access to licensed IgG anti-D antisera in Brazil.

  • RHD*weak D type 38: A family study

    2016, Revista Brasileira de Hematologia e Hemoterapia
  • Relevance of RH variants in transfusion of sickle cell patients

    2011, Transfusion Clinique et Biologique
    Citation Excerpt :

    There is a consensus to classify the variant based on the predicted localization of the substitution. The Rhesus index determining the antigen density can help also to decide [19]. Schematically, when localized at the outer surface of the membrane, the variant is considered partial, when localized in the intramembrane or in the intracellular domain, the variant is considered only weak.

View all citing articles on Scopus
1

WAF and FFW contributed equally to the work. The other authors are listed in alphabetical order and were the investigators of the participating laboratories.

View full text