Report on the 13th International Society of Blood Transfusion Platelet Immunology Workshop.

BACKGROUND AND OBJECTIVES: The aim of the 13th International Society of Blood Transfusion Platelet Immunology Workshop was to compare the sensitivity and specificity of the in-house method for the detection of human platelet antigen (HPA) antibodies currently used in participating laboratories with a modified rapid protocol for the monoclonal antibody (mAb) immobilization of platelet antigen (MR-MAIPA) assay. MATERIALS AND METHODS: Twenty-eight laboratories from 15 countries participated. A set of four freeze-dried minimum potency reference reagents with known single-specificity HPA antibodies were supplied for testing by titration with both assays and two coded freeze-dried plasma samples were provided for antibody specificity testing. Critical reagents and materials for the MR-MAIPA were provided including lyophilized panel platelets and five capture mAbs. RESULTS: Titration of the reference standards showed that the sensitivity of the MR-MAIPA was the same as the in-house methods. The proposed replacement anti-HPA-1a reference reagent 05/106 gave results that did not differ significantly from the current reference reagent 93/710. The results with the two blinded samples showed that in the first sample, 27 out of the 28 laboratories were able to correctly identify the anti-HPA-1a present when using their respective in-house methods, but only 23 correctly identified the antibody when using the rapid MAIPA method. The results from the second sample, which contained multispecificities, showed that only 50% of the participants correctly identified all five antibodies present using their in-house method. The results for the rapid MAIPA were lower, with only 32% identifying all specificities. The variability in the reconstitution of the freeze-dried platelets may have been one of the contributing factors to the poorer results. CONCLUSIONS: The sensitivity of the MR-MAIPA compared favourably with that of the in-house methods. Most laboratories were able to identify anti-HPA-1a alone in Sample 1 but more than half of the participants were not able to correctly assign the specificity of all HPA antibodies present in the second sample. The usefulness of the panel of freeze-dried platelets varied considerably between laboratories.

[Intravenous human immunoglobulin treatment for neonatal alloimmune thrombocytopenia due to anti-HPA-5b with severe previous history]. / Traitement par immunoglobulines intraveineuses dans 2 cas d'allo-immunisation foetomaternelle plaquettaire HPA-5b avec antécédents sévères.

Fetal and neonatal alloimmune thrombocytopenia due to mothers' anti-HPA-5b alloimmunization has generally a milder clinical presentation compared to anti-HPA-1a alloimmunization. Nevertheless, a case with infant's death probably due to intracranial haemorrhage has been reported. However, if platelet-specific alloimmunized mothers with prior fetal or neonate injury receive intravenous immunoglobulins during pregnancy, thrombocytopenia in heterozygous fetus and neonate may be prevented. Here are reported 2 cases of anti-HPA-5b fetal-maternal alloimmunization, one with prior fetal death, the other with prior severe fetal intracranial haemorrhage, which were successfully treated with intraveinous immunoglobulins alone during a second pregnancy with HPA-5b incompatibility.

New method for DNA microarrays development: applied to human platelet antigens polymorphisms.

DNA microarrays are a powerful experimental tool for the detection of specific genomic sequences and are invaluable to a broad array of applications: clinical diagnosis, personalized medicine, drug research and development, gene therapy, food control technologies, and environmental sciences. Alloimmunization to human platelet antigens (HPAs) is commonly responsible for neonatal alloimmune thrombocytopenia, post-transfusional purpura and platelet transfusion refractoriness. Using DNA microarrays, we developed a diagnosis to type the biallelic HPA-1 platelet group. The region for the human genomic DNA sequence that contains the polymorphism responsible for HPA-1 alleles was amplified by polymerase chain reaction (PCR). The expected DNA fragments were hybridized on DNA microarrays, and the data were analyzed using specially developed software. Our initial results show that the two HPA-1 antigens polymorphisms containing a single base difference were detected using DNA microarrays.

HLA antibodies and neonatal alloimmune thrombocytopenia.

A female baby with a severe thrombocytopenia at 18 x 10(9)/l was born to a 29-year-old (gestation 2/partum 2) mother. Scattered petechiae were present on her legs, arms, chest and face, but there was no bleeding, infection, fever or hepatosplenomegaly. A platelet antibody screening immunocapture test was positive, which was performed on the mother's serum 3, 12 and 38 days after delivery, but no platelet-specific antibodies were found by the monoclonal-antibody-specific immobilization of platelet antigen assay. The baby's platelets and lymphocytes and the father's platelets reacted strongly with the HLA antibodies present in the mother's serum. The neonate was treated with intravenous human immunoglobulin (Tegeline), 1 g/kg per day) 1, 2 and 3 days after delivery. The platelet count rose from 18 x 10(9)/l on day 0 to 37 x 10(9)/l on day 3 and to 227 x 10(9)/l on day 12. No platelet transfusion was needed. Several factors which developed hereafter lead us to think that this neonatal alloimmune thrombocytopenia is due to the transplacental passage of maternal HLA antibodies to the baby.

A rapid one-stage whole-blood HPA-1a phenotyping assay using a recombinant monoclonal IgG1 anti-HPA-1a.

Severe neonatal alloimmune thrombocytopenia and patients with HPA-1a-specific antibodies require transfusion of HPA-1a-negative platelets. Identifying HPA-1a-negative donors requires simple and reliable typing methods. Most existing techniques use polyclonal antibodies, are time consuming and involve platelet isolation. We have used a horseradish peroxidase (HRP)-conjugated recombinant IgG1 anti-HPA-1a (CAMTRAN007) to develop a rapid and reliable enzyme-linked immunosorbent assay (ELISA), which eliminates sample preparation and reduces the incubation and wash steps associated with traditional sandwich ELISAs. The assay uses simultaneous incubation of the monoclonal antibody RFGP56 to capture GPIIbIIIa from whole blood and the recombinant IgG1 antibody to detect captured HPA-1a antigen. It allows 96 samples to be typed in less than 1 h and can be used on stored samples. Initial testing of 85 samples of known HPA-1a genotype demonstrated that HPA-1a-negative samples had OD values of < 0.266, whereas HPA-1a-positive samples had OD values of > 0.6. Testing of 1862 random donor samples in two blood centres confirmed these OD cut-off values and identified 45 HPA-1a-negative samples (2.4%), all except one giving OD values of < 0.2. The remaining HPA-1a-negative sample had an OD value of 0.303. The HPA-1a status on all the negative samples and an equivalent number of randomly selected positive samples was confirmed by flow cytometry and polymerase chain reaction with sequence-specific primers (PCR- SSP).

Human platelet antigen frequencies of platelet donors in the French population determined by polymerase chain reaction with sequence-specific primers.

To prevent human platelet alloimmunization, Blood Transfusion Centres have to develop a strategy close to the erythrocytes' one. The first step of this strategy is to perform the HPA typing of donors with an accurate method. We applied the PCR-SSP to type 800 platelet donors in the HPA-1 and HPA-5 systems and 350 in the HPA-2 and HPA-3 ones. This study reports the human platelet antigen frequencies of four platelet-specific alloantigen systems in the French population. The results are quite similar to those currently published for Caucasian population frequencies. Low prevalences are observed for the HPA-1b, (2%), HPA-2b (0.6%) and HPA-5b (2%) groups. Furthermore, this study confirms the need to type donors and recipients in the HPA-1 system at least, in case of post-transfusion pupura and platelet refractoriness to platelet transfusion therapy.

[Contribution of dual CD13/CD14 markers in combination with CD34 for the collection of peripheral hematopoietic stem cells]. / Apport du double marquage CD13/CD14 associé au marquage CD34 pour le prélèvement des cellules souches hematopoïétiques périphériques.

We evaluated the reliability of a flow cytometry technique for counting mononuclear cells (MNCs) in cytapheresis products. Eighty freshly-prepared samples of peripheral stem cells were studied using a dual immunolabeling technique with antibodies to CD13/CD14, and were also labeled with anti-CD34. Results of this immunophenotype determination were compared to those of the conventional method for counting MNCs under the microscope. Dual CD13/CD14 labeling was found to be a simple and reliable method for counting MNCs in the presence of immature and stimulated cells. When used in combination with CD34 labeling, the dual immunolabeling method helped improve the evaluation of the quality of peripheral stem cell grafts.

Expression of a NK cell-restricted epitope on decidual large granular lymphocytes.

Decidual large granular lymphocytes (DLGL) are the most abundant lymphoid cell type found in the first trimester maternal decidua. The function of DLGL remains controversial, although freshly isolated DLGL have been shown to exert a weak NK activity. We report here the phenotypic characterization of two DLGL subpopulations by immunofluorescence, using mAb against CD56, PEN5 as well as adhesion molecules potentially involved in cell-cell contact between DLGL and trophoblasts. DLGL are CD56bright and express the CD2, CD11a, CD18, CD38 and CD50 molecules, dimly the CD54 molecule, and poorly the CD102 and CD69 molecules. A strong expression of the polysialylated form of N-CAM (or CD56) was also observed on the surface of DLGL. Finally, 40% of CD56bright DLGL cells express the PEN5 epitope, which is selectively expressed on CD56dim cells NK in the peripheral blood. No other phenotypic difference was detected between the CD56brightPEN5+ and the CD56brightPEN5- DLGL populations. Our results show that DLGL are heterogeneous and suggest that the CD56brightPEN5+ DLGL subset belongs to the classical NK cell lineage.

Strong increase in the percentage of the CD8bright+ CD28- T-cells and delayed engraftment associated with cyclosporine-induced autologous GVHD.

Four children with acute lymphoblastic leukaemia had autologous bone marrow (BM) or peripheral stem cell (PSC) transplantation with low dose of cyclosporine (CsA, img/kg/d i.v. during the first 28 d) to induce an autologous GVHD (auto-GVHD). Two children did not have clinical auto-GVHD and they relapsed 3 and 4 months after treatment. The 2 other children had clinical signs of auto-GVHD (grade I and grade II); they both are in complete remission but after a first normal haematological recovery they had a prolonged period of aplasia until month 9 for 1 patient and still persistent at month 7 in the other case. We studied lymphocyte subsets reconstitution after transplantation in these patients. All patients had an important decrease in the CD4/CD8 ratio related both to a strong decrease in the CD4+ cells and a strong increase in the CD8+ cells. Most of the CD8+ cells were of the CD8bright+ CD28- phenotype. These CD8bright+ CD28- T-cells represented from 33% to 68% of the total lymphocytes. We discuss the role of these cells after autologous transplantation with CsA, and wonder if these cells could mediate cytotoxicity. In conclusion, among 4 children who received autologous BM or PBC transplantation with low dose of CsA, we observed a complete remission after an auto-GVHD and a prolonged period of aplasia in 2 patients and a relapse of leukaemia in 2 other patients. All these 4 patients had an increase in the CD8bright+ CD28- T lymphocytes.

Allogenic BMT in children: differential lymphocyte subset reconstitution according to the occurrence of acute GVHD.

To acquire some biological markers associated with the occurrence of acute graft-versus-host disease (aGVHD) after allogeneic bone marrow transplant (BMT) in children, we have studied the lymphocyte subset reconstitution and the percentage of peripheral blood mononuclear cells bearing HLA-DR and HLA-DQ class II molecules. This study included 37 allogeneic BMT: either with (n = 17) or without (n = 20) aGVHD. Within 2 months after transplantation, we observed that patients with aGVHD had a unique mononuclear cell profile characterized by (i) a significant increase in the percentages of CD8bright+CD28- T cells (P = 0.05) and CD3+ T cells (P = 0.001), (ii) an important decrease in the percentage of CD56+ cells (P = 0.0001), and (iii) a decrease in the percentages of HLA-DQ+ and HLA-DR+ monocytes (P = 0.001) and HLA-DQ+ T lymphocytes (P = 0.0001), in comparison with patients without aGVHD. Moreover, statistical studies indicate that there was a positive correlation between CD8bright+CD28- and CD3+ T cells, whereas CD3+ T cells were negatively correlated to CD56+ cells. We did not find any statistical correlation between the percentages of HLA-DQ+ or HLA-DR+ cells and the percentages of these lymphocyte subsets. Therefore, in this study done in children, we suggest that patients with (i) less than 20% of DQ+ monocytes, (ii) less than 25% of CD56+ lymphocytes, and (iii) an enhanced percentage of CD8bright+CD28- T cells are strongly associated with aGVHD. Unfortunately, these biological markers of a GVHD may not precede the clinical manifestations of the disease.