Report on the 15th International Society of Blood Transfusion platelet immunology workshop.

BACKGROUND AND OBJECTIVES: The aim of the 15th ISBT Platelet Immunology Workshop was to evaluate the detection of free platelet-reactive autoantibodies from ITP patients by the use of a standardized MAIPA protocol, to compare sensitivity and specificity of antibody detection for anti-HPA-1a and serologically difficult-to-assess antibodies against HPA-3, to identify whether anti-HPA-1a titration results can be compared between laboratories, and to evaluate HPA genotyping methods. MATERIALS AND METHODS: Workshop materials were shipped from the organizing laboratory in Giessen, Germany. Thirty laboratories from 19 countries participated. RESULTS: Results for the detection of autoantibodies differed greatly between the laboratories and no consensus was reached for one of the two sera. Detection and titration of antibodies against HPA-1a, in contrast, gave largely congruent results. Serologically difficult-to-assess antibodies recognizing HPA-3a and HPA-3b were not detected by many laboratories. For genotyping, good agreement was achieved. CONCLUSIONS: Detection of HPA-1a antibodies, titration of anti-HPA-1a, and HPA genotyping are well performed in most participating laboratories. The workshop has identified two specific areas with room and need for improvement: the detection of autoantibodies and the detection of HPA-3 alloantibodies. Recommendations of the Working Party on techniques that can help to overcome these problems are desirable.

[Differential diagnosis and treatment of thrombocytopenia]. / Differenzialdiagnose und Differenzialtherapie der Thrombozytopenie.

Thrombocytopenia is usually acquired. The laboratory artefact of pseudothrombocytopenia should always be excluded. Bone marrow insufficiency with impaired platelet production results from infiltrating tumor cells or from a myelodsplastic syndrome. In patients with splenomegaly, platelets are trapped by the spleen. An increased platelet turnover is caused by activation of the clotting cascade, e.g. due to sepsis or malignancy. Platelet binding antibodies cause thrombocytopenia by increased platelet clearance. Important differential diagnoses in patients with severe thrombocytopenia are: acute leukemia, thrombotic thrombocytopenic purpura, autoimmune thrombocytopenia and drug-dependent thrombocytopenia. Multifactorial causes are thrombocytopenia associated with pregnancy, chronic alcohol abuse, and liver cirrhosis. Treatment should focus on the underlying disease. In regard to low platelet counts only clinical bleeding and not platelet count numbers should be treated.

Autologous blood donor screening indicated a lower prevalence of viral hepatitis in East vs West Germany: epidemiological benefit from established health resources.

Prevalence data concerning viral hepatitis and human immunodeficiency virus (HIV) in the general population are usually scarce. We aimed for a large cohort representative of the general population that required little funding. Autologous blood donors are relatively representative of the general population, and are tested for viral hepatitis and HIV in many countries. However, frequently these data are not captured for epidemiologic purposes. We analysed data from well over 35,000 autologous blood donors as recorded in 21 different transfusion centres for anti-hepatitis C virus (HCV), HBsAg and anti-HIV, as well as TPHA if available. We found a lower prevalence of hepatitis B virus and HCV in East vs West Germany, 0.2%vs 0.32% and 0.16%vs 0.32% respectively, which confirms earlier data in smaller cohorts, thus supporting the value of our approach. HIV was too rare to disclose significant differences, 0.01%vs 0.02%. TPHA was higher in East (0.34%) vs West Germany (0.29%) without significant differences. HCV was more frequent in women vs men. Transfusion institutes managing autologous blood donations should be used as a resource for epidemiological data relating to viral hepatitis and HIV, if such testing is performed routinely. This approach generates data relating to the general population with special emphasis on undiagnosed cases.

An in vitro system for the determination of individualized immunosuppression.

The risk of complications of immunosuppressive treatment in organ transplantation increases with the aggregate amount of immunosuppressive medication given to the patient. As the doses of immunosuppressive agents required to achieve comparable effects show considerable variability, methods to assess individual sensitivity toward immunosuppressive regimens are urgently needed. The aim of this study was to develop such an in vitro test system. As immunological model for allogeneic transplantation, individual pairs of recipient-derived lymphocytes and of donor-derived B lymphocytes mimicking HLA expression of cells in the transplanted organ were isolated and assessed in mixed-lymphocyte cultures (MLC). Alloreactivity was readily observed and MLC consisted of CD8(+) and CD4(+) T cells as well as CD56(+) natural killer cells. A proliferation assay to measure the response of individual MLC on the immunosuppression by cyclosporine (CsA) was developed. The concentrations of CsA leading to growth reductions by 50% (inhibitory concentration 50, IC(50)) were found between 110 and 220 ng/mL, which was near the trough whole blood levels for CsA. Accordingly, the IC(90) values (660 to 1760 ng/mL) were near the target values for peak whole blood levels. We believe that these data present a simple and potentially useful in vitro technology that allows for the prediction of individual responses to immunosuppressive therapeutic regimens.

Anti-HPA-1a in a case of post-transfusion purpura: binding to antigen-negative platelets detected by adsorption/elution.

Post-transfusion purpura (PTP) is a rare transfusion reaction almost exclusively observed in female patients. Affected patients develop severe immune-mediated thrombocytopenia in the course of a strong anamnestic alloimmune reaction against a platelet-specific antigen. The pathophysiology of thrombocytopenia has remained elusive. Immunological analysis in the HPA-1a-alloimmunized patient described in this report revealed an antibody with features considered typical of PTP: not only was anti-HPA-1a detectable in plasma, but it could also be eluted from the patients' (alloantigen negative) platelets, and anti-HPA-1a could be detected in eluates from both antigen positive and negative test platelets, which had been incubated in the patient's serum. This is in contrast to two sera with HPA-1a alloantibodies obtained from mothers of children with neonatal alloimmmune thrombocytopenia which were strictly HPA-1a specific. It is proposed that alloantibodies with HPA-1a-like specificity explain the patient's immune thrombocytopenia. The technique described in this report is proposed for further investigation, as it might be useful for discrimination of alloantibodes in PTP and alloantibodies of transfused thrombocytopenic patients.

Human platelet alloantigens.

Antibody formation against alloantigens of the human platelet membrane is responsible for clinical syndromes and transfusion related conditions as neonatal alloimmune thrombocytopenia (NAIT), post-transfusion purpura (PTP), platelet transfusion refractoriness (PTR) and passive alloimmune thrombocytopenia. Moreover, rare cases of alloimmune reactions involving platelets have been observed after transplantation of hematopoietic stem cells. Among alloantigens of the platelet membrane shared with other cells (type I alloantigens) are the glycoconjugates of the ABO system and class I human leukocyte antigen (HLA) antigens. Antibodies against these structures are responsible for PTR and for febrile nonhemolytic transfusion reactions. Antibodies against type II antigens (formerly termed "platelet specific antigens") have been observed in NAIT, PTP and passive alloimmune thrombocytopenia. ABH antigens have been identified on intrinsic platelet membrane glycoproteins. Moreover, it is now clear that HLA class I antigens are an integral part of the platelet membrane. The quantity of both HLA and ABH-antigen expression on the platelet membrane varies considerably. Single point mutations account for almost all platelet specific alloantigens, but most antigenic determinants seem to depend upon glycoprotein conformation: generally, platelet specific alloantibodies fail to recognize synthetic peptides encompassing the polymorphic residues. Restriction fragment polymorphism analysis and allele-specific PCR have been implemented for genotyping of platelet alloantigens in many laboratories. Antigen specific assays using monoclonal antibodies (MAIPA, immunobead assay) became de facto standard for diagnosis of platelet antibodies in serum/plasma samples. It can be expected that innovative techniques as human alloantibody fragments produced by phage display technique and the production of recombinant antigens will allow rapid and reliable phenotyping and antibody detection in the future.

Drug-induced and drug-dependent immune thrombocytopenias.

Thrombocytopenia is a frequent comorbid condition in many in hospital patients. In some patients, drugs are the cause of low platelet counts. While cytotoxic effects of anti-tumor therapy are the most frequent cause, immune mechanisms should also be considered. This review addresses thrombocytopenias in four groups. Heparin-dependent thrombocytopenia (HIT), by far the most frequent drug-induced immune-mediated type of thrombocytopenia, has a unique pathogenesis and clinical consequences. HIT is a clinicopathological syndrome in which antibodies mostly directed against a multimolecular complex of platelet factor 4 and heparin cause paradoxical thromboembolic complications. The mechanisms through which heparin can enhance thrombin generation are discussed and treatment alternatives for affected patients are presented in detail. It is of primary importance to recognize these patients as early as possible and to substitute heparin with a compatible anticoagulatory drug, such as hirudin, danaparoid or argatroban. Patients seem to benefit from therapeutic doses of alternative treatment rather than from low-dose prophylactic doses. With the increasing use of glycoprotein (GP) IIb/IIIa inhibitors in patients with acute coronary syndromes, thrombocytopenias are increasingly recognized as an adverse effect of these drugs. Up to 4% of treated patients are affected. Most important, pseudothrombocytopenia, a laboratory artefact, is as frequent as real drug-induced thrombocytopenia and must be excluded before changes in treatment are considered. The pathogenesis of these thrombocytopenias is still debated; an immune mechanism involving preformed antibodies is likely. However, since these antibodies are also detectable in a high percentage of normal controls and of patients not developing thrombocytopenia, their impact is still unclear. Patients with real thrombocytopenia are at an increased risk of bleeding; treatment consists of cessation of the GP IIb/IIIa inhibitor and platelet transfusions in cases of severe hemorrhage. Classic immune thrombocytopenia can be induced by some drugs, e.g. gold, which trigger anti-platelet antibodies indistinguishable from platelet autoantibodies found in autoimmune thrombocytopenia. Drug-induced and drug-dependent immune thrombocytopenia is induced by antibodies recognizing an epitope on platelet GP formed after binding of a drug to a platelet glycoprotein. Still unresolved is whether antibody binding is the consequence of a conformational change of the antigen, the antibody, or both. These antibodies typically react with monomorphic epitopes on platelet GP, but only in the presence of the drug or a metabolite. Although several platelet GP have been identified as antibody target (GPIb/IX, GPV, GP IIb/IIIa), antibodies in an individual patient are highly specific for a single GP. Clinically, these patients present with very low platelet counts and acute, sometimes severe, hemorrhage. Treatment is restricted to withdrawal of the drug and symptomatic treatment of bleeding.

Platelet alloantibodies in transfused patients.

BACKGROUND: Patients receiving cellular blood components may form HLA antibodies and platelet-specific alloantibodies. STUDY DESIGN AND METHODS: Serum samples from a cohort of 252 patients with hematologic or oncologic diseases who are receiving cellular blood components were studied for platelet-reactive antibodies. Specificity of platelet alloantibodies was determined with a panel of typed platelets RESULTS: Platelet-reactive antibodies were detected in the sera of 113 patients (44.8% of 252), HLA antibodies in the sera of 108 (42.9%), and platelet-specific antibodies in the sera of 20 (8%). The following platelet-specific antibodies were identified: anti-HPA-5b (n = 10), anti-HPA-1b (n = 4), anti-HPA-5a (n = 2), anti-HPA-1a (n = 1), anti-HPA-2b (n = 1), anti-HPA-1b+5b (n = 1), and anti-HPA-1b+2b (n = 1). Fifteen sera from the 108 patients with anti-HLA (13.9%) contained additional platelet-specific alloantibodies, while in 5 sera, platelet-specific alloantibodies only were detected: anti-HPA-5b (n = 4) and anti-HPA-1a (n = 1). Of the 108 sera with HLA antibodies, 29 (26.9%) showed discordant results when studied with the lymphocytotoxicity test and the glycoprotein-specific immunoassay. Ten sera contained panreactive antibodies against platelet glycoproteins (GP) IIb/IIIa, GPIa/IIa, and/or GPIb/IX. Alloimmunization occurred in 58.3 percent of female patients with previous pregnancies, but in only 23.3 percent of those without previous pregnancies (p = 0.0049). CONCLUSION: Platelet alloantibody specificities in transfused patients (predominantly anti-HPA-5b and -1b with antigen frequencies <30% among whites) differ significantly from those observed in patients with neonatal alloimmune thrombocytopenia or posttransfusion purpura, in whom anti-HPA-1a (antigen frequency >95%) is the most prevalent specificity. HLA antibody detection yields discordant results when the lymphocytotoxicity assay and a glycoprotein-specific immunoglobulin-binding assay are used.

Single amino acid substitution in human platelet glycoprotein Ibbeta is responsible for the formation of the platelet-specific alloantigen Iy(a).

We recently described a new low-frequency platelet alloantigen on the human platelet glycoprotein (GP) Ib-IX complex, termed Iy(a), which was implicated in a severe case of neonatal alloimmune thrombocytopenia. Immunoprecipitation studies with trypsin-treated platelets indicated that the Iy(a) alloantigenic determinants are formed by the membrane-associated remnant moiety of GP Ibalpha (GP Ibalpha(r)) together with GP Ibbeta and GP IX. To elucidate the molecular basis underlying the Iy(a) alloantigen, we amplified GPIbalpha(r), GPIbbeta, and GPIX genes by polymerase chain reaction (PCR). Nucleotide-sequence analysis of these 3 genes showed a G to A transition at position 141 on GPIbbeta gene in a subject positive for Iy(a). This transition resulted in a Gly(15)Glu dimorphism on the N-terminal domain of GPIbbeta. This finding was confirmed by genotyping analysis of 6 Iy(a)-positive subjects by restriction fragment length polymorphism (RFLP) studies using NarI endonuclease. In 300 randomly selected healthy blood donors, one Iy(a)-positive individual was found. Phenotypes determined by monoclonal antibody-specific immobilization of platelet antigens assay and genotypes determined by RFLP were identical in this population. Analysis of Iy(a)-positive platelets showed that the point mutation affected neither the degree of surface expression nor the function of the GP Ibalpha-GP Ibbeta-IX complex on the platelet surface. Transient expression of the GP Ib-IX complex in CHO cells using wild-type GP Ibbeta (Gly(15)) or mutant GP Ibbeta (Glu(15)) allowed us to demonstrate that this single amino acid substitution is sufficient to induce Iy(a) epitope(s). (Blood. 2000;95:1849-1855)

A point mutation Thr(799)Met on the alpha(2) integrin leads to the formation of new human platelet alloantigen Sit(a) and affects collagen-induced aggregation.

A new platelet-specific alloantigen, termed Sit(a), was identified in a severe case of neonatal alloimmune thrombocytopenia. The Sit(a) alloantigen is of low frequency (1/400) in the German population. Immunochemical studies demonstrated that the Sit(a) epitopes reside on platelet glycoprotein (GP) Ia. Nucleotide sequence analysis of GPIa cDNA derived from Sit(a)-positive platelets showed C(2531)-->T(2531) point mutation, resulting in Thr(799)Met dimorphism. Analysis of genomic DNA from 22 Sit(a)-negative normal individuals showed that the Thr(799) is encoded by ACG(2532) (90.9%) or ACA(2532) (9.1%). To establish a DNA typing technique, we elucidated the organization of the GPIa gene adjacent to the polymorphic bases. The introns (421 bp and 1.2 kb) encompass a 142-bp exon with the 2 polymorphic bases 2531 and 2532. Polymerase chain reaction-restriction fragment length polymorphism analysis on DNA derived from 100 donors using the restriction enzyme Mae III showed that the Met(799) form of GPIa is restricted to Sit(a) (+) phenotype. Analysis of stable Chinese hamster ovary transfectants expressing allele-specific recombinant forms of GPIa showed that anti-Sit(a) exclusively reacted with the Glu(505)Met(799), but not with the Glu(505)Thr(799) and the Lys(505)Thr(799) isoforms. In contrast, anti-Br(a) (HPA-5b) only recognized the Lys(505)Thr(799) form, whereas anti-Br(b) (HPA-5a) reacted with both Glu(505)Thr(799) and Glu(505)Met(799) isoforms. These results demonstrated that the Met(799) is responsible for formation of the Sit(a) alloantigenic determinants, whereas amino acid 505 (Lys or Glu) specifically controls the expression of Br(a) and Br(b) epitopes, respectively. Platelet aggregation responses of Sit(a) (+) individuals were diminished in response to collagen, indicating that the Thr(799)Met mutation affects the function of the GPIa/IIa complex.

Clinical aspects and typing of platelet alloantigens.

Platelet alloantigens can induce the formation of corresponding alloantibodies when exposed to phenotypically negative individuals. These antibodies are responsible for fetal and neonatal alloimmune thrombocytopenia, posttransfusion purpura, passive alloimmune thrombocytopenia and transplantation-associated thrombocytopenia and may contribute to platelet transfusion refractoriness together with HLA antibodies. Besides antibody detection laboratory diagnosis of the clinical syndromes requires alloantigen typing. Furthermore, typed platelet donors are a prerequisite for effective platelet transfusion therapy. Different techniques for phenotyping are well established and easy to perform but they rely on the availability of antisera. Since the molecular genetic background of the clinically most relevant alloantigens has been elucidated during the last years various genotyping methods have been applied to the platelet membrane polymorphisms and thus facilitated widespread platelet alloantigen typing. Generation of antibodies from phage display libraries and of lymphoblastoid cell lines from donors with all genetic variants will allow further developments.