Platelet refractoriness due to combined anti ... - Wiley Online Library

correspondence

Platelet refractoriness due to combined anti-HLA and anti-HPA5a alloantibodies: clinical management during myeloablative allogeneic HSCT and development of a quantitative MAIPA assay

Alloimmunization and consecutive platelet transfusion refractoriness due to multiple antibodies can represent a major obstacle or even a contraindication to myeloablative allogeneic haematopoietic stem cell transplantation (HSCT) when no compatible platelet donors are available. Logistics may be complicated when multiple alloantibodies, including those to high-frequency antigens such as human platelet antigen (HPA)-5a, are present. Only 1–2% of the Caucasian population have been reported to be HPA-5bb homozygous (Santoso et al, 1993). To date, anti-HPA-5a alloimmunization has been reported in neonatal alloimmune thrombocytopenia (MuellerEckhardt et al, 1989; Davoren et al, 2004) and post-transfusion purpura (Anolik et al, 2001) but has only rarely been described in adults treated for haematological malignancy (Kiefel et al, 2001). We report the case of a 50-year-old woman with acute myeloid leukaemia who became alloimmunized to multiple human leucocyte antigen (HLA)-antigens and to HPA-5a during cytarabine-based induction chemotherapy. Platelet transfusion refractoriness, including to 3/4 or 4/4 HLA classI matched products, occurred when the patient developed bilateral pneumonia with polymicrobial sepsis while in aplasia. Furthermore, diffuse gastrointestinal bleeding with haemorrhagic shock because of severe mucositis and refractory thrombocytopenia developed. A total of 23 single donor platelet products were administered. Anti-HLA class I, antiHLA class II and anti-HPA-5a alloantibodies were detected; the patient was genotyped as HPA-5bb. The anti-HPA-5a antibody was considered relevant because of persisting refractoriness including to HLA-matched HPA-5-mismatched products. The patient recovered completely from the complications, and achieved a first complete remission. Myeloablative allogeneic HSCT from her HLA-identical brother (genotyped HPA5ab) was planned. To monitor the anti-HPA-5a alloantibody during treatment, a quantitative method based on the monoclonal antibodyspecific immobilization of platelet antigens (MAIPA) procedure was developed (Bertrand et al, 2005, 2006). As there is no currently available international standard anti-HPA-5a serum, a pool of six sera was prepared and served as reference serum samples for quantification. They were collected from mothers immunized against HPA-5a fetal platelet alloantigen during

pregnancy, reacting very strongly in the MAIPA technique. An assigned potency of 1000 arbitrary units (AU)/ml was attributed to the pool. As the level of glycoprotein (GP)IaIIa expression on platelets is lower than GPIIbIIIa carrying the HPA-1 antigenic system, a higher quantity of platelets per test was used. Sixty million HPA-5aa platelets from healthy donors were incubated with 15 ll P16 monoclonal antibody (International Blood Group Reference Laboratory, Bristol, UK) and 40 ll of serum initially diluted from 1 to 1:128 (eight dilutions tested per sample). Optical density was measured at 495 nm (Multiskan RC, Life Sciences, Cergy, France). A standard curve was constructed with eight dilutions of the pooled reference serum, and gave the best correlation with a 4-parameter logistic regression (R2 ¼ 0Æ9989, Fig 1A). Quantification was performed using the Biolise software (Labsystems, Helsinki, Finland). In preparation for the transplant, HLA- and HPA-5compatible platelet donors were selected. Cross-match analyses identified a few compatible available donors. Because of the high-risk situation for refractory bleeding events, the patient was submitted to two thrombaphereses (TcPh) immediately prior to conditioning for cryopreservation of autologous platelets as backup (Valeri & Ragno, 2006). This procedure artificially induced a thrombocytopenia due to peripheral consumption, resulting in stimulated thrombopoiesis. The conditioning regimen consisted of cyclophosphamide/busulfan. Immunadsorption for reduction of circulating antibodies was attempted by transfusion of HLA- and HPA-incompatible platelet products on days )4 to )1 and a single plasmapheresis (PPh) on day )1. HSCT was performed on day 0. The posttransplant course was uneventful, platelet counts were maintained at levels >20 · 109/l with two compatible transfusions only (Fig 1B). The backup autologous platelet products were not used. The patient remains now, 1 year post-transplant, in complete remission. We believe that a multi-step approach in the management of the described patient was successful and enabled the patient to undergo curative myeloablative allogeneic HSCT. First, HLAand HPA-compatible platelet donors were identified within the registries of many blood donation centres in several countries. Because of the low frequency of the HPA-5bb genotype, only few donors were available. No donors identical for HLA class I

ª 2007 The Authors Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 139, 159–167

Correspondence (A)

(B)

Fig 1. Antibody concentrations and platelet counts in peripheral blood of the patient during treatment. (A) Validation of the monoclonal antibody-specific immobilization of platelet antigens method, standard curve. (B) Peripheral blood platelet counts (d) and antiHPA-5a antibody concentration (u). Thrombaphereses (TcPh) were performed on days )10 and )9. Conditioning regimen consisted of cyclophosphamide (CY) 60 mg/kg on days )7 and )6, and busulfan (BU) 0Æ8 mg/kg every 6 h on days )5 to )2 pretransplant. HLA- and HPA- incompatible platelet transfusions (Tcincomp) and plasmapheresis (PPh) for immunadsorption were performed shortly prior to haematopoietic stem cell transplant (HSCT). HSCT was performed with mobilized peripheral blood containing 12 · 106 CD34+ cells/kg body weight. Thereafter, platelet transfusions from HLA- and HPA- compatible donors (Tccomp) were given. Post-transfusion corrected count increments were 39 and 33, respectively. The two transfusions on days 3 and 8 were given even though the platelet count in the patient was well above 20 · 109/L due to logistic reasons of the directed donations. Graft-versus-host disease prophylaxis included methotrexate and cyclosporine A.

platelet rebound because of increased platelet production. Platelet levels increased during chemotherapy and remained unusually high early post-transplant. Third, HLA-and HPAincompatible platelet products were transfused for in vivo adsorption of circulating antibodies during the conditioning regimen. It can be assumed that new antibody production was largely prevented by the ongoing conditioning regimen. Fourth, ex vivo antibody reduction was performed the day before HSCT by a single PPh. The in vivo and ex vivo immunadsorption strategies were performed analogous to the standard procedures used in ABO-incompatible allogeneic HSCT where immunoglobulin (Ig)M- and IgG-isoagglutinin titres can be reduced significantly (Osterwalder et al, 1986; Stussi et al, 2006). Anti-HPA-5a antibody concentrations were monitored quantitatively during the treatment course. Transfusion of incompatible platelet products resulted in a significant reduction of alloantibody levels from 300 AU/ml to 209 AU/ml; however PPh on day )1 was the most efficient single intervention, reducing the titre to 66 AU/ml. Early posttransplant, antibody levels increased again, with persisting high levels at day 100 post-transplant. No platelet bound antibodies were detected at that time; platelet counts were normal (Fig 1B). This correlates with observations in ABO-incompatible HSCT where immune IgG can persist for up to 6–12 months post-transplant either with or without accompanying pure red cell aplasia (Stussi et al, 2006). For alloimmunized patients with haematological malignancies, HLA- and HPA-typing of platelet donors is of great importance. National registries allow rapid identification of potential donors for alloimmunized patients. Furthermore, we demonstrate that allogeneic myeloablative HSCT is feasible and safe even in a heavily alloimmunized patient when a combined antibody reduction strategy is pursued. Quantitative measurement of alloantibodies can provide clinicians with useful information concerning the efficiency of the antibody reduction during HSCT and can guide transfusion policies. Caroline Arber1* Gerald Bertrand2* Jo¨rg Halter,1 Franc¸oise Boehlen,3 Ce´cile Kaplan2* and Alois Gratwohl1* 1

Haematology, University Hospital Basel, Switzerland, 2Institut National

de la Transfusion Sanguine, Laboratoire d’ Immunologie Plaquettaire, Paris, France, and 3Haemostatis Unit, University Hospital Geneva, Switzerland.

and HPA-5 could be identified. However, nine potential HPA5bb donors were tested by cross-match analyses for anti-HLA alloreacitvity and four non-reactive donors were scheduled for directed platelet donations. Second, artificial platelet consumption was induced by two TcPh prior to start of the conditioning regimen. This resulted in an unexpectedly high 160

E-mail: [email protected]

Acknowledgements This work was supported by a grant of the Regional Cancer League Basel-Stadt (C.A.) and a grant of the Swiss National


Correspondence Research Foundation 3200B0-106105/1 (A.G.). We thank the blood donation centers in Switzerland, Germany and France for donor identification and cross-match analyses.

References Anolik, J.H., Blumberg, N., Snider, J. & Francis, C.W. (2001) Posttransfusion purpura secondary to an alloantibody reactive with HPA-5a [Br(b)]. Transfusion, 41, 633–636. Bertrand, G., Jallu, V., Gouet, M., Kjaer, K.M., Lambin, P., Husebekk, A. & Kaplan, C. (2005) Quantification of human platelet antigen-1a antibodies with the monoclonal antibody immobilization of platelet antigens procedure. Transfusion, 45, 1319–1323. Bertrand, G., Martageix, C., Jallu, V., Vitry, F. & Kaplan, C. (2006) Predictive value of sequential maternal anti-HPA-1a antibody concentrations for the severity of fetal alloimmune thrombocytopenia. Journal of Thrombosis and Haemostasis, 4, 628–637. Davoren, A., Curtis, B.R., Aster, R.H. & McFarland, J.G. (2004) Human platelet antigen-specific alloantibodies implicated in 1162 cases of neonatal alloimmune thrombocytopenia. Transfusion, 44, 1220–1225. Kiefel, V., Konig, C., Kroll, H. & Santoso, S. (2001) Platelet alloantibodies in transfused patients. Transfusion, 41, 766–770. Mueller-Eckhardt, C., Kiefel, V., Grubert, A., Kroll, H., Weisheit, M., Schmidt, S., Mueller-Eckhardt, G. & Santoso, S. (1989) 348 cases of

suspected neonatal alloimmune thrombocytopenia. Lancet, 1, 363–366. Osterwalder, B., Gratwohl, A., Nissen, C. & Speck, B. (1986) Immunoadsorption for removal of anti-A and anti-B blood group antibodies in ABO-incompatible bone marrow transplantation. Blut, 53, 379–390. Santoso, S., Kalb, R., Walka, M., Kiefel, V., Mueller-Eckhardt, C. & Newman, P.J. (1993) The human platelet alloantigens Br(a) and Brb are associated with a single amino acid polymorphism on glycoprotein Ia (integrin subunit alpha 2). Journal of Clinical Investigation, 92, 2427–2432. Stussi, G., Halter, J., Schanz, U. & Seebach, J.D. (2006) ABO-histo blood group incompatibility in hematopoietic stem cell and solid organ transplantation. Transfusion and Apheresis Science, 35, 59–69. Valeri, C.R. & Ragno, G. (2006) Cryopreservation of human blood products. Transfusion and Apheresis Science, 34, 271–287.

Keywords: transfusion refractoriness, alloimmunization, platelet antibodies, transplantation, human platelet antigen. *These authors contributed equally to this work.

doi:10.1111/j.1365-2141.2007.06753.x

Deregulation of a possible tumour suppressor gene, ZC3H12D, by translocation of IGK@ in transformed follicular lymphoma with t(2;6)(p12;q25)

Follicular lymphoma (FL) is characterized by the t(14;18) (q32;21) translocation, which juxtaposes the BCL2 gene (BCL2)at 18q21 with the immunoglobulin heavy locus gene (IGH@) at 14q32 (Knutsen, 1997). The resultant deregulation of BCL2 with subsequent inhibition of apoptosis is thought to contribute to the onset of FL. FL has an initial indolent course, but more rapid progression by transformation to diffuse large B-cell lymphoma (DLBCL) occurs in many cases. Although other additional chromosomal abnormalities, such as +7, del(1)(p36), del(6)(q) and del(10)(q22-24), have been shown to be associated with the morphological progression of FL (Horsman et al, 2001), their molecular significance remains to be completely elucidated. Deletions of the long arm of chromosome 6 (6q) have been assumed to play an important role in the pathogenesis of tumours including lymphoma, lung cancer, breast cancer and ovarian cancer (Tilly et al, 1994; Zhang et al, 1998; BaileyWilson et al, 2004). FL with 6q23-26 deletions was reported to have a high risk of transformation to DLBCL and to be a poor prognostic indicator (Tilly et al, 1994). We previously reported

a novel translocation t(2;6)(p12;q23) that appeared during the transformation of FL to DLBCL (Yamamoto et al, 2003). We suggested that t(2;6)(p12;q23) played a crucial role in the transformation, and we speculated that an unknown gene at 6q23 may be deregulated as a consequence of IGK@ translocation. Therefore, we initially performed Southern blot analysis of genomic DNA extracted from lymph nodes of the patient with t(2;6)(p12;q23) described previously (Yamamoto et al, 2003) by using Jj and Cj probes (Fig 1). Whereas there was no detectable rearranged band with the Jj probe, two extra bands in addition to a germline band were detected with the Cj probe. To clone the gene, tentatively called ‘transformed follicular lymphoma (TFL)’ at the translocation breakpoint of t(2;6)(p12;q23), the genomic DNA was completely digested by BamHI, and electrophoresed on a 0Æ8% agarose gel. A fraction between 7 and 12 kb, corresponding to the rearranged bands, was collected. The extracted DNA was ligated into the ZAP Express Predigested Vector. This genomic


161