ABO-incompatible Bone Marrow Transplantation Research Articles

The impact of infused red blood cell volume on major and bidirectional ABO-mismatched bone marrow transplantation

Background aimsABO incompatibility does not hinder bone marrow transplantation (BMT), but it has been associated with worse outcomes and additional adverse events. This study aimed to verify the impact of incompatible red blood cells (iRBCs) in allogeneic BMT and to determine a safe number of iRBCs to be infused. MethodsWe compared ABO-incompatible (iABO) allogeneic BMT (n = 42) with ABO-compatible allogeneic BMT (n = 44) and evaluated the impact of the number of infused iRBCs on outcomes and adverse events. ResultsThe iABO patients demonstrated delayed time to transfusion independence at 30 days and 60 days, increased requirement for red blood cell (RBC) transfusion and greater hemolysis signals and incidence of pure red cell aplasia. Neutrophil/platelet engraftment, length of hospitalization post-transplant, platelet units required, graft-versus-host disease occurrence and overall survival were similar in both groups. Patients in the iABO group received 1.03 × 1010 iRBCs/kg (range, 0.36–3.88). Infusion of iRBCs >1.0 × 1010 /kg was related to graft failure or death before neutrophil engraftment or platelet engraftment or both as well as increased plasma requirement and increased creatinine. Our results also suggest that antibody titers impact the transplantation scenario. ConclusionsThe iABO transplantation showed some unfavorable outcomes. It is important to monitor the value of iRBCs to be infused, considering the recipient antibody titers. We propose using the number of iRBCs (iRBCs/kg) as a dose parameter with regard to infused iRBCs. Further studies are necessary to clarify the maximum safe number of iRBCs in iABO transplants.

P‐BT‐19 | Red Cell Reduction of ABO Incompatible Bone Marrow Transplant Cells with Automated Separation Device

P‐BT‐19 | Red Cell Reduction of ABO Incompatible Bone Marrow Transplant Cells with Automated Separation Device

A comparison of two protocols for optimal red blood cell depletion using Sepax-2 device for ABO-major incompatible transplantation in adults

Purpose of the studyIn ABO-incompatible bone marrow transplantation, an efficient depletion of red blood cells (RBC) within the graft is mandatory to avoid adverse events in transplanted patients. Using non therapeutic products, we evaluated the substitution of the standard density gradient-based separation (DGBS) over Ficoll-Paque with the use of an automated procedure intended for buffy coat only (SmartRedux software) introducing modifications within the settings to achieve a drastic reduction of the initial volume of the product. Both methods were conducted on the Sepax-2 device. Samples and methodsRBC depletion rates and CD34+ cells recoveries from eight procedures with SmartRedux software using “in-house” settings (method A) were compared to those obtained from four procedures using NeatCell software, an automated DGBS over Ficoll-Paque (method B). ResultsMedian erythrocyte depletion of 95,4% (92,7%–99,0%) and 99,8% (99,0%–99,9%) were observed using methods A and B, respectively. Median residual RBC volumes in the final product were 19 mL (4,4 mL–31,2 mL) and 0,7 mL (0,4 mL–4,7 mL), respectively (p = 0,014). CD34+ cells recoveries of 90,9% (62,7%–102,1%) and 78,4% (64,1%–86,2%) were achieved for methods A and B. Median platelet depletion was 16,6% (10%–42,7%) and 89,8% (88,5%-92,4%) using methods A and B, respectively (p = 0,004). Processing duration was shorter using method A (168 ± 29 min) than method B (295 ± 21 min) (p = 0,004). ConclusionBoth methods achieved satisfactory erythrocyte depletion and CD34+ recovery. The use of Sepax-2 device in association with SmartRedux software could be extended to efficiently deplete RBC from large-volume BM in a raw instead of DGBS.

Erythrocyte depletion from bone marrow: performance evaluation after 50 clinical-scale depletions with Spectra Optia BMC

BackgroundRed blood cell (RBC) depletion is a standard graft manipulation technique for ABO-incompatible bone marrow (BM) transplants. The BM processing module for Spectra Optia, “BMC”, was previously introduced. We here report the largest series to date of routine quality data after performing 50 clinical-scale RBC-depletions.MethodsFifty successive RBC-depletions from autologous (n = 5) and allogeneic (n = 45) BM transplants were performed with the Spectra Optia BMC apheresis suite. Product quality was assessed before and after processing for volume, RBC and leukocyte content; RBC-depletion and stem cell (CD34+ cells) recovery was calculated there from. Clinical engraftment data were collected from 26/45 allogeneic recipients.ResultsMedian RBC removal was 98.2% (range 90.8–99.1%), median CD34+ cell recovery was 93.6%, minimum recovery being 72%, total product volume was reduced to 7.5% (range 4.7–23.0%). Products engrafted with expected probability and kinetics. Performance indicators were stable over time.DiscussionSpectra Optia BMC is a robust and efficient technology for RBC-depletion and volume reduction of BM, providing near-complete RBC removal and excellent CD34+ cell recovery.

1038-LBP: Incomplete reactions of mismatched donor HLA allele(s) in relapsed AML patient

Aim Major ABO incompatible bone marrow transplantation can cause Pure Red Cell Aplasia (PRCA), which can evolve into AML. We will present here incomplete reactions of missmatched donor allele(s) in relapsed AML patient with a history of ABO incompatible, haplo-identical HSCT. Methods High and low resolution HLA typing was performed by SBT and rSSO, respectively. Post-transplant engraftment analyses were conducted with STR. Results Patient is a 35 y/o, “O” blood group female with AML, who had HSCT from her “A” blood group sibling in Oct 2009. Patient’s initial HLA class I and II typing was homozygous. Her sibling was haplo-identical with 6/12 mismatch in host vs graft direction. Engraftment testing performed in Jan 2010 revealed more than 98% donor allele composition in un-fractioned whole blood, as well as in CD3+ and CD15+ populations, which was then decreased in Feb 2010 and recovered again after receiving donor lymphocyte infusion (DLI) in Aug 2010. BM biopsy of Feb 2010 revealed that 30–40% cellularity with M:E ratio greater than 10:1. Sustained erythroid hypoplasia with intact myelopoiesis and megakaryopoiesis raised a possibility of acquired PRCA. HLA typing in Jan 2010 showed full engraftment. However, following low resolution HLA verification typing performed by rSSO in Apr 2010 using buccal sample could not call mismatched alleles in HLA-A, B and DQB1 loci. In HLA-A and HLA-B loci, multiple beads presented weaker reactions than positive cut off value. To be heterozygous in DQA1 and DQB1 loci, 2 beads should be positive, however, they were too weak to be adjusted as positive. When HLA typing was tested by SBT in parallel, one DQB1 mismatched allele was still missing. Despite of DLI and other treatments, patient expired in a year after HSCT. Conclusions Summary: Both low and high resolution HLA typing methodologies revealed incomplete reactions of donor alleles (s) in relapsed AML patient, which supports other clinical findings of disease relapse.

Production of anti‐ABO blood group antibodies after minor ABO‐incompatible bone marrow transplantation in NOD/SCID/gamma(c)(null) mice

ABO incompatibility is a barrier for solid organ transplantation, but not for hematopoietic stem cell transplantation. To investigate tolerance induction, we enrolled patients who had undergone minor ABO-incompatible (O into A group, n = 6) and ABO-identical (O into O group, n = 4) bone marrow transplantation (BMT). None of the six O into A patients were positive for recipient-specific (anti-blood group A) isohemagglutinins, whereas all four O into O patients were. Peripheral blood mononuclear cells (PBMCs) were engrafted into NOD/SCID/gamma(c)(null) (NOG) mice, followed by sensitization of blood group A red blood cells. Anti-blood group A antibodies (Abs) in the sera of the patients and the human PBMC-engrafted NOG mice were measured by enzyme-linked immunosorbent assays. Anti-blood group A Abs in the patients' sera were significantly correlated with anti-A isohemagglutinin titers (p < 0.01). In the human PBMC-engrafted NOG mice, anti-blood group A Abs were significantly lower in the O into A group than in the O into O group (p < 0.05), despite ex vivo restimulation of B cells. The results of this study suggest that long after receiving minor ABO-incompatible BMT, B cells derived from newly engrafted donor precursor cells were induced tolerance to recipient-specific antigens.

Bone marrow processing for transplantation using Cobe Spectra cell separator

Concentration of bone marrow aspirates is an important prerequisite prior to infusion of ABO incompatible allogeneic marrow and prior to cryopreservation and storage of autologous marrow. In this paper we present our experience in processing 15 harvested bone marrow for ABO incompatible allogeneic and autologous bone marrow (BM) transplantation using Cobe Spectra® cell separator. BM processing resulted in the median recovery of 91.5% CD34+ cells, erythrocyte depletion of 91% and volume reduction of 81%. BM processing using cell separator is safe and effective technique providing high rate of erythrocyte depletion and volume reduction, and acceptable recovery of the CD34+ cells.

Major ABO不適合骨髄移植後に発症した赤芽球〓の1例

Major ABO不適合骨髄移植後に発症した赤芽球〓の1例

Bone marrow processing with the AMICUS™ separator system

ABO incompatible bone marrow transplantation (BMT) requires processing of the donated bone marrow (BM), either erythrocyte depletion, or also a volume reduction. The AMICUS™ system was introduced in the field of peripheral blood mononuclear cell collection, showing a good performance regarding efficiency and safety. To evaluate the performance of the MNC collection program of the Amicus device for BM, we analysed our data obtained from the Amicus and the Fenwal CS3000omnix™ plus device. From 2005 to 2008, we performed 22 automated erythrocyte depletions of BM for ABO mismatched BMT in 21 patients, 11 with the Amicus (A; 10 patients) and 11 with the CS3000 (F; 11) device. There were no statistical differences in donor age, recipient age, type of ABO mismatch, and CD34+ cell yield [group A pre 7.03 post 4.93 vs. group F pre 8.55 and post 6.2 × 10E06 cells per kilogram of bodyweight] for both devices. The efficiency for the CD34+ cell collection was lower, but not statistically significant, in the Amicus device (70% ± 12 vs. 84% ± 12; U-test P = 0.123). The erythrocyte volume in the final product was higher but not statistically significant different in the Amicus device (9.46 ± 2.3 vs. 6.98 ± 3.3 ml; U-test P = 0.17). During the evaluation period, no technical problems were observed. All patients but one, who died at d + 11, showed a sustained engraftment. We conclude that, in principle, the Amicus device can be used for MNC collection from BM to deplete erythrocytes from BM grafts in allogeneic stem cell transplantations.

Désérythrocytation de moelle osseuse par gradient de densité sur COBE 2991 avec le kit triple « bérets »

ABO-incompatible bone marrow transplantation requires red blood cell depletion. Lots of laboratory adopted the technique of density gradient centrifugation (Ficoll-hypaque) using the COBE 2991 cell processor with simple-bag processing set. However, tubing of this set is not adapted to the currently available peristaltic pumps. Moreover, two other sets are required: one for the buffy-coat and one for postgradient cell washing. We developed a method using triple-bag processing set to conduct whole-step procedure (concentration, Ficoll and washing). Peristaltic PVC tubing is provided in one line of the set allowing a safe processing without several connections thus reducing risks of microbial contamination. First, we used buffy-coat of total blood for training, then, we carried out red cell depletion of healthy bone marrow donors. The red blood cell depletion was 97.9 ± 1.1% and CD34+ recovery was 89.6 ± 8.7%. These results are very close to those obtained with the simple-bag set (red cell depletion. = 94.0 ± 6.8% and CD34+ recovery = 95.9 ± 20.3%). We conclude that the triple-bag system, very little used in France, is practical, simplified the manipulation and is more safety than the simple-bag set.

Non-ABO red blood cell alloantibodies following allogeneic hematopoietic stem cell transplantation.

Immune-mediated hemolysis is a well-recognized occurrence which complicates the period following a bone marrow transplant (BMT). However, although many studies have investigated the hemolytic anemia following ABO-incompatible BMT, data regarding the occurrence of alloantibodies against red blood cell (RBC) antigens other than ABO in patients undergoing hematopoietic stem cell transplantation are limited. In this review, we briefly analyze the most important non-ABO red blood cell (RBC) antigen systems involved in the development of post-BMT alloimmune hemolytic anemia, paying particular attention to the pathogenic mechanisms and the clinical significance of the alloantibodies involved. Among the non-ABO RBC antigens, RhD antigen is the one most frequently implicated in the development of post-BMT alloimmune hemolytic anemia. Although less frequent than hemolysis following transplants with ABO incompatibility, non-ABO-incompatible allograft hemolysis may severely complicate the post-BMT period creating difficult clinical management issues. For this reason, we advise careful pre-transplant donor and recipient checks for the most important RBC antigen systems and close post-BMT immunohematological monitoring in those patients undergoing allogeneic hematopoietic stem cell transplant with RBC antigen incompatibility.

ABO-incompatible organ and bone marrow transplantation: current status.

Despite the presence of preformed antibodies against AB oligosaccharide epitopes on the donor vascular endothelium, approximately one-third of ABO-incompatible organ allografts are not rejected by a humoral mechanism. With the growing immune-manipulation of the recipient, survival rates can be raised considerably, although they remain significantly inferior to those of ABO-compatible transplantation. Data from the Collaborative Transplant Study indicate a 1-year graft survival rate of approximately 50-60% following cadaveric ABO-incompatible kidney, liver or heart transplantation, compared with 70-80% for an ABO-compatible organ. The results for infants and young children, however, are very much better than those of adults, particularly for liver and heart transplantations, and the data suggest that B-cell tolerance can develop in the infant age group. We here review clinical and experimental experience with ABO-incompatible organ and bone marrow allotransplantation and address the mechanisms by which organs or cells survive in the presence of natural anti-carbohydrate antibodies.

Transfusion-related problems following major ABO-incompatible bone marrow transplantation

A case of acute myelocytic leukemia has been reported in which the patient’s surviving original B lymphocytes after pretransplant-conditioning chemotherapy probably reproduced hemagglutinins that reacted with red blood cells (RBCs) derived from engrafted donor marrow for a prolonged period of time. Although the direct antiglobulin test was negative and hemagglutinins were not detectable in the patient’s sera but only in the eluate, the antibodies reappeared in the sera. Therefore, it is important to confirm that the eluate does not contain antibodies that would react with donor-derived RBCs when the type of red cell used for transfusion is switched from the patient’s type to the donor’s type in a major ABO-mismatched bone marrow transplantation (BMT). Testing of ABO subgroups using lectins is also recommended to avoid a delayed hemolytic transfusion reaction following BMT.

ＡＢＯ不適合骨髄移植後の赤血球におけるＡＢＨ抗原型物質の解析

ABH-antigens and A-and B-glycosyltransferase activity were examined in four patients who received an ABO-incompatible bone marrow transplant from HLA-matched donors. ABO phenotype and ABO genotype were analyzed using flow cytometry and polymerase chain reaction with sequence specific reaction, respectively. On day 14 after bone marrow transplantation (BMT), the ABO genotype of erythroid burst-forming units was converted into the ABO genotype of the donors in all of the recipients. Afterward, ABO phenotype of red blood cells (RBC) in all recipients completely changed to that of the donors. Rh, P, Diego, Kidd, Duffy and MNSs phenotypes were also changed into the donor's phenotypes. However, in one patient the Lewis blood type did not change into the donor's phenotype. Similarly, using an elution method, ABH antigens of the recipients were detected in RBC in all of the recipients after BMT. In two recipients, the serum glycosyltransferase activities of the recipients were only detected after BMT. These findings indicate that ABH antigens may be released from non-blood cells and adsorbed by RBC, although the possibility of ABO chimeras or unknown mechanisms is not ruled out. Further studies are needed to determine the nature of ABH antigens eluted from RBC of recipients receiving an ABO-incompatible bone marrow transplant.

Concentration of Progenitor Cells Collected from Bone Marrow Fluid Using a Continuous Flow Cell Separator

In ABO major incompatibility on bone marrow transplantation (BMT), red cells must be removed from collected marrow fluid to prevent hemolysis. We report the concentration of progenitor cells collected using a continuous flow cell separator (Cobe Spectra). The average volume of concentrated bone marrow was 132 +/- 47 ml and that of red cells included was 5.1 +/- 2.4 ml. The red cell removal rate was 97.6%. The recovery rate was 40.6% for total nuclear cells, 77.9% for mononuclear cells, 100% for CD34+ cells, and 93.9% for colony forming unit granulocyte-macrophage. Eighteen patients undergoing allogeneic BMT showed no signs of fever or hemolysis during concentrated marrow fluid transfusion. The recovery rate of progenitor cells was high, indicating sufficient recovery of hemopoiesis. This technique is applicable in ABO-incompatible BMT and in frozen-storage stem cells.

Clinical application of therapeutic erythrocytapheresis (TEA)

Therapeutic erythrocytapheresis (TEA) has been used in different diseases such as polycythemia vera (PV), secondary erythrocytosis or hemochromatosis as a process of the less cumbersome but more expensive phlebotomy. TEA is preferred in emergency conditions such as thrombocytosis or in conditions such as porphyria cutanea tarda (PCT) or erythropoietic porphyria when plasma exchange (PEX) is often combined with TEA to reduce extracellular levels of uroporphyrin which contribute to plasma hyperviscosity. TEA is often combined with drug therapy that varies from etoposide in PV to EPO and desferoxamine which are used to mobilize and reduce iron stores in hemochromatosis. Benefits from this combination may be more long lasting than expected. Nonetheless for TEA, there is no standard protocol and, clinical experience with this therapy remains highly anecdotal. Therapeutic red cell-exchange (TREX) has been used with much interest over the years, starting with the management of hemolytic disease of the newborn and later used to correct severe anemia in thalassemia patients thereby preventing iron overload. It has also been used for the management of complications of sickle cell disease such as priapism, chest syndrome, stroke, retinal, bone, splenic and hepatic infarction or in preparation for surgery by reducing HbS to less than 30%. Automated apheresis has also favored the use of TREX in conditions such as paroxysmal nocturnal hemoglobinuria and aniline poisioning, arsenic poisioning, Na chlorate intoxicatons and CO intoxications, hemoglobinopathies, autoimmune hemolytic anemia, reactions due to ABO incompatibility, in preparation for ABO incompatible bone marrow transplantation or for preventing anti-D immunization after the transfusion of D + cells to D − recipients. Another field of application has been in the emergency management of intraerythrocytic parasite infections such as malaria and babesiosis. Application of TREX may be wide but its real use remains limited. In our personal experience, in 16 years, only 167 TREX procedures have been carried out in a total of 13,747 therapeutic procedures. This represents only 1.21% of the total.

Donor lymphocyte infusion as therapy for pure red cell aplasia following bone marrow transplantation

Pure red cell aplasia (PRCA) following allogeneic bone marrow transplantation (BMT) has been associated with major ABO incompatibility between donor and recipient red blood cells (3). Donor lymphocyte infusion (DLI) has been used with success following cytogenetic or haematologic relapse of chronic myeloid leukaemia (CML) after BMT (4). At our centre we initiated a study regarding the use of DLI as relapse prophylaxis in patients affected by acute lymphoblastic leukaemia (ALL) and submitted to BMT. We report the case of a patient with ALL who developed PRCA after ABO incompatible BMT and its unexpected successful treatment with DLI when other therapies failed. A 20-year-old female with ALL (L1, common) in first complete remission underwent an allogeneic BMT in October 1997. The donor was an HLA-compatible sister with blood group A+ and the recipient was O+. Busulphan (16 mg/kg) and cyclophosphamide (200 mg/kg) were used as conditioning therapy and 2.6 × 108/kg marrow cells were infused after removing red blood cells (RBC) by an automated cell separator (EXCEL-DIDECO, Mirandola, Italy). Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine-A (CyA) and short-course methotrexate (MTX). On day 10 the patient developed acute GVHD grade I (skin), and was treated successfully with prednisone 2 mg/kg. Myeloid engraftment (ANC > 0.5 × 109/l) was observed on day 20 and platelets recovered (>25 × 109/l) on day 32. Reticulocyte counts were measured with Becton Dickinson FACScan flow cytometer and the absolute reticulocyte count was persistently <0.02 × 1012/l. Repeated examinations of bone marrow revealed complete absence of RBC precursors in an otherwise normal marrow. There was no evidence of haemolysis, blood loss or infections. The anti-A isoagglutinin titre was 1 : 8. A diagnosis of PRCA was made and recombinant human erythropoietin (rh-EPO) was started at a dose of 50 U/kg three times per week. From day 70, immunoglobulin, 400 mg/kg, was infused daily for 5 d, but anaemia persisted and the patient required regular RBC transfusions. A total of five plasmapheresis was then performed from day 125 to day 135 and the anti-A isoagglutinins became undetectable. However, the patient remained reticulocytopenic and transfusion-dependent. On day 170, while still under rh-EPO therapy, the patient received the first infusion of donor lymphocytes at the dosage of 1 × 106/kg, according to the leukaemic relapse prophylaxis protocol in use at our centre. Within 8 d a marked increase of the absolute reticulocyte counts and Hb levels was observed, with an absolute reticulocyte count of 0.9 × 1012/l and an untransfused Hb concentration of 8.5 g/dl (Fig 1). Neither acute nor chronic GVHD was observed and full engraftment of myeloid and megakaryocyte series persisted. At this time the patient discontinued CyA. After 6 and 12 weeks, respectively, the patient received increased doses of donor lymphocytes, the reticulocyte counts and Hb levels returned to normal values, blood group converted to type A and no further treatment was necessary. At 12 months post transplant the patient is asymptomatic. . Response of reticulocyte counts after administration of erythropoietin (rhEPO), immunoglobulins (Ig), plasmapheresis (Pl-aph), and donor lymphocytes (DLI). In major ABO incompatible BMT the persistence of functional host-derived lymphocytes and plasma cells that have survived the BMT conditioning regimen can result in delayed erythroid engraftment and increased RBC transfusions. It has been reported that PRCA usually resolves spontaneously, coincidentally with the decrease of agglutinin titre to <1:16 (2), but this was not the case in our patient. In the post-transplant period anti-A isoagglutinin titre remained at 1 : 4 to 1 : 8 and no erythropoiesis was observed. Transfusion dependence remained unchanged even after depletive plasmapheresis, when the isoagglutinins became undetectable in the peripheral blood. The stimulatory effect of rh-EPO on bone marrow erythroid precursors was insufficient to counteract the possible cell-mediated suppression of erythropoiesis. Some patients have responded to immunosuppressive therapy such as anti-lymphocyte globulin (1) or prednisone (5). In our patient the donor's lymphocytes infused at 170 d from BMT proved to be effective immunosuppression and a consistent increment of reticulocyte count was observed after 8 d. Probably the donor lymphocytes served to reinforce the novel immunocompetent system in suppressing the host's residual lymphocytes responsible for the erythrocyte blockage. The donor lymphocytes overcame the immunomodulation of the graft derived immune system due to CyA, which leads to temporary but prolonged prevalence of host lymphoid cells. We think that donor's lymphocytes may be cautiously used in patients who have received BMT for leukaemia and develop PRCA when GVHD is absent, and myeloid and megakaryocyte engraftment is complete.

ｍｉｎｏｒ ＡＢＯ不適合骨髄移植後の正常同種凝集素および糖転移酵素の変化

Changes in anti-A and anti-B antibody titer and A- and B-glicosyltransferase activity were analyzed in five patients after minor ABO-incompatible bone marrow transplantation (BMT). All patients showed complete engraftment after BMT. Shortly after BMT, blood types of recipient's erythrocytes as examined by standard agglutination techniques changed into those of the donor's erythrocytes. Using the elution method, however, recipient's erythrocyte antigens were detected on erythrocytes from three of five patients with complete engraftment after BMT. Analysis of A- and B-transferase activity in serum demonstrated that four of five patients maintained pre-BMT transferase activity. The other patient, who showed no A- or B-transferase activity after BMT, had an inhibitor to this activity. These results suggest that in patients receiving BMT, a small amount of blood antigens on erythrocytes and serum transferase activity may be derived from the patient's tissue.

Pure red cell aplasia after major ABO-incompatible bone marrow transplantation: two case reports of treatment with recombinant human erythropoietin

Abstract A 34-year-old man with acute myelocytic leukemia (AML: MO) and a 32-year-old woman with AML: M2 developed pure red cell aplasia (PRCA) after receiving a major ABO-incompatible bone marrow transplant (BMT). The first patient responded to recombinant human erythropoietin (rhEPO) therapy, while the second did not. The second patient also received methylprednisolone (m-PSL) but developed reticulocytosis and hemolysis after the administration of m-PSL. Plasmapheresis was then performed and the patient promptly recovered from hemolysis and PRCA. We conclude that close attention must be paid when treating PRCA following major ABO-incompatible BMT with rhEPO and m-PSL, as there is always the potential for massive hemolysis.

Anti-A and/or anti-B is not detectable in some patients who underwent ABO-incompatible bone marrow transplantation.

Recently, anti-A and/or anti-B produced by B cells from donor marrow could not be detected for more than 20 weeks in some patients who had undergone ABO-incompatible bone marrow transplantation (BMT). Twelve to 72 weeks after 11 patients underwent ABO-incompatible BMT, titers of anti-A and anti-B were assayed, A and B antigens were identified by routine methods and flow cytometry, direct and indirect antiglobulin tests were performed, and the red cell antibody was eluted. In some patients who underwent ABO-incompatible BMT, anti-A and/or anti-B produced by the B cells from the donor marrow could not be detected after BMT when red cells taken from the patients before BMT carried the corresponding antigen--that is, when hematopoiesis had already changed the cells to the donor's type according to ABO blood typing. Furthermore, some blood samples from those patients gave positive results in direct antiglobulin tests. Blood typing of patients after BMT showed mixed-field agglutination. In one patient, the half-life of red cells assayed with 51Cr was 22.4 days (30.0 +/- 4.0 days for normal controls). Although many hypotheses could be considered to explain the present data, the possibility is proposed that anti-A and/or anti-B in the sera must have been consumed in some patients who underwent ABO-incompatible BMT. This may lead to problems such as difficulty of ABO typing, positive direct antiglobulin tests, and a relatively short life span of red cells.