Red Blood Cell-Incompatible Allogeneic Hematopoietic Progenitor Cell Transplantation

REVIEW Red blood cell-incompatible allogeneic hematopoietic progenitor cell transplantation

SD Rowley1, ML Donato1 and P Bhattacharyya2

1Adult Blood and Marrow Transplantation Program, John Theurer Cancer Center, Hackensack, NJ, USA and 2Department of Pathology, Hackensack University Medical Center, Hackensack, NJ, USA

Transplantation of hematopoietic progenitor cells from Introduction red cell-incompatible donors occurs in 30–50% of patients. Immediate and delayed hemolytic transfusion Upwards of 30% of allogeneic hematopoietic progenitor reactions are expected complications of red cell-disparate cell (HPC) transplants from related donors and over transplantation and both ABO and other red cell systems 50% of transplants from unrelated donors will involve such as Kidd and rhesus can be involved. The immunohe- ABO-disparate donor and recipients,1,2 and donor and matological consequences of red cell-incompatible trans- recipient pairs may differ for other RBC antigens, as well. plantation include delayed red blood cell recovery, pure HLA-identical siblings can differ for red cell antigens based red cell aplasia and delayed hemolysis from viable on the different genetic locations of the major and minor lymphocytes carried in the graft (‘passenger lympho- HLA loci and the genes that control the expression of the cytes’). The risks of these reactions, which may be abrupt over 30 defined groups of RBC antigens (Table 1).3 The in onset and fatal, are ameliorated by graft processing and transplantation of a red cell antigen-disparate HPC proper blood component support. Red blood cell antigens product is feasible, if attention is given to the immunohe- are expressed on endothelial and epithelial tissues in the matological considerations necessary in the management of body and could serve to increase the risk of GvHD. Mouse both immediate and delayed hemolytic reactions. Although models indicate that blood cell antigens may function as feasible, the clinical impact of red cell incompatibility on minor histocompatibility antigens affecting engraftment. transplant outcomes is uncertain, with various studies Similar observations have been found in early studies of showing positive, negative or no effect on important human transplantation for transfused recipients, although clinical end points such as neutrophil and platelet engraft- current conditioning and immunosuppressive regimens ments, acute and chronic GvHD, disease relapse and OS. appear to overcome this affect. No deleterious effects Definite short-term adverse consequences of red cell- from the use of red cell-incompatible hematopoietic grafts incompatible transplantation are reported, as will be on transplant outcomes, such as granulocyte and platelet discussed below, but these consequences can be minimized engraftments, the incidences of acute or chronic GvHD, with appropriate graft modifications to remove incompa- relapse risk or OS, have been consistently demonstrated. tible red cells and plasma, attention to the pre-transplant Most studies, however, include limited number of patients, conditioning and post-transplant immunosuppressive varying diagnoses and differing treatment regimens, regimens and prescription of proper blood product complicating the detection of an effect of ABO-incompa- support. tible transplantation. Classification of patients by ABO The major focus of blood cell-incompatible transplanta- phenotype ignoring the allelic differences of these antigens tion has been on the ABO system because of the also may obscure the effect of red cell-incompatible consequences of acute hemolysis resulting from the infusion transplantation on transplant outcomes. of incompatible RBCs or plasma, the delayed hemolytic Bone Marrow Transplantation (2011) 46, 1167–1185; reactions that may occur as a result of viable ‘passenger’ doi:10.1038/bmt.2011.135 lymphocytes infused as a component of the allograft and Keywords: ABO incompatibility; allogeneic transplant- the delayed RBC recovery resulting from persistent host ation; hemolytic transfusion reaction; red cell antigens; isoagglutinins directed against donor RBCs (Table 2). transfusion support; isoagglutinins Although there are also frequently disparate non-ABO genes in the population (Table 1),4 the primary risk of hemolysis arises from the nearly ubiquitous presence of antibody(s) among both the donor and recipient against ABO antigens that they lack. Thus, it is much more likely Correspondence: Dr SD Rowley, Adult Blood and Marrow Transplant- that donor/recipient pairs will have clinically significant ation Program, John Theurer Cancer Center, 92 Second Street, reactions to ABO antigens than to any other red cell Hackensack, NJ 07601, USA. E-mail: [email protected] antigen group. However, non-ABO antigen systems have Received and accepted 19 May 2011 been implicated in post-transplant reactions and red cell ABO-incompatible transplantation SD Rowley et al 1168 incompatibility encompasses ABO, rhesus (Rh) and other (for example, group A donor with a group B recipient) red cell antigen systems.5,6 raises the risk of both consequences.

Deﬁnition of RBC-incompatible transplants Clinical outcomes of RBC-incompatible transplants

Red cell incompatibility may be classiﬁed into two major Numerous reports of the effect (or lack thereof) of categories, one in which the recipient has antibodies transplantation using HPC from ABO-disparate donors directed against donor red cells with the potential of acute have been published (Tables 3a and b), including large hemolytic transfusion reaction upon infusion of the stem registry studies;2,7–9 studies limited to PBPCs,10–12 cell product and delayed recovery of red cell function after BM1,2,7,9,13–20 or cord blood sources of HPC;21 reduced- transplantation (major incompatibility), and one in which intensity conditioning regimens;8,10,11,22,23 T-cell-depleted the donor has antibodies against the recipient (minor grafts;17,19 and cohort studies from single or limited number incompatibility). Although the latter may rarely cause of institutions.24–33 Four large registry studies of HPC difﬁculty from infusion of incompatible plasma with high recipients including two studies of unrelated donor isoagglutinin titers, B-lymphocytes carried in the compo- transplantation showed inconsistent results with some nent (‘passenger lymphocytes’) can produce isoagglutinins deleterious effect on engraftment, GvHD or survival being resulting in a delayed transfusion reaction usually 7–12 observed in some patient groups but not in others days after transplantation. Bi-directional incompatibility (Table 3a). Registry data studies strictly limited to PBPC

Table 1 Genetic location of HLA and selected RBC antigens

Antigen Chromosome location3 Incidence in ethnic populations4 Incidence of corresponding antibodies Hemolysis risk

Ca AA A

Major HLA 6 NA NA NA ABO A1 9 33 19 27 Virtually all donors and recipients Hemolytic A2 10 8 Rare B 9 20 25 O444943 A1B 3 3 5 A2B 1 1 Rare

Rh DCe 1 42 17 70 Only after immunization Hemolytic Ce 2 2 2 DcE 14 11 21 cE 1 0 0 Dce 4 44 3 ce 37 26 3

Kidd Jk(a+bÀ) 18 26 51 23 Only after immunization Hemolytic Jk(aÀb+) 23 8 27 Jk(a+b+) 50 41 49 Jk(aÀbÀ) Rare Rare Rare

Abbreviations: A ¼ Asian; AA ¼ African-American; Ca ¼ Caucasian; NA ¼ not applicable; Rh ¼ rhesus. Shown are the chromosome locations of the major HLA and various RBC antigens for which post transplant hemolytic reactions have been reported. Shown also are the incidence of the relevant antigen for three different ethnic groups.

Table 2 Immunohematological consequences of ABO-incompatible transplantation

Incompatibility Consequence Cause

ABO major Acute hemolysis Infusion of incompatible red cells Delayed granulocyte and platelet engraftments Loss of HPC from processing to remove red cells. Expression of ABO antigens on granulocytes and platelets Delayed red cell engraftment Host anti-donor isoagglutinins Pure red cell aplasia Persistence of anti-donor isoagglutinins

ABO minor Acute hemolysis Donor plasma with high isoagglutinin titers Delayed hemolytic reaction Passenger lymphocytes producing anti-host isoagglutinins

Abbreviation: HPC ¼ hematopoietic progenitor cell.

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1169 or cord blood transplantation have not yet been reported. antigen-compatible related and unrelated donor trans- Most published studies involve retrospective analyses of plants. However, similar studies of the effect of incompat- data from single institutions with limited number of ibility for platelet alloantigens in human HPC patients. No consistent effects for a variety of outcomes transplantation have not yielded similar observations.48 It can be discerned from these reports, possibly a reflection of is reasonable to assume that current transplant condition- the varying number of patients studied; differing diagnoses; ing and post-transplant immunosuppressive regimens ob- differing donor and HPC product selection criteria; and use scure the effects of immunization to blood cell mHLAs and of various cell processing, transfusion support and con- that changes in these regimens could allow mHLAs to ditioning and GvHD prophylaxis regimens (Table 3b). emerge as a factor affecting transplant outcomes. Moreover, statistical analyses vary with some reports combining patient groups precluding the ability to discern consistent effects from either minor or major incompatible Does ABO incompatibility affect neutrophil and platelet transplants. Donor considerations have usually been engraftments? ignored, such as the effect of previous exposure of donors to blood cell antigens,34 although the deleterious effects of The presence of host–anti-donor isoagglutinins can delay exposure to antigens reflected by donor parity are well RBC recovery and increase post-transplant red cell recognized.9,35 Even dietary changes such as the ingestion transfusion requirements.1,2,7,10,11,13,18,21,22,25,27,28 No clear of probiotics can affect the isoagglutinin titers in a patient effects of ABO incompatibility have been found for or donor.36 neutrophil or platelet engraftments even though cells of It must also be recognized in reviewing the published these lineages express ABO antigens or can adsorb antigen reports of red cell-incompatible transplants that donors from the plasma of secretors.49 Two of the four registry and recipients are classified by the phenotype of the RBCs. studies correlated major ABO incompatibility with a Patient/donor pairs may have differing genotypes from median one or two day slower neutrophil engraftment allelic variations of isoantigens that could serve as minor after BMT,2,8 but a third equally large study involving histocompatibility antigens (mHLAs), leading to increased unrelated donors did not find a difference.9 One cohort risks of graft failure or of acute or chronic GvHD that study found slower neutrophil engraftment for ABO- would not be detected in the more broadly defined incompatible compared to ABO-compatible graft recipi- population.37,38 The red cell membrane contains numerous ents, but this analysis pooled all ABO-incompatible proteins of wide structural diversity, including carbohy- recipients and the specific effects of major or minor drate epitopes on glycoproteins and/or glycolipids, and incompatible transplantation were not analyzed.32 Seebach peptide antigens that serve as membrane transporters, et al.7 proposed a direct effect of recipient isoagglutinins, adhesion molecules and receptors, enzymes and structural possibly by adsorption onto the cell surface, on donor proteins.39,40 Polymorphisms among these antigens may granulopoiesis to explain the delayed recovery of neutro- lead to the production of clinically significant antibodies in phils after major ABO-incompatible transplantation, recipients after exposure to blood-containing products. although loss of hematopoietic progenitor cells (HPC) during processing of major ABO-incompatible grafts to remove incompatible RBCs may also be an explanation for Do blood cells present minor HLA antigens? this observed delay in neutrophil engraftment.2 A delay in platelet recovery was reported in one of the Given the diversity of RBC antigens and their associated registry studies for recipients of major ABO-incompatible polymorphisms, and the likely inability to match donors BM grafts.2 Two cohort studies involving related donor and recipients for all red cell antigen polymorphisms, most PBPC transplants or cord blood transplants reported patients outside syngeneic transplantation are likely reci- similar findings,10,21 as did a meta-analysis of seven cohort pients of HPC grafts that could give rise to red cells with groups (but only for recipients of unrelated vs related disparate antigens. Recognized early in the development of donors).33 Platelets express ABO antigens,49 and increased transplantation science was the enhanced host-vs-graft platelet transfusion requirements for recipients of major reaction as a consequence of blood transfusions given ABO-incompatible grafts have been reported.10,21 It is before transplantation, leading to a higher risk of engraft- conceivable that the lower platelet transfusion requirements ment failure for patients undergoing treatment in the after transplantation using non-myeloablative conditioning management of aplastic anemia.41,42 This enhanced host-vs- allowed the detection of an effect of residual host graft reaction, which was observed in recipients of HLA- isoagglutinins on thrombopoiesis reported in one cohort compatible HPC and explored in canine models of study,10 which would otherwise be obscured by the higher transplantation,43 likely is the result of immunization from transfusion requirements after transplantation using a mHLAs expressed on blood cells. In mouse models, pre- traditional conditioning regimen, but an increased platelet transplant exposure to red cells or platelets can induce transfusion requirement has not been observed in other rejection of transplanted HLA-compatible BM cells as a studies.11,22,50 In one study using non-myeloablative con- result of alloimmunization against mHLAs.44,45 Bean ditioning, recipients of minor ABO-incompatible trans- et al.46,47 reported that irradiation of blood products could plants required a greater number of platelet transfusions.11 prevent immunization to minor dog leukocyte antigens Remberger et al.51 reported a 14.9-fold higher risk of (minor canine histocompatibility antigen) with reduced risk engraftment failure for recipients of major ABO-incompa- of graft rejection in canine models of dog leukocyte tible unrelated donor BM or PBPC. Graft failure was

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1170 observed in 6 of 224 patients studied, including 4 of 67 a higher risk of aGvHD for recipients of ABO-incompa- major ABO-incompatible and 2 of 16 bi-directional tible compared to ABO-compatible grafts, but specifically incompatible transplants. Three of the six received cells noted a higher proportion of deaths attributed to acute or from donors mismatched at one or more HLA class one chronic GvHD for patients receiving ABO-incompatible antigens and HLA mismatch was also a significant grafts. Analyses of registry data are conflicting, with predictor of graft failure in the multivariate analysis. Three Seebach et al.7 reporting severe (grades III/IV) aGvHD of the six experienced primary graft failure and secondary limited to recipients of bi-directional incompatible grafts, graft failures were seen in the other three. Kimura et al.2 and Kimura et al.2 reporting severe (grades III/IV) aGvHD also found, in univariate analysis, an increased risk of of the liver only for recipients of major or minor ABO-, but secondary graft failure for recipients of major, minor and not bi-directional incompatible grafts (Table 3a). bi-directional ABO-incompatible unrelated donor BMTs, None of these reports of ABO-incompatible transplants but this factor became insignificant in multivariate analysis described a change (increased or decreased) in the risk of controlling for other variables. In a small study of 27 developing chronic GvHD (Tables 3a and b). patients undergoing reduced-intensity conditioning with PBPC transplantation, Badros et al.52 reported mixed chimerism at day 30 for three of six patients receiving Does ABO incompatibility affect the risk of disease major ABO-incompatible grafts compared with 2 of 18 relapse? receiving compatible grafts (P ¼ 0.01). One of the incompatible graft recipients experienced engraftment failure. Most studies demonstrated no effect on this outcome of The small number of patients in the reports of engraftment transplantation, either an increase or a decrease in the risk failure raises the question as to the biological relevance of (Tables 3a and b)11,20,22,27 including studies of patient the statistically significant correlation. None of the other cohorts receiving truly non-myeloablative conditioning in analyses of engraftment after ABO-incompatible trans- which any graft-vs-tumor effect may be more evident.11,22 plantation have found a higher risk of this complication Erker et al.26 reported a significantly increased risk of and ABO incompatibility cannot, at this time, be cited as a relapse for recipients of minor or bi-directional incompa- cause of primary or secondary graft failure.2,11–15,22,23 tible grafts compared with ABO-compatible and major incompatible grafts, but did not specifically report a comparison of these groups to patients receiving ABO- Does ABO incompatibility change the risks of GvHD? compatible grafts. They attributed this observation to the small number of patients studied, and possibly not of Presumably, the expression of blood cell antigens on the biological relevance. Mehta et al.20 reported in univariate endothelial and epithelial tissues in the body could be a analysis (P ¼ 0.028) a decreased risk of relapse for target for the GvH response, resulting in a higher incidence recipients of ABO-mismatched grafts compared with of GvHD and risk of transplant-related complications for recipients of ABO-matched transplants. This factor became recipients of red cell-incompatible grafts, although such an nonsignificant in multivariate analysis (P ¼ 0.086), effect has not been consistently demonstrated.53 Bacigalupo although in contrast to the report by Erker et al.,26 et al.16 proposed a potential increased risk of acute GvHD recipients of minor and bi-directional incompatible grafts for recipients of minor ABO-incompatible transplants had the lowest risk of relapse, which translated into compared with that faced by recipients of ABO-compatible improved OS. Similarly, Worel et al.11 reported a non- grafts, and a decreased risk for recipients of major ABO- significant (P ¼ 0.056) trend to lower relapse in recipients of incompatible transplants. This hypothesis is based on an ABO-incompatible compared to ABO-compatible grafts, as expected recognition by donor lymphocytes carried in the did Blin et al.27 The effect may be more obvious in patients graft of ABO antigens expressed on non-hematopoietic being treated for acute leukemia. tissues of the host leading to the initiation or increasing the ultimate severity of acute GvHD (aGvHD). These authors reported the development of aGvHD in 82% of recipients Does ABO incompatibility affect non-relapse mortality? of minor ABO-incompatible BM grafts compared with 54% of recipients of ABO matched and 39% of recipients The infusion of ABO-incompatible grafts may result in of major-incompatible grafts (P ¼ 0.006), supporting this severe immediate or delayed transfusion reactions, and hypothesis. Rh incompatibility was not predictive. Keever- numerous patient deaths have been attributed to the Taylor et al.17 also observed a higher risk of moderate to immunohematopoietic consequences of ABO-incompatible severe (grades II–IV) aGvHD for minor ABO-incompatible transplantation. Despite this increased risk, a consistent recipients of T-cell-depleted grafts that should not have effect of ABO-incompatible transplantation on non-relapse contained ‘passenger lymphocytes’. mortality, however, has not been found (Tables 3a and b). Others have found a correlation with minor ABO- The use of dose-intense conditioning regimens could incompatible transplants only for severe (grades III and obscure detection of such an effect and three of the four IV) aGvHD2,28 or, in one study, only with clinically reports that limited analysis to patients receiving reduced- insignificant (grade I) aGvHD.20 A decreased risk of intensity conditioning found increased mortality for re- aGvHD for recipients of major ABO-incompatible grafts cipients of ABO-incompatible grafts.8,11,22 Worel et al.11 was not observed in the other cohort studies (Table 3b). reported an increased risk of non-relapse mortality in a Resnick et al.23 found a statistically nonsignificant trend to group of patients conditioned for transplantation with a

Bone Marrow Transplantation Table 3a Registry reports of clinical outcomes after ABO-incompatible transplantation

Author ABO match (N) Donor Conditioning regimen HPC Engraftment* aGvHD* cGvHD* Relapse* Non-relapse OS* Compatible source product Neutrophil II–IV mortality* Minor GvHD regimen Major Platelet III/IV Bi-directional

Seebach et al.7 2108 RD Myeloablative BM o0.001 (major NS NS NS NS NS (CIBMTR) 451 slower) 430 CNI+MTX 0.006 (bi-direct 114 No data only) Kollman et al.9 2860 URD Myeloablative and BM NS No data NS NS NS NS (NMDP) 1802 reduced intensity 1670 No data NS 587 Various

Kimura et al.2 2820 URD Myeloablative and BM 0.004 (major No data No data No data o0.0001 (major ¼ 0.016 (major Rowley SD transplantation ABO-incompatible (JMDP) 1202 reduced intensity slower) higher) 0.009 lower) 1384 o0.001 (major (minor higher) NS for minor 143 CNI+MTX o0.001 (major or minor higher, NS bi-direct or bi-direct slower) NS bi-direct) al et

Michallet et al.8 716 URD Reduced intensity PBPC No data NS NS No data o0.01 (minor P ¼ 0.001 (lower (SFGM-TC) 205 RD BM higher) minor vs 187 Cord No data No data compatible) Not stated CNI±MTX/ NS for major MMF+other agents NS for major

Abbreviations: CIBMTR ¼ Center for International Blood and Marrow Transplant Research; CNI ¼ calcineurin inhibitor; JMDP ¼ Japan Marrow Donor Program; MMF ¼ mycophenolate mofetil; N ¼ number of patients; NMDP ¼ National Marrow Donor Program; NS ¼ not significant; RD ¼ related donor; SFGM-TC ¼ Socie´te´Française de Greffe de Moe¨lle et The´rapie Cellulaire; URD ¼ unrelated donor. Shown are analyses of outcomes of ABO-incompatible transplants from registry databases. *P-value is given where a significant correlation with outcome was found. oeMro Transplantation Marrow Bone 1171 1172 oeMro Transplantation Marrow Bone

Table 3b Cohort studies of clinical outcomes after ABO incompatible transplantation

Author ABO match (N) Donor Conditioning regimen HPC Engraftment* aGvHD* II– cGvHD* Relapse* Non-relapse Overall survival* Compatible source product neutrophil IV mortality* Minor GvHD regimen Major Platelet III/IV Bi-directional

Wang et al.22 292 RD Non-myeloablative PBPC No data NS NS NS NS NS 102 URD BM 89 CNI+MMF NS No data 20 Canals et al.10 52 RD Reduced intensity PBPC NS NS NS NS NS NS 15 8 CNI+MTX 0.005 (major NS transplantation ABO-incompatible 2 slower)

Worel et al.11 21 RD Non-myeloablative PBPC NS NS NS 0.056 (lower for o0.05 (higher for NS 9 URD incompatible) incompatible) 8 CNI+MMF NS No data

2 Rowley SD Stussi et al.25 361 RD Myeloablative and BM NS NS No data NS No data 0.0009 (lower 98 URD reduced intensity PBPC bi-direct)

86 NS No data NS (major or al et 17 minor) CNI±MTX Erker et al.26 79 RD Myeloablative and PBPC NS NS NS 0.002 (higher No data 0.006 (higher 32 URD reduced intensity BM minor/bi-direct) minor/bi-direct) 21 NS NS 11 CNI+MTX NS (major) Mielcarek et al.1 960 RD Myeloablative BM NS NS No data No data No data NS 299 URD 314 CSA+MTX NS No data 103

Keever-Taylor 266 RD Myeloablative BM No data o0.001 NS No data No data NS et al.17 96 URD CNI+MP±ATG (higher 90 No data minor) 29 No data Bacigalupo 124 RD Myeloablative BM No data 0.003 No data No data No data No data et al.16 27 (higher 23 CNI or MTX or No data minor) Excluded CNI+MoAb No data Blin et al.27 395 URD Myeloablative and BM NS 0.05 for NS NS No data NS 0 RD reduced intensity PBPC PBPC only 337 Cord NS 77 CNI+MTX or MMF No data Table 3b Continued

Ozkurt et al.28 80 RD Myeloablative and PBPC NS NS NS NS o0.01 (higher 0.02 (lower 30 URD reduced intensity BM minor) minor) 25 NS 0.04 (higher 12 CNI+MTX minor) Benjamin et al.18 153 RD Myeloablative BM NS No data No data No data No data 0.003 major 55 URD 0.05 minor 62 CNI+MTX±steroids NS No data 0.5 bi-direct 22 (AML only) Helming et al.19 121 RD Myeloablative and BM No data NS No data NS No data NS 40 URD reduced intensity 40 No data No data 15 CNI±MTX or MTX Mehta et al.20 76 RD Myeloablative BM No data NS NS 0.039 (lower 0.048 (lower 0.004 (higher 27 minor/bi-direct incompat.) incompat.) incompat.) 16 major CNI±MTX No data No data No data Resnick et al.23 127 RD Reduced intensity PBPC NS NS NS NS 0.045 (minor NS 38 minor URD BM higher)

56 major/bi-direct CNI NS NS Rowley SD transplantation ABO-incompatible 0.023 (major higher) Kalaycioglu 134 Not Myeloablative BM No data NS NS No data NS (AML) NS for AML et al.29 30 stated ¼ 0.04 (CML, ¼ 0.05 (CML al et 35 CNI+MTX or steroid No data No data lower incompat.) higher incomp.) Not speciﬁed Klumpp et al.30 148 RD Myeloablative and PBPC No data NS NS No data NS NS 40 URD reduced intensity BM 39 No data No data 13 CNI±MTX or MMF or other Kim et al.12 49 RD Myeloablative and PBPC NS NS NS NS NS NS 15 reduced intensity 20 NS No data 5 CNI±MTX or MMF Goldman et al.31 84 RD No data BM No data No data No data No data No data NS 29 URD PBPC oeMro Transplantation Marrow Bone 35 No data No data No data 5

Rozman et al.32 139 RD No data BM o0.037 (slower NS NS No data NS NS 32 incompat.) 34 No data 13 NS 1173 ABO-incompatible transplantation SD Rowley et al 1174 non-myeloablative regimen containing single fraction TBI (200 cGy), a regimen of very low regimen-related toxicity perhaps allowing this increased risk to be detected. In contrast, however, Wang et al.,22 using the same regimen but with much larger patient cohorts, did not detect an

Overall survival* increase in non-relapse mortality. An increased risk of non- relapse mortality was reported by the Japanese Marrow Donor Program for recipients of major or minor incompatible grafts in a study that included both standard and reduced-intensity conditioning regimens.2 Similar results were not found for analyses from other registries.9,13 mortality*

Does ABO incompatibility affect survival?

unrelated donor. Any effect of ABO-incompatible transplantation on OS is, ¼ likewise, inconsistently detected, but reflects the experience with non-relapse mortality. The Japanese Marrow Donor Program found a decreased survival for recipients of major, but not minor or bi-directional incompatible grafts,2 and the French registry reported lower survival for recipients of minor, but not major incompatible grafts.8 A deleterious related donor; URD cGvHD* Relapse* Non-relapse No data No data No data No data effect has been more consistently (although in only a few of ¼ the cohort studies) reported for recipients of minor or bi- directional incompatible grafts,14,26,28 or in analysis limited to subgroups of patients.18 A specific attribution of patient death to the immunohematological consequences of ABO- aGvHD* II– IV No data No data incompatible transplantation has not been published. not significant; RD ¼ Does Incompatibility for Non-ABO Red Cell Antigens Affect Transplantation? Engraftment* neutrophil NS 0.013 (slower major/bi-direct) Non-ABO red cell antigens are much less likely to present an increased risk of immediate or delayed hemolysis because of the lack of naturally formed antibodies found

number of patients; NS in the ABO system. Healthy donors are very unlikely to product Cord blood ¼ have been exposed to red cell antigens through blood transfusions. However, like minor ABO-incompatible transplantation, donor lymphocytes of a different phenotype may produce alloantibodies on engraftment resulting in hemolysis, possibly severe, of red cells in the recipient, which would be worsened by the transfusion of incompa- MTX tible red cells during the initial post transplant period. The ± development of non-ABO red cell antibodies after trans- mycophenolate mofetil; N Conditioning regimen HPC CNI

¼ plantation is an uncommon event, occurring in about 2–9% of patients.6,54–58 de la Rubia et al.56 studied 217 allograft recipients and identiﬁed eight recipients (3.7%) who Donor source URD Myeloablative developed non-ABO alloantibodies after HSC transplanta-

) tion at a median of 23 days after transplantation (range, N 16–672 days). The type of conditioning regimen, GvHD prophylaxis or source of cells did not predict for this event, 27 29 21 18

MinorMajor GvHD regimen Plateletbut development III/IV of alloantibodies occurred for 2 of 156

Compatible (2.1%) patients receiving ABO-compatible grafts and for 5 Bi-directional calcineurin inhibitor; MMF

¼ of 62 (9.6%) of patients receiving ABO-incompatible grafts, suggesting an enhanced immunological effect in 21 . Continued the ABO-incompatible setting. Four of six patients for

et al whom antibody speciﬁcity was identiﬁed developed anti-

-value is given where a signiﬁcant correlation with outcome was found. bodies directed against antigens absent in both the donor P and recipient, indicating that alloantibodies can be devel- Shown are analyses of*The outcomes of ABO-incompatible transplants from cohort studies. Tomonari Author ABO match ( Table 3b Abbreviations: CNI oped by host or donor lymphocytes against red cell

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1175 antigens transfused during the peri-transplant period and plasma contents of these various graft sources be despite the profound immunosuppression of the transplant considered in the management of the individual patient. recipient. Abou-Elella et al.55 studied 192 patients, of whom 148 received autologous transplantation. Seven patients had alloantibodies before transplantation and an addi- Management of red cell-incompatible transplants tional four (2.1%) developed new alloantibodies. In this same population, 8 of 180 patients developed anti-HLA Major ABO-incompatible transplantation antibodies. The immediate complication of major ABO-incompatible Despite the potential development of alloantibodies after transplantation is the necessity of infusing incompatible transplantation, few patients experience clinically relevant RBCs into the recipient with the potential for serious and adverse effects, with individual case reports of hemolysis life-threatening immediate hemolytic reaction. Although being published.57,59 Erker et al.26 studied the effects of the volume of incompatible cells infused predicts the differences in ABO, Rh and Kell systems on transplant likelihood of symptomatic reaction,66 unfortunately there outcomes for 143 patients. Fifteen patients had major Rh- is no ‘safe’ quantity of red cells below which hemolytic incompatible donors and 17 minor Rh-incompatible transfusion reactions will not occur. In a review of ABO- donors. An increased red cell transfusion requirement was incompatible RBC transfusions that classified patients into found for recipients of Rh-incompatible grafts in the receiving more or less than 50 mL of incompatible RBCs, second month after transplantation, but there was no effect 16 of 36 patients receiving the larger volume experienced on the total quantity of red cells transfused through the signs or symptoms of a hemolytic reaction, 10 experienced transplant course. Patients who received Rh-incompatible acute renal failure and 6 patients died.66 For the 12 patients grafts (either major or minor) had a significantly decreased receiving the smaller volume of incompatible cells, no probability of OS, but no effect was found on neutrophil deaths were attributed to the inappropriate transfusion, engraftment, acute or chronic GvHD or relapse. Kell only three patients experienced signs or symptoms of mismatching had no effect on any of the studied transplant hemolysis, and no patient developed renal failure. The outcomes in this limited group of patients. In contrast, a relationship of isoagglutinin titers of the transfusion larger study of 258 patients including 18 major and 27 recipient to the likelihood or severity of adverse sequelae minor Rh-mismatched donors found no effect of Rh of incompatible red cell infusions was not described. incompatibility on OS, TRM or acute or chronic GvHD.60 Accordingly, allograft recipients can experience serious hemolytic reactions to even the small volume of RBCs, typically less than 10 mL,67 contained in a PBPC or a BM Does cell source affect outcomes of ABO-incompatible product depleted of red cells using apheresis technol- transplants? ogy.66,68,69 The risk of acute hemolytic reactions can be decreased by The available sources of HPC differ with regard to the either reducing the red cell content of the graft or the volume of RBCs and plasma contained, but also the isoagglutinin titers of the recipient. Apheresis devices and number and maturity of lymphocyte populations.61 BM hydroxyethylstarch or dextran sedimentation are com- and cord blood products mirror the peripheral blood of the monly used to remove red cells from marrow components donor, with only modest dilution of RBCs and plasma for major ABO incompatibility.62,69,70,71 Braine et al.69 were from salt solutions and anti-coagulants added during the able to reduce the red cell contents of BM grafts from a collection procedures. Accordingly, a 1000 mL BM product mean of 354±115 to 21.0±9.5 mL by processing with a will usually contain about 350–450 mL of RBCs.62 Simi- Haemonetics (Braintree, MA, USA) apheresis device with larly, cord blood units typically of about 100 mL total manual collection of buffy-coat cells. Nine of 15 patients volume may contain proportionately large volumes of transplanted experienced symptoms of a hemolytic transfu- RBCs, although pre-storage processing is frequently sion reaction including fever, chills and hypertension, performed to reduce the overall volume and red cell although the reactions may have been ameliorated by pre- content.63 PBPCs collected by apheresis techniques are infusion medication with diphenhydramine, hydrocortisone characterized by low volumes of RBCs, but approximately and mannitol. Larghero et al.,62 using the continuous flow a 10-fold greater number of lymphocytes including B- Cobe Spectra (CaridianBCT, Lakewood, CO, USA) device, lymphocytes that may increase the risk of delayed were able to reduce the red cell content of BM grafts from hemolytic transfusion reactions (and aGvHD).61,64 Unique 309.9 to 4 mL while retaining 82.2% of CD34 þ cells, and to cord blood transplantation is the lack of immunization noted infusion-related symptoms that were not related to of lymphocytes to red cell antigens reducing the risk of the volume of red cells infused or recipient isoagglutinin delayed hemolytic transfusion reactions from passenger titers. Secondary processing of a PBPC component already lymphocytes.65 The immunohematological consequences of containing low quantities of RBCs is not likely to be major ABO-incompatible transplants using cord blood cells effective in further reducing the quantity of RBCs or risks were similar to that seen with other sources of HPC, of infusion reactions, and will likely result in HPC loss. including greater RBC transfusion requirements. Pure RBC Loss of HSCs is the major risk of red cell depletion of the aplasia was not observed, but the number of patients was graft and one management approach is to require red cell small. No comparative studies for these cell sources have depletion only if the recipient has a high anti-donor been published describing the effects of ABO incompat- isoagglutinin titer (Figure 1).15 Alternately, the risk of ibility on transplant outcomes. It is important that the RBC acute hemolytic reactions may be lessened by reduction of

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1176 Major ABO incompatible

Infuse without modification BC < 20 mL R Monitor for acute hemolytic reaction PBSC . 20 mL 1:32 RBC RBC depletion of component, or . BM isoagglutinin depletion of recipient Recipient anti- Monitor for acute hemolytic reaction donor titer - 1:16 BM Infuse without modification Monitor for acute hemolytic reaction PBSC

Minor ABO incompatible

Plasma depletion of component . 1:256 Donor anti- Monitor for delayed recipient titer hemolytic transfusion - reaction 1:128 Infuse without modification Figure 1 A schema for management of ABO-incompatible HPC transplants. Shown are schema for major ABO incompatible (a) and minor ABO incompatable (b) transplants. The isoagglutinin titer that may be considered ‘safe’ for any particular patient is not defined and should not be assumed from this figure. Each transplant program should define procedures for the management of ABO-incompatible transplants.

recipient isoagglutinin titers by plasma exchange or more rapid disappearance of isoagglutinins for recipients of immunoadsorption,15,72 by infusion of donor-type RBCs,68 unrelated donor cells (median, 46 vs 61 days, P ¼ 0.016). or combinations of both.73 The reduction of recipient Also, recipients of related donor cells who developed acute isoagglutinins by plasmapheresis or immunadsorption GvHD had a 2.2-fold greater likelihood of reaching allows ABO-incompatible kidney (and other solid organ) undetectable titers within 100 days compared with patients transplantation, and has been used in conjunction with without GvHD (1.12–4.39, 95% confidence interval; splenectomy or rituximab infusions to prevent post P ¼ 0.02). Similar findings were reported by Lee et al.77 in transplant rebound of isoagglutinin titers.74,75 Multiple an analysis of 62 patients undergoing ABO-incompatible procedures are usually necessary and isoagglutinins may transplantation. Blin et al.27 found faster disappearance of rebound rapidly after treatment requiring post-transplant anti-donor IgM hemagglutinins in unrelated recipients of plasmapheresis, as well. Critical isoagglutinin titers for BM (median time, 36 vs 44 days, P ¼ 0.03) and patients survival of the kidney graft,75 or for avoidance of reactions with severe aGvHD (35 vs 59 days, P ¼ 0.001), but did not during HPC infusion (Figure 1), remain undefined. find this difference for recipients of PBPC. Titers may Plasmapheresis is associated with frequent minor toxicities persist longer for patients treated with non-myeloablative and increases the cost of transplantation.74 Rebound of conditioning regimens that would be less immunosuppres- isoagglutinins after HPC transplantation, while described, sive.78 For example, Wang et al.22 found only pre- does not appear to cause significant toxicity.15,76 transplant titers predicted for the rate of fall for patients conditioned with a non-myeloablative regimen based on single-fraction TBI. Pure red cell aplasia and resolution of anti-donor antibodies A small proportion of patients will experience a much The presence of recipient isoagglutinins directed against slower decay of anti-donor isoagglutinin titers, resulting in donor RBC antigens results in delayed recovery of red cell normal myelopoiesis and thrombopoiesis, but ineffective function (pure red cell aplasia, PRCA) that may persist for erythropoiesis. BM samples may show normal early months or years after transplantation, resulting in in- erythroid precursors despite ongoing transfusion require- creased transfusion requirements and transfusion-related ments. This observation is explained by the development of hemosiderosis. In the vast majority of patients undergoing blood group antigens at different points in erythropoiesis. major incompatible transplantation, there is rapid clear- In vitro studies, for example, demonstrated the expression ance of anti-donor isoagglutinins resulting in a slight of ABO and Kell antigens during the EPO-independent increase in RBC transfusion requirements compared with phase of colony growth, with other antigens such as Rh recipients of ABO-compatible transplants.76 The gradual developing during the later EPO-dependent phase and Jk disappearance of host anti-donor isoagglutinins after major and Lu antigens appearing last.79,80 ABO transplantation is likely a reflection of graft-vs-host The primary consequence of delayed red cell recovery is reactivity depleting host mature B-lymphocytes and plasma the need for continued red cell transfusions and the cells. In a review of 383 patients undergoing related development of transfusional hemachromatosis. The (n ¼ 155) or unrelated (n ¼ 228) major ABO- or bi-direc- risk appears to be higher in recipients of older age, a tional incompatible transplants, Mielcarek et al.1 found a donor of blood group A, and use of certain (but not all)

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1177 non-myeloablative conditioning regimens.78,81–85 The in- patients developing acute GvHD suggest that mechanisms cidence of PRCA is likely inﬂuenced by the degree of to enhance the graft-vs-host response may be effective in suppression of recipient B cells achieved by various factors the management of patients with red cell aplasia. Verholen including the intensity of the conditioning regimen, post- et al.97 and Ebihara et al.98 described the successful transplant immunomodulations such as rapid withdrawal treatment with repetitive doses of donor lymphocytes of of immunosuppressive medications and the occurrence of two patients who failed previous treatments with changes in GvHD. Stussi et al.73 proposed pre-transplant plasmapher- immunosuppressive treatment, high-dose EPO, plasma esis and post-transplant infusion of donor type RBCs to exchange and rituximab therapy. Lymphocyte infusions decrease the risk of developing PRCA by reducing the titer may result in the development of GvHD that may be more of anti-donor isoagglutinins after transplantation. Plasma- difﬁcult to manage and of higher risk to the patient than pheresis is not a completely effective preventive strategy; it continued transfusion support until eventual resolution of is possible for host anti-donor isoagglutinin titers to anti-donor isoagglutinins. increase after transplantation.15,86

Minor ABO-incompatible transplantation Treatment of PRCA The primary risk of minor ABO incompatibility is delayed Any case reports describing the effective treatment of red cell hemolysis resulting from the infusion of immuno- PRCA must be interpreted with caution in that anti-donor competent lymphocytes and plasma cells. Passive hemolysis isoagglutinins will fall spontaneously over time. EPO has from infusion of anti-recipient isoagglutinins in the HPC been reported to be effective in limited case reports in the product is an unlikely event, except for use of donors with treatment of patients with post-transplant red cell apla- high isoagglutinin titers or for recipients of small blood sia.87–89 The mechanism of response is not clear in that volume. Delayed hemolytic reactions may be fatal if PRCA appears to be a result of destruction of red cell appropriate blood transfusion support (avoiding the precursors as isoantigens become expressed and not a lack infusion of donor-type red cells) is not initiated before the of production. Others found no effect of EPO on the transplant (see section on transfusion support). Delayed transfusion requirements of patients with PRCA.78,81,90 The hemolytic transfusion reactions typically occur with infu- patients reported to have responded to erythropoietic sion of incompatible RBCs into a previously exposed agents appear to have generally had lower isoagglutinin patient, in whom the titer of the ensuing serum antibody titers than patients who did not respond, although a critical against the incompatible antigen has fallen below detect- titer under which erythropoietic stimulation is likely to be able levels,99 resulting in the production of an amnestic effective is not defined. response 7–10 days later, with a subsequent rise of antibody Plasma exchange will transiently decrease the titer of titer and brisk hemolysis. In the setting of minor ABO- isoagglutinins in the peripheral blood. However, plasma incompatible transplantation, it is the infused donor B- exchange will not affect the ongoing production of lymphocytes (‘passenger lymphocytes’) reactive to host red isoagglutinins and high-affinity antigens can result in cell antigens that are responsible for the production of anti- hemolysis even at low titers. Case reports of successful host isoagglutinins and hemolysis of host RBCs. The usual plasmapheresis are published.81,90 time course to the first appearance of rising titers and Rituximab is a monoclonal anti-CD20 monoclonal hemolysis is about 7–10 days after transplantation (Tables antibody that targets mature B-lymphocytes, but not 4a and b), and the hemolysis is usually abrupt in onset, precursor B cells or mature plasma cells. Effective in the indicating that this is not a result of passive transfer of treatment of malignant disease such as B-cell lymphoma, isoagglutinins in the plasma of the HPC product. One rituximab is being explored in the treatment of autoimmune report describes no evidence of delayed anti-host isoagglu- diseases including autoimmune hemolytic anemia. Case tinins after cord blood transplantation,65 which would be reports describing the treatment of patients with red cell expected in that isoagglutinins develop after birth as a aplasia after major ABO-incompatible transplantation result of exposure to sources of antigens other than have been reported.27,81,91,92 Although this response is RBCs.100 likely a result of depletion of B cells producing isoagglu- Several case reports of severe red cell hemolysis, tinins, the mechanism of action and the overall efficacy of including fatal cases, after minor ABO-incompatible HPC this treatment approach cannot be determined from these transplantation have been reported (Tables 4a and b).101–118 limited reports. Red cell transfusion requirements during the period of Polyclonal anti-thymocyte globulins can induce apopto- active hemolysis may exceed the calculated blood volume of sis of naı¨ve and memory B cells and terminally differ- the recipient,104,109,118 probably a result of passive adsorp- entiated plasma cells.93 Case reports of the successful use of tion of antigen from destroyed red cell onto the surface of both anti-lymphocyte globulin and anti-thymocyte globulin transfused group O cells, allowing immunological destruc- have been published.94–96 The pre-transplant administration of these transfused cells, as well. The majority of tion of anti-thymocyte globulin is reported to reduce the patients reported were of ABO group A with a donor of risks of GvHD and to improve survival, but reports in the group O, suggesting that the recipient of group A is at management of ABO-incompatible transplants have not higher risk of this complication of transplantation. The been published. majority of reported cases also involved use of CYA alone The more rapid clearance of host anti-donor isoaggluti- for post transplant immunosuppression leading to the nins observed for recipients of unrelated donor cells and recommendation that agents effective in preventing B-cell

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1178 Table 4a Case reports of delayed hemolysis after ABO minor-incompatible transplants

Author Donor/recipient ABO and gender Conditioning regimen HPC source GvHD regimen RBC support Day hemolysis

Lauarencet et al.113 O/A, F/M Myeloablative PBPC CSA Recipient 12 Hows et al.107 O+/AÀ, M/F Myeloablative BM CSA/MP Recipient NS O+/A+, M/F Myeloablative BM CSA/MP Recipient NS O+/B+, M/F Reduced intensity BM CSA/MP Donor NS O+/A+, F/M Myeloablative BM CSA/MP Recipient NS B+/AB+, M/M Reduced intensity BM CSA/MP O+/B+ NS Toren et al.116 O+/A+, F/M NS PBPC CSA NS 8 Gajewski et al.118 O/A, NS Myeloablative BM CSA Group O NS O/A, NS Myeloablative BM CSA Group O NS A/B, NS Myeloablative BM CSA/MP/IT Group O NS A/B, NS Myeloablative BM CSA/MP/IT Group O NS O/B, NS Myeloablative BM CSA/MP/IT Group O NS O/A, NS Myeloablative BM CSA/MP/IT Group O NS O/A, NS Myeloablative BM CSA/MP/IT Group O NS Oziel-Taieb et al.117 O+/A+, F/M Myeloablative PBPC CSA/MP Recipient 9 Noborio et al.109 O+/B+, M/M Reduced intensity PBPC CSA Donor 8 Reed et al.105 O+/A+, F/M Myeloablative PBPC CSA Donor 8 Greeno et al.104 O+A+, F/F Myeloablative BM Fk Donor 8 Curtin and Schwarer106 O+/A+, M/M Reduced intensity PBPC CSA NS 11 Nair et al.108 O+/A+, F/F Myeloablative PBPC CSA/MTX NS 12 Worel et al.119 O/A, M/M Myeloablative PBPC CSA Donor 9 O/A, M/F Reduced intensity PBPC CSA/MMF Donor 10 O/A, M/M Reduced intensity PBPC CSA/MMF Donor 7 O/A, M/M Reduced intensity PBPC CSA/MMF Donor 8 Hoegler et al.103 O+/A+, M/M Myeloablative BM CSA/ATG NS 9 Bolan et al.115 O+/A+, F/M Reduced intensity PBPC CSA Donor 7 O+/B+, F/F Myeloablative PBPC CSA Donor 7 O+/A+, F/M Reduced intensity PBPC CSA Donor 10 Tiplady et al.109 O+/A+, F/M Myeloablative PBPC CSA Donor 9 Lee et al.111 O+/A+, NS/F Reduced intensity BM ATG/Fk/MTX Donor 6 Salmon et al.112 O+/A+, F/M Myeloablative PBPC CSA Donor 7 Moog et al.102 NS, M/M NS PBPC CSA NS 7 Bornhauser et al.114 B+/A+, NS/M Myeloablative PBPC CSA/MTX Group O 5

Abbreviations: F ¼ female; Fk ¼ tacrolimus; HPC ¼ hematopoietic progenitor cell; M ¼ male; MMF ¼ mycophenolate mofetil; NS ¼ not stated; Rh ¼ rhesus. Shown are individual case reports of hemolysis after ABO minor or bi-directional transplantation. Donor and patient ABO, Rh type and gender are given, when reported. Also given is RBC transfusion support provided to the recipient before hemolysis occurred.

Table 4b Case reports of delayed hemolysis after non-ABO minor-incompatible transplants

Author Donor/recipient Conditioning HPC GvHD RBC Antibody detected Day red cell phenotype regimen source regimen support (anti-) hemolysis

Young et al.59 A+/BÀ,JkaÀ/Jka+ CyTBI BM Not stated Not stated Jka 9 O+/O+, JkaÀ/Jka+ CyTBI PBPC Not stated Not stated Jka, RhE 11 Adams et al.34 AÀ/O+, FluMel PBPC CSA/MMF OÀ RhD 8 Leo et al.6 A+/A+, JkaÀ/Jka+ CyTBI PBPC CSA/MTX Not stated Jka 17 Hows et al.107 AÀ/A+ CyALG BM CSA/MP/Azathioprine AÀ RhD Not stated

Abbreviations: HPC ¼ hematopoietic progenitor cell; MMF ¼ mycophenolate mofetil; MP ¼ methylprednisolone; Rh ¼ rhesus. Shown are individual case reports of hemolysis after minor or bi-directional transplantation from donors incompatible for non-ABO RBC antigens. Donor and patient ABO, Rh type and Kidd phenotype are given, where applicable. Also given is RBC transfusion support provided to the recipient before hemolysis occurred. The ﬁrst day hemolysis was recognized is given.

proliferation would be protective. The choice of condition- post-grafting immunosuppression with CYA and myco- ing regimen should not have an effect on the incidence or phenolate mofetil, and the fourth patient was one of two severity of delayed hemolytic reactions, except that who received CYA alone after myeloablative conditioning. conditioning regimens are often linked to particular post- None of the 13 patients who received a myeloablative transplant immunosuppression regimens. Worel et al.119 conditioning regimen followed by post-grafting immuno- described 4 of 25 patients experiencing hemolysis after suppression with an MTX-containing regimen experienced PBPC transplantation, and noted that three of the patients hemolysis. received non-myeloablative conditioning regimens. How- It has been suggested that use of a reduced-intensity ever, these three patients were in a group of 10 who received conditioning regimen and PBPC with its larger quantities

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1179 of lymphocytes as the source of HPC is associated with a developing a robust anti-host response appears to be much higher risk of passenger lymphocyte syndrome. Analysis of reduced by administration of regimens that include donors and recipients of a multi-center study comparing MTX.118 However, awareness of possibility of a delayed outcomes of transplants for recipients of either PBPC or transfusion reaction is important for proper management BM grafts found increased isoagglutinin titers after minor of hemolysis that may be brisk and life-threatening. compared with ABO-compatible or major ABO-mis- Clinicians must be vigilant to recognize the onset of matched PBPC grafts (5 of 7 vs 3 of 14), but no difference hemolysis and immediately implement therapy to maintain for BM recipients (1 of 8 vs 3 of 20).120 This group also an appropriate hemoglobin level and avoid renal failure. reported that 3 of 21 PBPC recipients compared with 0 of Management of hemolysis is usually supportive care and 28 BM recipients developed non-ABO antibodies. It is not transfusion of group O red cells or red cells compatible with yet clear that delayed-type transfusion reactions will be the donor. Rituximab has been given to one patient with more likely or of greater severity in PBPC recipients and severe hemolysis.111 the frequent case reports reporting this event after PBPC Passive hemolysis directly resulting from the infusion of transplantation may be a result of publication bias. The HPC products from donors with high anti-host isoagglu- risk of delayed-type reaction may be related to the use of tinin titers may occur if large volumes of plasma are infused post transplant immunosuppressive regimens not contain- into the recipient. Clinically significant hemolysis is a rare, ing MTX.118 The Seattle transplant program found no but potentially severe complication of administering an delayed transfusion reactions in a large retrospective review ABO-mismatched platelet transfusion, particularly for of patients undergoing PBPC transplantation and who single donor platelet products containing a large volume received CYA and MTX.76 of plasma collected from high-titer group O donors.122 Individual case reports of non-ABO delayed hemolysis Unfortunately, a safe titer for platelet donors under which have been published, describing the post transplant passive hemolysis will not occur has not been defined and development of anti-D, anti-K, and anti-Jka or various technologies are used in measuring isoagglutinin -Jkb.6,34,54,58,102 Leo et al.6 described severe hemolysis titers, further complicating analysis of risk.122 In their occurring 17 days after ABO/Rh-compatible transplant- review of passive hemolysis after platelet transfusion, ation for a Jk(aÀb þ ) recipient receiving cells from a Josephson et al.122 collated case reports for 11 pediatric Jk(a þ b þ ) donor. Testing confirmed the presence of and 20 adults, all but two of whom were A or AB recipients alloantibody of anti-Jka specificity at a titer of 1:8. Young of group O platelets with a wide range of isoagglutinin et al.59 described two patients who developed alloantibodies titers. In a separate report, that group proposed defining of anti-Jka specificity after transplantation. high titers as being equal or greater than 1:64 for IgM isoagglutinins and 1:256 for IgG isoagglutinins.123 Addi- tional risk factors for passive hemolysis include a larger Management of minor ABO-incompatible transplants total volume of the platelet product infused, a smaller It is important that patients do not receive red cell blood volume of the recipient and group O donor with a transfusions of host type after transplantation with minor group A or AB recipient.124 In addition to giving attention ABO-incompatible grafts. It must be understood that the to post-transplant immunosuppression regimens, an appro- appropriate red cell type to be transfused cannot be priate management strategy for transplantation of minor determined by blood bank testing techniques—ABO typing ABO-incompatible HSC products is to reduce the volume will only identify host red cell type, not that of the donor, of plasma in the products being infused to avoid the risk of early after transplantation. Therefore, the blood bank must passive hemolysis, particularly for donors with high anti- be notified of the ABO-incompatible transplant in order host red cell isoagglutinin titers, and pediatric patients with that appropriate blood component support can be ar- small blood volumes. Particular attention should be given ranged. Transplant programs performing allogeneic trans- to recipients of blood group A or AB receiving HPC plantation must, as part of their quality program, have in products from a group O donor. Each transplant center place an error-proof mechanism for such notification. should define limits of donor isoagglutinin titer/plasma Pre-transplant red cell exchange should reduce the risk of volume permissible in a minor ABO-incompatible HPC severe hemolysis by reducing the quantity of target cells. product (Figure 1). Worel et al.121 described a lower incidence of delayed hemolysis at their one center after introducing a policy of pre-transplant red cell exchange. Only 1 of 20 patients Transfusion support for red cell-incompatible treated in this manner showed minor evidence of hemolysis. transplantation Interestingly, the authors also reported lower TRM (16 vs 53%, Po0.05) and an improved survival at 1 year after Transfusion support of the red cell-incompatible HSC transplantation (65 vs 40%, Po0.05) compared with a recipient must consider both the immunohematology of the historical control group not treated with red cell exchange. recipient and the donor.125 The reports of hemolysis Gajewski et al.118 similarly described less brisk period of involving non-ABO red cell antigens emphasize the hemolysis with lower red cell transfusion requirements for importance of screening both donor and recipient for the patients who underwent pre-transplant red cell exchange. presence of other antibodies and providing appropriate Post-transplant immunosuppressive regimens containing blood component support before and after transplantation. an anti-B-cell proliferative agent appear important in Further investigation of positive antibody screens for either preventing delayed hemolytic reactions, and the risk of the donor or recipient should be performed to determine

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1180 donor–patient red cell compatibility before transplantation, One group described favorable results of transfusing and patient records reviewed for a history of red cell only ABO-identical red cells and platelet products for antibodies. Routine donor–recipient serological cross- patients undergoing autologous or allogeneic transplanta- match testing for most healthy donors is unlikely to tion, claiming improved survivals compared with results improve patient safety,126–128 but each transplant program published by other transplant programs.133 A retrospective should establish policies regarding the appropriate testing study of pediatric patients found a correlation between of the donor–recipient pair. The transplant program ABO-incompatible plasma infusions and hepatic sinusoidal requesting blood components must have a mechanism syndrome, possibly a consequence of expression of ABO alerting the transfusion service of the need to use antigens in the liver.134 Multivariate analysis found only components that would not be first choice in the non- two factors that predicted for this event, use of melphalan transplant setting, but that may be critical for the patient in the pre-transplant conditioning (P ¼ 0.006) and trans- undergoing HPC transplantation. fusion of platelet concentrates containing ABO-incompa- The recipient should not receive incompatible red cells tible plasma (P ¼ 0.003). before or after transplantation, except as part of an intentional management strategy (for example, major ABO incompatibility); for red cell-incompatible trans- Transplant program standards plants, red cell compatibility with both the recipient and the donor must be accommodated to avoid both immediate Possible RBC disparities and the management of these and delayed hemolytic reactions. In general, red cell (or must be part of the evaluation of donors and patients being granulocyte) support of the major incompatible recipient offered allogeneic HPC transplantation. Published stan- should be of recipient type and of the donor type for minor dards for transplant programs such as those by the incompatible recipients (Table 5). For most patients, this Foundation for the Accreditation of Cellular Therapies translates into using group O red cells. Plasma product require ABO typing of both patients and donors, but do support (platelets, fresh–frozen plasma) is the opposite, not specifically require that blood group disparity be a with donor type being used for major incompatible selection criterion for proceeding to transplantation.135 The recipients and recipient type for minor incompatible Societe Francaise de Greffe de Moelle realized the existing recipients. The infusion of RhD-incompatible blood heterogeneity of transfusion practices and recognized the products will induce the development of RhD antibodies importance of standard management techniques for pa- in a small proportion of patients, possibly with an increased tients enrolled into a clinical study comparing PBPC and risk after non-myeloablative conditioning.129–131 Although marrow transplantation.125 For this study, patients and the incidence of antibody development is low, leading some donors underwent comprehensive phenotyping of red cell authors to suggest that use of RhD-incompatible blood antigens including Rh, Kell and Lewis, and required components is acceptable after transplantation, clinically transfusion support with red cells compatible to both relevant hemolysis has been reported. The administration donor and recipient beginning at day 0, and earlier if of appropriate doses of Rh immunoglobulin (20 mgofRh possible. Post-transplant identification and titration of immunoglobulin per 1 mL of RBCs given within 72 h of isoagglutinins was also required. Donors with ABO exposure is suggested) can prevent immunization of Rh- isoagglutinin titers greater than 1:32 or antibodies to other negative patients exposed to Rh-positive products.132 red cell antigens present on recipient red cells were excluded from this study. A survey by Gruppo Italiano Trapianto di Midollo Osseo of 34 transplant centers identified a variety of pre- and post-transplant testing protocols and trans- Table 5 Recommended transfusion support for recipients of ABO- incompatible HPC components fusion practices including some centers that allowed transfusion of recipient-type red cells after minor ABO- Recipient Donor RBC and Platelet and incompatible transplants.136 granulocyte plasma components components An appropriate quality management plan should address the possibility of red cell incompatibility in donor selection ABO major O A O A, AB policies. The transplant program may elect to modify OB O B,ABconditioning regimens, stem cell source, cell processing and OABO AB post-transplant immunosuppressive regimens for recipients AABA,OAB BABB,OAB of red cell-disparate HPC products, based on the deleter- ABO minor A O O A, AB ious effects that may be otherwise experienced. BO O B,AB The quality management procedures must also have a AB O O AB failsafe mechanism whereby the blood bank is notified of AB A A, O AB AB B B, O AB allogeneic HPC recipients in order that appropriate ABO major and minor A B O AB transfusion practices can be established for the individual BA O AB patient. In the minor ABO setting, typing of the patient will identify only patient red cells, and transfusion of red cells of Red cell components should be packed or washed to reduce plasma host type may greatly exacerbate the severity of delayed volume, or components from donors with low isoagglutinin titers. Shown are the first choice for platelet support; platelets from other blood groups hemolysis by passenger lymphocytes. In this setting, only can also be infused, but attention should be given to donor isoagglutinin by appropriate notification will the blood bank be aware of titers. the need to supply blood products of appropriate type.

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1181 Conflict of interest 13 Gale RP, Feig S, Ho W, Falk P, Rippee C, Sparkes R. ABO blood group system and bone marrow transplantation. Blood The authors declare no conflict of interest. 1977; 50: 185–194. 14 Buckner CD, Clift RA, Sanders JE, Williams B, Gray M, Storb R et al. ABO-incompatible marrow transplants. Transplantation 1978; 26: 233–238. Acknowledgements 15 Bensinger WI, Buckner CD, Thomas ED, Clift RA. ABO- incompatible marrow transplants. Transplantation 1982; 33: We thank Carolyn Magnani for her dedicated and skilled 427–429. assistance in preparing this manuscript. 16 Bacigalupo A, Van Lint MT, Occhini D, Ferrari G, Pittaluga PA, Frassoni F et al. ABO compatibility and acute graft-versus-host disease following allogeneic bone marrow transplantation. Transplantation 1988; 45: 1091–1094. References 17 Keever-Taylor CA, Bredeson C, Loberiza FR, Casper JT, Lawton C, Rizzo D et al. Analysis of risk factors for the 1 Mielcarek M, Leisenring W, Torok-Storb B, Storb R. Graft- development of GvHD after T cell-depleted allogeneic BMT: versus-host disease and donor-directed hemagglutinin titers effect of HLA disparity, ABO incompatibility, and method after ABO-mismatched related and unrelated marrow allo- of T-cell depletion. Biol Blood Marrow Transplant 2001; 7: grafts: evidence for a graft-versus-plasma cell effect. Blood 620–630. 2000; 96: 1150–1156. 18 Benjamin RJ, McGurk S, Ralston MS, Churchill WH, Antin 2 Kimura F, Sato K, Kobayashi S, Ikeda T, Sao H, Okamoto S JH. ABO incompatibility as an adverse risk factor for et al. Impact of ABO-blood group incompatibility on the survival after allogeneic bone marrow transplantation. outcome of recipients of bone marrow transplants from Transfusion 1999; 39: 178–187. unrelated donors in the Japan Marrow Donor Program. 19 Helming AM, Brand A, Wolterbeek R, van Tol MJ, Egeler Haematologica 2008; 93: 1686–1693. RM, Ball LM. ABO incompatible stem cell transplantation in 3 Logdberg L, Reid ME, Zelinski T. Human blood group genes children does not influence outcome. Pediatr Blood Cancer 2010: chromosomal locations and cloning strategies revisited. 2007; 49: 313–317. Transfus Med Rev 2011; 25: 36–46. 20 Mehta J, Powles R, Sirohi B, Treleaven J, Kulkarni S, Saso R 4 Ried ME, Lomas-Francis C. The Blood Group Antigen Facts et al. Does donor–recipient ABO incompatibility protect Book, 2nd edn. Elsevier Academic Press: New York, 2004. against relapse after allogeneic bone marrow transplantation 5 Franchini M, de Gironcoli M, Gandini G, Vassanelli A, in first remission acute myeloid leukemia? Bone Marrow Rocca P, Benedetti F et al. Transmission of an anti-RhD Transplant 2002; 29: 853–859. alloantibody from donor to recipient after ABO-incompatible 21 Tomonari A, Takahashi S, Ooi J, Tsukada N, Konuma T, BMT. Bone Marrow Transplant 1998; 21: 1071–1073. Kobayashi T et al. Impact of ABO incompatibility on 6 Leo A, Mytilineos J, Voso MT, Weber-Nordt R, Liebish P, engraftment and transfusion requirement after unrelated cord Lensing C et al. Passenger lymphocyte syndrome with severe blood transplantation: a single institute experience in Japan. hemolytic anemia due to an anti-Jka after allogeneic PBPC Bone Marrow Transplant 2007; 40: 523–528. transplantation. Transfusion 2000; 40: 632–636. 22 Wang Z, Sorror ML, Leisenring W, Schoch G, Maloney DG, 7 Seebach JD, Stussi G, Passweg JR, Loberiza FR, Gajewski Sandmaier BM et al. The impact of donor type and ABO JL, Keating A et al. ABO blood group barrier in allogeneic incompatibility on transfusion requirements after nonmye- bone marrow transplantation revisited. Biol Blood Marrow loablative haematopoietic cell transplantation. Br J Haematol Transplant 2005; 11: 1006–1013. 2010; 149: 101–110. 8 Michallet M, Le Q-H, Mohty M, Prebet T, Nicolini R, 23 Resnick IB, Tsirigotis PD, Shapira MY, Aker M, Bitan M, Boiron JM et al. Predictive factors for outcomes after Samuel S et al. ABO incompatibility is associated with reduced intensity conditioning hematopoietic stem cell increased non-relapse and GvHD mortality in patients with transplantation for hematological malignancies: a 10-year malignancies treated with a reduced intensity regimen: a retrospective analysis from the Societe Francaise de Greffe de single center experience of 221 patients. Biol Blood Marrow Moelle et de Therapie Cellulaire. Exp Hematol 2008; 36: Transplant 2008; 14: 409–417. 535–544. 24 Storb R, Prentice RL, Thomas ED. Marrow Transplantation 9 Kollman C, Howe CW, Anasetti C, Antin JH, Davies SM, for treatment of aplastic anemia. An analysis of Filipovich AH et al. Donor characteristics as risk factors factors associated with gaft rejection. N Engl J Med 1977; in recipients after transplantation of bone marrow from 296: 61–66. unrelated donors: the effect of donor age. Blood 2001; 98: 25 Stussi G, Muntwyler J, Passweg JR, Seebach L, Schanz U, 2043–2051. Gmur J et al. Consequences of ABO incompatibility in 10 Canals C, Muniz-Diaz E, Martinez C, Martino R, Moreno I, allogeneic hematopoietic stem cell transplantation. Bone Ramos A et al. Impact of ABO incompatibility on allo- Marrow Transplant 2002; 30: 87–93. geneic peripheral blood progenitor cell transplantations 26 Erker CG, Steins MB, Fischer R-J, Kienast W, Berdel WE, afterreduced intensity conditioning. Transfusion 2004; 44: Sibrowski W et al. The influence of blood group differences in 1603–1611. allogeneic hematopoietic peripheral blood progenitor cell 11 Worel N, Kalhs P, Keil F, Prinz E, Moser K, Schulenberg A transplantation. Transfusion 2005; 45: 1382–1390. et al. ABO mismatch increases transplant-related morbidity 27 Blin N, Traineau R, Houssin S, Peffault de LaTour R, and mortality in patients given nonmyeloablative allogeneic Petropoulou A, Robin M et al. Impact of donor–recipient HPC transplantation. Transfusion 2003; 43: 1153–1161. ABO mismatch on allogeneic transplantation outcome 12 Kim JG, Sohn SK, Kim DH, Baek JH, Lee KB, Min WS according to stem cell source. Biol Blood Marrow Transplant et al. Impact of ABO incompatibility on outcome after 2010; 16: 1315–1323. allogeneic peripheral blood stem cell transplantation. Bone 28 Ozkurt ZN, Yegin ZA, Yenicesu I, Aki SZ, Yagci M, Sucak Marrow Transplant 2005; 35: 489–495. GT. Impact of ABO-incompatible donor on early and late

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1182 outcome of hematopoietic stem cell transplantation. Trans- transplant rejection in a mouse model. Blood 2009; 114: plant Proc 2009; 41: 3851–3858. 2315–2322. 29 Kalacioglu M, Copelan E, Avalos B, Klein J, Goormastic M, 45 Patel SR, Cadwell CM, Medford A, Zimring JC. Transfusion Bolwell B. Survival after ABO-incompatible allogeneic bone of minor histocompatibility antigen-mismatched platelets marrow transplant after a preparative regimen of busulfan induces rejection of bone marrow transplants in mice. J Clin and cyclophosphamide. Bone Marrow Transplant 1995; 15: Invest 2009; 119: 2787–2794. 105–110. 46 Bean MA, Storb R, Graham T, Raff T, Sale GE, Schuening F 30 Klummp TR, Herman JH, Ulicny J, Emmons RVB, Martin et al. Prevention of transfusion-induced sensitization to minor ME, Mangin KF. Lack of effect of donor–recipient ABO histocompatibility antigens on DNA-identical canine marrow mismatching on outcome following allogeneic hematopoietic grafts by gamma irradiation of marrow donor blood. stem cell transplantation. Bone Marrow Transplant 2006; 38: Transplantation 1991; 52: 956–960. 615–620. 47 Bean MA, Graham T, Appelbaum FR, Deeg HJ, Schuening 31 Goldman J, Liesveld J, Nichols D, Heal J, Blumberg N. ABO F, Sale GE et al. Gamma-irradiation of pretransplant blood incompatibility between donor and recipient and clinical transfusions from unrelated donors prevents sensitization to outcomes in allogeneic stem cell transplantation. Leuk Res minor histocompatibility antigens on dog leukocyte antigen- 2003; 27: 489–491. identical canine marrow grafts. Transplantation 1994; 57: 32 Rozman P, Kosir A, Bohinjec M. Is the ABO incompatibility 423–426. a risk factor in bone marrow transplantation? Transplant 48 Rozman P, Karas M, Kosir A, Labar B, Madrigal A, Immunol 2005; 14: 159–169. Middleton D et al. Are human platelet alloantigens (HPA) 33 Kanda J, Ichinohe T, Matsuo K, Benjamin RJ, Klumpp TR, minor transplantation antigens in clinical bone marrow Rozman P et al. Impact of ABO mismatching on the transplantation? Bone Marrow Transplant 2003; 31: 497–506. outcomes of allogeneic related and unrelated blood and 49 Cooling LL, Kelly K, Barton J, Hwang D, Koerner TA, marrow stem cell transplantations for hematologic malig- Olson JD. Determinants of ABH expression on human blood nancies; IPD-based meta-analysis of cohort studies. Transfu- platelets. Blood 2005; 105: 3356–3364. sion 2009; 49: 624–635. 50 Weissinger F, Sandmaier BM, Maloney DG, Bensinger WI, 34 Adams BR, Miller AN, Costa LJ. Self-limited hemolysis due Golley T, Storb R et al. Decreased transfusion requirements to anti-D passenger lymphocyte syndrome in allogeneic for patients receiving nonmyeloablative compared with hematopoietic stem cell transplantation. Bone Marrow conventional peripheral blood stem cells transplants from Transplant 2010; 45: 772–773. HLA-identical siblings. Blood 2001; 98: 3584–3588. 35 Loren AW, Bunin GR, Boudreau C, Champlin RE, Cnaan A, 51 Remberger M, Watz E, Ringden O, Mattson J, Shanwell A, Horowitz MM et al. Impact of donor and recipient sex and Wikman A. Major ABO blood group mismatch increases the parity on outcomes of HLA-identical sibling allogeneic risk for graft failure after unrelated donor hematopoietic stem hematopoietic stem cell transplantation. Biol Blood Marrow cell transplantation. Biol Blood Marrow Transplant 2007; 13: Transplant 2006; 12: 758–769. 675–681. 36 Daniel-Johnson J, Leitman S, Klein H, Alter H, Lee-Stroka 52 Badros A, Tricot G, Toor A, Morris C, Guo C, Munshi N A, Scheinberg P et al. Probiotic-associated high-titer anti-B et al. ABO mismatch may affect engraftment in multiple in a group A platelet donor as a cause of severe hemolytic myeloma patients receiving nonmyeloablative conditioning. transfusion reactions. Transfusion 2009; 49: 1845–1849. Transfusion 2002; 42: 205–209. 37 Olsson MJ, Chester MA. Polymorphism and recombination 53 Ravn V, Dabelsteen E. Tissue distribution of histo-blood events at the ABO locus: a major challenge for genomic group antigens. APMIS 2000; 108: 1–28. ABO blood grouping strategies. Transfus Med 2001; 11: 54 Ting A, Pun A, Dodds AJ, Atkinson K, Biggs JC. Red cell 295–313. alloantibodies produced after bone marrow transplantation. 38 Eiz-Vesper B, Seltsam A, Blasczyk R. ABO glycosyltrans- Transfusion 1987; 27: 145–147. ferases as potential source of minor histocompatibility 55 Abou-Elella AA, Camarillo TA, Allen MB, Barclay S, Pierce antigens in allogeneic peripheral blood progenitor cell JA, Holland HK et al. Low incidence of red cell and HLA transplantation. Transfusion 2005; 45: 960–968. antibody formation by bone marrow transplant patients. 39 Daniels G. Functions of red cell surface proteins. Vox Sang Transfusion 1995; 35: 931–935. 2007; 93: 331–340. 56 de la Rubia J, Arriaga F, Andreu R, Sanz G, Jimenez C, 40 Cartron JP, Colin Y. Structural and functional diversity of Vicente A et al. Development of non-ABO RBC alloanti- blood group antigens. Transfus Clin Biol 2001; 8: 163–199. bodies in patients undergoing allogeneic HPC transplanta- 41 Weiden PL, Storb R, Thomas ED, Graham TC, Lerner KG, tion. Is ABO incompatibility a predisposing factor? Buckner CD et al. Preceding transfusions and marrow Transfusion 2001; 41: 106–110. graft rejection in dogs and man. Transplant Proc 1976; 8: 57 Zupanska B, Saucha MJ, Michalewska B, Malinowska A, 551–554. Brojer E, Hellmann A. Multiple red cell alloantibodies, 42 Storb R, Thomas ED, Buckner CD, Clift RA, Deeg HJ, Fefer including anti-Dib, after allogeneic ABO-matched peripheral A et al. Marrow transplantation in thirty ‘untransfused’ blood progenitor cell transplantation. Transfusion 2005; 45: patients with severe aplastic anemia. Ann Intern Med 1980; 16–20. 92: 30–36. 58 Lopez A, de la Rubia J, Arriaga F, Jimenez C, Sanz GF, 43 Storb R, Weiden PL, Deeg HJ, Graham TC, Atkinson K, Carpio N et al. Severe hemolytic anemia due to multiple red Slichter SJ et al. Rejection of marrow from DLA-identical cell alloantibodies after an ABO-incompatible allogeneic canine littermates given transfusions before grafting: antigens bone marrow transplant. Transfusion 1998; 38: 247–251. involved are expressed on leukocytes and skin epithelial 59 Young PP, Goodnough LT, Westervelt P, Dipersio JF. cells but not on platelets and red blood cells. Blood 1979; 54: Immune hemolysis involving non-ABO/RhD alloantibodies 477–484. following hematopoietic stem cell transplantation. Bone 44 Desmerets M, Cadwell CM, Peterson KR, Neades R, Zimring Marrow Transplant 2001; 27: 1305–1310. JC. Minor histocompatibility antigens on transfused leukor- 60 Wirk B, Klumpp TR, Ulicny J, Herman JH, Gajewski JL, educed units of red blood cells induce bone marrow Martin ME et al. Lack of effect of donor–recipient Rh

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1183 mismatch on outcomes after allogeneic hematopoietic stem antigen-specific immunoadsorption and rituximab—insights cell transplantation. Transfusion 2008; 48: 163–168. and uncertainties. Contrib Nephrol 2009; 162: 47–60. 61 Korbling M, Huh YO, Durett A, Mirza N, Miller P, Engel P 76 Rowley SD, Liang PS, Ulz L. Transplantation of ABO- et al. Allogeneic blood stem cell transplantation: peripher- incompatible bone marrow and peripheral blood stem cell alization and yield of donor-derived primitive hematopoietic components. Bone Marrow Transplant 2000; 26: 749–757. progenitor cells (CD34+Thy-1dim) and lymphoid subsets, 77 Lee JH, Lee JH, Choi SJ, Kim S, Seol M, Kwon SW et al. and possible predictors of engraftment and graft-versus-host Changes of isoagglutinin titres after ABO-incompatible disease. Blood 1995; 86: 2842–2848. allogeneic stem cell transplantation. Br J Haematol 2003; 62 Larghero J, Rea D, Esperou H, Biscay N, Maurer MN, 120: 702–710. Lacassangne M et al. ABO-mismatched marrow processing 78 Bolan CD, Leitman SF, Griffith LM, Wesley RA, Procter JL, for transplantation: results of 114 procedures and analysis of Stroncek DR et al. Delayed donor red cell chimerism and immediate adverse events and hematopoietic recovery. pure red cell aplasia following major ABO-incompatible Transfusion 2006; 46: 398–402. nonmyeloablative hematopoietic stem cell transplantation. 63 Rubinstein P, Dobrila L, Rosenfield RE, Adamson JW, Blood 2001; 98: 1687–1694. Migliaccio G, Migliaccio AR et al. Processing and cryopre- 79 Southcott MJ, Tanner MJ, Anstee DJ. The expression of servation of placental/umbilical cord blood for unrelated human blood group antigens during erythropoiesis in a cell bone marrow reconstitution. Proc Natl Acad Sci USA 1995; culture system. Blood 1999; 93: 4425–4435. 92: 10119–10122. 80 Bony V, Gane P, Bailly P, Cartron JP. Time-course 64 Rowley SD, Prather K, Bui KT, Appel M, Felt T, Bensinger expression of polypeptides carrying blood group antigens WI. Collection of peripheral blood progenitor cells with during human erythroid differentiation. Br J Haematol 1999; an automated leukapheresis system. Transfusion 1999; 39: 107: 263–274. 1200–1206. 81 Helbig G, Stella-Holowiecka B, Krawczyk-Kulis M, Wojnar 65 Snell M, Chau C, Hendrix D, Fox R, Downes KA, Creger R J, Markiewicz M, Wojciechowska-Sadus M et al. Successful et al. Lack of isohemagglutinin production following minor treatment of pure red cell aplasia with repeated, low doses of ABO incompatible unrelated HLA mismatched umbilical rituximab in two patients after ABO-incompatible allogeneic cord blood transplantation. Bone Marrow Transplant 2006; haematopoietic stem cell transplantation for acute myeloid 38: 135–140. leukemia. Haematologica 2005; 90: e93–e96. 66 Janatpour KA, Kalmin ND, Jensen HM, Holland PV. 82 Dahl D, Hahn A, Koenecke C, Heuft HG, Dammann E, Clinical outcomes of ABO-incompatible RBC transfusions. Stadler M et al. Prolonged isolated red blood cell transfusion Am J Clin Pathol 2008; 129: 276–281. requirement after allogeneic blood stem cell transplantation: 67 Stroncek DF, Clay ME, Smith J, Herr S, Ilstrup SJ, Kunkel identification of patients at risk. Transfusion 2010; 50: LA et al. Composition of peripheral blood progenitor cell 649–655. components collected from healthy donors. Transfusion 1997; 83 Griffith LM, McCoy JP, Bolan CD, Stroncek DF, Pickett 37: 411–417. AC, Linton GF et al. Persistence of recipient plasma cells and 68 Scholl S, Klink A, Mugge L-O, Schilling K, Hoffken K, Sayer anti-donor isohaemagglutinins in patients with delayed donor HG. Safety and impact of donor-type red blood cell erythropoiesis after major ABO incompatible non-myeloa- transfusion before allogeneic peripheral blood progenitor cell blative haematopoietic cell transplantation. Br J Haematol transplantation with major ABO mismatch. Transfusion 2005; 2005; 128: 668–675. 45: 1676–1683. 84 Schetelig J, Breitschaft A, Kroger N, Zabelinn T, Ebell W, 69 Braine HG, Sensenbrenner LL, Wright SK, Tutschka PJ, Saral Bornhauser M et al. After major ABO-mismatched allogeneic R, Santos GW. Bone marrow transplantation with major ABO hematopoietic progenitor cell transplantation, erythroid blood group incompatibility using erythrocyte depletion of engraftment occurs later in patients with donor blood marrow prior to infusion. Blood 60; 1982: 420–425. group A than donor blood group B. Transfusion 2005; 45: 70 Warkentin PI, Hilden JM, Kersey JH, Ramsay NK, 779–787. McCullough J. Transplantation of major ABO-incompatible 85 Maciej Zaucha J, Mielcarek M, Takatu A, Little MT, Gooley bone marrow depleted of red cells by hydroxyethyl starch. T, Baker J et al. Engraftment of early erythroid progenitors is Vox Sang 1985; 48: 89–104. not delayed after non-myeloablative major ABO-incompati- 71 Tsang KS, Li CK, Wong AP, Leung Y, Lau TT, Li K et al. ble haematopoietic stem cell transplantation. Br J Haematol Processing of major ABO-incompatible bone marrow for 2002; 119: 740–750. transplantation by using dextran sedimentation. Transfusion 86 Gmur JP, Burger J, Schaffner A, Neftel K, Oelz O, Frey D 1999; 39: 1212–1219. et al. Pure red cell aplasia of long duration complicating 72 Bensinger WI, Buckner CD, Clift RA, Thomas ED. Plasma major ABO-incompatible bone marrow transplantation. exchange and plasma modification for the removal of anti-red Blood 1990; 75: 290–295. cell antibodies prior to ABO-incompatible marrow trans- 87 Heyll A, Aul C, Runde V, Arning M, Schneider W, Wernet P. plant. J Clin Apher 1987; 3: 174–177. Treatment of pure red cell aplasia after major ABO- 73 Stussi G, Halter J, Bucheli E, Valli PV, Seebach L, Gmur J incompatible bone marrow transplantation with recombinant et al. Prevention of pure red cell aplasia after major or erythropoietin. Blood 1991; 77: 906. bidirectional ABO blood group incompatible hematopoietic 88 Santamaria A, Sureda A, Martino R, Domingo-Albos A, stem cell transplantation by pretransplant reduction of Muniz-Diaz E, Brunet S. Successful treatment of pure red cell host anti-donor isoagglutinins. Haematologica 2009; 94: aplasia after major ABO-incompatible T cell-depleted bone 239–248. marrow transplantation with erythropoietin. Bone Marrow 74 Tobian AAR, Shirey RS, Montgomery RA, Tisch DJ, Ness Transplant 1997; 20: 1105–1107. PM, King KE. Therapeutic plasma exchange reduces ABO 89 Paltiel O Cournoyer D, Rybka W. Pure red cell aplasia titers to permit ABO-incompatible renal transplantation. following ABO-incompatible bone marrow transplantation: Transfusion 2009; 49: 1248–1254. response to erythropoietin. Transfusion 1993; 33: 418–421. 75 Geyer M, Fischer K-G, Drognitz O, Walz G, Pisarski P, 90 Benjamin RJ, Connors JM, McGurk S, Churchill WH, Antin Wilpert J. ABO-incompatible kidney transplantation with JH. Prolonged erythroid aplasia after major ABO-mis-

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1184 matched transplantation for chronic myelogenous leukemia. 107 Hows J, Beddow K, Gordon-Smith E, Branch DR, Spruce W, Biol Blood Marrow Transplant 1998; 4: 151–156. Sniecinski I et al. Donor-derived red blood cell antibodies and 91 Maschan AA, Skorobogatova EV, Balashov DN, Pashanov immune hemolysis after allogeneic bone marrow transplant- ED, Trakhtman PE, Schipitzina IP et al. Successful treatment ation. Blood 1986; 67: 177–181. of pure red cell aplasia with a single dose of rituximab in a 108 Nair V, Sharma A, Ratheesh J, Das S, Mishra DK, Kale R child after major ABO incompatible peripheral blood et al. Severe intravascular haemolysis following minor group allogeneic stem cells transplantation for acquired aplastic mismatched peripheral blood stem cell transplantation. Bone anemia. Bone Marrow Transplant 2002; 30: 405–407. Marrow Transplant 2007; 39: 805–806. 92 Sora F, De Matteis S, Piccirillo N, Chiusolo P, Laurenti L, 109 Tiplady CW, Fitzgerald JM, Jackson GH, Conn JS, Proctor Putzulu et al. Rituximab for pure red cell aplasia after ABO- SJ. Massive haemolysis in a group A recipient of a group O mismatched allogeneic peripheral blood progenitor cell peripheral blood stem cell allogeneic transplant. Transfus transplantation. Transfusion 2005; 45: 643–644. Med 2001; 11: 455–458. 93 Zand MS. B-cell activity of polyclonal antithymocyte 110 Noborio K, Muroi K, Izumi T, Toshima M, Kawano- globulins. Transplantation 2006; 82: 1387–1395. Yamamoto C, Otsuki T et al. Massive immune hemolysis 94 Labar B, Bogdanic V, Nemet D, Kovacevic-Metelko J, Mrsic after non-myeloablative allogeneic peripheral blood stem cell M, Pavletic Z et al. Antilymphocyte globulin for treatment of transplantation with minor ABO-incompatibility. Leuk pure red cell aplasia after major ABO incompatible marrow Lymph 2003; 44: 357–359. transplant. Bone Marrow Transplant 1992; 10: 471–472. 111 Lee HJ, Gulbis A, DePadua Silva L, Hosing C, Khouri I, 95 Bierman PJ, Warkentin P, Hutchins MR, Klassen LW. deLima M et al. Rituximab for passenger lymphocyte Pure red cell aplasia following ABO mismatched marrow syndrome associated with allogeneic SCT. Bone Marrow transplantation for chronic lymphocyte leukemia: response Transplant 2008; 42: 67–69. to antithymocyte globulin. Leuk Lymphoma 1993; 9: 112 Salmon JP, Michaux S, Hermanne JP, Baudoux E, Gerard C, 169–171. Sontag-Thull D et al. Delayed massive immune hemolysis 96 Roychowdhury DF, Linker CA. Pure red cell aplasia mediated by minor ABO incompatibility after allogeneic complicating an ABO-compatible allogeneic bone marrow peripheral blood progenitor cell transplantation. Transfusion transplantation, treated successfully with antithymocyte 1999; 39: 369–373. globulin. Bone Marrow Transplant 1995; 16: 471–472. 113 Laurencet FM, Samii K, Bressoud A, Tajeddin M, Easton J, 97 Verholen F, Stalder M, Heig C, Chalandon Y. Resistant pure Stelling MJ et al. Massive delayed hemolysis following red cell aplasia after allogeneic stem cell transplantation with peripheral blood stem cell transplantation with minor ABO major ABO mismatch treated by escalating dose donor incompatibility. Hematol Cell Ther 1997; 39: 159–162. leukocyte infusion. Eur J Haematol 2004; 73: 441–446. 114 Bornhauser M, Ordemann R, Paaz U, Schuler U, Kompf J, 98 Ebihara Y, Manabe A, Tsuruta T, Ishikawa K, Haseqawa D, Holig K et al. Rapid engraftment after allogeneic ABO- Ohtsuka Y et al. The effect of donor leukocyte infusion on incompatible peripheral blood progenitor cell transplantation refractory pure red blood cell aplasia after allogeneic stem cell complicated by severe hemolysis. Bone Marrow Transplant transplantation in a patient with myelodysplastic syndrome 1997; 19: 295–297. developing from Kostmann Syndrome. Int J Hematol 2007; 115 Bolan CD, Childs RW, Procter JL, Barrett AJ, Leitmann SF. 86: 446–450. Massive immune haemolysis after allogeneic peripheral blood 99 Ramsey G, Smietana SJ. Long-term follow-up of red cell stem cell transplantation with minor ABO incompatibility. Br alloantibodies. Transfusion 1994; 34: 122–124. J Haemataol 2001; 112: 787–795. 100 Happ WM. Appearance of isoagglutinins in infants and 116 Toren A, Dacosta Y, Manny N, Varadi G, Or R, Nagler A. children. J Exp Med 1920; 29: 313–314. Passenger B-lymphocyte-induced severe hemolytic disease 101 Rowley S, Braine H. Probable hemolysis following minor after allogeneic peripheral blood stem cell transplantation ABO incompatible marrow transplantation. Blood 1982; 60: (letter). Blood 1996; 87: 843–844. 171a. 117 Oziel-Taieb S, Faucher-Barbey C, Chabannon C, Ladaique P, 102 Moog R, Melder C, Prumbaum M, Muller N, Schaefer UW. Saux P, Souin F et al. Early and fatal immune haemolysis Rapid donor type isoagglutinins production after allogeneic after so-called ‘minor’ ABO-incompatible peripheral blood peripheral progenitor cell transplantation. Beitr Infusions stem cell allotransplantation. Bone Marrow Transplant 1997; Ther Transfus Med 1997; 34: 150–152. 19: 1155–1156. 103 Hoegler W, Nachbauer D, Kropshofer G, Wehl G, Schoe- 118 Gajewski JL, Petz LD, Calhoun L, O’Rourke S, Landaw EM, nitzer D, Nussbaumer W. Abrupt onset of severe passenger Lyddane NR et al. Hemolysis of transfused group O red lymphocyte syndrome after HLA-mismatched unrelated bone blood cells in minor ABO-incompatible unrelated-donor bone marrow transplantation in a seven-year-old boy. Med Pediatr marrow transplants in patients receiving cyclosporine without Oncol 2002; 38: 143–144. posttransplant methotrexate. Blood 1992; 79: 3076–3085. 104 Greeno EW, Perry EH, Ilstrup SJ, Weisdorf DJ. Exchange 119 Worel N, Greinix HT, Keil F, Mittenbauer M, Lechner K, transfusion the hard way: massive hemolysis following Fischer G et al. Severe immune hemolysis after minor ABO- transplantation of bone marrow with minor ABO incompat- mismatched allogeneic peripheral blood progenitor cell ibility. Transfusion 1996; 36: 71–74. transplantation occurs more frequently after nonmyeloabla- 105 Reed M, Yearsley M, Krugh D, Kennedy M. Severe tive than myeloablative conditioning. Transfusion 2002; 42: hemolysis due to passenger lymphocyte syndrome after 1293–1301. hematopoietic stem cell transplantation from an HLA- 120 Lapierre V, Oubouxar N, Auperin A, Tramalloni D, Tayebi matched related donor. Arch Pathol Lab Med 2003; 127: H, Robinet E et al. Inﬂuence of the hematopoietic stem cell 1366–1368. source on early immunohematologic reconstitution after 106 Curtin NJ, Schwarer AP. Nonmyeloablative peripheral blood allogeneic transplantation. Blood 2001; 97: 2580–2586. stem cell transplant for T-cell prolymphocytic leukaemia 121 Worel N, Greinix HT, Supper V, Leitner G, Mitterbauer M, complicated by fulminant haemolysis and acute renal failure Rabitsch W et al. Prophylactic red blood cell exchange for at engraftment secondary to minor ABO incompatibility. Clin prevention of severe immune hemolysis in minor ABO- Lab Haem 2005; 27: 206–208. mismatched allogeneic peripheral blood progenitor cell

Bone Marrow Transplantation ABO-incompatible transplantation SD Rowley et al 1185 transplantation after reduced-intensity conditioning. Transfu- 129 McLeod BC, Piehl MR, Sassetti RJ. Alloimmunizaation to sion 2007; 47: 1494–1502. RhD by platelet transfusion in autologous bone marrow 122 Josephson CD, Castillejo MI, Grima K, Hillyer CD. ABO- transplant recipients. Vox Sang 1990; 59: 185–189. mismatched platelet transfusions: strategies to mitigate 130 Asfour M, Narvios A, Lichtiger B. Transfusion of RhD- patient exposure to naturally occurring hemolytic antibodies. incompatible blood components in RhD-negative blood Transfus Apher Sci 2010; 42: 83–88. marrow transplant recipients. Med Gen Med 2004; 6: 22. 123 Josephson CD, Mullis NC, Van Demark C, Hillyer CD. 131 Mijovic A. Alloimmunization to RhD antigen in RhD- Signiﬁcant number of apheresis-derived group O platelets incompatible haemopoietic cell transplants with non-myeloa- units have ‘high titer’ anti-A/A,B: Implications for transfu- blative conditioning. Vox Sang 2002; 83: 358–362. sion policy. Transfusion 2004; 44: 805–808. 132 Avache S, Herman JH. Prevention of D sensitization after 124 Garraty G. Problems associated with passively transfused mismatched transfusion of blood components: toward blood group alloantibodies. Am J Clin Pathol 1998; 109: optimal use of RhIG. Transfusion 2008; 48: 1990–1999. 769–777. 133 Heal JM, Liesveld JL, Phillips GL, Blumberg N. What 125 Lapierre V, Kuentz M, Tiberghien P, for Societe Francaise de would Karl Landsteiner do? The ABO blood group and Greffe de Moelle. Allogeneic peripheral blood hematopoietic stem cell transplantation. Bone Marrow Transplant 2005; 36: stem cell transplantation: guidelines for red blood cell 747–755. immuno-hematological assessment and transfusion practice. 134 Lapierre V, Mahe C, Auperin A, Stambouli F, Oubouzar N, Bone Marrow Transplant 2000; 25: 507–512. Tramalloni D et al. Platelet transfusion containing ABO- 126 Heddle NM, O’Hoski P, Singer J, McBride JA, Ali MA, incompatible plasma and hepatic veno-occlusive disease after Kelton JG. A prospective study to determine the safety of hematopoietic transplantation in young children. Transplant- omitting the antiglobulin crossmatch from pretransfusion ation 2005; 80: 314–319. testing. Br J Haematol 1992; 81: 579–584. 135 International Standards for Cellular Therapy Product Collec- 127 Judd WJ, Fullen DR, Steiner EA, Cavenport RD, Knaﬂ PC. tion. Processing, and Administration, 4th edn. Foundation for Revisiting the issue: can the reading for serologic reactivity the Accreditation of Cellular Therapy: Omaha, NE. following 37 degrees C incubation be omitted? Transfusion 136 Raimondi R, Soli M, Lamparelli T, Bacigalupo A, Arcese W, 1999; 39: 295–299. Belloni M et al. ABO-incompatible bone marrow transplan- 128 Chapman JF, Milkins C, Voak D. The computer crossmatch: tation: a GITMO survey of current practice in Italy and a safe alternative to the serological crossmatch. Transfus Med comparison with the literature. Bone Marrow Transplant 2000; 10: 251–256. 2004; 34: 321–329.

Bone Marrow Transplantation