- Email: [email protected]
Transfusion-related problems following major ABO-incompatible bone marrow transplantation
Transfusion and Apheresis Science 28 (2003) 155–161 www.elsevier.com/locate/transci
Transfusion-related problems following major ABO-incompatible bone marrow transplantation Tetsunori Tasaki a,*, Shyukuko Satoh a, Kenji Gotoh a, Kieko Fujii a, Sakiko Sasaki a, Junko Takadate a, Mihoko Tachibana a, Yoko Ono b, Yoji Ishida b b
a Department of Clinical Pathology, Iwate Medical University 19-1, Uchimaru, Morioka City Iwate 020-8505, Japan Hematology Division, Department of Medicine 3, School of Medicine, Iwate Medical University, Morioka City Iwate 020-8505, Japan
Received 5 November 2002; accepted 11 November 2002
Abstract A case of acute myelocytic leukemia has been reported in which the patientÕs surviving original B lymphocytes after pretransplant-conditioning chemotherapy probably reproduced hemagglutinins that reacted with red blood cells (RBCs) derived from engrafted donor marrow for a prolonged period of time. Although the direct antiglobulin test was negative and hemagglutinins were not detectable in the patientÕs sera but only in the eluate, the antibodies reappeared in the sera. Therefore, it is important to confirm that the eluate does not contain antibodies that would react with donorderived RBCs when the type of red cell used for transfusion is switched from the patientÕs type to the donorÕs type in a major ABO-mismatched bone marrow transplantation (BMT). Testing of ABO subgroups using lectins is also recommended to avoid a delayed hemolytic transfusion reaction following BMT. Ó 2003 Elsevier Science Ltd. All rights reserved. Keywords: Pure red cell aplasia; ABO-incompatible BMT; ABO subgroup
1. Introduction Delayed onset of erythropoiesis and pure red cell aplasia are sometimes encountered in major ABO-incompatible bone marrow transplantation (BMT), especially in cases in which the blood type of the donor is A and that of the recipient is O [1–3]. We report here on a rare case in which autologous immunologically competent cells were
* Corresponding author. Tel.: +81-19-651-5111x3249; fax: +81-19-624-5030. E-mail address: [email protected] (T. Tasaki).
maintained in spite of premedication during BMT using myelosuppressive drugs. This produced hemagglutinins reacting with red blood cells (RBCs) derived from the engrafted marrow and transfused red cells.
2. Case report A 53-year-old man with acute myelocytic leukemia and an original blood type of O(+) following conditioning chemotherapy received bone marrow from his HLA-identical younger sister whose blood type was A(+). Pretransplant anti-A
1473-0502/03/$ - see front matter Ó 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S1473-0502(03)00014-4
156
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
IgG and IgM antibody titers in the patient were 1024 and 64, respectively. Prompt engraftment was confirmed by the rise of granulocytes to over 0:5 109 /L on day 28 and by the change of chromosome type from 46,XY to 46,XX; however, high titers of hemagglutinin persisted for more than 7 months and no donor-type RBCs were observed. Prednisolone and tacrolimus hydrate were administered to the patient for the treatment of chronic graft-versus-host disease (GVHD), then anti-A agglutinin titers began to decrease and reticulocytes started to increase on day 376. Since anti-A agglutinins in the recipientÕs sera disappeared and a direct antiglobulin test (DAT) showed a negative result thereafter, we changed the blood type of the transfusion from the recipientÕs type (i.e., O(+)) to the donorÕs type (i.e., A(+)) as therapy for gastro intestinal (G-I) bleeding, although anti-A antibodies were dissociated from the recipientÕs RBCs. After that, the anti-A IgG titer increased to 1 again, and mild hemolysis was observed. Also, mixed-field agglutination with anti-A reagent was temporarily observed. This suggested the coexistence of a small number of the patientÕs original type O RBCs with a large number of type A RBCs derived from the engrafted donor marrow. Subsequently, type O RBCs were used when blood transfusion was needed. As of June 4, 2001, the patient was leading a peaceful life without apparent clinical problems or signs of leukemic relapse, although the DAT is weakly positive and mild anemia persists [4,5].
3. Materials and methods Blood types were determined using commercially available reagent antisera. The quantity of A antigens of the red cells was evaluated through the strength of the reaction between serial dilutions of lectins (anti-A1 , Dolichos biflorus; anti-H, Ulex europaeus) and the patientÕs RBCs as well as the reagent red cells with phenotypes of A1 and A2 . Scores, the sums of all of the degrees of RBC agglutination (described from 0 to 4) caused by antisera diluted stepwise, were also employed to determine the characteristics of A antigens [6].
An antibody screening test for RBCs was carried out using a set of three vials of reagent RBCs (SURGESCREN, Ortho, USA) and Di(a) RBCs. In the antiglobulin phase of the crossmatch testing, polyethylene glycol (PEG) was used to enhance the reaction between antibodies and RBCs. Screening for antiplatelet antibodies was performed by a mixed passive hemagglutinin method. A DAT was performed using polyspecific antihuman globulin reagent including anti-IgG, antiC3d and anti-C3b. The change of anti-A in the patientÕs sera after BMT was carefully monitored periodically. When anti-A was not detectable and the DAT was negative, a heat elution procedure (56 °C for 5 min) was performed to examine for antibodies still attached to red cells; i.e. one drop of 2% saline suspension of A RBCs was added to two drops of the eluate and the agglutination reaction was observed with the naked eye by shaking the button of the RBCs formed by centrifugation (3400 rpm, 15 s). To determine the anti-A IgM titers, the immediate-spin saline method was employed; i.e. agglutination between 2% suspension of A1 red cells and serial dilutions of the patientÕs sera was observed after centrifugation (3400 rpm, 15 s) at room temperature. For anti-A IgG titers, the standard PEG/IAT method was used. The activities of A and B glycosyltransferases were measured using a commercially available reagent kit (GALSERVE, Sanko-Junyaku, Tokyo, Japan). The patientÕs sera with UDP-N-acethylgalactosamine or UDP-galactose was incubated at 37 °C for 5 min, then a 50% suspension of type O RBCs was added. After that, serial dilutions of anti-A or -B were added to a 2% suspension of RBCs, centrifuged, and judged. To identify a small number of type O RBCs among those of type A and to speculate on the quality of A antigens, flow cytometric analysis was carried out as follows: a 5% suspension of the patientÕs RBCs (50 ll) treated with glutalaldehyde was added to 100 ll of a 32-fold dilution of monoclonal anti-A and then stored in a refrigerator overnight. After being washed with phosphate buffer solution, each sample was stained with fluorescein isothiocyanate (FITC)-conjugated anti-
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
mouse IgM antibodies by incubation at 4 for an hour. The fluorescence intensity of each red cell was evaluated using flow cytometry (FACSCalibur, BECTON DICKINSON, USA) and compared with that of type O RBCs as a control. Hematological and biological tests were performed by standard methods. To prove the engraftment of hematopoietic stem cells, the changes of Rh-phenotype and sex chromosome were examined.
4. Results After premedication (cyclophosphamide, busulfan, cytarabine), the patient received a total of 3:3 105 CD34+ cells/kg body weight from the donor following the removal of incompatible red
157
cells using an apheresis machine (Spectra, COBE Laboratories, USA). Pretransplant anti-A antibody titers (IgM/IgG) were high (64= 1024). Fig. 1 shows the change in antibody titers. High titers of anti-A agglutinin persisted for more than 7 months; however, the level began to decrease gradually with the use of prednisolone and tacrolimus hydrate for the treatment of chronic GVHD. At 14 months after BMT, hemagglutinins were undetectable in the patientÕs sera and the DAT was negative, although the eluate reacted against type A red cells. About 3 months later, anti-A IgG, probably derived from the patientÕs surviving original B lymphocytes, reappeared, and the DAT was weakly positive as a result of a reaction between those antibodies and type A red cells produced by engrafted donor marrow and transfused as a therapy for GI bleeding. The titer
Fig. 1. Time-course change in antibody titers DAT: direct antiglobulin test; mf: mixed-field agglutination; :weak agglutination.
158
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
has persisted at 1 to date. Incidentally, the anti-B IgG and IgM antibody titers also decreased gradually but never disappeared during the posttransplant period. Figs. 2 and 3 show the changes in the hematological and biological data. The transfusion of blood components, principal treatment regimen and clinical features were also described in the figures. Although prompt engraftment was confirmed by changes in the chromosome type on day 28, a marked delay in red cell recovery was observed. The patientÕs strong anti-A agglutinins were considered to be the main reason for the impaired erythroid recovery, but an additional therapy such as plasma exchange was not performed. The patient was therefore transfusiondependent for about a year and received a significant amount of group O(+) leukocyte-poor red cells (LPRCs). On day 376, the onset of erythro-
poiesis was confirmed by an elevation of the reticulocyte count of >1%. The Rh phenotype also changed from recipient type (CcDee) to donor type (CcDEe). The patient received no transfusion support for about 2 months thereafter. On day 425, the patient became more anemic due to bleeding from a gastric ulcer and required a blood transfusion. Type A(+) LPRCs were used in this transfusion since anti-A agglutinins were not detected in the sera or DAT was negative, although the eluate still reacted with type A RBCs. After that point, anti-A hemagglutinins reappeared weakly, and a slight increase in the levels of serum LDH, AST, and bilirubin, possibly due to red cell hemolysis, were observed. On day 490, there was a marked fall in the Hb level due again to G-I bleeding, so group O(+) LPRCs were transfused this time. The period of required RBC transfusion support was 517 days, and more than four years
Fig. 2. Time-course change in Hb and reticulocyte counts One bar in ÔRBC transfusionÕ indicates two units of LPRCs transfused. One bar in ÔPC transfusionÕ indicates 15 or 20 units of platelet concentrates (PC) transfused.
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
159
Fig. 3. Biochemical date after BMT m-Psl: methylprednisolone; CyA: cyclosporin A; T-B: total bilirubin; AST: aspartate aminotransferase; LDH: lactate dehydrogenase. Normal range for T–B is 0.2–1.2 mg/dL, for LDH 118-257 IU/L, for AST 10–32 IU/L.
have passed since the patient received his last blood transfusion. No activity of A glycosyltransferase has been observed so far.
Table 1 shows the result of the serological examination of the patientÕs blood obtained December 11, 2000. The patientÕs RBCs, like type
Table 1 Reactions between antisera and RBCs Phenotype of RBCs Antisera
A1
a
4+ 2+ 4+ 0 1+
Anti-A1 [IS] Anti-Hb [IS] Anti-Ac [IS] PatientÕs sera [IS] [PEG/IAT]
(32, 16.5) (8, 4.5) (512, 29.0) (1, 0.5)
A2
PatientÕs RBCs
0 4+ ð32, 16.5) 3+ (32, 7.5) 0 0
1+ (2, 1.5) 1+ (4, 2.0) 4+ (256, 25.5)
Each figure (from 0 to 4+) indicates the degree of red cell agglutination. The former figure in parentheses indicates the highest dilution of antisera that agglutinated red cells, and the latter indicates a score which is the sum of all of the degrees of red cell agglutination by serial dilutions of antisera. IS: immediate spin; PEG/IAT: polyethylene glycol/indirect antiglobulin test. a Dolichos biflorus. b Ulex europaeus. c Polyclonal antisera.
160
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
Fig. 4. Flow cytometric analysis results using type O, A1 and patientÕs red cells are shown in a, b and c, respectively.
Aint , reacted weakly with anti-A1 . The degree of red cell agglutination against anti-H was almost the same as that of type A1 red cells, both of which were apparently weakened in comparison with those using type A2 red cells. Anti-A antibodies were not detected in the patientÕs sera by the immediate-spin saline method, but were detected by the PEG/IAT technique. Fig. 4 shows the results of flow cytometric analysis on the same specimen. Type O RBCs were substantially undetectable, although the figure in the Ô% TotalÕ column was described as 0.01. The fluorescence intensity of the patientÕs RBCs was, however, apparently weaker than that for type A1 red cells. In any case, the moderately high reticulocyte count has continued, but overt hemolysis has not been observed to date.
5. Discussion We encountered a case with donor-type RBCs about one year after a major ABO-mismatched BMT. The interesting findings were that the eluate contained antibodies reacting with RBCs derived from engrafted donor marrow, although serial chimerism studies or Gm allotyping had not been performed [7,8]. Those antibodies were not detectable in the patientÕs sera and DAT was negative, but reappeared in the patientÕs sera with
positive results for DAT, and mild hemolysis was observed following transfusions of type A RBCs as a therapy for G-I bleeding. Also, it seemed that a small number of the patientÕs original type O RBCs still existed from the 4+mf result of the reaction between the patientÕs red cells and anti-A, although the influence of transfused type O RBCs should be taken into account. It was obscure whether, in the presence of plenty of type A RBCs derived from the engrafted marrow, transfused type A RBCs further stimulated the patientÕs residual original B lymphocytes to produce anti-A antibodies. No association between those two matters may exist; however, it is noteworthy that A antigens of RBCs derived from the engrafted marrow were wholly weakened like Aint [9] as shown in the Table 1 and Fig. 4. This might be due to residual anti-A, which has precluded the maturation of progenitor cells from engrafted donor marrow to produce erythrocytes with incomplete antigens or has shortened the life span of those red cells. In any case, anti-A IgG weakly produced by the patientÕs surviving B lymphocytes irrespective of intensive chemotherapy reacted at first with Aint -like RBCs without apparent clinical problems. The production of those antibodies might then be strengthened by the type A red cells transfused for the treatment of G-I bleeding, and the hemolysis of those transfused RBCs might occur thereafter [10].
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
As of June 4, 2001, the titer of anti-A (IgG) is 1 and DAT is weakly positive due to the reaction between those antibodies from the patientÕs surviving original B lymphocytes and impaired A antigens of RBCs, like type Aint , from engrafted donor marrow, but no clinically significant hemolysis has been observed.
[4]
[5]
Acknowledgement The authors thank Ryuichi Yabe of the Tokyo Western Blood Center for his helpful suggestions.
[6] [7]
References [1] Blacklock HA, Katz F, Hichalevicz R, Hazlehurst GRP, Davies L, Prentice HG, et al. A and B blood group antigen expression on mixed colony cells and erythroid precursors: relevance for human allogeneic bone marrow transplantation. Br J Haematol 1984;58:267–76. [2] Sniecinski IJ, Oien L, Petz LD, Blume KG. Immunohematologic consequences of major ABO-mismatched bone marrow transplantation. Transplantation 1988;45: 530–4. [3] Gm€ ur JP, Burger J, Schaffner A, Neftel K, Oelz O, Frey D, et al. Pure red cell aplasia of long duration complicating
[8]
[9]
[10]
161
major ABO-incompatible bone marrow transplantation. Blood 1990;75:290–5. van Leeuwen JE, van Tol MJ, Joosten AM, Wijnen JT, Verweij PJ, Khan PM, et al. Persistence of host-type hematopoiesis after allogeneic bone marrow transplantation for leukemia is significantly related to the recipientÕs age and/or the conditioning regimen, but it is not associated with an increased risk of relapse. Blood 1994;83:3059– 67. Bader P, Holle W, Klingebiel T, Handgreitinger R, Benda N, Schlegel PG. Mixed hematopoietic chimerism after allogeneic bone marrow transplantation: the impact of quantitative PCR analysis for prediction of relapse and graft rejection in children. Bone Marrow Transplant 1997;19:697–702. Brain P. Subgroups of A in the South African Bantu. Vox Sang 1966;11:686–98. Tasaki T, Sasaki S, Gotoh K, Itoh C, Itoh S, Kuriya S. Multiple red blood cell antibodies produced by donor B lymphocytes after ABO-matched allogeneic bone marrow transplantation. Transfus Sci 1999;20:121–7. Ishikura H, Endo J, Saito Y, Kosaka Y, Matsuo S, Nagasue N. Graft-versus-host antibody reaction causing a delayed hemolytic anemia after blood transfusion. Blood 1993;82:3222–3. Yoshida A, Dave V, Hamilton HB. Imbalance of blood group A subtypes and the existence of superactive B gene in Japanese in Hiroshima and Nagasaki. Am J Hum Genet 1988;43:422–8. Hazlehurst MCGR, Armitage SE. Development of autoanti-A1 antibodiesÕ following alloimmunization in an A2 recipient. Br J Haematol 1983;55:657–63.
Transfusion-related problems following major ABO-incompatible bone marrow transplantation Tetsunori Tasaki a,*, Shyukuko Satoh a, Kenji Gotoh a, Kieko Fujii a, Sakiko Sasaki a, Junko Takadate a, Mihoko Tachibana a, Yoko Ono b, Yoji Ishida b b
a Department of Clinical Pathology, Iwate Medical University 19-1, Uchimaru, Morioka City Iwate 020-8505, Japan Hematology Division, Department of Medicine 3, School of Medicine, Iwate Medical University, Morioka City Iwate 020-8505, Japan
Received 5 November 2002; accepted 11 November 2002
Abstract A case of acute myelocytic leukemia has been reported in which the patientÕs surviving original B lymphocytes after pretransplant-conditioning chemotherapy probably reproduced hemagglutinins that reacted with red blood cells (RBCs) derived from engrafted donor marrow for a prolonged period of time. Although the direct antiglobulin test was negative and hemagglutinins were not detectable in the patientÕs sera but only in the eluate, the antibodies reappeared in the sera. Therefore, it is important to confirm that the eluate does not contain antibodies that would react with donorderived RBCs when the type of red cell used for transfusion is switched from the patientÕs type to the donorÕs type in a major ABO-mismatched bone marrow transplantation (BMT). Testing of ABO subgroups using lectins is also recommended to avoid a delayed hemolytic transfusion reaction following BMT. Ó 2003 Elsevier Science Ltd. All rights reserved. Keywords: Pure red cell aplasia; ABO-incompatible BMT; ABO subgroup
1. Introduction Delayed onset of erythropoiesis and pure red cell aplasia are sometimes encountered in major ABO-incompatible bone marrow transplantation (BMT), especially in cases in which the blood type of the donor is A and that of the recipient is O [1–3]. We report here on a rare case in which autologous immunologically competent cells were
* Corresponding author. Tel.: +81-19-651-5111x3249; fax: +81-19-624-5030. E-mail address: [email protected] (T. Tasaki).
maintained in spite of premedication during BMT using myelosuppressive drugs. This produced hemagglutinins reacting with red blood cells (RBCs) derived from the engrafted marrow and transfused red cells.
2. Case report A 53-year-old man with acute myelocytic leukemia and an original blood type of O(+) following conditioning chemotherapy received bone marrow from his HLA-identical younger sister whose blood type was A(+). Pretransplant anti-A
1473-0502/03/$ - see front matter Ó 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S1473-0502(03)00014-4
156
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
IgG and IgM antibody titers in the patient were 1024 and 64, respectively. Prompt engraftment was confirmed by the rise of granulocytes to over 0:5 109 /L on day 28 and by the change of chromosome type from 46,XY to 46,XX; however, high titers of hemagglutinin persisted for more than 7 months and no donor-type RBCs were observed. Prednisolone and tacrolimus hydrate were administered to the patient for the treatment of chronic graft-versus-host disease (GVHD), then anti-A agglutinin titers began to decrease and reticulocytes started to increase on day 376. Since anti-A agglutinins in the recipientÕs sera disappeared and a direct antiglobulin test (DAT) showed a negative result thereafter, we changed the blood type of the transfusion from the recipientÕs type (i.e., O(+)) to the donorÕs type (i.e., A(+)) as therapy for gastro intestinal (G-I) bleeding, although anti-A antibodies were dissociated from the recipientÕs RBCs. After that, the anti-A IgG titer increased to 1 again, and mild hemolysis was observed. Also, mixed-field agglutination with anti-A reagent was temporarily observed. This suggested the coexistence of a small number of the patientÕs original type O RBCs with a large number of type A RBCs derived from the engrafted donor marrow. Subsequently, type O RBCs were used when blood transfusion was needed. As of June 4, 2001, the patient was leading a peaceful life without apparent clinical problems or signs of leukemic relapse, although the DAT is weakly positive and mild anemia persists [4,5].
3. Materials and methods Blood types were determined using commercially available reagent antisera. The quantity of A antigens of the red cells was evaluated through the strength of the reaction between serial dilutions of lectins (anti-A1 , Dolichos biflorus; anti-H, Ulex europaeus) and the patientÕs RBCs as well as the reagent red cells with phenotypes of A1 and A2 . Scores, the sums of all of the degrees of RBC agglutination (described from 0 to 4) caused by antisera diluted stepwise, were also employed to determine the characteristics of A antigens [6].
An antibody screening test for RBCs was carried out using a set of three vials of reagent RBCs (SURGESCREN, Ortho, USA) and Di(a) RBCs. In the antiglobulin phase of the crossmatch testing, polyethylene glycol (PEG) was used to enhance the reaction between antibodies and RBCs. Screening for antiplatelet antibodies was performed by a mixed passive hemagglutinin method. A DAT was performed using polyspecific antihuman globulin reagent including anti-IgG, antiC3d and anti-C3b. The change of anti-A in the patientÕs sera after BMT was carefully monitored periodically. When anti-A was not detectable and the DAT was negative, a heat elution procedure (56 °C for 5 min) was performed to examine for antibodies still attached to red cells; i.e. one drop of 2% saline suspension of A RBCs was added to two drops of the eluate and the agglutination reaction was observed with the naked eye by shaking the button of the RBCs formed by centrifugation (3400 rpm, 15 s). To determine the anti-A IgM titers, the immediate-spin saline method was employed; i.e. agglutination between 2% suspension of A1 red cells and serial dilutions of the patientÕs sera was observed after centrifugation (3400 rpm, 15 s) at room temperature. For anti-A IgG titers, the standard PEG/IAT method was used. The activities of A and B glycosyltransferases were measured using a commercially available reagent kit (GALSERVE, Sanko-Junyaku, Tokyo, Japan). The patientÕs sera with UDP-N-acethylgalactosamine or UDP-galactose was incubated at 37 °C for 5 min, then a 50% suspension of type O RBCs was added. After that, serial dilutions of anti-A or -B were added to a 2% suspension of RBCs, centrifuged, and judged. To identify a small number of type O RBCs among those of type A and to speculate on the quality of A antigens, flow cytometric analysis was carried out as follows: a 5% suspension of the patientÕs RBCs (50 ll) treated with glutalaldehyde was added to 100 ll of a 32-fold dilution of monoclonal anti-A and then stored in a refrigerator overnight. After being washed with phosphate buffer solution, each sample was stained with fluorescein isothiocyanate (FITC)-conjugated anti-
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
mouse IgM antibodies by incubation at 4 for an hour. The fluorescence intensity of each red cell was evaluated using flow cytometry (FACSCalibur, BECTON DICKINSON, USA) and compared with that of type O RBCs as a control. Hematological and biological tests were performed by standard methods. To prove the engraftment of hematopoietic stem cells, the changes of Rh-phenotype and sex chromosome were examined.
4. Results After premedication (cyclophosphamide, busulfan, cytarabine), the patient received a total of 3:3 105 CD34+ cells/kg body weight from the donor following the removal of incompatible red
157
cells using an apheresis machine (Spectra, COBE Laboratories, USA). Pretransplant anti-A antibody titers (IgM/IgG) were high (64= 1024). Fig. 1 shows the change in antibody titers. High titers of anti-A agglutinin persisted for more than 7 months; however, the level began to decrease gradually with the use of prednisolone and tacrolimus hydrate for the treatment of chronic GVHD. At 14 months after BMT, hemagglutinins were undetectable in the patientÕs sera and the DAT was negative, although the eluate reacted against type A red cells. About 3 months later, anti-A IgG, probably derived from the patientÕs surviving original B lymphocytes, reappeared, and the DAT was weakly positive as a result of a reaction between those antibodies and type A red cells produced by engrafted donor marrow and transfused as a therapy for GI bleeding. The titer
Fig. 1. Time-course change in antibody titers DAT: direct antiglobulin test; mf: mixed-field agglutination; :weak agglutination.
158
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
has persisted at 1 to date. Incidentally, the anti-B IgG and IgM antibody titers also decreased gradually but never disappeared during the posttransplant period. Figs. 2 and 3 show the changes in the hematological and biological data. The transfusion of blood components, principal treatment regimen and clinical features were also described in the figures. Although prompt engraftment was confirmed by changes in the chromosome type on day 28, a marked delay in red cell recovery was observed. The patientÕs strong anti-A agglutinins were considered to be the main reason for the impaired erythroid recovery, but an additional therapy such as plasma exchange was not performed. The patient was therefore transfusiondependent for about a year and received a significant amount of group O(+) leukocyte-poor red cells (LPRCs). On day 376, the onset of erythro-
poiesis was confirmed by an elevation of the reticulocyte count of >1%. The Rh phenotype also changed from recipient type (CcDee) to donor type (CcDEe). The patient received no transfusion support for about 2 months thereafter. On day 425, the patient became more anemic due to bleeding from a gastric ulcer and required a blood transfusion. Type A(+) LPRCs were used in this transfusion since anti-A agglutinins were not detected in the sera or DAT was negative, although the eluate still reacted with type A RBCs. After that point, anti-A hemagglutinins reappeared weakly, and a slight increase in the levels of serum LDH, AST, and bilirubin, possibly due to red cell hemolysis, were observed. On day 490, there was a marked fall in the Hb level due again to G-I bleeding, so group O(+) LPRCs were transfused this time. The period of required RBC transfusion support was 517 days, and more than four years
Fig. 2. Time-course change in Hb and reticulocyte counts One bar in ÔRBC transfusionÕ indicates two units of LPRCs transfused. One bar in ÔPC transfusionÕ indicates 15 or 20 units of platelet concentrates (PC) transfused.
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
159
Fig. 3. Biochemical date after BMT m-Psl: methylprednisolone; CyA: cyclosporin A; T-B: total bilirubin; AST: aspartate aminotransferase; LDH: lactate dehydrogenase. Normal range for T–B is 0.2–1.2 mg/dL, for LDH 118-257 IU/L, for AST 10–32 IU/L.
have passed since the patient received his last blood transfusion. No activity of A glycosyltransferase has been observed so far.
Table 1 shows the result of the serological examination of the patientÕs blood obtained December 11, 2000. The patientÕs RBCs, like type
Table 1 Reactions between antisera and RBCs Phenotype of RBCs Antisera
A1
a
4+ 2+ 4+ 0 1+
Anti-A1 [IS] Anti-Hb [IS] Anti-Ac [IS] PatientÕs sera [IS] [PEG/IAT]
(32, 16.5) (8, 4.5) (512, 29.0) (1, 0.5)
A2
PatientÕs RBCs
0 4+ ð32, 16.5) 3+ (32, 7.5) 0 0
1+ (2, 1.5) 1+ (4, 2.0) 4+ (256, 25.5)
Each figure (from 0 to 4+) indicates the degree of red cell agglutination. The former figure in parentheses indicates the highest dilution of antisera that agglutinated red cells, and the latter indicates a score which is the sum of all of the degrees of red cell agglutination by serial dilutions of antisera. IS: immediate spin; PEG/IAT: polyethylene glycol/indirect antiglobulin test. a Dolichos biflorus. b Ulex europaeus. c Polyclonal antisera.
160
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
Fig. 4. Flow cytometric analysis results using type O, A1 and patientÕs red cells are shown in a, b and c, respectively.
Aint , reacted weakly with anti-A1 . The degree of red cell agglutination against anti-H was almost the same as that of type A1 red cells, both of which were apparently weakened in comparison with those using type A2 red cells. Anti-A antibodies were not detected in the patientÕs sera by the immediate-spin saline method, but were detected by the PEG/IAT technique. Fig. 4 shows the results of flow cytometric analysis on the same specimen. Type O RBCs were substantially undetectable, although the figure in the Ô% TotalÕ column was described as 0.01. The fluorescence intensity of the patientÕs RBCs was, however, apparently weaker than that for type A1 red cells. In any case, the moderately high reticulocyte count has continued, but overt hemolysis has not been observed to date.
5. Discussion We encountered a case with donor-type RBCs about one year after a major ABO-mismatched BMT. The interesting findings were that the eluate contained antibodies reacting with RBCs derived from engrafted donor marrow, although serial chimerism studies or Gm allotyping had not been performed [7,8]. Those antibodies were not detectable in the patientÕs sera and DAT was negative, but reappeared in the patientÕs sera with
positive results for DAT, and mild hemolysis was observed following transfusions of type A RBCs as a therapy for G-I bleeding. Also, it seemed that a small number of the patientÕs original type O RBCs still existed from the 4+mf result of the reaction between the patientÕs red cells and anti-A, although the influence of transfused type O RBCs should be taken into account. It was obscure whether, in the presence of plenty of type A RBCs derived from the engrafted marrow, transfused type A RBCs further stimulated the patientÕs residual original B lymphocytes to produce anti-A antibodies. No association between those two matters may exist; however, it is noteworthy that A antigens of RBCs derived from the engrafted marrow were wholly weakened like Aint [9] as shown in the Table 1 and Fig. 4. This might be due to residual anti-A, which has precluded the maturation of progenitor cells from engrafted donor marrow to produce erythrocytes with incomplete antigens or has shortened the life span of those red cells. In any case, anti-A IgG weakly produced by the patientÕs surviving B lymphocytes irrespective of intensive chemotherapy reacted at first with Aint -like RBCs without apparent clinical problems. The production of those antibodies might then be strengthened by the type A red cells transfused for the treatment of G-I bleeding, and the hemolysis of those transfused RBCs might occur thereafter [10].
T. Tasaki et al. / Transfusion and Apheresis Science 28 (2003) 155–161
As of June 4, 2001, the titer of anti-A (IgG) is 1 and DAT is weakly positive due to the reaction between those antibodies from the patientÕs surviving original B lymphocytes and impaired A antigens of RBCs, like type Aint , from engrafted donor marrow, but no clinically significant hemolysis has been observed.
[4]
[5]
Acknowledgement The authors thank Ryuichi Yabe of the Tokyo Western Blood Center for his helpful suggestions.
[6] [7]
References [1] Blacklock HA, Katz F, Hichalevicz R, Hazlehurst GRP, Davies L, Prentice HG, et al. A and B blood group antigen expression on mixed colony cells and erythroid precursors: relevance for human allogeneic bone marrow transplantation. Br J Haematol 1984;58:267–76. [2] Sniecinski IJ, Oien L, Petz LD, Blume KG. Immunohematologic consequences of major ABO-mismatched bone marrow transplantation. Transplantation 1988;45: 530–4. [3] Gm€ ur JP, Burger J, Schaffner A, Neftel K, Oelz O, Frey D, et al. Pure red cell aplasia of long duration complicating
[8]
[9]
[10]
161
major ABO-incompatible bone marrow transplantation. Blood 1990;75:290–5. van Leeuwen JE, van Tol MJ, Joosten AM, Wijnen JT, Verweij PJ, Khan PM, et al. Persistence of host-type hematopoiesis after allogeneic bone marrow transplantation for leukemia is significantly related to the recipientÕs age and/or the conditioning regimen, but it is not associated with an increased risk of relapse. Blood 1994;83:3059– 67. Bader P, Holle W, Klingebiel T, Handgreitinger R, Benda N, Schlegel PG. Mixed hematopoietic chimerism after allogeneic bone marrow transplantation: the impact of quantitative PCR analysis for prediction of relapse and graft rejection in children. Bone Marrow Transplant 1997;19:697–702. Brain P. Subgroups of A in the South African Bantu. Vox Sang 1966;11:686–98. Tasaki T, Sasaki S, Gotoh K, Itoh C, Itoh S, Kuriya S. Multiple red blood cell antibodies produced by donor B lymphocytes after ABO-matched allogeneic bone marrow transplantation. Transfus Sci 1999;20:121–7. Ishikura H, Endo J, Saito Y, Kosaka Y, Matsuo S, Nagasue N. Graft-versus-host antibody reaction causing a delayed hemolytic anemia after blood transfusion. Blood 1993;82:3222–3. Yoshida A, Dave V, Hamilton HB. Imbalance of blood group A subtypes and the existence of superactive B gene in Japanese in Hiroshima and Nagasaki. Am J Hum Genet 1988;43:422–8. Hazlehurst MCGR, Armitage SE. Development of autoanti-A1 antibodiesÕ following alloimmunization in an A2 recipient. Br J Haematol 1983;55:657–63.