Impacts of ABO-Blood Type Incompatibility on Outcome of Unrelated Bone Marrow Transplantation through the Japan Marrow Donor Program.

ABO incompatibility between donor and recipient is not a barrier for successful allogeneic hematopoietic stem cell transplantation, but conflicting data still exist concerning its influence on transplant outcome, graft-versus-host disease (GVHD), relapse, and survival. We retrospectively analyzed the data of patients who underwent UR-BMT through the Japan Marrow Donor Program between January 1993 and September 2005, with complete data on ABO-blood group compatibility, age, and gender in donors and recipients.

A total of 4,970 patients were transplanted with marrow from ABO-matched (M; n=2,513, 50.6%), major incompatible (MA; n=1,254, 25.2%), minor incompatible (MI; n=1,081, 21.8%), and bidirectional incompatible donors (IA; n=122, 2.5%), and were followed up over a median period of 325 days. Among these four groups, excluding age, there was no significant difference in the gender of patients and donors, number of transplantations, conditioning regimen, GVHD prophylaxis, and performance status before transplantation by the likelihood ratio test. The 5-year overall survival of any ABO-incompatible group was significantly lower compared to an identical group (Wilcoxon test, p<0.0001); the estimates for each group were 50.0% (M), 44.7% (MA), 46.7% (MI), and 41.3% (IA). Even in HLA-matched transplantation (n=2,608), a similar difference in overall survival was observed among the four groups (p=0.0124).

In ABO-mismatched transplantation, the processing of bone marrow is necessary to prevent hemolysis of donor or recipient red blood cells as a result of the infusion of ABO-incompatible red blood cells or plasma contained within it. This procedure may reduce the number of hematopoietic stem cells. In fact, the mean number of total infused cells in each group was 3.10 (M), 1.52 (MA), 2.87 (MI), and 1.33 (IA) x10⁸ per patient body weight (kg), with a significant difference in 4,210 patients in which data on the infused cell number were available (M; n=2,310, MA; n=996, MI; n=802, IA; n=102). To examine whether the difference in overall survival depended on the transplanted cell number, we used time-dependent Cox proportional hazards modeling to compare identical and major incompatible groups in terms of overall survival. Whereas the disease (standard and high-risk malignant disease, and benign disease; p=0.0000), patient age (p=0.0000), and ABO compatibility (p=0.0311) were elucidated to be significant risk factors, the number of infused cells was not (p=0.0603).

Engraftment of red blood cells, white blood cells, and platelets were significantly delayed in major ABO mismatch in comparison with ABO identity (p<0.0001). Univariate analysis revealed a small but significant difference in the rate of grade III and IV GVHD among the four groups (p=0.0204). Patients with major and minor ABO incompatibility had a higher incidence of severe GVHD compared to ABO identity (21.9%, 20.4% vs 16.2%). There was no significant difference in GVHD of the skin and gut, but major and minor mismatch developed a higher incidence of moderate to severe hepatic GVHD compared to ABO match (p<0.0001, p=0.0010, respectively). ABO incompatibility had no significant effect on relapse, but the incidence of rejection was significantly higher with ABO-incompatible transplantation (p=0.0219).