Abstract 959

Survival after bone marrow transplantation (BMT) in children with aplastic anemia (AA) has improved markedly over the last 3 decades. However, we have experienced a certain number of patients who presented with bone marrow aplasia with full donor chimerism after BMT (donor-type aplasia), especially in the recent years. Since the 2000s, fludarabine (FLU)-based conditioning regimen has been often used for Japanese children with AA. The present study therefore evaluated whether patient and transplantation characteristics (especially the FLU regimen) could associate with donor-type aplasia in a large population of children with AA.

To identify the risk factors for donor-type aplasia, we reviewed the clinical data of 660 patients (< 20 years) with AA who received BMT from 1980 to 2010 and whose records were available in the Japan Society for Hematopoietic Cell Transplantation Registry. The influence of potential risk factors on donor-type aplasia was assessed according to patient and transplantation characteristics, including the conditioning regimen. The median age at transplantation was 11 years, and 238 patients received immunosuppressive therapy (IST) before BMT. The median interval between diagnosis and BMT was 9 months. Regarding transplantation, 419 patients received BMT from related donors, whereas 241 received BMT from unrelated donors. Totally, 220 patients received a regimen that included FLU.

Primary graft failure was observed in 25 patients. Of the 635 patients who achieved primary engraftment, 46 developed secondary graft failure or relapse: 11 were of the recipient-type and 35 were of the donor-type. The cumulative incidence of donor-type aplasia was 5.7%, and the 10-year overall survival of patients who presented with donor-type aplasia was 87%. Notably, the incidence of donor-type aplasia was significantly higher in FLU regimen group than in non-FLU regimen group (11.6% vs. 2.7%; P < 0.0001), and multivariate analysis confirmed that the FLU regimen was an independent risk factor for donor-type aplasia. Low infused cells (≤ 3×108/kg) (P = 0.008) and IST before BMT (P = 0.04) were also associated with a high incidence of donor-type aplasia. Age at BMT, gender, etiology, severity, interval between diagnosis and BMT, transfusion times, donor type (related or unrelated), human leukocyte antigen disparity, acute or chronic graft versus host disease (GVHD), GVHD prophylaxis, and viral reactivation or infection did not influence the incidence of donor-type aplasia in multivariate analysis.

When FLU was introduced in the regimen for Japanese children with AA, the dose of cyclophosphamide (CY) was reduced by half, to reduce the toxicity; the conventional dose of CY was 200 mg/kg before the introduction of FLU. Therefore, we investigated the impact of the dose reduction of CY in the FLU regimen group. In patients receiving the FLU regimen, the incidence of donor-type aplasia in the CY half-dose group (n = 175) was 14%, whereas donor-type aplasia was not observed in the CY full-dose group (n = 35) (P = 0.04). In contrast, the addition of antithymocyte globulin or irradiation did not influence the incidence of donor-type aplasia in the FLU regimen group. Of note, the incidence of heart failure as a result of CY administration was higher in the CY full-dose group (5.7%) than in the CY half-dose group (0.6%), although this difference was not statistically significant.

In conclusion, the FLU regimen was identified as an independent risk factor for donor-type aplasia in children with AA, which could possibly be ascribed to the CY dose reduction. However, the high incidence of heart failure associated with the regimen including a full dose of CY is an important issue that needs to be resolved. To develop optimal treatment, the conditioning regimen for children with AA needs to be reconsidered.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution