Introduction: The impact of birth order of patient and donor in matched sibling hematopoietic stem cell transplantation (HSCT) has not been fully defined. It has been postulated that as a result of fetomaternal trafficking donors who are younger than the transplant patient may develop in utero tolerance to patient cells. A large registry study of adult and pediatric patients found that using a donor younger than the recipient (R>D) was associated with decreased acute GVHD; single center studies found this to be correlated with decreased relapse and improved overall survival as well. We hypothesized that in pediatric patients with hematologic malignancies and a presumed low incidence of acute GVHD that birth order would not have an impact on relapse incidence.

Methods: We performed an IRB -approved retrospective chart review of consecutive patients with acute leukemia undergoing matched sibling transplant bone marrow transplantation at Dana Farber/ Boston Children’s Hospital between 1/1/2001 and 10/31/ 2013. All patients were conditioned ablatively with a Busulfan or TBI based regimen. No product underwent T cell depletion. Patients with twin donors or with secondary leukemia were excluded. Basic patient demographics were collected including development of Grade II-IV acute GVHD, relapse and transplant-related mortality (TRM). Patients were assigned a Low Risk (HSCT in 1st remission) or High Risk (all others) disease status. Patients were then divided into 2 groups: those with donor older than recipient (D>R) and those with donor younger (R>D). Chi-Square test was used to assess for significant differences in relapse between patients.


We reviewed 91 patients with AML or ALL. Age range was 7 months to 24 years (median 11 years). 43 patients were D>R and 48 were R>D. The overall incidence of acute GVHD (Grade II-IV) was 16 % and TRM 4%; neither was significantly different between the two cohorts. Only 1 patient experienced graft rejection (R>D/ALL). For the population as a whole the relapse incidence was not statistically different - 35% in the R>D cohort vs 25% in the D>R cohort (p=0.46). There was also no difference detected in relapse rate when AML and ALL were evaluated separately (p=0.45 and p=0.57 respectively). We also compared the relapse rate by disease risk status. While previous data (Dobbelstein, ASH abstract 2008) suggested fewer relapses in R>D patients, our data revealed a similar trend but only in high risk patients (24% vs 41%), and statistical significance was not achieved (p=0.25). For low risk patients birth order had no effect on relapse incidence (p=0.88).


Previous studies, primarily involving adult patients, have demonstrated an effect of birth order in matched sibling transplants for hematologic malignancies. A decrease in acute GVHD when the donor in younger than recipient (R>D) has been the most consistent finding although single center studies have shown an effect on relapse rate and overall survival as well. In our single institution retrospective study of a similarly treated cohort of pediatric patients we saw no effect of birth order on GVHD, TRM or relapse. Perhaps in the pediatric setting when the incidence of significant GVHD and TRM is low, maternally induced donor tolerance or sensitization to patient cells has less impact and other factors including disease status or patient age predominate. Thus our results suggest that birth order need not be considered as an important factor in matched sibling donor selection for pediatric patients with acute leukemias.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.