BACKGROUND: There have been few reports on outcome of adolescent and young adults (AYAs) with acute myeloid leukemia (AML) who received allogeneic hematopoietic cell transplantation (allo-HCT). We performed a retrospective analysis of nationwide registration data of the Japan Society of Hematopoietic Cell Transplantation collected between 1990 and 2013 to assess the allo-HCT outcomes of AYA patients with AML in Japan.

PATIENTS & METHODS: 2973 patients with de novo AML (excluding acute promyelocytic leukemia, myeloid leukemia associated with Down syndrome, and secondary AML) whose age was 0-29 years old at their first allo-HCT either in first or second remission (CR), primary induction failure, or first relapse were identified. Outcomes including overall survival rate (OS), disease-free survival rate (DFS), cumulative incidence of relapse (CIR), and treatment-related mortality (TRM) were compared between children (0-14 years old at HCT, N=1123) and AYAs (15-29 years old, N=1850).

RESULTS: AYA patients had male predominance, higher incidence of intermediate risk cytogenetics but lower incidence of high-risk cytogenetics [-7/7q-, -5/5q-, complex karyotype, t(6;9), or t(16;21)], and higher prevalence of French-American-British M0, M1, and M2 morphology, while M5 and M7 were lower. Compared to children, AYA patients were transplanted more frequently by HLA-matched-related or unrelated donors, while less frequently by HLA-mismatched related or unrelated cord blood donors. As for conditioning regimen, proportion of myeloablative conditioning (MAC) using total-body-irradiation (TBI) was higher in the AYAs, while that of non-TBI MAC and reduced-intensity-conditioning (RIC) were lower. Cyclosporine was more frequently used than tacrolimus as graft-versus-host-disease (GVHD) prophylaxis in the AYAs.

Five-year OS and DFS rates were significantly poorer in the AYAs: 58% vs 54% (p<0.01) and 53% vs 48% (p=0.03), respectively. Five-year CIR did not differ between the two groups; 33% vs 34% (p=0.99). However, 5-year TRM was significantly higher in the AYAs; 13% vs 16% (p=0.02). Multivariate analysis for both OS and TRM showed significant negative impact of AYAs, but not for DFS or CIR. Subgroup analyses showed that impact of AYAs for OS was greater in male, patients with low-risk cytogenetic abnormalities [t(8;21) or inv(16)], transplanted in CR, transplanted from HLA-matched related donor, received non-TBI MAC conditioning, or received cyclosporine-based GVHD prophylaxis.

Finally, we analyzed the impact of transplant center types on HCT outcomes among the 317 adolescent patients (15-18 years old) who were transplanted at 1CR or 2CR, because they could be transplanted either by pediatricians or by adult hematologists according to their referral pattern. 92 cases were transplanted in pediatric centers, 203 in adult centers, and 22 in combined pediatric and adult centers. Basic characteristics and method of HCT did not differ significantly among the three groups. Interestingly, there was no difference in OS, DFS, CIR, or TRM.

CONCLUSIONS: AYAs with AML showed inferior post-transplant survival, which was mainly due to higher TRM in the AYAs.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.