Immunosuppressive therapy (IST) has been extensively used as first line treatment in children with severe aplastic anemia (SAA), who do not have a suitable donor for hematopoietic stem cell transplantation (HSCT). Several studies have reported that androgens may enhance the response to IST. We retrospectively investigated the effects of androgen on survival. The patient population consisted of 112 children with diagnosis with SAA at St. Mary’s Hospital between 1991 and 2004. All patients received horse type anti-lymphocyte (ALG) or rabbit type anti-thymocyte globulin (ATG), and methylprednisolone, cyclosporin A without G-CSF. Forty-seven children began oral administration of oxymetholone (OMT, 2mg/kg/day) at post-IST 4 weeks for 3 months. We assessed response at post-IST 4 weeks, 6 months on the basis of Camitta’s criteria. Treatment failure was defined as transfusion dependency, death, clonal evolution, and progression to HSCT. We analyzed failure free survival (FFS) and overall survival (OS) in childhood SAA treated with IST.

There were no differences in patient characteristics but children treated with ALG received more OMT than those treated with ATG (P < 0.01). The response rate at post-IST 6 months was 57.1%, 24.1% showing a complete response (CR) and 33.0% a partial response (PR). A high CR rate was identified in those treated with OMT administration (P = 0.04). Factors associated with response to therapy were type of anti-globulin utilized, response at post-IST 4 weeks, and the administration of OMT (P < 0.01). The relapse rate was 40.6%, and was noted to be higher in those given ALG. The OS and FFS in this study was 77.6 ± 4.3% and 54.5 ± 5.1% respectively at 8 year. Factors associated with FFS were response at post-IST 6 months (P < 0.01, RR: 2.155, 1.246 ~ 3.728) and the administration of OMT (P = 0.04, RR: 1.830, 1.026 ~ 3.265) in multivariate analysis, although type of anti-globulin was also included in univariate analysis. The FFS was higher in those treated with OMT (74.3 ± 6.4%) than those who were not given OMT (33.5 ± 7.3%) at 8 year. With regards to the type of response, the FFS was higher in those administered OMT (80.0 ± 10.3%) than those not given OMT (26.4 ± 14.9%) (P = 0.02, RR: 2.273, 1.1624 ~ 12.064) amongst patients who showed PR, although there was no significant difference in FFS amongst patients who showed CR with OMT (94.7 ± 9.0%) or without OMT (83.3 ± 15.2%) at 8 year (P = 0.76). The factor associated with OS was response at post-IST 6 months (P = 0.013, RR: 3.30, 1.28 ~ 8.52) in multivariate analysis, although severity at diagnosis was also included in univariate analysis.Our data revealed that OMT improved FFS in childhood SAA responding to IST, especially in those with PR. In this retrospective study, we concluded that the addition of short term OMT to IST as first line therapy significantly improved CR rate and FFS in childhood SAA patients without a suitable HLA-matched donor.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author