Abstract

Corticosteroid is often used after allogeneic hematopoietic stem cell transplantation (AHCT) to control graft-versus-host disease (GVHD). However, it is concerned that corticosteroid therapy may suppress graft-versus-leukemia (GVL) effect and may increase the risk of leukemia relapse after AHCT. The aim of the present study is to determine whether the corticosteroid administration increases the relapse after AHCT for acute myelogeneous leukemia (AML) and myelodysplastic syndrome overt leukemia (MDS overt leukemia). We analyzed 98 patients who received their 1st AHCT for AML (n=78; 47 1st or 2nd complete remission, 31 advanced stage) or MDS overt leukemia (n=20) from related (n=45; 37 HLA matched, 8 HLA mismatched) or unrelated (n=53; 37 HLA matched, 16 HLA mismatched) donors at our institution between 1997 - 2006. The median age was 42 years (range, 18–64). Conventional conditioning regimen was used in 71 patients and reduced-intensity conditioning regimen was used in 27 patients. GVHD prophylaxis was cyclosporine and methotrexate in 48 patients, tacrolimus and methotrexate in 49 patients, and cyclosporine alone in 1 patient. Univariate Cox regression analysis showed AML other than 1st or 2nd complete remission (CR) and HLA mismatch were the significant risk factors for relapse but corticosteroid administration (analyzed in time-dependent fashion) was not: AML other than 1st or 2nd CR (p=0.0230, hazard ratio(HR) 2.373, 95% confidence interval (CI) 1.126–4.999), HLA mismatch (p=0.0045, HR 2.776, 95% CI 1.373–5.613), and corticosteroid administration (p=0.8483, HR 1.067, 95% CI 0.551–2.065). In multivariate analysis, corticosteroid administration did not significantly increase the risk for relapse: HR was 0.760 (95% CI, 0.379–1.524, p=0.4394) after adjusting for disease status and HLA disparity. Corticosteroid is concerned to hamper GVL effect, resulting in increased relapse rate after AHCT. However, in this study, corticosteroid administration after AHCT was not a risk factor for relapse. This may be explained by the fact that among the 54 patients who had corticosteroid administration 45(88.9%) had acute and/or chronic GVHD, resulting in relatively strong GVL effect. On the other hand, there were 7 patients who had neither grade II to IV acute GVHD nor chronic GVHD but had corticosteroid administration. Among them 2 patients relapsed. In conclusion, the result of this study suggests that corticosteroid therapy does not increase relapse rate after AHCT at least in the presence of GVHD. However, larger number of patients is needed to evaluate the effect of corticosteroid therapy on relapse in the absence of GVHD.

Author notes

Disclosure: No relevant conflicts of interest to declare.