Outcome of 107 HSCT (91 children) performed between March 1986 and November 2011 in 18 French centres was retrospectively studied. Overall Survival (OS), Non Relapse Mortality (NRM), Graft Versus Host Disease (GVHD) and relapse cumulative incidences were analyzed per patient from the first transplantation. Multivariate analyses were performed to assess risk factors. Statistical analyses were carried out according to guidelines of European group for Blood and Marrow Transplantation (EBMT).
Ninety-one children (58 males), median age 1.4 yrs (range 0.0-15,7 yrs) with JMML underwent 107 allogeneic HSCT (14 second graft and a third one). The median follow-up was 47 months (1-166). At 72 months, OS was 59% (95% CI: 47-74). Cumulative incidence of aGVHD was 48% (95% CI: 37-58). The 6-year cumulative incidence of relapse and NRM was 34% (95% CI: 23-44) and 21% (95% CI: 13-32), respectively. The median delay of relapse was 80 days (range 15-1096). In multivariate analysis, age at HSCT older than 2 years, female donor to male recipient sex-mismatch, matched unrelated or mismatched non cord donor, total body irradiation and no serotherapy in conditioning regimen predicted poorer outcomes. Age at HSCT older than 2 years was an increased risk of relapse whereas Busulfan/Cyclophosphamide/Melphalan conditioning regimen was a decreased risk. Serotherapy was associated with a decreased risk of NRM. Age at HSCT older than 2 years and male recipient were associated with higher probability of dying of NRM causes.
Our results show that allogeneic HSCT may cure approximately 60% of patients with JMML and are similar to best results published by other groups. Relapse represents the main cause of treatment failure and a second HSCT should be proposed. Novel strategies to decrease the risk of relapse are needed.