Natural History of AML / MDS Relapsing after Allogeneic Hematopoietic Stem Cell Transplantation (HSCT) Using Reduced Intensity (RIC) Preparative Regimens.

Patients relapsing after HSCT preceded by myeloablative preparative regimens have in general a poor prognosis. Less is known of the natural history of relapses after RIC for AML/MDS. Here we analyzed the outcomes of patients relapsing after RIC.

Patients and Methods: 208 patients with a diagnosis of AML/MDS treated from August 1996 to December 2003 with fludarabine 100–150mg/m² combined to IV busulfan 6.4–11.2mg/Kg (FluBu; n=53), to melphalan 100–180mg/m² (FM; n=122), or to AraC 4 g/m² and idarubicin 36 mg/m² (Flagida; n=34) and unmanipulated HSCT. Donors were HLA matched related (MRD; n=111), unrelated (MUD; n=78) or mismatched related (MMRD; n=19). Graft-versus-host disease (GVHD) prophylaxis was tacrolimus based in all but 8 pts that received cyclosporin-based regimens. Anti-thymocyte globulin was used in 61 MUD or MMRD HSCT.

Results: 64 of 176 patients in complete remission (CR) 30 days post HSCT relapsed (FM, n=29; FluBu, n=17; Flagida, n=18) with one-year relapse-free survival (RFS) of 65%. Among the 64 relapsed patients, disease status at first HSCT was CR (n=13), relapsed/refractory (n=31), primary induction failure (PIF) (n=18) or untreated (n=2). Median time to relapse was 110 days (95% CI 89–131 days). Initial treatment was immunossupression withdrawal in 51 pts who had no ongoing GVHD; two achieved a transient CR for 1.3 and 8 months. 39 patients received further salvage therapy, 18 patients preferred palliative care, and in 7 cases no data was available.

Median overall survival (OS) after relapse was 34 days (range, 12–246) on palliative care. Among pts that received salvage therapy, those achieving a CR had an OS of 27 months (range 2.6–68.2). OS was 94 days (range, 22–290) after failing any type of salvage therapy. Although one patient survived more than one year in CR after DLI +/− chemo salvage, all other long-term survivors received a second HSCT (Table).

Conclusion: The likelihood of achieving a CR after failure of the first HSCT was greater among patients treated in CR at the time of the first HSCT. In this high-risk population, only a second HSCT provided prolonged relapse-free survival.