Effect of Conditioning Regimen Intensity on Acute Myeloid Leukemia Outcomes after Umbilical Cord Blood Transplantation

Reduced-intensity conditioning (RIC) umbilical cord blood (UCB) transplantation is increasingly used in hematopoietic stem cell transplantation (HCT) for older and medically unfit patients. Data on the efficacy of HCT after RIC relative to myeloablative conditioning (MAC) are limited. We compared the outcomes of acute myeloid leukemia (AML) patients >18 yrs who received UCB grafts after either RIC or MAC. One hundred nineteen adult patients with AML in complete remission (CR) underwent an UCB transplant after RIC (n =74, 62%) or MAC (n = 45, 38%) between January 2001 and December 2009. Conditioning was either reduced intensity and consisted of cyclophosphamide 50 mg/kg, fludarabine 200 mg/m2, and total-body irradiation (TBI) 200 cGy or myelablative and consisted for cyclophosphamide 120 mg/kg, fludarabine 75 mg/m2, and TBI 1200-1320 cGy. All patients received cyclosporine (day −3 to day +180) and mycophenolate mofetil (day −3 to day +45) post-HCT immunosuppression and hematopoietic growth factor. Use of RIC was reserved for patients >45 years (n = 66, 89%) or preexisting severe comorbidities (n = 8, 11%). The 2 groups were similar except for preceding myelodysplastic syndrome (RIC = 28% versus MAC = 4%, P < .01) and age that was dictated by the treatment protocols (median, RIC = 55 years versus MAC = 33years; P < .01). The incidence of neutrophil recovery at day +42 was higher with RIC (94% versus MAC = 82%, P < .1), whereas platelet recovery at the sixth month was similar (RIC = 68% versus MAC = 67%, P = .30). Incidence of grade II-IV acute graft-versus-host disease (aGVHD) (RIC = 47% versus MAC = 67%, P < .01) was decreased with similar incidence of chronic GVHD (cGVHD) (RIC = 30% versus MAC = 34%, P = .43). Median follow-up for survivors was 3.8 and 4.5 years for RIC and MAC, respectively (P = .4). Using RIC, 3-year leukemia-free survival (LFS) was decreased (31% versus MAC = 55%, P = .02) and 3-year relapse incidence was increased (43% versus MAC = 9%, P < .01). Two-year transplant-related mortality (TRM) was similar (RIC = 19% versus MAC = 27%; P = .55). In multivariate analysis, RIC recipients and those in CR2 with CR1 duration <1 year had higher risk of relapse and poorer LFS with no independent predictors of TRM. UCB with RIC extends the use of allogeneic HCT for older and frail patients without excessive TRM with greater benefit for patients in CR1 and CR2 with longer CR1.