Favorable Leukemia-Free Survival (LFS) for Adults and Children Undergoing Myeloablative Umbilical Cord Blood (UCB) Transplantation with Cyclophosphamide (CY), Fludarabine (FLU) and Total Body Irradiation (TBI): A Single Center Analysis of 194 Patients

Abstract

We evaluated the outcomes in 194 consecutive patients (pts) transplanted with UCB after a myeloablative therapy for acute myeloid leukemia (AML,n=86), acute lymphoblastic leukemia (ALL,n=89), chronic myeloid leukemia (CML,n=13), and myelodysplastic syndrome (MDS,n=6) between 1995 and 2007. Methods: Pts were assigned to 3 treatment groups (Table 1): Cy 120mg/kg, TBI 1320 cGy and equine anti-thymocyte globulin (ATG) 90mg/kg with one UCB unit (CyTBI, 1995–2000; n=65; median age 9 yrs, r 0.5–52); CyTBI and Flu 75mg/m2/day with one UCB unit (CYFluTBI-1, 2000–2007; n=36; median age 8yrs, r 2–44); and Cy/Flu/TBI with two UCB units (CyFluTBI-2, 2001–2007; n=93; median age 25yrs, r 9–45). All received cyclosporine A (CsA) with either short course methylprednisolone (2 mg/kg/days; days +5 to +19) or mycophenolate mofetil (days -3 to +30). Pts were classified as standard (acute leukemia in CR1-2 or CML in first chronic phase, n=138) or high risk (n=56) for relapse. Median follow-up for survivors is 2.8 years (r: 0.7–9.2). Pts received a 0–1 HLA-mismatched graft in 81% for CyFluTBI-1, 47% for CyTBI, and 36% for CyFluTBI-2 (p<.01). The total nucleated cell dose infused in recipients of CyFluTBI-1 was 3.6 (2.1–13.9), CyTBI 3.0 (0.9–14.0), and CyFluTBI-2 3.6 (1.7–6.5) x 107/kg (p=.01). Results: Engraftment was not significantly different for the 3 groups (CyFluTBI-1 97% (92–100) vs. CyTBI 89% (82–96) vs. CyFluTBI-2 87% (80–94) (p=.06). Patients receiving a total CD34+ cell dose > 2.5 x 106/kg had significantly better engraftment ( 6.5, 98% [94–100], p<.01). Incidence of grades II-IV GVHD was higher after CyFluTBI-2 (51% [40–62]) as compared to CyFluTBI-1 (25% [11–39]) or CyTBI 37% (25–49) (p<.01) in univariate, but not in multivariate analysis after adjusting for age and disease risk. While treatment-related mortality (TRM) at 2 yrs was similar in univariate analysis (CyFluTBI-1, 11% [1–21] vs. CyFluTBI–2, 32% [22–42] vs. CyTBI 32% [22–42], p=.07), after adjusting for age and disease risk TRM was higher for the CyTBI (RR 3.5 [1.1–9.2], p=.03) vs. CyFluTBI-1. The relapse risk at 4 yrs for CyFluTBI-1 was 37% (19–55), CyFluTBI-2 19% (10–28), and CyTBI 28% (18–38) (p=.19). The LFS at 4 yrs for CyFluTBI-1 was 52% (34–70), CyFluTBI-2 49% (38–60), and CyTBI 37% (25–49) (p=.46). In multivariate analysis, only high risk disease was associated with higher relapse risk (RR 2.0;95%CI,1.0–3.9;p=.05) and mortality (RR 2.2;95%CI,1.4–3.4;p<.01). Notably, cell dose and HLA match had no effect on any outcome in multivariate analysis. Conclusions: Use of two partially HLA matched UCB units has increased the applicability of UCB particularly in adult recipients. As a result all UCB grafts contain >2.5 x 107 nucleated cells/kg at cryopreservation, limiting the negative effect of low cell dose on engraftment and survival and reducing the negative impact of HLA mismatch previously reported.