Abstract 662

For patients with hematologic malignancies who lack a matched or mismatched unrelated donor, UCBT is an alternative. However, only a minority of larger adolescents and adults are able to find a single UCB with an adequate cell dose. Double UCB unit grafts (DUCB) have previously been shown to extend this HSC option. In order to understand the relative merits of each HSC source, we combined the datasets at the University of Minnesota and Fred Hutchinson Cancer Research Center. Between 2001- and 2008, 536 patients with leukemia (AML, n=211; ALL, n=236; CML, n=70 and MDS, n=19) were transplanted with HSC from a matched sibling (SIB, n=204), 8/8- (MUD, n=152) or 7/8-HLA (MMUD, n=52) matched unrelated donor or DUCB (n=128). All patients received myeloablative conditioning with cyclophosphamide 120 mg/kg and total body irradiation 1200-1320cGy with fludarabine 75mg/m2administered prior to DUCB. All patients received GVHD immunosuppression with a calcineurin inhibitor and either methotrexate (SIB, MUD and MMUD) or mycophenolate mofetil (DUCB). While patient weight and sex distribution and proportion with standard risk disease (STD, AML and ALL in CR1-2, and CML in CP1) were similar, DUCB pts were younger (median age 25 yrs, SIB 40 yrs, MUD 31 yrs, MMUD 31 yrs; p<.01). The proportion of AML and ALL was similar among groups, although more CML patients received a MUD or MMUD. Recipient CMV seropositivity was lower among MUD (44%; SIB 58%; MMUD 60; DUCB 62%, p<.01). The median follow-up of survivors was 3.1 yrs (0.3-8.1). The univariate point estimates at the stated timepoints and multivariate outcomes are summarized in the Table.

Outcomes after SIB vs. MUD vs. MMUD vs. DUCB

SIB* Ref.MUD* RR (95%CI)MVA p-value†MMUD* RR (95%CI)MVA p-value†DUCB* RR (95%CI)MVA p-value
Neutrophil Recovery at day 45 98% 1.0 97% 0.62 (0.50-0.78) . . <0.01 98% 0.63 (0.46-0.86) . . <0.01 87%; 0.29 (0.23-0.37) . . <0.01 
Platelet Recovery at day 100 86% 1.0 81% 0.83 (0.64-1.09) . . 0.18 75% 0.71 (0.48-1.04) . . 0.08 45% 0.22 (0.17-0.29) . . <0.01 
Grade II-IV aGVHD at day 100 65% 1.0 80% 1.68 (0.32-2.14) . . <0.01 85% 2.17 (1.46-3.23) . . <0.01 60% 1.01 (0.76-1.35) . . 0.94 
cGVHD at 2 year 47% 1.0 43% 1.04 (0.75-1.46) . . 0.80 48% 1.24 (0.77-2.0) . . 0.37 26% 0.73 (0.47-1.13) . . 0.16 
TRM at 3 years 24% 1.0 14% 0.92 (0.53-1.60) . . 0.76 27% 1.83 (0.97-3.42) . .0.06 34% 2.86 (1.81-4.51) . . <0.01 
Relapse at 5 years 43% 1.0 37% 0.83 (0.58-1.19) . . 0.32 35% 0.68 (0.39-1.20) . . 0.19 15% 0.27 (0.16-0.47) . . <0.01 
PFS at 5 years 33% 1.0 48% 0.83 (0.62-1.11) . . 0.20 38% 1.04 (0.70-1.53) . . 0.85 51% 1.00 (0.73-1.37) . . 0.99 
SIB* Ref.MUD* RR (95%CI)MVA p-value†MMUD* RR (95%CI)MVA p-value†DUCB* RR (95%CI)MVA p-value
Neutrophil Recovery at day 45 98% 1.0 97% 0.62 (0.50-0.78) . . <0.01 98% 0.63 (0.46-0.86) . . <0.01 87%; 0.29 (0.23-0.37) . . <0.01 
Platelet Recovery at day 100 86% 1.0 81% 0.83 (0.64-1.09) . . 0.18 75% 0.71 (0.48-1.04) . . 0.08 45% 0.22 (0.17-0.29) . . <0.01 
Grade II-IV aGVHD at day 100 65% 1.0 80% 1.68 (0.32-2.14) . . <0.01 85% 2.17 (1.46-3.23) . . <0.01 60% 1.01 (0.76-1.35) . . 0.94 
cGVHD at 2 year 47% 1.0 43% 1.04 (0.75-1.46) . . 0.80 48% 1.24 (0.77-2.0) . . 0.37 26% 0.73 (0.47-1.13) . . 0.16 
TRM at 3 years 24% 1.0 14% 0.92 (0.53-1.60) . . 0.76 27% 1.83 (0.97-3.42) . .0.06 34% 2.86 (1.81-4.51) . . <0.01 
Relapse at 5 years 43% 1.0 37% 0.83 (0.58-1.19) . . 0.32 35% 0.68 (0.39-1.20) . . 0.19 15% 0.27 (0.16-0.47) . . <0.01 
PFS at 5 years 33% 1.0 48% 0.83 (0.62-1.11) . . 0.20 38% 1.04 (0.70-1.53) . . 0.85 51% 1.00 (0.73-1.37) . . 0.99 
*

Results are shown as pointwise estimates, relative risk (RR) and the 95% confidence interval. SIB group was the reference (Ref) group for the multivariate analysis (MVA).

P-value shown relates to multivariate analysis.

In summary, after a Cy/TBI based myeloablation, DUCB is associated with slower hematopoietic recovery compared to other HSC sources. While incidence of acute GVHD was similar to that observed after SIB, DUCB was associated with less acute and trend toward less chronic GVHD relative to unrelated donors. Despite reduced risk of GVHD after DUCBT, risk of relapse was remarkably low while TRM was elevated, resulting in similar PFS. In the absence of a SIB donor, using either MUD or DUCB yield encouraging PFS and are promising donor options. These results support a) the front line use of HLA 0-2 antigen mismatched DUCB in patients with hematological malignancy, b) development of new strategies to enhance hematopoietic recovery after DUCBT, potentially reducing the risk of TRM, c) studies to understand why relapse rates are reduced with DUCBT and d) quality of life assessments long term between HSC groups, as there appears to be less risk of acute and chronic GVHD after DUCBT.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution