Abstract 652


Double cord blood transplantation (dCBT) has emerged as a feasible alternative for the treatment of adults and children with hematologic malignancies. In the great majority of cases after dCBT, one unit emerges as the sole source of long-term hematopoiesis. However, the determinants and mechanism of this single unit dominance remain poorly understood. To investigate this dominance phenomenon we have analyzed the cell subset content of the cord blood (CB) units after dCBT. Methods: All CB units were thawed and washed by centrifugation prior to infusion, and a small sample removed from the final products post-wash for multicolor flow cytometric assessment of graft composition; viable cell subsets included stem/progenitors (CD34+), monocytes (CD14+), B (CD20+), NK (CD3CD56+), NK/T (CD3+CD56+), and T-cell subsets (CD3+CD4+, CD3+CD8+, Memory, Naïve, Regulatory) expressed as infused cell subset/kg of actual recipient weight (in some cases insufficient sample was available for complete cell subset analysis on both CB units (see Table 1). Analysis of donor chimerism was performed on sorted peripheral blood CD3+, CD56+, CD14+ or CD33+ fractions on days 7, 14, 21, 28, 56, and 80 after transplantation. Single donor dominance was defined as more than 95% in all cell fractions. In the rare cases of both units engrafting, the unit contributing >60% hematopoiesis was considered the dominant one. A one-sample test of the null hypothesis that a binomial proportion is equal to 0.5 was used to test the association of concordance of higher cell dose with unit dominance. A one-sample paired t-test was also used to compare the mean differences in cell dose between winning and losing units. A total of 72 patients received dCBT between January 2008 and May 2011. Twelve patients (16%) were excluded because of death before informative chimerism data or due to graft failure. Sixty patients were included in the final analysis: 32 (54%) were conditioned with cyclophosphamide (CY) 60 mg/kg, fludarabine (FLU) 75 mg/m2 and total body irradiation (TBI) 1200–1320 cGy; 17 (28%) with Treosulfan 14 mg/m2, FLU 200 mg/m2 and TBI 200 cGy; and 11 (18%) with CY 50 mg/kg, FLU 200 mg/m2 and TBI 200 cGy. GVHD prophylaxis consisted of cyclosporine and mycophenolate mofetil. Results: Single unit dominance was observed in all recipients of myeloablative conditioning within 21 days of dCBT. Recipients of non-ablative conditioning initially had mixed donors-host chimerism but usually were 100% single donor by day 100. Similar to the findings of others, the unit containing the greater CD3 dose was statistically concordant as the dominant unit (p<0.0001). Of interest, further T-cell subset analyses showed dominance was more correlated with higher CD3+CD8+ (p<0.0001) than with CD3+CD4+ (p=0.001), and more with naive CD3+CD8+ (p=0.009) as compared with naive CD3+CD4+ (p=0.10). The CD3+CD56+ NKT cell content also showed a significant concordance (p=0.02). Furthermore, the mean difference in both naive CD8+ and NKT between what became the dominant and the non-dominant unit was 0.41 × 105/Kg and 0.11 × 104/Kg, respectively (p=0.008 and p=0.02). In contrast to other published reports, no significant associations were found with the infused dose of total cells or CD34+ cells, and no correlations with the infused dose of monocytes, B-cells, NK, memory CD3+CD4+ or CD3+CD8+, or regulatory T-cells were seen (Table1). Conclusion: These data indicate that higher T-cell content, in particular the naïve CD3+CD8+ T-cell subset, is crucial in determining which cord blood unit will ultimately predominate. Our findings provide further insight into the biology of dCBT, and support the hypothesis that dominance is an immune-mediated process in which naïve CD8+ T-cells and perhaps NKT cells within the graft play an important role. Our findings may have important implications for optimal unit selection, and for ex-vivo graft manipulations to enhance engraftment, immune reconstitution, and disease control.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution