Outcomes of Second Unrelated Donor Cord Blood Transplants (UCBT) Performed in Children with Graft Failure or Autologous Recovery Following the First UCBT

Background: In the past 20 years, use of UCBT has improved access to transplantation, particularly in children and ethnic minorities. Rapid donor availability, partial HLA matching and low incidence of graft versus host disease (GVHD) are important advantages. UCBT outcomes have significantly improved in the last decade due to better graft availability and donor selection. However, a small fraction of recipients develop auto-recovery or graft failure and are candidates for a second transplant. In most cases, due to urgency and original limitations in donor choices, another cord blood unit (CBU) is the best option for these children.

Methods: Retrospective analysis of all pediatric patients (<21 years) undergoing UCBT at Duke between 1995 and 2013 who subsequently received a second UCBT due to graft failure or auto-recovery after the first UCBT (n=36) was conducted. Patients requiring second transplants for relapse were excluded. Kaplan-Meier estimates of overall survival (OS) and cumulative incidence (CI) of engraftment and GVHD were calculated. All received reduced intensity conditioning for the second transplant. The most common prep regimen was cyclophosphamide 60mg/kg x 2 and equine ATG 30mg/kg x 3 (n=15). Single cord (n=28), double cord (n=3) or cord +haplo graft (n=5) were the donor sources. Demographic, disease and graft characteristics are shown in the table.

Results: The OS at 100 days, 1 year and 5 years was 50.0%, 37.1% and 33.3% respectively. Mortality was lower for patients transplanted after 2007, but the difference was not statistically significant (p=0.33). Median time to neutrophil (ANC 500) and platelet (50K) engraftments were 27 (range, 12-94) and 98 days (range, 37-434) respectively. The probability of neutrophil engraftment at day 42 was 64.3%. The CI of grades II-IV and grades III-IV acute GVHD, at 100 days was 47.1%.and 31.3% respectively. Extensive/chronic GVHD was seen in 3 patients (CI=13.6%).

Conclusions: Second UCBT resulted in acceptable survival particularly in the recent years when used as a rescue for failed initial UCBT. Use of a CBU allows quick second transplants and thus decreases the overall duration of cytopenias. Second UCBT should be offered as a viable option for children with a failed first UCBT.

Characteristics of 2nd UCBT
Gender	N	%
Male	30	83.3
Primary Disease
Heme Malignancy	16	44.4
Inherited Metabolic Disease	10	27.8
Marrow failure	4	11.1
Immunodeficiency	3	8.3
Hemoglobinopathy	3	8.3
Reason for 2nd UCBT
Primary graft failure	25	69.4
Late graft failure	3	8.33
Autologous recovery	8	22.2
Demographics	median	range
Age at 2nd UCBT (years)	7.4	1.5-20.9
Weight at 2nd UCBT (kg)	22.9	9.0-108.0
Time from 1st to 2nd UCBT (days)	61.5	42-3454
Cell Dose (n=39)
Cryopreserved TNCs x 107/kg	5.6	1.3-21.9
Reinfused TNCs x 107/kg	4.6	1.1-18.1
Reinfused CD34+ x 105/kg	1.4	0.1-15.0
Reinfused CD3+ x 106/kg	9.1	1.5-34.4
Reinfused CFUs x 104/kg	5.7	0.0-175.4