Abstract
Abstract 836
Patients with high-risk acute myeloid leukemia (AML) have few chances of cure without allogeneic hematopoietic stem cell transplantation (HSCT). HSCT can be used in first remission to cure patients with poor-risk cytogenetics, or as rescue for patients refractory to chemotherapy. Umbilical cord blood (UCB) is one acceptable stem cell source for adult lacking a suitable HLA matched donor.
We retrospectively analyzed 134 adults (>18 years) with de novo AML in first complete remission (CR1), who received UCBT as first transplant after a myeloablative conditioning regimen (MAC). Patients were transplanted from 2000–2010 in EBMT centers. Median age at UCBT was 33 years (range 18–63) and median weight was 65 Kg (range 46–112). Based on available cytogenetic and molecular markers at diagnosis (n=111) 45% of patients were classified in the intermediate risk and 31% in unfavorable risk group. Twenty six patients had a FLT3/ITD-positive mutation status. The median time from diagnosis to UCBT was 5.9 months, and the median duration of remission prior to transplant was 104 days.
Grafts were composed of 1 (sUCBT) (n=100) or 2 (dUCBT) (n=34) CB units. Thirty two percent of CB units were identical to recipient or had 1 HLA disparity (antigen level for HLA-A and B allelic level for DRB1), while 68% had 2–3 HLA disparities. Median TNC cell dose at infusion was 2.7×10^7/kg in sUCBT and 3.9×10^7/kg in dUCBT.All patients were transplanted using a MAC, from which 60% were busulfan based (n=80). The most frequent regimens used were busulfan+fludarabine+thiotepa (n=56) and cyclophosfamide+TBI12Gy+fludarabine (n=40). ATG was used as part of the conditioning regimen in 73% of patients. GVHD prophylaxis consisted either of CSA ± MMF or CSA ± steroids in 45% and 32% of patients, respectively. Overall, the median follow-up was 22 months (range 3–120), (median follow up 28 months for sUCBT and 22 months for dUCBT).
The cumulative incidence (CI) of 60 days neutrophil recovery was 90 ± 3%, (median time of 23 days) and it was 87% for sUCBT and it was 90% dUCBT.At day 100, chimerism test showed full donor chimerism in 92% of patients (data available for 70% of patients who engrafted). At day 100, CI of acute GVHD (grade II–IV) was 19 ± 3%. and it was 12% and 38% for patients transplanted with sUCBT and dUCBT, respectively (p=0.001). Among 34 patients who received a dUCBT, 14 experienced grade II–IV acute GVHD (grade II (n=10), grade III (n=3), grade IV (n=1)) and of those transplanted with a sUCBT, 10 out of 17 had grade II, 4 had grade III, and 3 grade IV.
CI of chronic GvHD at 1 year was 20 ± 3%. CI of TRM at 1 year was 32 ± 3%. In the multivariable analysis, TNC dose at infusion >2.9×10^7/kg (p=0.02) was associated with decreased TRM, however in spite of higher incidence of aGVHD dUCBT was not associated with higher TRM. Fifty eight patients died. The causes of death were infection, other transplant-related events (n=45) or relapse (n=13).
CI of 2y relapse was 16 ± 4%. It was 15% for those patients transplanted with sUCBT (n=100) and 18% with dUCBT (n=34). Out of 26 patients FLT3/ITD positive, 7 relapsed at a median time of 96 days after UCBT.
The 2y probability of leukemia-free-survival (LFS) was 52 ± 2%. There was a trend towards a higher LFS in patients with a longer duration of remission (>66 days, 27% versus 61%, p=0.07). In multivariable model, the use of busulfan in conditioning regimen (HR 2.2, p=0.006) and the use of double UCBT (HR 1.9, p 0.06) were independently associated with better LFS. In fact, LFS was 58% for those patients receiving Busulfan based compared to 42% receiving a TBI based, and it was 58% for dUCBT and 50% for sUCBT.
In conclusion, UCBT using one or two CB units is a valid treatment option for adults with high risk AML in CR1 without an HLA identical donor. The best conditioning regimen for single or double UCBT has to be assessed in prospective clinical trials.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
