Effect of Conditioning Regimen Intensity
on Acute Myeloid Leukemia Outcomes
after Umbilical Cord Blood
Betul Oran, John E. Wagner, Todd E. DeFor, Daniel J.
Weisdorf, Claudio G.Brunstein
Biology of Blood and Marrow Transplantation 01/2011
• Unrelated umbilical cord blood (UCB) has emerged as an alternative
source for HCT.
Recent studies have demonstrated similar leukemia free survival
(LFS) after UCB and unrelated donor (URD) transplantation after MAC
in patients with acute leukemia.
Our group has reported high rates of engraftment, low TRM and
promising event-free survival for hematological malignancies with
UCB even after RIC HCT.
We retrospectively studied the outcomes in adults with AML patients
in complete remission (CR) who underwent UCB transplantation after
MAC or RIC regimens.
PATIENTS AND METHODS
• Patients ≥18 years with AML in first (CR1) or second (CR2) ,between January 2001
and December 2009.
86% the graft consisted of 2 UCB units to achieve a combine cell dose ≥ 2.5 x107/kg.
Seventy-four patients received a RIC regimen consisting of fludarabine (40mg/m2
intravenously daily for 5 days) and 200 cGy total body irradiation (TBI) with
cyclophosphamide (50 mg/kg intravenously for 1 day) ;22 (30%) patients received ATG
Forty-five patients received a MAC regimen consisting of cyclophosphamide (60 mg/kg
intravenously daily for 2 days), 1320 cGy TBI +fludarabine (25 mg/m2 daily for 3).
Graft versus host disease (GVHD) prophylaxis : cyclosporine+ mycophenolate mofetil
Endpoints and Definitions : LFS ,cumulative incidence of relapse, TRM,
neutrophil recovery, platelet recovery, acute and chronic GVHD.
Statistical Considerations :Patient and transplant characteristics by conditioning
intensity were compared using the chi-square test for categorical data
and the Wilcoxon rank-sum test for continuous data.
Cox proportional hazards model or the Fine &Gray method for competing hazards was
used for multivariate regression of leukemia free survival and incidences of relapse and
transplant related mortality respectively.
Patient and Grafts Characteristics
Patient and Grafts Characteristics
Patient and Grafts Characteristics
• The two groups were similar for patient and graft characteristics
except for age ; as age > 45 years (n=66) were only eligible for RIC
Secondary AML or MPD 28 % (RIC) , 4 % (MAC) p< 0,01.
The majority of patients received a transplant of 2 partially HLA-
matched UCB units.
Donor-recipient HLA disparity was similar between groups.
RESULTS : LFS
The probability of LFS at 3-years after RIC was 31% and after MAC 55 %(figure 1A)
Considering all patients, the LFS for patients in CR1 was 43% (95%CI, 31-54%), in
CR2w/CR1<1y was 26% (95%CI, 10-45%) and in CR2w/CR1≥1y was 54% (95%CI, 30-
For recipients of MAC the LFS for patients transplanted in CR1 was 52% (95%CI, 31-
69%), in CR2w/CR1<1y was 61% (95%CI, 27-84%), and in CR2w/CR1≥1y was
65%(95%CI, 25-87%) (Figure 1B).
LFS after RIC was 38% (95%CI, 24-52%) in CR1, 7% (95%CI, 1-28%) in
CR2w/CR1<1y, and 47% (95%CI, 18-72%) in CR2w/CR1≥1y (Figure 1C).
In multivariate analysis, both the intensity of conditioning and disease status at the time
of transplantation were independent predictors of LFS (Table 3).
Recipients of RIC had a 2.3-fold higher risk of relapse or death as compared to MAC
The median follow-up of survivors for RIC was 3.8 years and forMAC 4.5 years (p=0.4).
The most frequent causes of death after RIC (n=44) were relapse 61%, infection 22%
and GVHD 5%, while after MAC (n=19) were relapse 32%,infection 36% and GVHD
• The incidence of relapse at 3-years after RIC was 43% (95%CI, 31%-55%) and after
MAC was 9% (95%CI, 5%-18%; p<0.01) (Figure 2A).
The overall incidence of relapse for patients in CR1 was 30% (95%CI, 19-41%), in CR2
w/CR1<1y was 48% (95%CI, 27-69%) and in CR2 w/CR1≥1y was 21% (95%CI, 3-
In recipients of MAC the relapse incidence in CR1 was 12% (95%CI, 0-24%), in
CR2w/CR1<1y was 19% (95%CI, 0-42%), and in CR2w/CR1≥1y was 13% (95%CI, 0-
33%) (Figure 2B).
patients with RIC had relapse incidence of 40% (95%CI, 25-55%) in CR1, 71%
(95%CI, 21-99%) in CR2w/CR1<1y, and 27% (95%CI, 2-52%) in CR2w/CR1≥1y
Median time to relapse was 1.8 years in RIC and not reached in MAC group.
In multivariate analysis, after adjusting for disease status at transplantation, RIC had a
4.7 fold higher risk of relapse as compared to MAC recipients (table 3).
• TRM at 2 years was similar in the two groups (19% [95%CI, 10-28%]
in RIC vs.27% [95%CI, 14-40%] in MAC, p=0,5.
TNC cell dose or CD34+ cell not affected the incidence of TRM.
The incidence of TRM for patients with HCT-CI score of 0, 1-2 or ≥3
was 17% (95%CI, 2-32%), 16% (95%CI, 4-28%) and 29%(95%CI, 14-
41%) respectively (p=.31)
TRM was also not significantly influenced by disease status at
transplant , Table 3
No variables were independent predictors of TRM in the multivariate
RESULTS : Hematopoietic Recovery
• The incidence of neutrophil recovery at day 42 after RIC was 94%
(95%CI, 89-99%) at a median of 10 days (range, 5-39 days).
MAC was 82% (95%CI 71- 93%) at a median of 23 days (range, 13-
38 days) (p<.01).
In contrast, the proportion with full chimerism at day 21 among
evaluable patients (n=105) after RIC was 20%(95%CI,10-30%) and
after MAC was 60% (95%CI, 45-75) (p<.01).
The cumulative incidence of platelet recovery ≥ 50,000/μL after RIC
was 68% (95%CI, 51-85%) at a median time of 55 days (range 0-181
days) and after MAC was 67% (95% CI, 50-84%) at a median of 77
days (range 42-177 days) (p=0.30).
RESULTS : Graft-vs.-Host Disease
• The incidences of grade II-IV acute GVHD after RIC was
47% (95%CI, 35-59%) and after MAC was 67% (95%CI,
• Similarly the incidence of gradeIII-IV acute GVHD was
lower after RIC as compared to MAC (Table 2).
• However,subgroup analysis evaluating risk of grade II-IV
acute GVHD in recipients of ATG vs.no-ATG after RIC
showed lower incidence with ATG (27% [95%CI, 9-45%]
vs. 56%[95%CI, 41-71%], p=0.04).
• The 2-year incidence of chronic GVHD was similar in RIC
and MAC recipients (30 [95%CI, 19-41%] vs. 34%
[95%CI, 19-49%], p=0.43).
• The main observations of this study were that :
1) there was a lower risk of relapse and superior LFS after MAC,
with similar TRM regardless of the intensity of the conditioning
2) Lower risk of acute GVHD after RIC.
3) similar rates of sustained donor engraftment.
This study is unique as it is focused on recipients of UCB grafts with
all patients receiving a uniform conditioning regimen (within their
respective groups) and supportive care.
In this study, the risk of relapse after MAC UCB HCT was relatively
low (9% at 3 years), and compared favorably to that of other donor
In this retrospective study we had insufficient molecular data to
further define leukemia phenotypes and the risk of relapse.
Discussion Download full-text
• Considering RIC patient were older compared with MAC in our study,
it is possible that poor prognosis inherent with older age could not be
overcome by the GVL effect of RIC UCB transplantation.
A larger study is required….
Our institutional defined age cut off for offering RIC rather than MAC
UCB transplantation may need to be revised, in particular for older
patients who are in good clinical condition and are likely to tolerate
more intensive therapy.
We did not observe a significant difference for 2-year TRM between
conditioning groups. This result was seen despite the possible bias to
select older and less fit patients with comorbidities for RIC
Disease control was especially poor for patients in CR2 with short C1