Effect of Graft Source on Unrelated Donor Hemopoietic Stem
Cell Transplantation in Adults with Acute Myeloid Leukemia
after Reduced-Intensity or Nonmyeloablative Conditioning:
A Study from the Société Francaise de Greffe de Moelle et de
, Noel Milpied3
, Didier Blaise4
, Patrice Chevallier1
, Mauricette Michallet5
, Laurence Clément7
, Yosr Hicheri8
, Catherine Cordonnier9
, Anne Huynh10
, Regis Peffault de Latour12
, Mohamad Mohty2,13,*
1Hematology Department, Nantes Hospital and University Center (CHU), Nantes, France 2INSERM UMR 938 and Pierre et Marie Curie University, Paris, France
3Hematology and Cell Therapy Department, University Hospital and University of Bordeaux, Bordeaux, France 4Blood and Marrow Transplantation Program, Hematology Department, Institut Paoli-Calmettes, Marseille, France
5Hematology Department, Groupement Hospitalier Sud, Hospices Civils de Lyon, Claude Bernard Lyon EST University, Pierre Bénite, France 6Department of Hematology, Hautepierre Hospital, Strasbourg, France
7Department of Hematology, Nancy CHU, UTMA, Vandoeuvre-lès-Nancy, France 8Hematology Department, Montpellier CHU, Montpellier, France
9Department of Hematology, AP-HP, Henri Mondor Hospital, and University Paris-Est-Creteil, Créteil, France 10Hematology Department, Institut Universitaire du Cancer Toulousee Oncopole, Toulouse, France 11Bone Marrow Transplantation Unit, Lille CHU, Lille, France
12Blood and Marrow Transplantation Unit, Saint-Louis Hospital, AP-HP, Paris, France 13Hematology Department, AP-HP, Saint Antoine Hospital, Paris, France
Received 30 December 2014 Accepted 14 February 2015 Key Words:
Allogeneic stem cell transplantation Acute myeloid leukemia Reduced-intensity conditioning Nonmyeloablative conditioning Umbilical cord blood 9/10 Mismatched unrelated donor
10/10 Matched unrelated donor
a b s t r a c t
This retrospective report compared the 4-year outcomes of allogeneic stem cell transplantation (allo-SCT) in 651 adult patients with acute myeloid leukemia receiving a reduced-intensity (RIC) or nonmyeloablative conditioning (NMA) regimen according to the type of unrelated donors. These were either umbilical cord blood (UCB, n¼ 205), a 9/10 mismatched unrelated donor (MisMUD, n ¼ 99), or a 10/10 matched unrelated donor (MUD, n¼ 347) graft. Neutrophil recovery was slower in UCB (74.5% by day 42) compared with MisMUD (94.8%) and MUD (95.6%) (P< .001). There was no signiﬁcant difference in nonrelapse mortality between UCB and both MUD (hazard ratio [HR], 1.05; 95% conﬁdence interval [CI], .62 to 1.78; P ¼ .85) and MisMUD (HR, 1.58; 95% CI, .88 to 2.83; P¼ .13) The relapse/progression was similar between UCB and Mis-MUD (HR, .62; 95% CI, .37 to 1.03; P¼ .07), but was signiﬁcantly lower in MUD compared with UCB (HR, .60; 95% CI, .39 to .92; P¼ .02). The rate of extensive chronic graft-versus-host disease (GVHD) was similar be-tween UCB and both MUD (HR, 2.15; 95% CI, .93 to 4.97; P¼ .08) and MisMUD (HR, 1.84; 95% CI, .68 to 4.95; P¼ .23). The rate of severe grade III and IV acute GVHD was signiﬁcantly increased in MisMUD compared with UCB (HR, 2.61; 95% CI, 1.30 to 5.23; P¼ .007). There was no signiﬁcant difference in overall survival between UCB and both MisMUD (HR, .98; 95% CI, .66 to 1.45; P¼ .92) and MUD (HR, .74; 95% CI, .52 to 1.03; P ¼ .08). These data suggest that in the setting of RIC/NMA, allo-SCT UCB is a valid alternative graft source, with signiﬁcantly less chronic GVHD, compared with MisMUD, when there is no MUD available or when urgent transplantation is needed.
Ó 2015 American Society for Blood and Marrow Transplantation.
Allogeneic stem cell transplantation (allo-SCT) is a well-established therapy for the treatment of acute myeloid leu-kemia (AML) in adult patients. Although HLA-matched
Financial disclosure: See Acknowledgments on page 1065.
* Correspondence and reprint requests: Mohamad Mohty, MD, PhD, Department of Hematology, Hopital Saint Antoine, 184 Rue du Faubourg Saint Antoine, 75571 Paris cedex 12, France.
E-mail address:email@example.com(M. Mohty).
1083-8791/Ó 2015 American Society for Blood and Marrow Transplantation.
Biology of Blood and
sibling donor (MSD) is the preferred donor, only about 30% of patients have such a donor available. The lack of MSD leads transplantation centers to search for alternative donors, which now account for more than one half of trans-plantations performed[1,2]. Currently, for AML patients who lack an MSD but require allo-SCT, the choice will primarily be an HLA-matched unrelated donor (MUD). However, the likelihood of identifying a MUD is, at most, 75% and it could be much lower depending of the patient’s ethnic origin. Therefore, umbilical cord blood (UCB) and mismatched un-related donor (MisMUD) have emerged as effective stem cell sources for allo-SCT candidates who cannot beneﬁt from an MSD or a MUD. Several studies found a comparable leukemia-free survival between UCB, HLA 8/8 MUD, and HLA 7/8 MisMUD in the setting of myeloablative conditioning (MAC)[4,5], reduced-intensity conditioning (RIC), and non-myeloablative (NMA) regimens . One study evaluating alternative donors was limited to elderly AML patients in ﬁrst complete remission. However, in spite of the patient population comprising only those over the age of 50, both MAC and RIC/NMA regimens were included. MAC regi-mens are limited to younger patients and to patients in good medical condition because of the high incidence of non-relapse mortality (NRM). RIC/NMA regimens have been developed in an attempt to decrease NRM in elderly patients, in heavily pretreated patients, or in those with medical comorbidities precluding the use of standard MAC regimens
[9,10]. RIC and NMA regimens aim to take advantage of the
graft-versus-leukemia effect mediated by the donor’s
immunocompetent cells rather than the upfront cytoreduc-tive effect of the conditioning chemotherapy. Evaluating the effect of alternative donors in patients who received a RIC/NMA regimen seems more relevant than only in patients over age 50, whatever their conditioning.
Furthermore, in many non-European centers, HLA matching was performed at the HLA-A, -B, -C, and -DRB1 high-resolution level (8/8 HLA matching) whereas, in many European centers, unrelated donor choice includes also HLA-DQB1 (10/10 HLA matching) for unrelated donors, as similar outcomes between MUD and 10/10 HLA MUD were reported
Taking all this into consideration, the aim of our study was to evaluate outcomes of AML adult patients after 10/10 HLA MUD peripheral blood stem cell (PBSC), HLA 9/10 Mis-MUD PBSC, and UCB allo-SCT in the RIC/NMA setting.
PATIENTS AND METHODS Study Design
This was a retrospective and multicenter study of the Société Française de Greffe de Moelle et de Therapie Cellulaire (participating centers listed in
Supplemental Data). For the purpose of this analysis, all patients diagnosed with AML (except acute promyelocytic leukemia) who were at least 18 years of age at the time of transplantation, and who received a granulocyte colonyestimulating factoremobilized PBSC MUD (MUD group), a gran-ulocyte colonyestimulating factoremobilized PBSC 1 allele or antigen-mismatched unrelated donor (MisMUD group), or 1 or 2 UCB (UCB group) transplantation after a RIC or a NMA conditioning were included. These included 651 patients who underwent transplantation between April 2002 and December 2010. There were 99 patients in the MisMUD, 205 in the UCB, and 347 in the MUD groups. Patients without a suitable HLA 10/10-matched related or unrelated donor were considered for a 9/10 MisMUD. However, if there was no sufﬁcient time for an unrelated donor search because of a high-risk disease, or if the preliminary search indicated a lower likelihood of ﬁnding an unrelated donor within a reasonable time frame, the use of UCB was preferred.
The eligibility criteria for RIC or NMA allo-SCT included the following: (1) patients above age 50, (2) heavily pretreated patients who received an autologous stem cell transplant (auto-SCT) or more than 2 lines of chemo-therapy before allo-SCT, and (3) patients with poor performance status
because of signiﬁcant medical comorbidities as described by Sorror et al.. Written informed consent was obtained from each patient and donor. For the purpose of this study, patients who received a MAC regimen were excluded; all patients received either a RIC or a NMA regimen. The following regimens were considered as NMA: low-dose (2 Gy) total body irradiation (TBI),ﬂudarabine (Flu) and cyclophosphamide (Cy); 2 Gy TBI and Flu; Flu and Cy; low-dose busulfan (3.2 mg/kg) and Flu. All other regimens were considered RIC regimens.
All donor-recipient pairs were typed at the allelic level. They wereﬁrst typed at the 2-digit level for HLA class I (HLA-A, HLA-B, and HLA-C) and class II (HLA-DRB1 and HLA-DQB1) using published HLA class I PCR sequence-speciﬁc oligonucleotide and/or PCR sequence-sequence-speciﬁc primers typing protocols. Per study deﬁnition, all donor-recipient pairs were matched at HLA-A, -B, -C, -DRB1, and -DQB1 at the allele level (10/10 matched) in the MUD group, and in the MisMUD group, all donor/recipient pairs presented a single HLA mismatch (9/10 matched). Such mismatch was at HLA-A for 24 patients (24%), HLA-B for 16 patients (16%), HLA-C for 36 patients (37%), and HLA-DQB1 for 23 patients (23%), with no donor-recipient pair presenting a mismatch at HLA-DRB1 (Table 1). HLA-DP typing was not routinely per-formed during the study period. HLA typing was perper-formed according to the current protocol of the European Federation for Immunogenetics Histo-compatibility Laboratory standards. HLA typing in UCB recipients was per-formed at antigen level for HLA class I A and B loci and at allelic level with high-resolution typing for HLA class II DRB1 locus. Matching at HLA-C or allele level for HLA-A and -B were not considered, as these data were not available at the time analysis. The majority (n¼ 178; 87%) of UCB donor-recipient pairs were mismatched at 1 or 2 HLA loci, 12 (6%) presented a mismatched at 3 HLA loci, 6 (3%) presented no HLA mismatch, and in 9 (4%), the donor-recipient HLA mismatch was unknown. Of the 205 patients who received UCB grafts, 99 (48%) received 1 UCB unit and 106 (52%) received 2 units. We decided to group together UCB transplantation using 1 or 2 UCB units, as no signiﬁcant differences in survival are reported between the 2
[14,15]. Furthermore, univariate analysis was performed to compare patient outcomes after single and double UCB in our cohort: there was no difference in term of overall survival (OS), progression-free survival (PFS), NRM, relapse/progression, acute and chronic graft-versus-host disease (GVHD) (data not shown). HLA match between UCB units (n ¼ 106) was not considered. All patients received the preparative regimen as inpatients in single rooms and remained hospitalized until hematopoietic and clinical recovery. Supportive care and antimicrobial prophylaxis were left to the initiative of the centers[16,17]. All blood products wereﬁltered, irradiated, and screened for cytomegalovirus (CMV). In theﬁrst 100 days after allo-SCT, patients were assessed at least once per week for CMV infection to initiate preemptive therapy. For GVHD prophylaxis, most patients received either cyclosporine A (CsA) alone or in combination with methotrexate or myco-phenolate mofetil (MMF). CsA was administered at a dose of 3 mg/kg/day by continuous intravenous infusion starting from day -3 or -2, and changed to twice daily oral dosing as soon as tolerated. Methotrexate was admin-istered as a short course (15 mg/m2at dayþ1, 10 mg/m2at dayþ3 and þ6).
MMF was given at aﬁxed oral dose of 2 g/day. No treatment adjustment was performed for MMF. MMF was tapered over 4 weeks starting from day 60, and CsA from day 90 if no GVHD appeared.
Recorded clinical outcomes after transplantation included time to neutrophil and platelet engraftment, time of onset and severity of acute GVHD (aGVHD), chronic GVHD (cGVHD), disease relapse or progression, PFS, and OS. Time to neutrophil recovery was deﬁned as the ﬁrst of 3 consecutive days in which the absolute neutrophil count exceeded .5 109/L, and time to
platelet recovery as theﬁrst of 3 days with 50 109/L or higher without a
need for platelet transfusion during a 5-day period. Acute and chronic GVHD were evaluated according to standard criteria[19,20], as data to determine National Institutes of Health cGVHD classiﬁcation were not available for most patients. NRM was deﬁned as death occurring in the absence of leu-kemia relapse. OS was the primary endpoint. Secondary endpoints included overall PFS, relapse rate, NRM, aGVHD and cGVHD, and engraftment. Statistical Methods
OS and PFS were calculated by the Kaplan-Meier method and subgroups were compared using the log-rank test. For categorical variables, compari-son between the MisMUD, the MUD, and the UCB groups was carried out using the chi-squared test. For continuous data, comparison between regi-mens was performed using the Mann-Whitney test. Probabilities of relapse/ progression, NRM, GVHD, and neutrophil and platelet recovery were calculated using the cumulative incidence procedure and comparisons were performed using the Gray test. Univariate analysis was performed to
compare inﬂuence of graft type, patient’s age at transplantation (greater or younger than 40 years), sex of the donor, year of transplantation (2002 to 2006 versus 2007 to 2010), disease status at time of transplantation, cyto-genetic status, conditioning regimen, use of TBI, use of antithymocyte globulin (ATG), GVHD prophylaxis, and time between diagnosis and trans-plantation. Also, these parameters were evaluated in a multivariate analysis for outcome. The only covariate with missing data was cytogenetic status (missing in 13.7% of patients) for which an“unknown” group was created. To study the association between graft type and outcomes, multivariate Cox regressions models were built. The variable for graft type as the main study question was retained in all steps of model building. All P values are 2-sided and values .05 were considered signiﬁcant. Data were computed using the
R package (R Foundation for Statistical Computing, Vienna, Austria) and GraphPad Prism 5.0 (GraphPad Software, San Diego, CA).
Patients and Transplantation Characteristics
Patients and transplantation characteristics for the whole study population and for each group are detailed inTable 1. Compared with PBSC recipients, UCB recipients were younger. The median age at transplantation was 49 years in the UCB group, compared with 55 years in the MisMUD and
Study Population and Transplantation Characteristics
Characteristic UCB (n¼ 205) MisMUD (n¼ 99) MUD (n¼ 347) P Value Patient age, median (range), yr 49.0 (19.3-69.1) 55.1 (19.0-68.2) 57.1 (19.0-70.3) <.001 Gender (female) 117 (57.1%) 56 (57.0%) 166 (47.8%) .07 CMV-seronegative donor-recipient pair 44 (21.5%) 21 (21.0%) 120 (34.6%) <.001
Disease status <.001 CR1 100 (48.8%) 47 (48.0%) 193 (57.1%) CR> 1 83 (40.5%) 30 (30.0%) 92 (26.5%) Relapse/progression 22 (10.7%) 22 (22.0%) 57 (16.4%) Cytogenetic risk .31 Low 14 (6.8%) 5 (5.0%) 28 (8.1%) Intermediate 115 (56.1%) 63 (64.0%) 225 (64.8%) High 42 (20.5%) 16 (16.0%) 54 (15.6%) Unknown 34 (16.6%) 15 (15.0%) 40 (11.5%)
Stem cell source NA
PBSC - 99 347
1 UCB 99 (48.3%) -
-2 UCB 106 (51.7%) -
-HLA matching: 1 UCB
6/6 4 (2.0%) -
-6/5 26 (12.7%) -
-4/6 61 (29.8%) -
-3/6 6 (2.9%) -
-Missing data 2 (1.0%) -
-HLA matching: dUCB NA
6/6þ 6/6 2 (1.0%) - -6/6þ 5/6 6 (2.9%) - -6/6þ 4/6 2 (1.0%) - -5/6þ 5/6 20 (9.8%) - -5/6þ 4/6 23 (11.2%) - -4/6þ 4/6 40 (19.5%) - -4/6þ 3/6 5 (2.4%) - -3/6þ 3/6 1 (.5%) - -Missing data 7 (3.4%) - -HLA matching: 9/10 NA Mismatch at HLA-A - 24 (24.0%) -Mismatch at HLA-B - 16 (16.0%) -Mismatch at HLA-C - 36 (37.0%) -Mismatch at HLA-DRB1 - 0 (0%) -Mismatch at HLA-DQB1 - 23 (23.0%) -Cell dose, median (range)*
TNC 108/kg 0.31 (0.02-0.82) 8.70 (2.49-18.24) 8.48 (.05-19.23) <.001 CD34þcells 106/kg 0.1 (.04-0.9) 5.71 (1.85-20.66) 6.40 (.20-21.4) <.001 Conditioning regimen <.001 NMA 179 (87.0%) 27 (27.0%) 108 (31.1%) RIC 26 (13.0%) 72 (73.0%) 239 (68.9%) TBI-based regimen <.001 Yes 194 (95.0%) 28 (28.0%) 83 (23.9%) No 11 (5.0%) 71 (72.0%) 264 (76.1%) ATG <.001 Yes 9 (4.0%) 80 (81.0%) 265 (76.4%) No 196 (96.0%) 19 (19.0%) 82 (23.6%) GVHD prophylaxis <.001 CsAþ MMF 160 (78.0%) 52 (53.0%) 178 (51.3%) CsAþ MTX 2 (1.0%) 21 (21.0%) 49 (14.1%) CsA 8 (3.9%) 18 (18.0%) 103 (29.7%) Others 4 (2.0%) 6 (6.0%) 10 (2.9%) Unknown 31 (15.1%) 2 (2.0%) 7 (2.0%)
Median follow-up, (range), mo 48.6 (12.0-97.8) 42.8 (18.7-104.9) 43.2 (12.1-126.7) .02 NA indicates not available; dUCB, double umbilical cord blood; TNC, total nucleated cells; MTX, methotrexate.
Data presented are n (%), unless otherwise indicated.
57 years in the MUD group. In the latter, fewer patients were female and more donor-recipient pairs were CMV seroneg-ative. Although the majority of patients underwent trans-plantation in complete remission (CR) in all groups, there were more patients inﬁrst CR in the MUD group and fewer patients in relapse or progression in the UCB group at the time of transplantation. Whereas cytogenetic data were available for the majority of patients (86.3%), molecular ge-netic data were missing for many (36.6%) and, for this reason, only cytogenetic data were considered to classify patients according to the European Leukemia Net criteria . Favorable risk cytogenetics was similar across the 3 groups, ie, 6.8% in UCB, 5.0% in MisMUD, and 8.1% in MUD, but the frequency of high cytogenetic risk was increased in the UCB group: 20.5% versus 16% and 15.6% in the MisMUD and MUD groups, respectively. The median numbers of total nucleated and CD34þcells were lower in the UCB group compared with the MisMUD and the MUD groups. Whereas only 27.0% and 31.1% of patients received an NMA regimen in the MisMUD and MUD groups, respectively, this type of conditioning was used in the majority (87.0%) of patients in the UCB group. This difference was due to the frequent use of the classical NMA Flu, Cy, and 2 Gy TBI conditioning regimen in the UCB setting. Similarly, whereas TBI was applied to 95.0% of patients in the UCB group, only 28.0% and 23.9% of patients received TBI in the MisMUD and MUD groups, respectively. Only 4.0% of patients who were not heavily pretreated in the UCB group received ATG (Thymoglobulin; Genzyme, Lyon, France), whereas 81.0% and 76.4% received ATG in the Mis-MUD and the Mis-MUD groups, respectively, because of the ex-pected higher risk of GVHD in these groups. The combination of CsA and MMF was used in the majority of patients in all groups for GVHD prophylaxis, but it was more frequent in the UCB cohort. The median follow-up among surviving patients was longer in the UCB group (48.6 months) than in the MisMUD (42.8 months) and MUD (43.2 months) groups.
Neutrophil and Platelet Recovery
Univariate analysis of transplantation-related events is summarized inTable 2. Overall hematopoietic recovery was signiﬁcantly slower in the UCB compared with the MisMUD and the MUD groups. Median time to neutrophil recovery was signiﬁcantly longer in the UCB group, 19 (range, 0 to 106) days, compared with 15 (range, 0 to 31) days in the MisMUD and 16 (0 to 52) days in the MUD group (P< .001 for both). The same was observed for median time to platelet recovery, which was signiﬁcantly longer in the UCB group, 41 (range, 1 to 222) days compared with 13 (range, 4 to 73) days in the MisMUD and 12 (range, 0 to 371) days in the MUD group (P< .001 for both). The day 42 cumulative incidences of neutro-phil recovery was signiﬁcantly lower in the UCB group, 74.5% (95% conﬁdence interval [CI], 70.9% to 78.1%) compared with 94.8% (95% CI, 92.4% to 97.2%) in the MisMUD and 95.6% (95% CI, 94.0% to 96.4%) in the MUD group (P< .001 for both) (Figure 1A). The day 180 cumulative incidence of platelet recovery was signiﬁcantly lower in the UCB group: 56.0% (95% CI, 52.2% to 59.8%) versus 75.3% (95% CI, 70.7% to 79.9%) in the MisMUD and 84.2% (95% CI, 82.1% to 86.3%) in the MUD group (P< .001 for both) (Figure 1B).
GVHD and NRM
Multivariate analysis of transplantation-related events is summarized inTable 3. The cumulative incidence of severe grade III and IV aGVHD was 11.7% in the UCB group compared with 20.2% in the MisMUD (P¼ .05) and 13.0% in the MUD group (P¼ .71) (Figure 2A). In multivariate analysis, the rate of severe grade III and IV aGVHD was similar between UCB and MUD groups (hazard ratio [HR], 1.72; 95% CI, .93 to 3.19; P¼ .08). However, the rate of severe grade III and IV aGVHD was signiﬁcantly increased in the MisMUD group compared with the UCB group (HR, 2.61; 95% CI, 1.30 to 5.23; P¼ .007). Graft type was the only parameter with a signiﬁcant impact on grade III and IV aGVHD in multivariate analysis.
Transplantation-Related Events Univariate Analysis
Outcome UCB (n¼ 205) MisMUD (n¼ 99) MUD (n¼ 347) P Value (CB versus MisMUD) P Value (CB versus MUD) OS At 2 Yr (95% CI) 47.9 (40.9-54.6) 50.5 (40.3-59.8) 56.7 (51.2-61.8) .81 .007 At 5 Yr (95% CI) 36.1 (28.7-43.5) 32.6 (23.6-44.0) 48.6 (42.2-54.7) PFS At 2 Yr (95% CI) 42.0 (35.2-48.7) 49.5 (39.3-58.9) 54.0 (48.6-59.1) .53 .001 At 5 Yr (95% CI) 28.3 (21.2-35.8) 33.7 (23.3-44.4) 47.4 (41.5-53.0) NRM At 2 Yr (cumulative incidence) 18.6% (15.9-21.3) 23.2% (18.9-27.5) 16.6% (14.6-18.6) .08 .73 At 5 Yr (cumulative incidence) 21.0% (17.9-24.1) 30.5% (25.7-35.7) 20.5% (18.1-22.9) Acute GVHD at days 100 Grade 2-4 32.2% (28.9-35.5) 39.4% (34.5-44.3) 31.1% (28.6-33.6) .21 .73 Grade 3-4 11.7% (9.4-14.0) 20.2% (16.1-24.3) 13.0% (11.2-14.8) .05 .71 Overall chronic GVHD At 2 Yr (cumulative incidence) 16.1% (13.5-18.7) 29.3% (24.7-33.9) 36.7% (34.1-39.3) .004 <.001 At 5 Yr (cumulative incidence) 16.6% (14.0-19.2) 30.5% (25.8-35.2) 39.1% (36.4-41.8) Extensive chronic GVHD At 2 Yr (cumulative incidence) 5.8% (4.2-7.4) 10.1% (7.0-13.2) 11.5% (9.8-13.2) .18 .02 At 5 Yr (cumulative incidence) 5.8% (4.2-7.4) 10.1% (7.0-13.2) 12.2% (10.4-14.0) Relapse/progression At 2 Yr (cumulative incidence) 39.6% (36.2-43.0) 27.3% (22.8-31.8) 29.6% (27.1-32.1) .03 .002 At 5 Yr (cumulative incidence) 50.8% (46.9-54.7) 35.7% (30.2-41.2) 32.2% (29.6-34.8)
ANC> .5 109/L, median (range), d 19 (0-106) 15 (0-31) 16 (0-52) <.001 .001
At Day 28 (cumulative incidence) 72.8 (69.5-76.1) 94.8 (92.4-97.2) 95.6 (94.0-96.4) <.001 <.001 At Day 42 (cumulative incidence) 74.5 (70.9-78.1) 94.8 (92.4-97.2) 95.6 (94.0-96.4)
Platelet> 50 109/L, median (range), d 41 (1-222) 13 (4-73) 12 (0-371) <.001 <.001
Platelet> 50 109/L at d 180
56.0 (52.2-59.8) 75.3 (70.7-79.9) 84.2 (82.1-86.3) <.001 <.001
At 2 years, the cumulative incidence of extensive cGVHD was 5.8% in the UCB group, 10.1% in the MisMUD (P¼.18), and 11.5% in the MUD group (P¼ .02) (Figure 2B). In multivariate analysis, the rate of extensive cGVHD was similar between UCB group and both the MUD (HR, 2.15; 95% CI, .93 to 4.97; P¼ .08) and the MisMUD group (HR, 1.84; 95% CI, .68 to 4.95; P¼ .23). No parameter, including the use of ATG, had a sig-niﬁcant impact on extensive cGVHD in multivariate analysis.
NRM, Relapse Incidence, PFS, and OS
At 2 years, the cumulative incidence of NRM was 18.6% in the UCB group compared with 16.6% in the MUD group (P¼ .73) and 23.2% in the MisMUD group (P¼ .08) (Figure 3A). In multivariate analysis, there was no signiﬁcant difference in NRM between the UCB group and both the MUD (HR, 1.05;
95% CI, .62 to 1.78; P¼ .85) and the MisMUD group (HR, 1.58; 95% CI, .88 to 2.83; P¼ .13). At 2 years, the cumulative inci-dence of relapse/progression was 39.6% in the UCB group and 27.3% and 29.6% in the MisMUD and MUD groups, respec-tively (P ¼ .03 and P ¼ .002) (Figure 3B). In multivariate analysis, the rate of relapse/progression was similar between the UCB and the MisMUD groups (HR, .62; 95% CI, .37 to 1.03; P¼ .07), but it was signiﬁcantly lower in the MUD group than in the UCB group (HR, .60; 95% CI, .39 to .92; P ¼ .02). Furthermore, the rate of relapse/progression was signi ﬁ-cantly increased in patients in relapse/progression at trans-plantation and in those with intermediate or high-risk cytogenetic.
The Kaplan-Meier estimate of PFS at 2 years was 42.0% (95% CI, 35.2% to 48.7%) in the UCB group versus 49.5% (95% CI, 39.3% to 58.9%) in the MisMUD group (P¼ .53) and 54.0%
(95% CI, 48.6% to 59.1%) in the MUD group (P ¼ .001)
(Figure 3C). In multivariable analysis, there was no signi ﬁ-cant difference in PFS between the UCB group and either the MisMUD (HR, 1.17; 95% CI, .86 to 1.65; P¼ .92) and the MUD groups (HR, 1.10; 95% CI, .84 to 1.42; P¼ .49). Similar to the cumulative incidence of relapse/progression, PFS was signiﬁcantly decreased in patients in relapse/progression at transplantation and in those with intermediate or high-risk cytogenetic. The 2-year OS was 47.9% (95% CI, 40.9% to 54.6%), 50.5% (95% CI, 40.3% to 59.8%), and 56.7% (95% CI, 51.2% to 61.8%) in the UCB, MisMUD, and MUD groups, respectively (P¼ .81 and P ¼ .007) (Figure 3D). In multivar-iable analysis, there was no signiﬁcant difference in OS be-tween UCB group and both MisMUD (HR, .98; 95% CI, .66 to 1.45; P¼ .92) and MUD groups (HR, .74; 95% CI, .52 to 1.03; P¼ .08). In contrast, OS was signiﬁcantly decreased in pa-tients who underwent transplantation in
relapse/progres-sion compared with patients who underwent
transplantation in CR (HR, 3.15; 95% CI, 2.40 to 4.14; P< .001) and in patients over 40 years of age at time of transplantation (HR, 1.35; 95% CI, 1.02 to 1.79; P¼ .04).
The use of RIC and NMA regimens before allo-SCT has widely expanded over the past decade, allowing patients with AML and comorbidities or older age to beneﬁt from a potential immune graft-versus-leukemia effect. The primary objective of this retrospective study was to assess the relative
efﬁcacy of UCB compared with HLA-matched and 9/10
Figure 1. Engraftment after allo-SCT in the MUD (green line), the MisMUD (blue line) and the UCB groups (red line). (A) Shows cumulative incidence of neutrophil recovery to more than .5 109/L. (B) Shows cumulative incidence of platelet recovery to more than 50 109/L.
Transplantation-Related Events Multivariate Analysis
P Value OS
9/10 HLA-matched PBSC versus UCB .98 (.66-1.45) .92 10/10 HLA-matched PBSC versus UCB .74 (.52-1.03) .08 Relapse/progression versus CR 3.15 (2.40-4.14) <.001 Age> 40 yr versus < 40 yr 1.35 (1.02-1.79) .04 PFS
9/10 HLA-matched PBSC versus UCB 1.17 (.86-1.65) .29 10/10 HLA-matched PBSC versus UCB 1.10 (.84-1.42) .49 Relapse/progression versus CR 2.94 (2.26-3.83) <.001 Cytogenetic adverse versus favorable 2.67 (1.55-4.61) <.001 Cytogenetic Intermediate versus favorable 2.15 (1.29-3.59) .003 NRM
9/10 HLA-matched PBSC versus UCB 1.58 (.88-2.83) .13 10/10 HLA-matched PBSC versus UCB 1.05 (.62-1.78) .85 Relapse/progression versus CR 2.25 (1.41-3.58) <.001 Age> 40 yr versus < 40 yr 1.89 (1.10-3.24) .02 Relapse/progression
9/10 HLA-matched PBSC versus UCB .62 (.37-1.03) .07 10/10 HLA-matched PBSC versus UCB .60 (.39-.92) .02 Relapse/progression versus CR 3.64 (2.61-5.06) <.001 Cytogenetic adverse versus favorable 4.47 (2.02-9.85) <.001 Cytogenetic intermediate versus favorable 3.43 (1.60-7.36) .002 Grade III-IV acute GVHD
9/10 HLA-matched PBSC versus UCB 2.61 (1.30-5.23) .007 10/10 HLA-matched PBSC versus UCB 1.72 (.93-3.19) .08 Extensive chronic GVHD
9/10 HLA-matched PBSC versus UCB 1.84 (.68-4.95) .23 10/10 HLA-matched PBSC versus UCB 2.15 (.93-4.97) .08
mismatched unrelated donor PBSC in adults with AML treated with RIC or NMA conditioning regimens.
In this study, with a median follow-up of more than 3.5 years, UCB and 9/10 HLA MUD PBSC appear to be suitable alternative stem cell sources when there is no 10/10 HLA MUD PBSC available: in multivariate analysis OS, PFS, and NRM were comparable between the 3 stem cells sources. The ﬁnding of a comparable NRM between the 3 stem cells sources is 1 of the most interesting discoveries of our study. Indeed, patients eligible to RIC/NMA allo-SCT are frequently elderly, heavily pretreated, or impaired by comorbidities, and most physicians hesitate to use UCB to perform RIC/NMA
allo-SCT in them when there is no matched related or un-related donor available. The main cause of NRM after UCB transplantation is infectious complications [23-25], but in our retrospective study, we were unable to analyze the impact of infection-related NRM. The cumulative incidence of grade III and IV aGVHD and cGVHD was lower in the UCB group compared with the MisMUD. In contrast, despite only PBSC being used in the MisMUD and MUD groups[26,27], the cumulative incidence of extensive cGVHD was similar be-tween UCB and both MisMUD and MUD. Nevertheless, the frequent use of ATG in those groups, which has increasingly proven to exert a protective effect against severe cGVHD
Figure 2. Outcome after allo-SCT of the MUD (green line), the MisMUD (blue line) and the UCB groups (red line). (A) Shows cumulative incidence of grade III and IV acute GVHD, and (B) shows cumulative incidence of extensive chronic GVHD.
Figure 3. Outcome after allo-SCT of the MUD (green line), the MisMUD (blue line) and the UCB groups (red line). (A) Shows cumulative incidence of nonrelapse mortality; (B) shows cumulative incidence of relapse; (C) shows progression-free survival; and (D) shows overall survival.
[28-30], probably limited the higher rate of extensive cGVHD, particularly in the MisMUD group.
Our results are comparable to those reported in ran-domized studies evaluating ATG in MUD and MisMUD allo-SCT. Finke et al. reported a cumulative incidence of extensive cGVHD of 12.2% at 2 years in the ATG group, comparable to our study, against 42.6% in the non-ATG group
. Walker et al. reported an even lower cumulative inci-dence of extensive cGVHD of 4.4% at 2 years in the ATG group, against 431% in the non-ATG group . The cumulative incidence of neutrophil recovery was signiﬁcantly lower af-ter UCB transplantation, consistent with most previous re-ports[6,7,32].
In AML, the main cause of treatment failure after RIC/ NMA allo-SCT is leukemia progression or relapse. The cu-mulative incidence of relapse/progression was signiﬁcantly increased with UCB compared with MUD, including in multivariate analysis. The incidence in our UCB group (39.6% at 2 years [95% CI, 36.2% to 43.0%]) is similar to that reported in a similar setting by Chen et al.(42.7% at 2 years) or Brunstein et al. (49% at 2 years). Furthermore, Peffault de Latour et al. recently described an increased cumulative incidence of relapse with UCB compared with sibling and unrelated donors after RIC allo-SCT. In our study, there was a trend towards a larger number of patients with high-risk cytogenetics in the UCB group and, in multivariate analysis, high-risk cytogenetics was associated with a signiﬁcantly increased risk of relapse/progression and a worse PFS. Thus, the higher relapse/progression rate in the UCB patients might be related to the over-representation of high-risk cytogenetics in this group. As to disease status at transplantation, we acknowledge that there is a trend to-wards a lower number of patients with AML in relapse/pro-gression at transplantation in the UCB group. However, intensity of conditioning regimen is an important determi-nant in AML control, particularly in the setting of progressive disease at transplantation, and patients in the UCB group received signiﬁcantly less cytotoxic regimen, with a majority of NMA, TBI-based conditioning regimen, compared with the MisMUD and MUD groups. The increased risk of relapse in AML patients after NMA low-dose TBI-based conditioning regimen has been previously reported , and our study conﬁrmed such a trend (HR, 1.25; 95% CI, .81 to 1.75; P ¼ .20). The overall use of less cytotoxic conditioning regimen in the UCB group, may have contributed to the increased cumula-tive incidence of relapse/progression. These results highlight the issue of AML control in the setting of RIC and NMA conditioning regimen, making the development of new conditioning regimens to increase cytotoxicity and improve disease control without increasing NRM of paramount importance. Thus, we recently reported that a reduced-toxicity regimen based on Flu, ATG, and 3 days of busulfan appeared to be safe, with a low NRM at 2 years (11%) in high-risk patients, and efﬁcient for disease control . Such conditioning regimens should be developed not only for matched related donors and unrelated donors, but also for UCB and haploidentical family donors. Similarly, the devel-opment of several strategies to reduce relapse after allo-SCT is ongoing, such as hypomethylating agents aloneor in combination with prophylactic delayed lymphocyte infusion or target therapy with FLT3 inhibitors for FLT3-positive AML
Recent studies demonstrated that the UCB trans-plantation outcomes could be improved when matching at the HLA-C locus is consideredand when HLA typing is
performed at the allele level for HLA-A, -B, and -C[39,40]. At this level, mismatch at 1 or 2 alleles was better tolerated, with an improved neutrophil recovery and a decreased NRM
. In our study for UCB selection, HLA typing was per-formed at the antigen level for HLA-A and -B and at the allele level for HLA-DRB1, and data were not available to retro-spectively evaluate the impact of high-resolution HLA class I matching on the outcome after UCB allo-SCT. It seems likely that, in comparison to other alternative donors, further improvement of the UCB allo-SCT outcome can be achieved when class I HLA-matching is performed at the allele level.
It would be very interesting to analyze the impact of the type of mismatch in the MisMUD group. Although it is well established that a single HLA mismatch, even at the allele level, has a signiﬁcant impact on patients’ outcome, the inﬂuence of each HLA locus mismatch is variable. Previous reports suggested that mismatches at HLA-B or -C allele level were better tolerated than those at HLA-A or -DRB1. In daily practice, HLA-DRB1 mismatches are avoided and no patient in our study received a transplant mismatch at the HLA-DRB1 level. However, the type of HLA-A, -B, -C, and -DQB1mismatch was very disparate and, unfortunately, the number of patients in the MisMUD group was too small to allow us to analyze the impact of type of allelic mismatch.
Besides UCB and MisMUD, a third alternative donor has recently emerged: a haploidentical family donor. The results are very promising, especially when GVHD prophy-laxis included the use of post-transplantation Cy . Brunstein at al. reported a similar outcome for lymphoma and leukemia patients after UCB allo-SCT and haploidentical allo-SCT with post-transplantation Cy in the setting of RIC/ NMA. Therefore, comparative studies restricted to AML are necessary before drawing deﬁnitive conclusion. Given the novelty of this option, haploidentical allo-SCT was not routinely used in the Société Française de Greffe de Moelle et de Therapie Cellulaire center during the study period (2002 to 2010), and could not be included in the current analysis.
In conclusion, our results suggest that UCB is a valid alternative to the use of PBSC single allele mismatched un-related donor in the setting of RIC/NMA and compared favorably with PBSC 10/10 MUD. One must acknowledge that UCB is an expensive alternative, but it has the advantage of being readily available compared to sources from living do-nors. Thus, the cost must be weighed carefully relative to the urgency of the transplantation. No prospective comparative study has been performed so far, given the difﬁculty of including enough patients to be able to draw conclusions. Therefore, international prospective comparative studies evaluating alternative donors with HLA-A, -B, and -C typing performed at the allele level for UCB selection are warranted. These studies should also include haploidentical donors and, for 9/10 HLA MisMUD, mismatches should be restricted to those better tolerated (HLA-B and -C at the allele level). Only such restrictive approaches would allow us to draw de ﬁni-tive conclusions regarding the best alternaﬁni-tive donors for AML patients.
The authors thank Nicole Raus for data management. F.M. thanks Dr. M. Labopin (Paris, France) for her advice regarding statistical analysis. M.M. thanks Prof. J.V. Melo (Adelaide, Australia) for critical reading of the manuscript.
Financial disclosure statement: F.M. was supported by educational grants from the Association for Training, Edu-cation and Research in Hematology, Immunology and
Transplantation (ATERHIT). The authors also thank the Ré-gion Pays de Loire, the Association pour la Recherche sur le Cancer (ARC; grant #3175 to M.M. and B.G.), the Fondation de France, the Fondation contre la Leucémie, the Agence de Biomédecine, the Association Cent pour Sang la Vie, the As-sociation Laurette Fuguain, and the International Research Group on Unrelated Hematopoietic Stem Cell Trans-plantation for their generous and continuous support to our clinical and basic research work. Our group is supported by several grants from the Hospital Clinical Research Program from the French National Cancer Institute to MM.
Conﬂict of interest: The authors have nothing to disclose in relation with the content of this analysis.
Supplementary data related to this article can be found online athttp://dx.doi.org/10.1016/j.bbmt.2015.02.014. REFERENCES
1. Malard F, Chevallier P, Guillaume T, et al. Continuous reduced non-relapse mortality after allogeneic hematopoietic stem cell trans-plantation: a single-institution’s three decade experience. Biol Blood Marrow Transplant. 2014;20:1217-1223.
2. Hahn T, McCarthy PL Jr, Hassebroek A, et al. Signiﬁcant improvement in survival after allogeneic hematopoietic cell transplantation during a period of signiﬁcantly increased use, older recipient age, and use of unrelated donors. J Clin Oncol. 2013;31:2437-2449.
3. Gragert L, Eapen M, Williams E, et al. HLA match likelihoods for he-matopoietic stem-cell grafts in the U.S. registry. N Engl J Med. 2014;371: 339-348.
4. Brunstein CG, Gutman JA, Weisdorf DJ, et al. Allogeneic hematopoietic cell transplantation for hematologic malignancy: relative risks and beneﬁts of double umbilical cord blood. Blood. 2010;116:4693-4699. 5. Eapen M, Rocha V, Sanz G, et al. Effect of graft source on unrelated
donor haemopoietic stem-cell transplantation in adults with acute leukaemia: a retrospective analysis. Lancet Oncol. 2010;11:653-660. 6. Brunstein CG, Eapen M, Ahn KW, et al. Reduced-intensity conditioning
transplantation in acute leukemia: the effect of source of unrelated donor stem cells on outcomes. Blood. 2012;119:5591-5598. 7. Weisdorf D, Eapen M, Ruggeri A, et al. Alternative donor
trans-plantation for older patients with acute myeloid leukemia inﬁrst complete remission: a Center for International Blood and Marrow Transplant Research-Eurocord analysis. Biol Blood Marrow Transplant. 2014;20:816-822.
8. Sorror ML, Giralt S, Sandmaier BM, et al. Hematopoietic cell trans-plantation speciﬁc comorbidity index as an outcome predictor for pa-tients with acute myeloid leukemia inﬁrst remission: combined FHCRC and MDACC experiences. Blood. 2007;110:4606-4613.
9. Champlin R, Khouri I, Komblau S, et al. Reinventing bone marrow transplantation. Nonmyeloablative preparative regimens and induction of graft-vs-malignancy effect. Oncology (Williston Park). 1999;13: 621-628. discussion 631, 635-638, 641.
10. Mohty M, Malard F, Blaise D, et al. Reduced-toxicity conditioning with ﬂudarabine, once-daily intravenous busulfan, and antithymocyte globulins prior to allogeneic stem cell transplantation: results of a multicenter prospective phase 2 trial. Cancer. 2015;121:562-569. 11. Horowitz MM, Gale RP, Sondel PM, et al. Graft-versus-leukemia
re-actions after bone marrow transplantation. Blood. 1990;75:555-562. 12. Yakoub-Agha I, Mesnil F, Kuentz M, et al. Allogeneic marrow stem-cell
transplantation from human leukocyte antigen-identical siblings versus human leukocyte antigen-allelic-matched unrelated donors (10/10) in patients with standard-risk hematologic malignancy: a prospective study from the French Society of Bone Marrow Trans-plantation and Cell Therapy. J Clin Oncol. 2006;24:5695-5702. 13. Bacigalupo A, Ballen K, Rizzo D, et al. Deﬁning the intensity of
condi-tioning regimens: working deﬁnitions. Biol Blood Marrow Transplant. 2009;15:1628-1633.
14. Ruggeri A, Sanz G, Bittencourt H, et al. Comparison of outcomes after single or double cord blood transplantation in adults with acute leu-kemia using different types of myeloablative conditioning regimen, a retrospective study on behalf of Eurocord and the Acute Leukemia Working Party of EBMT. Leukemia. 2014;28:779-786.
15. Scaradavou A, Brunstein CG, Eapen M, et al. Double unit grafts suc-cessfully extend the application of umbilical cord blood transplantation in adults with acute leukemia. Blood. 2013;121:752-758.
16. Mohty M, Jacot W, Faucher C, et al. Infectious complications following allogeneic HLA-identical sibling transplantation with antithymocyte globulin-based reduced intensity preparative regimen. Leukemia. 2003; 17:2168-2177.
17. Bay JO, Peffault de Latour R, Bruno B, et al. Diagnosis and treatment of CMV and EBV reactivation as well as post-transplant lymphoprolifer-ative disorders following allogeneic stem cell transplantation: an SFGM-TC report. Pathol Biol (Paris). 2013;61:152-154.
18. Malard F, Szydlo RM, Brissot E, et al. Impact of cyclosporine-A con-centration on the incidence of severe acute graft-versus-host disease after allogeneic stem cell transplantation. Biol Blood Marrow Transplant. 2010;16:28-34.
19. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant. 1995;15:825-828. 20. Flowers ME, Kansu E, Sullivan KM. Pathophysiology and treatment of
graft-versus-host disease. Hematol Oncol Clin North Am. 1999;13: 1091-1112. viii-ix.
21. Dohner H, Estey EH, Amadori S, et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an inter-national expert panel, on behalf of the European LeukemiaNet. Blood. 2010;115:453-474.
22. Brunstein CG, Barker JN, Weisdorf DJ, et al. Umbilical cord blood transplantation after nonmyeloablative conditioning: impact on transplantation outcomes in 110 adults with hematologic disease. Blood. 2007;110:3064-3070.
23. Cahu X, Rialland F, Touzeau C, et al. Infectious complications after unrelated umbilical cord blood transplantation in adult patients with hematologic malignancies. Biol Blood Marrow Transplant. 2009;15: 1531-1537.
24. Peric Z, Cahu X, Chevallier P, et al. Features of EBV reactivation after reduced intensity conditioning unrelated umbilical cord blood trans-plantation. Bone Marrow Transplant. 2012;47:251-257.
25. Malard F, Furst S, Loirat M, et al. Effect of graft source on mismatched unrelated donor hemopoietic stem cell transplantation after reduced intensity conditioning. Leukemia. 2013;27:2113-2117.
26. Gallardo D, de la Camara R, Nieto JB, et al. Is mobilized peripheral blood comparable with bone marrow as a source of hematopoietic stem cells for allogeneic transplantation from HLA-identical sibling donors? A case-control study. Haematologica. 2009;94:1282-1288.
27. Schmitz N, Eapen M, Horowitz MM, et al. Long-term outcome of pa-tients given transplants of mobilized blood or bone marrow: a report from the International Bone Marrow Transplant Registry and the Eu-ropean Group for Blood and Marrow Transplantation. Blood. 2006;108: 4288-4290.
28. Bacigalupo A, Lamparelli T, Barisione G, et al. Thymoglobulin prevents chronic graft-versus-host disease, chronic lung dysfunction, and late transplant-related mortality: long-term follow-up of a randomized trial in patients undergoing unrelated donor transplantation. Biol Blood Marrow Transplant. 2006;12:560-565.
29. Mohty M, Labopin M, Balere ML, et al. Antithymocyte globulins and chronic graft-vs-host disease after myeloablative allogeneic stem cell transplantation from HLA-matched unrelated donors: a report from the Societe Francaise de Greffe de Moelle et de Therapie Cellulaire. Leukemia. 2010;24:1867-1874.
30. Finke J, Bethge WA, Schmoor C, et al. Standard graft-versus-host disease prophylaxis with or without anti-T-cell globulin in haematopoietic cell transplantation from matched unrelated donors: a randomised, open-label, multicentre phase 3 trial. Lancet Oncol. 2009;10:855-864. 31. Walker I, Schultz KR, Toze CL, et al. Thymoglobulin decreases the need
for immunosuppression at 12 months after myeloablative and non-myeloablative unrelated donor transplantation: CBMTG 0801, a ran-domized, controlled trial. Blood. 2014;124 [Abstract 38].
32. Chen YB, Aldridge J, Kim HT, et al. Reduced-intensity conditioning stem cell transplantation: comparison of double umbilical cord blood and unrelated donor grafts. Biol Blood Marrow Transplant. 2012;18: 805-812.
33. Peffault de Latour R, Brunstein CG, Porcher R, et al. Similar overall survival using sibling, unrelated donor, and cord blood grafts after reduced-intensity conditioning for older patients with acute myelog-enous leukemia. Biol Blood Marrow Transplant. 2013;19:1355-1360. 34. Gyurkocza B, Storb R, Storer BE, et al. Nonmyeloablative allogeneic
hematopoietic cell transplantation in patients with acute myeloid leukemia. J Clin Oncol. 2010;28:2859-2867.
35. Goodyear OC, Dennis M, Jilani NY, et al. Azacitidine augments expan-sion of regulatory T cells after allogeneic stem cell transplantation in patients with acute myeloid leukemia (AML). Blood. 2012;119: 3361-3369.
36. Chen YB, Li S, Lane AA, et al. Phase I trial of maintenance sorafenib after allogeneic hematopoietic stem cell transplantation for patients with Fms-like Tyrosine kinase 3 internal tandem duplication acute myeloid leukemia. Biol Blood Marrow Transplant. 2014;20:2042-2048. 37. Sandmaier BM, Khaled SK, Oran B, et al. Results of a phase 1study of
quizartinib (AC220) as maintenance therapy in subjects with acute myeloid leukemia in remission following allogeneic hematopoietic cell transplantation. Blood. 2014;124 [Abstract 428].
38. Eapen M, Klein JP, Sanz GF, et al. Effect of donor-recipient HLA matching at HLA A, B, C, and DRB1 on outcomes after umbilical-cord blood transplantation for leukaemia and myelodysplastic syndrome: a retrospective analysis. Lancet Oncol. 2011;12:1214-1221.
39. Eapen M, Klein JP, Ruggeri A, et al. Impact of allele-level HLA matching on outcomes after myeloablative single unit umbilical cord blood transplantation for hematologic malignancy. Blood. 2014;123:133-140. 40. Malard F, Gagne K, Chevallier P, et al. HLA-A, -B, -C and -DRB1 high resolution matching can improve patient’ outcome after double um-bilical allogeneic stem cell transplantation (allo-SCT). ASH Annual Meeting Abstracts. 2013;122 [Abstract 414].
41. Lee SJ, Klein J, Haagenson M, et al. High-resolution donor-recipient HLA matching contributes to the success of unrelated donor marrow transplantation. Blood. 2007;110:4576-4583.
42.Luznik L, O’Donnell PV, Symons HJ, et al. HLA-haploidentical bone marrow transplantation for hematologic malignancies using nonmyeloablative conditioning and high-dose, posttransplantation cyclophosphamide. Biol Blood Marrow Transplant. 2008;14: 641-650.
43.Brunstein CG, Fuchs EJ, Carter SL, et al. Alternative donor transplantation after reduced intensity conditioning: results of parallel phase 2 trials using partially HLA-mismatched related bone marrow or unrelated double umbilical cord blood grafts. Blood. 2011;118:282-288.