A Meta-Analysis of Unrelated Donor Umbilical Cord Blood Transplantation versus Unrelated Donor Bone Marrow Transplantation in Acute Leukemia Patients

A Meta-Analysis of Unrelated Donor Umbilical Cord Blood Transplantation versus Unrelated Donor Bone Marrow Transplantation in Acute Leukemia Patients Haoran Zhang, Junmin Chen, Wenzhong Que Umbilical cord blood has emerged as an alternative stem cell source to bone marrow or peripheral blood stem cells. Umbilical cord blood transplantation (UCBT) is also potentially curative for acute leukemia. However, the effect of unrelated donor bone marrow transplantation (UBMT) and UCBT on the outcome of patients with acute leukemia has not been systematically reviewed. In the present meta-analysis, we systematically searched Cochrane Library, MEDLINE, EMBASE, and CNKI up to May 2011. Two reviewers extracted data independently. Seven studies totaling 3389 patients have been assessed. Pooled results found that the incidence of engraftment failure and transplantation-related mortality were higher in UCBT than in UBMT, and relative risks (RRs) were 4.27 (95% confidence interval [CI], 2.94-6.21) and 1.27 (95% CI, 1.01-1.59), respectively. The rates of acute and chronic graft-versus-host disease (GVHD) in the UCBT group were significantly lower than that in the UBMT group, and RRs were 0.71 (95% CI, 0.65-0.79) and 0.69 (95% CI, 0.52-0.91), respectively. The relapse rate was similar between the UCBTand UBMT group. The leukemiafree survival (LFS) and overall survival (OS) were significantly lower in the UCBT group than in the UBMT group; RRs were 1.14 (95% CI, 1.07-1.22) and hazard ratios (HRs) were 1.31 (95% CI, 1.16-1.48), respectively. Subgroup analysis showed that in patients with acute lymphoblastic leukemia (ALL), the survival was similar between UCBT and UBMT. Biol Blood Marrow Transplant 18: 1164-1173 (2012) Ó 2012 American Society for Blood and Marrow Transplantation KEY WORDS: Cord blood transplantation, Bone marrow transplantation, Unrelated donor, Acute leukemia, Meta-analysis INTRODUCTION Bone marrow transplantation (BMT) is a widely accepted treatment for many hematologic malignancies, including leukemia, lymphoma, and multiple myeloma. Current bone marrow (BM) from HLAmatched unrelated donors has been a major alternative graft source [1,2]. However, many patients are unable to proceed to an allogeneic transplantation because of the lack of a suitable BM donor. In recent years, umbilical donor cord blood (UCB) has emerged as a feasible alternative source of hematopoietic From the Department of Hematology and Rheumatology, The First Affiliated Hospital of Fujian Medical University, 20 Chazhong Road, Fuzhou, Fujian, China. Financial disclosure: See Acknowledgments on page 1172. Correspondence and reprint requests: Junmin Chen, MD, Department of Hematology and Rheumatology, The First Affiliated Hospital of Fujian Medical University, 20 Chazhong Road, Fuzhou, Fujian 350005, China (e-mail: yulongcd@163.com). Received November 5, 2011; accepted January 24, 2012 Ó 2012 American Society for Blood and Marrow Transplantation 1083-8791/$36.00 doi:10.1016/j.bbmt.2012.01.015 progenitors (CD341) for allogeneic stem cell transplantation in patients who lack HLA-matched marrow donors. The first successful umbilical donor cord blood transplantation (UCBT) was performed in 1988 in a child with Fanconi anemia who underwent UCBT from his HLA-identical newborn sister [3]. Since that time, more than 20,000 UCBTs have been performed throughout the world [4-7]. The advent of UCBT has significantly improved allogeneic transplantation as a treatment option by expanding the available donor pool. This increase has been stimulated by the recognition that UCBT has certain advantages relative to unrelated donor bone marrow transplantation (UBMT), including rapid availability, low risk of infection transmission, absence of donor risk, and the relatively lower risk of graft-versus-host disease (GVHD). At the same time, the disadvantages of UCBT are the limited cell dose, delayed engraftment, and lack of additional immune cells if donor lymphocytes are needed [8]. Thus, large variation in practice and considerable uncertainty exists with respect to the relative effects of UCBT versus UBMT on the outcomes of patients with acute leukemia. However, differences in the 1164

Biol Blood Marrow Transplant 18:1164-1173, 2012 UCBT versus UBMT 1165 patient populations and transplantation protocols may have also contributed. In order to address this question, several comparative clinical studies [9-15] have been conducted to evaluate the benefit of UCBT for acute leukemia compared with UBMT. However, the effect of UCBT and UBMT on the outcome of patients with acute leukemia has not been systematically reviewed. Herein, we report the differences in the outcomes between UCBT and UBMT as therapy for acute leukemia. This analysis anticipated that results for patients with acute leukemia might differ from those for patients with other diseases. The main objective of this systematic literature review was to evaluate whether UCBT is equivalent to UBMT in treating patients with acute leukemia. METHODS This review was performed according to the standard described in the Preferred reporting items for systematic reviews and meta-analyses statement [16]. Selection Criteria All comparative studies of UCBT versus UBMT were selected whether or not randomization was used. Patients were children and adults requiring allogeneic hematopoietic stem cells transplantation to treat acute leukemia. Data for engraftment failure, GVHD, transplantation-related mortality (TRM), relapse, and survival had to be available. Each study was critically appraised for validity based on consistency, accuracy, and balance between treatment groups. Studies without comparable patient demographics between the 2 comparative groups were excluded. The studies with case reports, poor quality, and little information were excluded. There were far more review articles than primary studies on cord blood transplantation (CBT), and all review articles reiterating previous data were excluded. Search Strategy The following expanded Medical Subject Headings (MeSH) terms were used for the initial literature search: cord blood, bone marrow, transplant, transplantation, and transplants. The electronic databases searched were: MEDLINE (1966 to May 2011), EMBASE (1980 to 2011 week 20), Cochrane Library (issue 2, 2011), and Chinese National Knowledge Infrastructure (CNKI; 1994 to May 2011). No language or date restrictions were applied. Reference lists of all included studies and reviews related to the topic of the present meta-analysis were manually searched for other potentially eligible studies. Full text articles were obtained to extract the data for this analysis. Methods of the Review Potential studies for inclusion were identified from the search results. Two reviewers (H.Z. and W.Q.) independently applied the search strategy to selected references from databases. The titles and abstracts of the articles were reviewed independently. When there was a doubt, the full article was retrieved for further scrutiny. The 2 authors independently assessed each full study report to see whether it met the review s inclusion criteria, and authors were contacted for more information and clarification of data as necessary. Disagreements on inclusion of studies were resolved by the senior reviewer (J.C.) until consensus was obtained. The authors of individual studies were contacted directly to provide further information when necessary. We systematically reviewed all data on comparative studies of UCBT versus UBMT in which survival was the key outcome measure. Unrelated donor searches were eligible for analysis of graft availability and outcome. To obtain reliable evidence on the relative effect of UCBT versus UBMT in the primary treatment of adults and children with acute leukemia, results from independent and comparable studies were integrated to increase statistical power. The primary outcomes were estimated survival; secondary outcomes studied included engraftment, GVHD, TRM, and relapse. Statistical Analysis Dichotomous data for each arm in a particular study were expressed as proportions or risks, and the treatment effect as relative risks (RRs), with their respective 95% confidence intervals (CIs). Hazard ratios (HRs) were the preferred form of data for calculating overall survival (OS). Heterogeneity was explored by the chi-squared test with a significance set at a P value of.10, and the quantity of heterogeneity was measured by I 2 [17]. When significant heterogeneity was found with the fixed-effects model, a random-effects model was used to estimate the overall treatment effect. Forest plots were used to summarize the results of included studies, and inverted funnel plots and the Egger s test were used to examine publication bias. All analyses were done in Review Manager (version 5.1.2; The Cochrane Collaboration, Oxford, England) and STATA software package (version 11.0; Stata- Corp LP, College Station, TX). RESULTS Description of Studies Bibliographic searches from the Cochrane Library, MEDLINE, EMBASE, and CNKI databases yielded 387 articles, of which 359 were excluded. The rest of the records were assessed, and a further 21 were excluded for various reasons according to the inclusion and exclusion criteria (Figure 1). Finally,

1166 H. Zhang et al. Biol Blood Marrow Transplant 18:1164-1173, 2012 Figure 1. Flow chart of included and excluded studies. 7 studies [9-15] totaling 3389 patients were assessed (Table 1). Four studies were performed on pediatric patients and 3 studies were on adults. These studies on children involved a total of 649 patients receiving UCBT and 601 patients receiving UBMT. A total of 550 undergoing UCBT and 1589 UBMT adult patients were included. The majority of included studies were studies with between 49 and 785 participants. However, non-randomized and non-blinded comparative studies were selected for this meta-analysis. We verified that the patients included in 7 studies came from completely separate, not overlapping cohorts. The cohort in the Eapen et al. [10] study included both CBT and BMT from unrelated donors. Although no language restrictions were imposed initially, for the full-text review and final analysis, our resources only permitted the review of articles published in the English language. Effects of Interventions Engraftment failure Five studies [9-12,14] reported the rate of engraftment failure and these included 1635 patients Table 1. Characteristics of Included Studies Studies Publication Year Study Population Study Arm Number of Patients Median Age Years (Range) HLA Matching Rocha et al. [9] 2001 Children with acute leukemia UCBT 99 6 (2.5-10) 8% 6/6 HLA-matched 43% 1 antigen-mismatched 41% 2 antigen-mismatched 8% >3 antigen-mismatched UBMT 262 8 (5-12) 80.5% 6/6 HLA-matched 17.6% 1 antigen-mismatched 0.4% 2 antigen-mismatched Eapen et al. [10] 2007 Children with acute leukemia UCBT 503 Not available 7% 6/6 HLA-matched 40% 1 antigen-mismatched 53% 2 antigen-mismatched UBMT 282 Not available 41% 6/6 HLA-matched 59% antigen-mismatched Rocha et al. [11] 2004 Adults with acute leukemia UCBT 98 24.5 (15-55) 6% 6/6 HLA-matched 51% 1 antigen-mismatched 39% 2 antigen-mismatched 4% 3 antigen-mismatched UBMT 584 32 (15-59) 100% 6/6 HLA-matched Jacobsohn et al. [12] 2004 Children with ALL UCBT 26 3.9 (0.3-11.9) 15.4% 6/6 HLA-matched 38.5% 1 antigen-mismatched 42.3% 2 antigen-mismatched 3.8% 3 antigen-mismatched UBMT 23 2.5 (0.3-15.4) 87% 6/6 HLA-matched 13% 1 antigen-mismatched Atsuta et al. [13] 2009 Adults with AML UCBT 173 38 (16-69) 7% 6/6 HLA-matched 20% 1 antigen-mismatched 73% 2 antigen-mismatched UBMT 311 38 (16-60) 100% 6/6 HLA-matched Atsuta et al. [13] 2009 Adults with ALL UCBT 114 34 (16-58) 7% 6/6 HLA-matched 22% 1 antigen-mismatched 71% 2 antigen-mismatched UBMT 222 32 (16-59) 100% 6/6 HLA-matched Smith et al. [14] 2009 Children with ALL UCBT 21 7.0 (2.1-17.9) 9.5% 6/6 HLA-matched 52.4% 1 antigen-mismatched 38.1% 2 antigen-mismatched UBMT 34 7.0 (0.9-17.7) 100% 6/6 HLA-matched Eapen et al. [15] 2010 Adults with acute leukemia UCBT 165 Not available 6% 6/6 HLA-matched 24% 1 antigen-mismatched 70% 2 antigen-mismatched UBMT 472 Not available 100% 6/6 HLA-matched UCBT indicates umbilical cord blood transplantation; UBMT, unrelated donor bone marrow transplantation; ALL, acute lymphoblastic leukemia; AML, acute myelogenous leukemia.

Biol Blood Marrow Transplant 18:1164-1173, 2012 UCBT versus UBMT 1167 Figure 2. Meta-analysis of the rates of engraftment failure for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom. (Figure 2). The result of the heterogeneity test in 2 groups showed P 5.18 and I 2 5 37%, indicating no evidence of heterogeneity in the 5 studies; hence, the fixed effect model was used. The rate of engraftment failure in the UCBT group was significantly higher than in the UBMT group (RR 5 4.27; 95% CI 5 2.94-6.21; P \.00001). Acute and chronic GVHD All 7 studies reported the occurrence of GVHD. The rates of acute GVHD (agvhd) and chronic GVHD (cgvhd) in the UCBT group were significantly lower than those in the UBMT group (RR 5 0.71; 95% CI 5 0.65-0.79; P \.00001 versus RR 5 0.69; 95% CI 5 0.52-0.91; P 5.009, respectively; Figure 3, Figure 4). Relapse All 7 studies reported the relapse rate after transplantation, and the result is shown in Figure 5. The rate of relapse in the UCBT group was similar to that in the UBMT group (RR 5 0.98; 95% CI 5 0.77-1.24; P 5.87). Non-relapse mortality All 7 studies reported the TRM, and the result is shown in Figure 6. The TRM in the UCBT group was significantly higher than that in the UBMT group (RR 5 1.27; 95% CI 5 1.01-1.59; P 5.04). Leukemia-free survival All 7 studies reported the leukemia-free survival (LFS) between the 2 groups, and the result is shown in Figure 7. There was a significant difference in LFS when studies were pooled. The LFS was lower in UCBT compared with UBMT (RR 5 1.14; 95% CI 5 1.07-1.22; P \.0001). Overall survival Four studies [9,11-13] reported the OS between the 2 groups, and the result is shown in Figure 8. There was a significant difference in OS when studies were pooled. The OS in the UCBT group was significantly lower than that in the UBMT group (HR 5 1.31; 95% CI 5 1.16-1.48; P 5.000). Subgroup Analyses Children and adults The rate of TRM and LFS in pediatric patients in the UCBT group were non-significantly different from that in the UBMT group (RR 5 1.51; 95% CI 5 0.98-2.34 and RR 5 1.10; 95% CI 5 0.99-1.22, respectively). This meta-analysis showed that in adults, the TRM was similar between UCBT and UBMT (RR 5 1.13; 95% CI 5 0.99-1.29). Acute lymphoblastic leukemia (ALL) The rate of engraftment failure, cgvhd, and TRM in the UCBT group were non-significantly different from that in the UBMT group (RR 5 1.57; 95% CI 5 0.34-7.26 versus RR 5 0.84; 95% CI 5 0.61-1.17 versus RR 5 0.85; 95% CI 5 0.65-1.12, respectively). The LFS and OS in the UCBT group were also nonsignificantly different from that in the UBMT group

1168 H. Zhang et al. Biol Blood Marrow Transplant 18:1164-1173, 2012 Figure 3. Meta-analysis of incidence of acute graft-versus-host disease (agvhd) for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom. (RR 5 1.07; 95% CI 5 0.89-1.29 versus HR 5 1.18; 95% CI 5 0.89-1.57, respectively). Publication bias Funnel plots and the Egger s test were performed to access the publication bias in studies. The shape of the funnel plot (Figure 9) revealed obviously symmetrical distribution, which means no publication bias. The Egger s test did not suggest any evidence of publication bias (Egger s regression test P values..10). DISCUSSION We performed a systematic review and metaanalysis of clinical studies comparing UCBT to UBMT in patients with acute leukemia. The results showed that UCBT was associated with less GVHD, more engraftment failure and TRM, similar relapse, and shorter survival. There was no difference in LFS and OS in patients with acute lymphoblastic leukemia (ALL) when studies were pooled. We applied a systematic methodology in which studies were searched, selected, and analyzed, but because allocation to a particular treatment group was based on the availability of suitable unrelated BM or CB donors, randomized and blinded controlled studies were not available. So, the studies are not randomized controlled studies. It is true that patients baseline characteristics were similar in these studies, and the lack of availability of fully matched unrelated donors often predicated the need for less well-matched CB grafts. In addition, the degree of match is less stringent for UCBT [18]. There were, however, explicit statements of intent, as evident in the studies of Rocha et al. [11], Atsuta et al. [13], Smith et al. [14], and Eapen et al. [15], that patients were eligible for UCBT if fully matched siblings or fully matched unrelated BM donors were not available immediately or within a satisfactory time frame. Therefore, we cannot fully exclude the possibility of center biases in the treatment and selection of patients for UCBT or UBMT. These studies demonstrated significant differences in graft failure between UCBT and UBMT recipients. The higher risk of graft failure may be related to several factors, including immaturity of stem cells (which might need more cell divisions before differentiation to marrow progenitors), the lack of subpopulations to facilitate engraftment [19], and the lower nucleated cell and CD341 cell doses with the UCB grafts as compared with the marrow allografts [20]. One study described a threshold of 1.7 10 5 CD341 cells/kg and suggested that cord blood units containing less than this CD341 cell dose should be considered deficient [7]. So, the main problem with using UCB for transplantation is the relatively low number of hematopoietic progenitor cells and hematopoietic stem cells in UCB compared with bone marrow or mobilized peripheral blood grafts, which translates into increased risk of graft failure, delayed hematopoietic engraftment [10,21], and delayed immune reconstitution [22,23].

Biol Blood Marrow Transplant 18:1164-1173, 2012 UCBT versus UBMT 1169 Figure 4. Meta-analysis of incidence of chronic graft-versus-host disease (cgvhd) for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom. This meta-analysis was undertaken to determine whether there is a relationship between stem cell source and the development of acute and chronic GVHD after allogeneic transplantation. This metaanalysis showed that the rate of acute and chronic GVHD were higher in UBMT than in UCBT. The incidence of GVHD among UCBT recipients has been lower than recipients of matched UBMT, because the UCB stem cells are immunologically na ıve [24-26], with low T cell mediated cytotoxic capacity Figure 5. Meta-analysis of relapse rate after transplantation for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom.

1170 H. Zhang et al. Biol Blood Marrow Transplant 18:1164-1173, 2012 Figure 6. Meta-analysis of the rates of non-relapse mortality for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom. and, in vitro, markedly reduced responsiveness to allogeneic stimuli in the secondary mixed lymphocyte reaction [27,28]. Lu et al. [29] recently demonstrated a marked increased in CD161CD56 natural killer cells in the peripheral blood after UCBT. The decreased incidence of GVHD with UCBT despite HLA mismatch may be attributable in part to the immaturity and lower number of immunocompetent donor T cells [18]. Although most patients were mismatched with their cord blood donors at 1 to 3 HLA loci, the incidence of either grades II to IV agvhd or cgvhd was low and compares favorably with that reported in several studies on UBMT. But, the cgvhd in UCBT in the patients with ALL group was non-significantly different from that in the UBMT group. These studies demonstrated no significant differences in relapse between UCBT and UBMT recipients. Because of the decreased incidence of GVHD in patients undergoing UCBT and the delayed T cell recovery, one concern has been the potential for a decreased graftversus-leukemia effect. However, studies have not demonstrated an increased risk of relapse after UCBT [30]. Concerns raised about the possibility of an increased risk of leukemia recurrence in UCBT recipients are derived from the incidence and severity of both agvhd and cgvhd, which seemed to be less, and immaturity and delayed engraftment, which impair the immunemediated antimalignancy effect after UCBT. The probability of relapse after CBT was similar to that in the BMT group, but suggested that the donor-derived graft-versus-tumor effect might be preserved [31,32]. Further optimization of cord selection, ex vivo expansion, and biologic manipulation of UCB and the immune system should help to decrease the risk of recurrent leukemia [33]. This meta-analysis demonstrated significant differences in TRM between UCBT and UBMT recipients. TRM remains as the main obstacle for successful UCBT. The disadvantage of a higher TRM will generate an adverse effect on survival. Compared with patients transplanted from matched unrelated BM donors, those given CBT from an unrelated donor experience a lower incidence of GVHD and, due to a higher risk of severe infectious complications [34,35], may be exposed to an increased risk of TRM. A major limitation to the use of UCB is the availability of sufficient numbers of hematopoietic progenitor cells for hemopoietic recovery [9-11,13,21,36,37]. There are significant differences in failure to engraft between adults and children. The rate of adults and children engraftment failure in the UCBT group was significantly higher than in the UBMT group, although shortening the period of neutropenia might lead to lower early mortality [38]. Maybe one relevant factor influencing the risk of TRM is the rate of engraftment failure. Our analyses demonstrated significant differences in survival between UCBT and UBMT recipients. The advantage of a lower GVHD rate without any apparent increase in relapse in UCBT transplantation will offset any adverse effect of reduced cell dose on

Biol Blood Marrow Transplant 18:1164-1173, 2012 UCBT versus UBMT 1171 Figure 7. Meta-analysis of leukemia-free survival for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom. survival. When UCBT was administrated in patients with acute leukemia, impaired survival was observed in this meta-analysis. The LFS and OS in the UCBT group were significantly lower than that those in the UBMT group. However, there was no difference in LFS and OS in patients with ALL when studies were pooled. Although it is true that the analysis did show that some of the confidence intervals are wide, and the possibility of large differences has not been excluded, some studies thus far suggest that UBMT in Figure 8. Meta-analysis of overall survival for umbilical cord blood transplantation (UCBT) and unrelated donor bone marrow transplantation (UBMT). M-H, Mantel-Haenszel; CI, confidence interval.

1172 H. Zhang et al. Biol Blood Marrow Transplant 18:1164-1173, 2012 CONCLUSION Figure 9. Funnel plot of the 7 studies. P 5.401 for the Egger s test. Each point represents an individual study for the indicated association. patients with ALL can result in long-term survival in many patients. The results thus far suggest that UCBT can result in long-term LFS and OS in patients with ALL. Some studies have reported OS in the range of 49% to 65%, which is similar to UBMTs. Results of the current study are consistent with those of 3 previous meta-analyses. Two previously published meta-analyses included patients with hematological diseases or hematological malignancies [39,40]. Another meta-analysis, published by Xu et al. [41] (including 1453 patients in 7 studies), was limited to children with hematological diseases or hematological malignancies. However, in our meta-analysis, we only analyzed clinical studies that only included adult and pediatric patients with acute leukemia. To our knowledge, this is the first meta-analysis that focused on this comparison in patients with acute leukemia. At the same time, our review was performed according to the standard described in the Preferred reporting items for systematic reviews and meta-analyses statement [16]. Since that time, additional clinical studies determining the effects of UCBT have also been reported. At the same time, we should address the concern that the results with CBT compared to BMT could have improved over time. Our analysis does have several limitations. First, the studies are not randomized controlled studies. The levels of evidence are moderate [42]. Second, the comparison here is between UCBT and UBMT; peripheral blood stem cell transplantation is not included in the analysis. Third, duration of follow-up was mostly short-term. Fourth, there were manuscripts included that used registry data. In this type of analysis, choice of intervention is often governed by a complex list of unmeasured factors that can potentially affect outcome, such as the choice of alternative donor and the time to transplantation. Fifth, acute myeloid leukemia and ALL are different disease entities that require different chemotherapy regimens for treatment. However, most comparison studies have included both diseases because of the limitation in the number of UCBTs given to patients with acute leukemia. We conclude that UCBT was associated with less GVHD, more engraftment failure and TRM, similar relapse, and shorter survival. At the same time, there was no difference in LFS and OS in patients with ALL when studies were pooled. Taken together, these results suggest that UCBT led to inferior outcomes than UBMT in patients with acute leukemia. UCBT could be considered as an alternative stem cell source for patients with acute leukemia, especially when an unrelated marrow donor is not available in a timely fashion. The present report suggests that large samples and well-designed studies to test these observations are warranted. ACKNOWLEDGMENTS Financial disclosure: The authors have nothing to disclose. REFERENCES 1. Petersdorf EW, Gooley TA, Anasetti C, et al. Optimizing outcome after unrelated marrow transplantation by comprehensive matching of HLA class I and II alleles in the donor and recipient. Blood. 1998;92:3515-3520. 2. Hansen JA, Gooley TA, Martin PJ, et al. Bone marrow transplants from unrelated donors for patients with chronic myeloid leukemia. N Engl J Med. 1998;338:962-968. 3. Gluckman E, Broxmeyer HA, Auerbach AD, et al. Hematopoietic reconstitution in a patient with Fanconi s anemia by means of umbilical cord blood from an HLA-identical sibling. N Engl J Med. 1989;321:1174-1178. 4. Rocha V, Locatelli F. Searching for alternative hematopoietic stem cell donors for pediatric patients. Bone Marrow Transplant. 2008;41:207-214. 5. Gluckman E, Rocha V. Cord blood transplantation: state of the art. Haematologica. 2009;94:451-454. 6. Cornetta K, Laughlin M, Carter S, et al. Umbilical cord blood transplantation in adults: results of the prospective Cord Blood Transplantation (COBLT). Biol Blood Marrow Transplant. 2005;11:149-160. 7. Wagner JE, Barker JN, DeFor TE, et al. Transplantation of unrelated donor umbilical cord blood in 102 patients with malignant and nonmalignant diseases: influence of CD34 cell dose and HLA disparity on treatment-related mortality and survival. Blood. 2002;100:1611-1618. 8. Brunstein CG. Umbilical cord blood transplantation for the treatment of hematologic malignancies. Cancer Control. 2011; 18:222-236. 9. Rocha V, Cornish J, Sievers EL, et al. Comparison of outcomes of unrelated bone marrow and umbilical cord blood transplants in children with acute leukemia. Blood. 2001;97:2962-2971. 10. Eapen M, Rubinstein P, Zhang MJ, et al. Outcomes of transplantation of unrelated donor umbilical cord blood and bone marrow in children with acute leukaemia: a comparison study. Lancet. 2007;369:1947-1954. 11. Rocha V, Labopin M, Sanz G, et al. Transplants of umbilicalcord blood or bone marrow from unrelated donors in adults with acute leukemia. N Engl J Med. 2004;351:2276-2285. 12. Jacobsohn DA, Hewlett B, Ranalli M, Seshadri R, Duerst R, Kletzel M. Outcomes of unrelated cord blood transplants and allogeneic-related hematopoietic stem cell transplants in

Biol Blood Marrow Transplant 18:1164-1173, 2012 UCBT versus UBMT 1173 children with high-risk acute lymphocytic leukemia. Bone Marrow Transplant. 2004;34:901-907. 13. Atsuta Y, Suzuki R, Nagamura-Inoue T, et al. Disease-specific analyses of unrelated cord blood transplantation compared with unrelated bone marrow transplantation in adult patients with acute leukemia. Blood. 2009;113:1631-1638. 14. Smith AR, Baker KS, Defor TE, Verneris MR, Wagner JE, Macmillan ML. Hematopoietic cell transplantation for children with acute lymphoblastic leukemia in second complete remission: similar outcomes in recipients of unrelated marrow and umbilical cord blood versus marrow from HLA matched sibling donors. Biol Blood Marrow Transplant. 2009;15:1086-1093. 15. 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. 16. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement. BMJ. 2009;339:b2535. 17. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21:1539-1558. 18. Liao C, Wu JY, Xu ZP, et al. Indiscernible benefit of highresolution HLA typing in improving long-term clinical outcome of unrelated umbilical cord blood transplant. Bone Marrow Transplant. 2007;40:201-208. 19. Sierra J, Storer B, Hansen JA, et al. Transplantation of marrow cells from unrelated donors for treatment of high-risk acute leukemia: the effect of leukemic burden, donor HLA-matching, and marrow cell dose. Blood. 1997;89:4226-4235. 20. Mavroudis D, Read E, Cottler-Fox M, et al. CD341 cell dose predicts survival, posttransplant morbidity, and rate of hematologic recovery after allogeneic marrow transplants for hematologic malignancies. Blood. 1996;88:3223-3229. 21. Laughlin MJ, Eapen M, Rubinstein P, et al. Outcomes after transplantation of cord blood or bone marrow from unrelated donors in adults with leukemia. N Engl J Med. 2004;351: 2265-2275. 22. Niehues T, Rocha V, Filipovich AH, et al. Factors affecting lymphocyte subset reconstitution after either related or unrelated cord blood transplantation in children a Eurocord analysis. Br J Haematol. 2001;114:42-48. 23. Komanduri KV, St John LS, de Lima M, et al. Delayed immune reconstitution after cord blood transplantation is characterized by impaired thymopoiesis and late memory T-cell skewing. Blood. 2007;110:4543-4551. 24. Madrigal JA, Cohen SB, Gluckman E, Charron DJ. Does cord blood transplantation result in lower graft-versus-host disease? It takes more than two to tango. Hum Immunol. 1997; 56:1-5. 25. Chalmers IM, Janossy G, Contreras M, Navarrete C. Intracellular cytokine profile of cord and adult blood lymphocytes. Blood. 1998;92:11-18. 26. Rocha V, Wagner JE Jr, Sobocinski KA, et al. Graft-versus-host disease in children who have received a cord-blood or bone marrow transplant from an HLA-identical sibling. Eurocord and International Bone Marrow Transplant Registry Working Committee on Alternative Donor and Stem Cell Sources. N Engl J Med. 2000;342:1846-1854. 27. Moretta A, Locatelli F, Mingrat G, et al. Characterisation of CTL directed towards non-inherited maternal alloantigens in human cord blood. Bone Marrow Transplant. 1999;24:1161-1166. 28. Risdon G, Gaddy J, Horie M, Broxmeyer HE. Alloantigen priming induces a state of unresponsiveness in human umbilical cord blood T cells. Proc Natl Acad Sci U S A. 1995;92:2413-2417. 29. Lu X, Kondo Y, Takmatsu H, et al. CD161 CD56- NK cells in the peripheral blood of cord blood transplant recipients: a unique subset of NK cells possibly associated with graft-versusleukemia effect. Eur J Haematol. 2008;81:18-25. 30. Takahashi S, Ooi J, Tomonari A, et al. Comparative singleinstitute analysis of cord blood transplantation from unrelated donors with bone marrow or peripheral blood stem-cell transplants from related donors in adult patients with hematologic malignancies after myeloablative conditioning regimen. Blood. 2007;109:1322-1330. 31. Parkman R, Cohen G, Carter SL, et al. Successful immune reconstitution decreases leukemic relapse and improves survival in recipients of unrelated cord blood transplantation. Biol Blood Marrow Transplant. 2006;12:919-927. 32. Haut PR, Gonzalez-Ryan L, Wang LJ, Olszewski M, Morgan E, Kletzel M. Induction of a transient graft vs. leukemia effect following unrelated cord blood transplantation. Pediatr Transplant. 2002;6:348-351. 33. Advani AS, Laughlin MJ. Umbilical cord blood transplantation for acute myeloid leukemia. Curr Opin Hematol. 2009;16: 124-128. 34. Barker JN, Hough RE, van Burik JA, et al. Serious infections after unrelated donor transplantation in 136 children: impact of stem cell source. Biol Blood Marrow Transplant. 2005;11:362-370. 35. Narimatsu H, Matsumura T, Kami M, et al. Bloodstream infection after umbilical cord blood transplantation using reducedintensity stem cell transplantation for adult patients. Biol Blood Marrow Transplant. 2005;11:429-436. 36. van Heeckeren WJ, Fanning LR, Meyerson HJ, et al. Influence of human leukocyte antigen disparity and graft lymphocytes on allogeneic engraftment and survival after umbilical cord blood transplants in adults. Br J Haematol. 2007;139:464-474. 37. Rocha V, Gluckman E; Eurocord-Netcord registry and European Blood and Marrow Transplant group. Improving outcomes of cord blood transplantation: HLA matching, cell dose and other graft- and transplantation-related factors. Br J Haematol. 2009;147:262-274. 38. Rocha V, Broxmeyer HE. New approaches for improving engraftment after cord blood transplantation. Biol Blood Marrow Transplant. 2009;16(1 Suppl):S126-S132. 39. Wang J, Zhan P, Ouyang J, Chen B, Zhou R, Yang Y. Unrelated donor umbilical cord blood transplantation versus unrelated donor bone marrow transplantation in adult and pediatric patients: a meta-analysis. Leuk Res. 2010;34:1018-1022. 40. Hwang WY, Samuel M, Tan D, Koh LP, Lim W, Linn YC. A meta-analysis of unrelated donor umbilical cord blood transplantation versus unrelated donor bone marrow transplantation in adult and pediatric patients. Biol Blood Marrow Transplant. 2007;13:444-453. 41. Shi-Xia X, Xian-Hua T, Xiang-Feng T. Unrelated umbilical cord blood transplantation and unrelated bone marrow transplantation in children with hematological disease: a metaanalysis. Pediatr Transplant. 2009;13:278-284. 42. van Tulder M, Furlan A, Bombardier C, Bouter L; Editorial Board of the Cochrane Collaboration Back Review Group. Updated method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group. Spine (Phila Pa 1976). 2003; 28:1290-1299.