If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The number of umbilical cord blood transplants is increasing worldwide. At this time, it is important to evaluate their results and to compare the outcome of umbilical cord blood transplants with allogeneic bone marrow transplants.
Data Sources
Data have been reported to the Eurocord Registry by multiple transplant centers. Close links have been established with the cord blood banks through Netcord. Bone marrow transplant data have been provided by transplant centers and through the European Blood and Marrow Transplant (EBMT) and International Bone Marrow Transplant Registries (IBMTR).
Results
Eurocord has analyzed the outcomes of 527 umbilical cord blood transplants from 121 transplant centers and 29 countries. The donor was related in 138 cases and unrelated in 399 cases. The results showed that survival with umbilical cord blood transplants was comparable to that with related or unrelated bone marrow transplants. Engraftment with cord blood was delayed resulting in an increased incidence of early transplant complications. The incidence of acute and chronic graft-vs-host disease was reduced with cord blood grafts even in HLA-mismatched transplants and in adults. In patients with leukemia, the rate of relapse was similar to the rate of relapse after bone marrow transplant. The overall event-free survival with umbilical cord blood transplantation was not statistically different compared to bone marrow transplants.
Conclusions
This large registry study confirms the potential benefit of using umbilical cord blood hematopoietic stem cells for allogeneic transplants.
The first allogeneic umbilical cord blood transplant (UCBT) was performed successfully in 1988 to treat a child with Fanconi's anemia; the umbilical cord blood donor was his HLA-identical sister [
]. Twelve years later, this patient is doing well, with full donor hematopoietic and lymphoid reconstitution. This first success opened the way to an entire new field in the domain of allogeneic hematopoietic stem cell transplant, as it showed that a single umbilical cord blood unit contained sufficient numbers of hematopoietic stem cells to reconstitute definitively the host lymphohematopoietic compartment, that an umbilical cord blood unit could be collected at birth without any harm to the newborn infant, and that an umbilical cord blood hematopoietic stem cell graft could be cryopreserved and transplanted to a myeloablated host after thawing without losing its repopulating ability. Since then, our knowledge of the biologic characteristics of umbilical cord blood cells has increased, and the advantages of using umbilical cord blood stem cells for transplantation have become apparent. Simultaneously, umbilical cord blood banks have been established for related or unrelated UCBTs, with >30,000 units currently available and >1,500 UCBTs having been performed in children and increasing numbers in adults with malignant and nonmalignant diseases [
Successful transplantation of HLA matched and HLA mismatched umbilical cord blood from unrelated donors analysis of engraftment and acute graft versus host disease.
The principal limitations of allogeneic bone marrow transplantation (BMT) are the lack of suitable HLA-matched donors and complications due to graft-vs-host disease (GVHD) that are more severe with increasing HLA disparities [
]. In the absence of a suitable HLA-identical sibling donor, alternative donors, such as mismatched related or matched unrelated donors, are sought. Past results have been disappointing, because major and minor histocompatibility differences often were not detected due to the inadequacy of standard typing techniques, explaining the relatively high frequency of posttransplant complications, graft failure, GVHD, and delayed immune reconstitution [
]. More recently, molecular techniques have improved the sensitivity of HLA typing and, therefore, the quality of donor selection, which has resulted not only in decreasing complications but also the probability of finding a fully matched donor for class I and class II HLA antigens [
]. Despite the increasing number of bone marrow donor registries, which now contain >5 million bone marrow donors worldwide, some patients cannot be transplanted because of the lack of an HLA-identical donor. In these cases, alternative approaches have been used with T-cell depletion in unrelated matched or partially matched donors [
A phase I-II clinical trial to evaluate removal of CD4 cells and partial depletion of CD8 cells from donor marrow for HLA-mismatched unrelated recipients.
Children with acute lymphoblastic leukemia who receive T-cell depleted HLA mismatched marrow allografts from unrelated donors have an increased incidence of primary graft failure but a similar overall transplant outcome.
Successful engraftment of T-cell depleted haploidentical three loci incompatible transplants in leukemia patients by addition of recombinant human granulocyte colony stimulating factor mobilized peripheral blood progenitor cells to bone marrow inoculum.
Umbilical cord blood cells have many theoretical advantages as grafts for stem cell transplantation because of the immaturity of newborn cells. Hematopoietic progenitors from umbilical cord blood are enriched for in vivo long-term repopulating stem cells. Compared to adult cells, umbilical cord blood hematopoietic stem cells produce larger hematopoietic colonies in vitro, have different growth factors requirements, are able to expand in long-term culture in vitro, engraft SCID-human mice in the absence of additional human growth factors, and have longer telomeres [
]. These properties theoretically should compensate for the relatively low number of cells contained in a single umbilical cord blood donation and, through rapid expansion, reconstitute myeloablated patients. The second advantage of umbilical cord blood grafts relates to the immaturity of the immune system at birth. This property should decrease the alloreactive potential of the lymphocytes within a cord blood graft and consequently should reduce the incidence and severity of GVHD after HLA-matched or HLA-mismatched transplants. Cord blood lymphocytes are naive and immature, are enriched in double-negative CD3+ cells, and produce fewer cytokines. Cord blood cells express mRNA transcripts for interferon-γ, interleukin-4 (IL-4), and IL-10, but very little IL-2; they have a fully constituted polyclonal T-cell repertoire; and they can be protected from apoptosis because of low levels of CD95 [
]. Most of these functions are inducible through in vitro or in vivo activation; as a consequence, early NK and T-cell cytotoxicity is impaired but secondary activation can occur. One can speculate that despite the reduction of GVHD, a graft-vs-leukemia (GVL) effect can still be observed with umbilical cord blood cells [
Recipient tumor necrosis factor-alfa and interleukin-10 gene polymorphisms associate with early mortality and acute graft versus host disease severity in HLA-matched sibling bone marrow transplants.
]. But because acute GVHD is an early event after allogeneic BMT and because it is in part triggered by the release of cytokines, it is reasonable to postulate that umbilical cord blood grafts might induce less frequent and less severe acute and chronic GVHD than adult hematopoietic stem cell transplants that contain a greater number of activated T cells. These properties should lead to less stringent criteria for HLA donor recipient selection.
Netcord experience in cord blood banking
The increasing success of reported UCBTs resulted in establishment of cord blood banks [
]. Their number has been increasing, with >30,000 units collected and available for unrelated donor hematopoietic stem cell transplant searches. The main practical advantages of using umbilical cord blood as an alternative source of stem cells are the relative ease of procurement, absence of risk to donors, reduced risk of transmitting infection, and prompt availability of cryopreserved samples to transplant centers. These advantages were first recognized in UCBT using related donors. Secondarily, large unrelated umbilical cord blood banks established criteria for standardization of umbilical cord blood collection, banking, processing, and cryopreservation for unrelated UCBTs for treatment of various hematologic malignant and nonmalignant diseases. The other advantages of UCBT are the large donor pool, faster allocation process (the units are readily available, having been previously tested and cryopreserved), decreased risk of viral transmission, and low incidence of GVHD due to the immaturity of the newborn's immune system. Compared to unrelated BMTs where complete HLA identity for class I and class II antigens is required, most UCBTs have been performed with donors having one, two, or three HLA antigen mismatches. Clinical analyses have shown that the most important factor in predicting a positive outcome for transplant is that the number of nucleated cells infused be >3 × 107/kg [
Because umbilical cord blood banking for unrelated transplantation is a relatively new field, it was important to set up minimum standards and to reach an international agreement on aspects that protect the infant donor and the mother and that optimize the chances of finding a suitable donor for the recipient. For this purpose, Netcord was founded in 1998. This cooperative network of large experienced umbilical cord blood banks includes cord blood banks in the United States, Europe, Japan, and Australia. The joint inventory available on a single research file contained 28,650 validated units in October 1999. Over the last decade, 1,017 units have been transplanted to pediatric patients and 272 to adult patients. Netcord's board of directors has developed a detailed set of standards for umbilical cord blood banking, which take into account respective national and international regulations and have submitted documents to the Foundation for the Accreditation of Hematopoietic Cell Therapy (FACHT, Omaha, NE, USA) for accreditation. Furthermore, a joint allocation system using the most recent Internet technology has been implemented to facilitate rapid allocation (within an average time of 48 hours) of umbilical cord blood units according to the histocompatibility status and the number of nucleated cells contained within the graft.
Results of cord blood transplantation from the Eurocord registry
Eurocord is an international registry operating on behalf of the European Blood and Marrow Transplant (EBMT) group. Participation is open to European and non-European centers conducting UCBT. Eurocord works in close collaboration with Netcord banks. Centers carrying out transplants with umbilical cord blood units from Netcord banks were invited to report their cases to Eurocord. From October 1988 to March 2000, 700 UCBTs were reported to Eurocord from 121 transplant centers in 29 countries in and outside Europe. The donors were related in 150 cases (145 children and 5 adults) and unrelated in 534 cases (378 children and 156 adults) (Fig. 1). Sixteen patients were transplanted with cord blood cells in other situations (8 patients received cord blood in addition to bone marrow cells from the same donor, 1 patient received autologous cells for gene therapy, 1 an autologous transplant, 4 received ex vivo expanded cells from a part of the cord blood graft, and 2 patients received simultaneously 2 cord blood units). The number of cells collected and infused is shown in Table 1.
Figure 1Number of related and unrelated cord blood transplants performed per year and reported to Eurocord
Table 1Median number of nucleated and progenitor cord blood cells collected and infused per recipient's weight (kg) in related and unrelated cord blood transplants (CBT)
Results in children transplanted with a related donor umbilical cord blood
The diagnoses of the recipients of the reported cord blood transplants are shown in Table 2. The donor was HLA-identical in 114 cases and was mismatched for one, two, or three HLA antigens in 24 cases. The conditioning varied according to age and diagnosis. Most often, GVHD prophylaxis consisted of only cyclosporin A. Median follow-up time was 41 months, and median age was 5 years (range 0–14).
Table 2Diagnosis of children receiving a related or an unrelated cord blood transplant (CBT)
Diagnosis
Related CBT in children [N = 138]
Unrelated CBT in children [N = 291]
Malignancies
74
202
Acute leukemia
Myeloid
18
50
Lymphoid
39
107
MDS
6
25
CML
6
14
NHL
2
5
Neuroblastoma
3
1
Bone marrow failure syndromes
25
28
SAA
8
4
Fanconi's anemia
11
18
Blackfan Diamond anemia
2
1
Dyskeratosis congenita
2
1
Amegakaryocytic thrombocytopenia
1
2
Kostmann's syndrome
1
2
Hemoglobinopathies
24
—
Sickle cell disease
3
Thalassemia
21
Inborn errors of metabolism
15
61
Immunodeficiencies
7
25
Metabolic disorders
8
23
Histiocytic disorders
13
CML = chronic myeloid leukemia; MDS = myelodysplastic syndrome; NHL = non-Hodgkin's lymphoma; SAA = severe aplastic anemia.
As of January 2000, the 2-year probability of survival was 46% in patients with malignancies, 76% in patients with aplastic anemia, 100% in hemoglobinopathies of whom 35% underwent autologous reconstitution, and 79% in patients with inborn errors of metabolism (Fig. 2). The probability of neutrophil engraftment was 83% ± 4%, with a median time of 26 days (range 8–60) to reach >500/μL neutrophils. The probability of developing acute GVHD ≥II was 20%, and the probability of developing chronic GVHD was 6%. The major cause of death was relapse in patients with leukemia who received the transplant in an advanced stage of the disease [
Comparison of graft-versus-host disease in children transplanted with HLA identical sibling umbilical cord blood versus HLA identical sibling bone marrow transplant.
We then performed a study in collaboration with the International Bone Marrow Transplant Registry (IBMTR) where we compared the outcome of HLA-identical sibling transplants using either an umbilical cord blood or a bone marrow graft as a source of hematopoietic stem cells. The primary endpoint was to compare the incidence of acute and chronic GVHD; the secondary endpoints were to compare engraftment and survival. We studied 113 children receiving an HLA-identical UCBT and 2,052 children receiving an HLA-identical sibling BMT from 1990–1997. We used the Cox proportional regression hazard model to adjust for other factors that potentially could influence GVHD risk. Time to neutrophil and platelet recovery and overall survival also were compared.
Compared with BMT recipients, recipients of UCBTs were younger (median 5 vs 8 years, p < 0.001), weighed less (median 17 vs 26 kg, p < 0.001), more frequently had major ABO incompatibility with their donor (25% vs 15%, p < 0.001), and less frequently received methotrexate for GVHD prophylaxis (28% vs 65%, p < 0.001) than BMT recipients. The interval from diagnosis to transplant was shorter for BMT recipients (median 10 vs 25 months, p < 0.001). The median number of nucleated umbilical cord blood cells infused was 0.47 × 108/kg as compared with 3.5 × 108/kg nucleated bone marrow cells (p < 0.001). Multivariate analysis demonstrated lower risks of grade II–IV acute GVHD (relative risk [RR] 0.40; 95% confidence interval [CI] 0.24–0.70; p = 0.001) and chronic GVHD (RR 0.35; 95% CI 0.14–0.85; p = 0.02) in UCBT recipients. Neutrophil recovery was significantly delayed after UCBT (RR 0.40; 95% CI 0.32–0.51; p < 0.001), as was platelet recovery in the first month after transplant (RR 0.20; 95% CI 0.13–0.29; p < 0.001). Survival was similar in the two groups (RR 1.15; 95% CI 0.81–1.65; p = 0.43). This study demonstrated that recipients of HLA-identical sibling UCBTs have less acute and chronic GVHD than recipients of HLA-identical sibling BMTs. Survival following cord blood and BMT was similar, suggesting that umbilical cord blood was an acceptable source of hematopoietic stem cells for transplantation [
Results in children transplanted with unrelated donor umbilical cord blood
The diagnosis of this group of patients is shown in Table 2. In patients with malignancies, 15 patients had previously received an autologous and 7 an allogeneic BMT. Most of the donors (83%) had 1–4 HLA mismatches and 50 only were matched by serology and low-resolution DRB1 typing. Median follow-up time was 21 months (range 1–64), median age was 5 years (0.2–15), median weight was 19 kg (range 4–83), and median number of nucleated cell dose infused was 4.5 × 107/kg (range 0.6–36).
The 2-year event-free survival (EFS) was 36% in patients with malignant diseases, 21% in patients with aplastic anemia, and 51% in patients with inborn errors of metabolism (Fig. 3). The probability of granulocyte engraftment by day 60 was 82%, with a median time of 29 days (range 10–60) to reach >500/μL granulocytes. The rate of engraftment was reduced to only 50% in aplastic anemia. The incidence of acute GVHD ≥II was 39%.
Figure 3Unrelated cord blood transplant in children: event-free survival according to diagnosis. BMFS = bone marrow failure syndromes
]. Children given UCBT during first or second complete remission were considered as belonging to the good-risk group, whereas those transplanted in more advanced stages of the disease were assigned to the poor-risk group (Fig. 4). Kaplan-Meier estimates for neutrophil recovery at day 60 were 79% ± 6% following UCBT. In multivariate analyses, the most important factor influencing neutrophil engraftment was a nucleated cell dose infused >3.7 × 107/kg (RR 1.85; 95% CI 0.98–3.4; p = 0.05). The incidence of grade II–IV GVHD was 37% ± 6%. Kaplan-Meier estimate of 2-year EFS was 30% ± 7%. In multivariate analyses, the most important factor influencing EFS was disease status at time of transplantation: good-risk patients had a 2-year EFS of 49% ± 7% compared to 8% ± 5% in patients with chemotherapy-resistant or more advanced disease (RR 0.40; 95% CI 0.24–0.65; p = 0.0003). This was a consequence of both an increased 1-year transplant-related mortality and a higher 2-year relapse rate in the poor-risk group (65% ± 9% and 77% ± 14%, respectively) compared to good-risk patients (34% ± 6% and 31% ± 9%, respectively). Notably, there were no major differences in outcomes between related and unrelated UCBT. These data confirm that allogeneic UCBT from an unrelated donor is a feasible procedure able to cure a significant proportion of children with acute leukemia, especially if they undergo transplantation while they are in a favorable phase of disease.
Figure 4Cord blood transplant in children with acute leukemia: event-free survival according to leukemia status at transplant
We then performed a multicenter analysis comparing the outcome of unrelated cord blood transplant to unrelated BMTs in children with acute leukemia. We analyzed 515 UCBTs in children with acute leukemia transplanted either with an unrelated UCBT (n = 99) or an unrelated BMT (UBMT) (n = 416) between January 1994 and May 1998. The median number of nucleated cells infused was 4 × 108/kg in patients receiving UBMT and 0.38 × 108/kg in patients receiving UCBT. In unrelated BMT, 262 children received an unmanipulated BMT and 180 children received a T-cell–depleted BMT (T-UBMT) with Campath-1M being the most frequent means of T-cell depletion. On January 1, 1999, median follow-up was 29 months (range 7–60); it was significantly shorter in the UCBT group (p < 0.001) because most of UCBTs (90%) were performed after January 1996.
The major differences among the three groups were the higher number of HLA mismatches (defined by serology for class I and molecular typing for DRB1) in the UCBT group. The donor was HLA-mismatched in 92% of UCBT, 18% of UBMT, and 43% of T-UBMT (p < 0.001). Other significant differences were observed in pretransplant disease characteristics, preparative regimens, GVHD prophylaxis, and number of cells infused. Table 3 lists the probability of neutrophil and platelet recovery, acute and chronic GVHD, early transplant-related mortality, relapse, and overall survival by transplant type without adjustment for differences in prognostic variables. A significant delay in neutrophil and platelet recovery was observed in the UCBT group compared to the UBMT and the T-UBMT groups (p = 0.001). A significant reduction of acute GVHD >II and of chronic GVHD was observed in T-UBMT and UCBT compared to UBMT (p = 0.001). Early transplant-related mortality was higher in the UCBT group compared to the other groups (p = 0.01). Two-year overall survival and EFS were better in the UBMT and similar in the T-UBMT and the UCBT groups.
Table 3Univariate analysis of outcomes (95% confidence interval) by transplant group unadjusted for differences in prognostic factors
After adjustment for prognostic variables, differences in outcomes appeared during the first 100 days after transplant. Compared to UBMT, UCBT recipients had delayed hematopoietic recovery (hazard rate [HR] 0.37; 95% CI 0.27–0.52; p < 0.001), increased 100-day transplant-related mortality (HR 2.13; 95% CI 1.20–3.76; p < 0.01), and decreased acute GVHD (HR 0.50; 95% CI 0.34–0.73; p < 0.001). T-UBMT recipients had decreased acute GVHD (HR 0.25; 95% CI 0.17–0.36; p < 0.0001) and increased rate of relapse (HR 1.96; 95% CI 1.11–3.45; p = 0.02). After day 100 posttransplant, the three groups achieved similar results in terms of relapse. Chronic GVHD was decreased after T-UBMT (HR 0.21; 95% CI 0.11–0.37; p 0.0001) and UCBT (HR 0.24; 95% CI 0.01–0.66; p = 0.002), and overall mortality was higher in T-UBMT recipients (HR 1.39; 95% CI 0.97–1.99; p < 0.07).
The main findings that emerged from these adjusted comparisons were the poor results associated with UCBT concerning neutrophil and platelet recovery and early transplant-related mortality. In contrast, UCBT and T-cell depleted BMT resulted in less acute GVHD than the unmanipulated BMT group. Finally, the unmanipulated BMT and the UCBT groups had a lower incidence of leukemic relapse than the T-cell–depleted group. Considering the long-term outcome, the unmanipulated BMT group had more chronic GVHD than the T-cell–depleted BMT and the UCBT groups. Furthermore, although the outcomes of the three groups were comparable for long-term relapse, mortality after day 100 was increased in the T-cell–depleted BMT group mostly because of the occurrence during the first 100 days after BMT of early relapse, GVHD, and lack of engraftment.
In summary, the main differences in adjusted outcomes among the three transplant groups occurred during the first 100 days after transplant, but no group had any advantage. The delay of engraftment and increased treatment-related mortality observed after UCBT must be balanced with the higher risk of acute GVHD after unmanipulated BMT and the higher risk of relapse after T-cell–depleted BMT. In contrast, after day 100 posttransplant, the three groups achieved similar results in terms of relapse, yet chronic GVHD occurred more frequently with unmanipulated BMT and death with T-cell–depleted BMT.
We conclude that use of UCBT, as a source of hematopoietic stem cells, is a reasonable option for children with acute leukemia lacking an acceptably matched unrelated donor.
Results in adults transplanted with unrelated donor umbilical cord blood
In 108 adults who received an unrelated UCBT for malignant disorders, median follow-up was 20 months (range 0.6–60), median age was 26 years (range 15–53), median weight was 60 kg (range 35–110), and median number of nucleated cells infused was 1.7 × 107/kg (range 0.2–6). The diagnosis was 32 cases of acute lymphoblastic leukemia (ALL), 23 acute myeloid leukemia (AML), 37 chronic myeloid leukemia (CML), 12 myelodysplastic syndrome, and 4 non-Hodgkin's lymphoma. Prior to UCBT, an autologous BMT had failed in 20 patients. The donor was mismatched for 1–3 HLA antigens in 102 cases, only 6 donors were considered as matched (A and B by serology and DRB1 allelic typing). The overall 1-year survival was 27%, being 39% in patients transplanted for CML in chronic phase or acute leukemia in first or second complete remission and 17% in patients transplanted in more advanced stage of the disease. The 60-day probability of neutrophil engraftment was 81% at a median time of 32 days (range 13–60). Neutrophil recovery was improved in patients with other diseases than CML (p = 0.08) and in patients who received >1.7 × 107 nucleated cells/kg (p = 0.01). The incidence of acute GVHD ≥II was 38%. The number of relapses was low, but the follow-up was too short and most of the deaths were related to infection or GVHD. Factors associated with decreased 100-day transplant-related mortality were disease in chronic phase or remission (p = 0.016), number of nucleated cells infused/kg ≥2.0 × 107 (p = 0.004) (Fig. 5), and transplant performed after January 1998.
Figure 5Unrelated cord blood transplant in adults with malignancies: 100-day transplant-related mortality according to nucleated cells infused per kilogram
These results show that UCBT can be a good option for adults who lack an unrelated matched donor, indicating that this approach should be further investigated.
Discussion and conclusion
This registry-based analysis confirms other reported results and supports the concept of establishing cryopreserved cord blood banks for clinical use. Several questions have been answered by these analyses, but others remain to be investigated.
Engraftment has been, and remains, the major concern, as all studies showed that the time to neutrophil and platelet recovery was delayed whereas long-term engraftment was similar after UCBT and BMT. This result was expected, as it has been shown that the number of infused cells predicted the outcome after BMT [
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.
Successful transplantation of HLA matched and HLA mismatched umbilical cord blood from unrelated donors analysis of engraftment and acute graft versus host disease.
]. In our study, we found a correlation between the time to hematopoietic recovery and the infusion of >3.7 × 108 marrow nucleated cells per recipient's weight (one log higher than cord blood cells). Our results confirm our previous recommendation that units should be selected on the basis of a number of nucleated cells collected >4 × 107/kg recipient body weight before thawing. The cause for delayed recovery after UCBT might be due to the low number of cells infused or to other factors such as the immaturity of stem cells that might require more cell divisions before differentiation to myeloid progenitors or the lack of subpopulations facilitating engraftment. We also can speculate that other confounding factors increase the incidence of TRM after UCBT, such as toxicity of intensive preparative regimens and delayed immune reconstitution. Whether current approaches being explored to speed hematopoietic recovery after cord blood transplantation, such as ex vivo expansion, will result in decreased TRM is unknown. It is interesting to note that the number of cells infused is one log less than a standard allogeneic BMT and 10 times less than a standard peripheral blood stem cell transplantation. The observation that the number of cells infused is far below the recommended dose of bone marrow cells is in favor of the hypothesis that cord blood cells have a selective qualitative advantage over adult bone marrow cells. We also have shown that other factors interfere with engraftment, such as patient diagnosis and HLA incompatibility. Patients with severe aplastic anemia or with hemoglobinopathies had a higher incidence of graft failure than patients with leukemia or inborn errors of metabolism. Similarly with allogeneic BMT, this failure is due to the addition of several factors, including alloimmunization and reduction of the conditioning myeloablative regimens used for transplantation in nonmalignant disorders. The role of HLA disparities still is unclear in related and unrelated situations. Our analyses as well as the New York cord blood bank analysis [
] have documented a relation between the speed of neutrophil and platelet recovery and the degree of HLA mismatch. However, we could not find any correlation between the degree of HA mismatch and the severity of GVHD or survival in UCBT. More patients and better typing technology are needed for definitive conclusions.
To improve the speed of engraftment, the use of hematopoietic growth factors, including granulocyte colony-stimulating factor, stem cell factor, or thrombopoietin, after transplant has been investigated. At this stage, the usefulness of these factors has not been demonstrated and deserves further investigation. Another approach being explored is the expansion of cord blood progenitors in vitro to improve short-term engraftment. This area of investigation seems particularly interesting, as in vitro studies have shown that the stem cell yield was increased by ex vivo expansion with cord blood compared to bone marrow cells. The use of cytokine cocktails, including thrombopoietin and FLT3, are particularly effective, and preliminary clinical observations are encouraging [
]. Another avenue of research is the possibility of infusing several cord blood units to increase stem cell yield. From a practical point of view, these findings form the basis of recommendations to umbilical cord blood banks to collect as many cells as possible and to improve the technique of processing and volume reduction to produce units with high cellular counts.
One of the first concerns raised by the use of umbilical cord blood for allogeneic transplant was the possibility of inadvertently transplanting cells of maternal origin. We and other authors have shown that maternal cells were always present in cord blood, but that their number was insufficient to induce GVHD [
]. Their presence in large numbers can be detected in cord blood by high-molecular-resolution HLA typing that can detect additional lymphocyte populations of maternal origin. As in previously published UCBT series, we observed GVHD, but it usually was not severe and the incidence of chronic GVHD was low. This is remarkable in highly mismatched transplants without T-cell depletion. In recent analyses of unrelated BMT, incompatibilities detected by high-resolution HLA typing for class I and class II were predictive of GVHD and survival [
]. In one study, the overall incidence of grade II–IV GVHD was 73%, and grade III–IV GVHD was estimated at 21% in patients receiving a high-resolution HLA-matched unmanipulated unrelated BMT and 47% in patients receiving a mismatched unrelated BMT. This must be compared to the incidence of 21% grade III–IV acute GVHD in unrelated mismatched CBT.
The incidence of acute and chronic GVHD was lower after related HLA-identical UCBT than after related HLA-identical BMT. In unrelated transplants for acute leukemia in children, the difference was even more striking, as the incidence of acute GVHD ≥II was lower after HLA-mismatched UCBT compared to unmanipulated unrelated BMT and intermediate with T-cell depleted UBMT. Chronic GVHD was identical for both UCBT and T-cell depleted unrelated BMT and significantly lower than in unmanipulated, unrelated BMT. This trend toward decrease of the incidence and severity of both acute and chronic GVHD also was found in the group of adults transplanted with a mismatched, unrelated cord blood donor. This study shows that the decreased incidence of acute and chronic GVHD after UCBT still is observed in the presence of major class I and class II HLA mismatches. This observation lends support to the hypothesis that umbilical cord blood differs from bone marrow in its alloreactive potential. Studies directly comparing graft composition of umbilical cord blood and BMTs for lymphoid numbers and proportion of different subsets of T, B, and NK cells are available, but these cell types are expected to be present in lower absolute numbers due to the low total cell dose of cord blood grafts. The hypothesis that reduced GVHD results from fewer T cells infused is plausible, because T-cell depletion of BMTs leads to a lower GVHD risk. However, the number of T cells infused with UCBTs is on the order of 8 × 106/kg, and it is known that GVHD can be induced by as few as 1 × 106 CD3+ cells/kg and even fewer in HLA-mismatched situations. Because acute GVHD results from activation, clonal expansion, and proliferation of donor-derived T lymphocytes that recognize alloantigens presented by either host or donor antigen-presenting cells, the lower GVHD risk after UCBT might be due to impairment of these functions in umbilical cord blood cells. Therefore, looking for complete HLA identity does not seem to be an absolute prerequisite for a successful UCBT, as we did not find any correlation between the number of HLA mismatches and survival after UCBT. HLA compatibility has been difficult to evaluate, and discrepant results have been obtained from different analyses. The methods of HLA typing have changed over time and new polymorphisms have been described, including minor histocompatibility antigens and cytokines polymorphisms. Among other confounding factors, the age of donor and recipient, cytomegalovirus serology of the recipient, sex of the donor, and number of pregnancies for female donors must be included in any multivariate analysis. This complexity emphasizes the need for large multi-institutional studies through the Registry. At this stage, we can provisionally conclude that HLA compatibility is not the major factor for selecting a donor and that the primary criteria of choice should be based on the number of cells infused.
As the interaction between lower risk of GVHD and higher risk of leukemic relapse is known, we expected a higher risk of relapse in UCBT compared to unmanipulated BMT, but we did not find any difference between the adjusted risk of relapse in the unmanipulated BMT and the UCBT. The number of patients with early relapse was higher in the T-cell–depleted BMT. If confirmed with more patients and more follow-up, this observation could indicate that NK or other cells present in cord blood have an antileukemic potential despite the immaturity of the immune system at birth [
In conclusion, the accumulation of clinical data indicates that UCBT is a good option for patients who require allogeneic hematopoietic stem cell transplant. Search for an unrelated umbilical cord blood donor should be used routinely at the same time as a bone marrow donor search. The future objectives of Eurocord are to compare the results of UCBT to BMT in each indication including unrelated BMT and haplo-identical related BMT in order to analyze prognostic factors associated with each transplant source and to publish recommendations for each of the following treatment strategies. These studies will be performed with Netcord, EBMT, and IBMTR. In addition, we plan to perform prospective studies analyzing the different factors associated with outcome, including the role of major and minor histocompatibility antigens, the methods of conditioning and GVHD prevention, and the study of immune reconstitution after UCBT.
Acknowledgements
This work was supported by an EEC grant Eurocord BIOMED II BMH-CT96-0833, by Etablissement français des greffes and PHRC 96, French Ministry of Health. The author thanks members of Eurocord and Netcord for providing their data to the Registry as well as the following staff members of Eurocord for their assistance: Vanderson Rocha, M.D., medical coordinator; Sylvie Chevret, biostatistician; and Federico Garnier and Irina Ionescu, data managers.
References
Gluckman E.
Broxmeyer H.E.
Auerbach A.D.
et al.
Hematopoietic reconstitution in a patient with Fanconi's anemia by means of umbilical cord blood from an HLA-identical sibling.
A phase I-II clinical trial to evaluate removal of CD4 cells and partial depletion of CD8 cells from donor marrow for HLA-mismatched unrelated recipients.
Children with acute lymphoblastic leukemia who receive T-cell depleted HLA mismatched marrow allografts from unrelated donors have an increased incidence of primary graft failure but a similar overall transplant outcome.
Successful engraftment of T-cell depleted haploidentical three loci incompatible transplants in leukemia patients by addition of recombinant human granulocyte colony stimulating factor mobilized peripheral blood progenitor cells to bone marrow inoculum.
Recipient tumor necrosis factor-alfa and interleukin-10 gene polymorphisms associate with early mortality and acute graft versus host disease severity in HLA-matched sibling bone marrow transplants.
Comparison of graft-versus-host disease in children transplanted with HLA identical sibling umbilical cord blood versus HLA identical sibling bone marrow transplant.
00540-3&id=gr1.gif){kind=link}
00540-3&id=gr2.gif){kind=link}
00540-3&id=gr3.gif){kind=link}
00540-3&id=gr4.gif){kind=link}
00540-3&id=gr5.gif){kind=link}