Cord Blood Transplantation

Article Outline

• Introduction
• Address
• Ten most important publications of Eurocord
• Copyright

Key words:  Cord blood , Transplantation

Back to Article Outline

Introduction 

Professor Eliane Gluckman, MD, was this year’s honored E. Donnall Thomas lecturer at the Tandem Bone Marrow Transplant (BMT) Meetings, held on February 18, 2006, in Honolulu, Hawaii. Dr Gluckman was born and trained in Paris, France, and is currently Head of Faculty in the Department of Hematology at the Hospital St Louis, Paris. After her residency, she was a fellow of Professor Thomas in Seattle, Washington. From that point on, Dr Gluckman has always been very active in the field of allogeneic stem cell transplantation, with a focus on advancing methods for the prevention and treatment of graft-versus-host disease (GVHD) and the development of programs to facilitate alternative donor transplants for those patients without an HLA-identical sibling. In 1988, she established a phenomenal and productive initial collaboration with Dr Hal Broxmeyer and A. D. Auerbach to introduce the concept of umbilical cord blood transplantation (CBT) to the field. Since then, Dr Gluckman has been at the forefront of CBT research that has seen the approach become very much a standard of practice for children who are not fortunate enough to have an HLA-identical sibling donor. The practical use of unrelated donor CBT has also been gaining recognition for use in adults and may be considered a better option for patients than transplantation of matched unrelated blood or bone marrow cells. The following text is a modified transcribed version of the presentation made by Dr. Gluckman.

Back to Article Outline

Address 

I will discuss today the current results and future for CBT. We know that, in cord blood, there is a high number of hematopoietic stem cells that can be transplanted to treat a large variety of hematopoietic disorders. More recently, and this is a new exciting finding, we found that cord blood also contains nonhematopoietic stem cells that might be used in the future for tissue and organ replacements.

The story of the clinical application of cord blood began in 1988. I’m very grateful that Dr Hal Broxmeyer, in New York and now in Indianapolis, called me in Paris and asked me if I wanted to perform a CBT for a patient with Fanconi anemia. We had already been collaborating with Dr Arleen Auerbach at Rockefeller University on the diagnosis and treatment of Fanconi anemia. Our group was the first to describe the successful use of a nonmyeloablative conditioning in Fanconi anemia, demonstrating that the DNA repair defect in this disease made the patients unable to tolerate high-dose cyclophosphamide. For this reason, I was asked to perform this first CBT to optimize the chance of success.

This transplant was really straightforward. The only difference between this transplant and a normal bone marrow transplant (BMT) was that, instead of using bone marrow cells from the HLA-matched sibling donor, we used the sibling cord blood that had been cryopreserved at birth. We knew before birth that this donor was HLA-identical and not affected by the disease. The patient had a very uneventful engraftment; he is now followed by Dr Joanne Kurtzberg at Duke University and is still doing well. It was the first demonstration that cord blood could generate a long-term hematopoietic and immunologic reconstitution.

After the publication of this case in the New England Journal of Medicine, several questions were frequently asked. First, what was the general applicability of CBT? Was it only for a rare congenital disease with an HLA-identical sibling, or would it be applicable for all BMT indications? Nowadays, we can say yes, CBT has been used in a large variety of malignant and nonmalignant diseases. The second question was: Would it engraft properly? The question was asked because we know that in a single cord, there is one-tenth the number of cells as in a marrow collection. We can say yes, it can engraft children and adults, but engraftment is delayed. Would HLA incompatible grafts be feasible? The immunologic system is immature at birth, and we know that lymphoid cells from cord blood are naive, immature, and have decreased reactivity. We have shown that mismatched transplants were possible and that there was a decreased incidence and severity of acute and chronic GVHD. Another question that is not discussed anymore but was important at that time was the possibility that there would be maternal cells in cord blood that could engraft and give rise to severe acute GVHD. We conducted a study of this issue and found that, indeed, there were maternal cells in cord blood, but not in very high numbers. However, even if they were detectable, we have not seen a single case of engraftment of these cells, nor of GVHD related to this contamination. So we think it is irrelevant. Finally, it was also questioned whether it was safe to collect cord blood at birth. Would it deprive the infant of important cells? This has been thoroughly studied, and we can say that it is safe and that the amount of 100 mL of cord blood collected after birth is harmless to the child.

After the first CBT was performed in 1988, the next step was the clinical observation that GVHD was reduced compared with HLA-matched BMT. The first paper we published with Center for International Bone Marrow Transplant Research (CIBMTR) showed that, in children receiving an HLA-identical sibling transplants, either from bone marrow or from cord blood, the outcomes were the same in terms of survival and engraftment, but the big difference was that the level of acute and chronic GVHD was reduced in the latter. This observation was also reproduced in unrelated HLA-mismatched CBT. Based on these findings, cord blood banks were established for unrelated transplants, and the 4 major banks were in New York, Paris, Milan, and Düsseldorf. Since then, there has been extensive cord blood banking activity, and now there are more than 200 000 U stored worldwide.

In 1995, we established Eurocord whose aim was to collect data on cord blood transplants worldwide. Our group and others showed that the nucleated cell count was the most important factor for engraftment and survival and that there was some influence of HLA on engraftment. More recently, it has been shown that CBT can be performed not only in children but also in adults, the results of which are comparable to unrelated BMT. New exciting results coming from the Minnesota group showed that double CBT can give good results with a decrease of relapse rates, and that it is possible to condition with a nonmyeloablative regimen.

To review CBT results, the Eurocord registry has collected data from 2310 CBTs performed from 1988 to October, 2005, provided by 318 transplantation centers in 39 countries. Of these, 1500 blood cases came from European Blood and Marrow Transplant (EBMT) centers and 742 came from non-EBMT centers. Overall, the reports indicated that the number of related CBT performed annually has really not increased over the years, although probably not all of them have been reported. In contrast, there was a sharp increase in the number of unrelated CBTs in 2004 and 2005. The proportion of children has changed with time, with more adults transplanted recently.

Concerning the 261 cases of related CBT, their number did not increase with time; all were children, 60% were transplanted for malignant diseases (Acute Lymphoblastic Leukemia (ALL) and Acute Myeloblastic Leukemia (AML)) and 40% for nonmalignant diseases, mostly bone marrow failure syndromes, metabolic disease, and hemoglobinopathies. For these latter recipients, there was 100% survival and 86% event-free survival. Due to these extremely good results, we are trying to promote—and we are doing that extensively in Paris—cord blood collection in siblings of patients with sickle cell anemia, which is relatively easy to do in families of African origin. For patients with bone marrow failure syndromes, including Fanconi anemia, constitutional aplastic anemia and idiopathic aplastic anemia, CBT resulted in 63% survival and 51% event-free survival.

For unrelated CBT, we collected data from 1320 children and 693 adults, the majority with malignant diseases (962), mostly acute and chronic leukemia, lymphoma. and myeloma. In 270 cases of nonmalignant diseases, we collected 113 cases of aplastic anemia, 124 cases of severe combined immunodeficiency, and 89 cases of metabolic diseases. Survival for ALL was 50% in first complete remission (CR1), 45% in second complete remission (CR2), with no difference between CR1 and CR2, and 29% in more advanced disease. For AML, survival was 70% in CRI, 55% in CR2, and 27% for more advanced disease. These results are quite interesting for totally mismatched transplants and quite comparable with the results of matched unrelated bone marrow transplants. Survival following CBT for children with congenital disease was 68%, metabolic disease 63%, and congenital bone marrow failure syndrome 43%. For the latter group, separation into patients with Fanconi anemia, 49%, and congenital marrow failure syndrome is around 70% survival. For adult CBT before 1998, there was only 20% survival. In CBT performed after 1988, there was 30% survival, because both the quality of the transplant and the selection of the patients improved.

Within the Netcord organization, we were able to set up standards with the FACT/Netcord accreditation to ensure good quality units and to be able to exchange units all around the world. Of the more than 100 000 units that have been collected, 3942 have been used for transplant, 2300 have been used for children, and 1500 for adults.

There has been discussion about the autologous cord blood banking and the relation between public and private banks. Most of these banks are located in the United States, some are in Europe, but some countries do not authorize it. In Asia, the proportion of private banks is high. There have been a lot of ethical discussions on this topic because the rationale for autologous cord banking is not clearly established. The approach has been advertised as biological insurance for the future of the newborn, in that nonhematopoietic stem cells, that are present in cord blood, might later be useful. Although there is extensive research on nonhematopoietic stem cells in cord blood, it’s not yet ready for clinical use, so there is a lot of controversy about selling a product for which the clinical use has not yet been proven.

During this period, most of our work had been to compare the result of unrelated mismatched CBT with the results of unrelated matched BMT. In a study that we did in 2001, in children with acute leukemia, we compared outcomes for 3 groups of patients—those who had received a T-replete unrelated BMT, a T-depleted unrelated BMT, or an unrelated CBT. We found that there was a delay of engraftment in the CBT recipients, although it was the same as in T-replete BMT. There was increased transplant-related mortality and this was due to the delay of engraftment. However, the number of earlier relapse in the T-depleted group was higher, and for long-term outcomes, the proportion of chronic GVHD was the same between the CBT and T-depleted BMT groups, compared with the higher proportion in the T-replete BMT group. Overall, there was no difference between the unrelated BMT and CBT in terms of survival and event-free survival.

In another study by Juliet Barker, published in Blood in 2001, and 2 other studies, published in Bone Marrow Transplantation in 2004, the same results were found: delay of engraftment, diminishment of acute and chronic GVHD, same or higher transplant-related morbidity, same relapse rate, and same survival. The implication for unrelated donor selection in children is that the choice depends on the clinical situation. Cord blood is quickly available, especially in patients who need an urgent transplant or do not have an HLA-identical donor. In a combined study by CIBMTR and Cord Blood New York (CBNY) on CBT for children, which was presented at the recent American Society of Hematology (ASH) meeting, better results with matched CBT compared with matched BMT were observed.

The same comparison in adults with acute leukemia receiving either an unrelated CBT or a BMT was published in the New England Journal of Medicine in 2004. We showed that neutrophils and platelets engraftment was delayed. Acute GVHD was reduced, and chronic GVHD was the same. In regard to long-term survival, transplant-related mortality was the same, relapse was the same, and leukemia-free survival was the same in both groups. Our conclusion was that matched unrelated BMT had results that were similar to HLA-mismatched CBT.

In the same issue of the New England Journal of Medicine, Mary Laughlin et al compared BMT and CBT in adults with leukemia. The results were not exactly similar, but comparable to our study. The best results were obtained with HLA-identical unrelated BMT and CBT with 1 antigen mismatch and were comparable to BMT with 1 antigen mismatch. At the same time, the Japanese group published in Blood in 2004 a single-institution comparative analysis of unrelated matched BMT and CBT for patients with hematological malignancies. The results were quite different, with CBT leading to 70% survival compared with 40% after BMT. The cause of the difference in survival seemed due to higher levels of GVHD and infection in the BMT group.

To summarize, we can say that, in adults, mismatched CBT compared with HLA-matched unrelated BMT gives delayed engraftment, decreased or same acute GVH, decreased or same chronic GVHD, increased or same transplant-related mortality, and similar degree of survival. The conclusion is that CBT is an alternative source of stem cells in adults with high-risk hematological disorders, without a matched bone marrow donor.

We conducted a new study in which we looked at the impact of HLA compatibility and cell dose on outcomes of unrelated CBT for patients with malignant and nonmalignant disorders. We reviewed all patients transplanted from 1994 to 2005 and analyzed separately malignant (929 patients) and nonmalignant (268 patients) disorders. In patients with malignant diseases, the median follow-up time was 28 months, the median age was 11 years (range, 2 months to 56 years), and the disease characteristics were: AML 30%; ALL 47%; chronic leukemia 9%; Myelodysplastic Syndrome (MDS) 10%; other 4%. The status was 23% early; 46% intermediate; and 31% advanced disease. With HLA compatibility determined by antigen typing for class I antigens and allelic typing for class II antigens, 9% were HLA identical, 42% had 1 difference, 40% had 2 differences, and 9% had 3 or 4 HLA differences. The median number of nucleated cells (NCs) at collection was 4.4 × 10⁷/kg. The median dose infused was 3.1 × 10⁷ NCs/kg and the median CD34⁺ cells infused was 1.4 × 10⁵ NCs/kg.

First, we confirmed the role of cell dose on engraftment. The minimum cell dose collected was less than 3 × 10⁷/kg, and the minimum dose of cells infused was 2 × 10⁷ NCs/kg. HLA number of mismatches was also an important factor for engraftment with increased delay of engraftment according to the number of HLA mismatches (0-4 mismatches). The question was whether this difference was related to the number of cells infused. For 0-1 mismatch, there was a difference, meaning that if there were more than 2 × 10⁷/kg infused cells, there was a better engraftment. For 2-4 HLA differences, the effect on engraftment was abrogated by increasing cell dose. There was no difference according to either HLA class I or class II mismatches. For blood platelet recovery, the effect of cell dose was even more important, and there was also no difference between class I and class II mismatches. For II-IV acute GVHD, we did not find any effect of cell dose, nor any effect of the number of HLA mismatches. However, it seemed that class II mismatches gave more GVHD. For chronic GVHD, we found a correlation of higher incidence with the fewer cells transplanted, and we found a significant effect of HLA mismatches. Cord blood cell dose influenced transplant-related mortality. For HLA, the more matched, the better the survival, but increasing the cell dose abrogated the effect of HLA mismatches. For example, in a group with 3-4 mismatches, the transplant-related mortality was 20% versus 40% with an increased dose of cells for the same number of HLA mismatches. It was worse to be transplanted with 2 HLA class II mismatches rather than 2 class I mismatches. Relapse was higher with 0-1 HLA mismatch compared with 2-4 HLA mismatches. This explains why survival was not influenced by the number of HLA mismatches. The survival for early and intermediate stages of transplant was 38%, whereas it was 20% in advanced disease. Increasing cord blood cell dose improved survival. There was no effect of HLA on disease-free survival, although it was slightly better with 2 class II mismatches. This effect on the disease-free survival was due to the reduction of relapse rate according to the number of HLA mismatches, resulting in the same survival. When looking at the effect of the cell dose and the number of the HLA mismatches, a 3-4 HLA antigen mismatched CBT with a high cell dose gave better results than a 3-4 HLA antigen mismatched CBT with a low cell dose.

To summarize, increasing the number of HLA mismatches delayed engraftment, increased transplant-related mortality, chronic GVHD, decreased relapse, and resulted in the same overall and disease-free survival. Considering the interaction between HLA and the number of cells, the worst group was 3-4 antigen HLA mismatches, with less than 2 × 10⁷/kg mononuclear cells infused, because it delayed engraftment and increased transplant-related mortality. It had no effect on acute and chronic GVHD, and relapse was decreased by lowering the number of mismatches and increasing the number of cells. Overall survival was decreased with more HLA mismatches and fewer cells.

In conclusion, cell dose is the most important factor for outcome—requiring a minimum of 3 × 10⁷ NCs at collection, or 2 × 10⁷ NCs/kg at infusion. HLA mismatches increase the risk of delays of engraftment and increase transplant mortality and chronic GVHD, and decrease the risk of relapse. The type of HLA mismatch did not seem to have an effect, but matching for HLA class II seems better. Increasing cell dose overcomes the effect of HLA mismatches.

For nonmalignant disorders, we analyzed 268 patients with a median follow-up time of 32 months; median age was 3 years (range, 22 days to 53 years); 40% of patients had aplastic anemia, 36% had severe combined immunodeficiency, and 24% a metabolic disorder; 18% were HLA identical CBT, 43% had 1 HLA mismatch, 35% had 2 HLA mismatches, and 4% had 3 and 4 antigen mismatches. Because most of the patients were children, the number of cells infused was higher than in malignant disease. It was 8.5 × 10⁷ NCs/kg before freezing and 6.4 × 10⁷ NCs/kg infused, and the median number of CD34⁺cells was 2 × 10⁶/kg. The median cell loss was 27%.

The cell requirement was higher, and fewer than 4.9 × 10⁷ NCs/kg collected gave a low rate of engraftment. For the infused cells, it was fewer than 3.5 × 10⁷/kg compared with 2.0 × 10⁷/kg for malignant diseases. We observed that increasing the cell dose enhanced engraftment and overcame effects of HLA mismatches. Acute and chronic GVHD were influenced by the number of HLA mismatches. We found the same result for transplant-related mortality, with the best outcome for 0-1 HLA mismatch and a high cell dose. The overall survival for the group of patients with primary immunodeficiency was 65%, metabolic disease 50%, and bone marrow failure syndrome 35%. Looking at survival according to the number of cells, we found that for the high cell dose it was 60% and less than 40% in the low cell dose group. A cell dose of less than 3.5 × 10⁷ NCs/kg gave a very poor survival. Survival was affected by the number of HLA mismatches and the cell dose; it was different from the malignant diseases where HLA mismatches did not influence survival, because there was no need for a graft-versus-leukemia response.

In conclusion, requirements for cell dose and HLA mismatching are different in malignant and nonmalignant diseases. In nonmalignant diseases, minimum cell dose is higher, 4 × 10⁷/kg at collection and 3.5 × 10⁷/kg at infusion. In addition, more than a 2 HLA-antigen mismatch CBT affects engraftment, GVHD, and survival, but can be partially abrogated by increasing cell dose.

There are many new CBT protocols under investigation, but I would like to focus on one of the more exciting advances in the field, and that is the double cord transplant program pioneered by the Minnesota group. Because cell dose is very important to the success of CBT, the idea was to increase the dose by giving 2 separate HLA-mismatched CBT. This attitude was really against the dogma because the risk of giving 2 different transplants was that each transplant would reject the other, and if eventually it would take, some allogeneic reaction could impair the immune reconstitution capacity of this transplant. Surprisingly, there was rejection of 1 donor’s stem cells, but the other 1 always survived, and this conflict facilitated engraftment by a still unknown mechanism.

Another possibility to improve engraftment was to use cytokine factors and give ex vivo expanded cells. There are currently several studies with this approach, but it is too early to know if it will work. Other options are to give the cells directly into the bone to avoid loss of the cells before homing or to add mesenchymal cells to improve engraftment. An alternative new and very attractive approach, also pioneered by the Minnesota group, is to use a nonmyeloablative conditioning regimen. Preliminary results indicate that this approach works, so we will have to follow that very carefully.

An exciting area that may have an effect on CBT is the prospect of regenerative medicine. To use stem cells for cell therapy, cell repair, or “regenerative” medicine, there are different sources, ie, embryonic, cord blood, or adult stem cells, all of which have specific advantages and disadvantages. Umbilical cord is highly proliferative, can divide into almost all different cell types, and can be autologous or allogeneic. However, it is only available once in a lifetime and it is limited in quantity. The hope is that embryonic properties are retained in cord blood stem cells, making these cells the ideal source for all possible clinical applications in the future. We can thus conclude that cord blood is a unique biological resource for hematopoietic transplantation, regenerative medicine, and scientific research, and, therefore, there is a need for increasing the number of units stored.

Back to Article Outline

Ten most important publications of Eurocord 

1. Gluckman E , Rocha V , Boyer-Chammard A , et al.   Outcome of cord blood transplantation from related and unrelated donors . N Engl J Med. . 1997;337:373–381
2. Eurocord-Cord Blood Transplant Group Locatelli F , Rocha V , Chastang C , et al.   Factors associated with outcome after cord blood transplantation in children with acute leukemia . Blood . 1999;93:3662–3671
3. Rocha V , Wagner JE , Sobocinski K , et al.   Comparison of graft versus host disease in children transplanted with HLA-identical sibling umbilical cord blood versus HLA-identical sibling bone marrow . N Engl J Med . 2000;342:1846–1854
4. 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
5. Burgio GR , Gluckman E , Locatelli F . Ethical reappraisal of 15 years of cord-blood transplantation . Lancet . 2003;361:250–252
6. Locatelli F , Rocha V , Reed W , et al.   Related umbilical cord blood transplant in patients with thalassemia and sickle cell disease . Blood . 2003;101:2137–2143
7. Michel G , Rocha V , Chevret S , et al.   Unrelated cord blood transplantation for childhood acute myeloid leukemia (a Eurocord group analysis) . Blood . 2003;102:4290–4297
8. Gluckman E . Hematopoietic stem-cell transplants using umbilical-cord blood . N Engl J Med. . 2001;344:1860–1861
9. Eurocord Group Gluckman E , Rocha V , Arcese W , et al.   Factors associated with outcomes of unrelated cord blood transplant (guidelines for donor choice) . Exp Hematol. . 2004;32:397–407
10. Acute Leukemia Working Party of European Blood and Marrow Transplant Group Eurocord-Netcord Registry Rocha V , Labopin M , Sanz G , et al.   Transplants of umbilical-cord blood or bone marrow from unrelated donors in adults with acute leukemia . N Engl J Med . 2004;351:2276–2285