For patients without an HLA-identical sibling donor, alternative sources of stem cells are partially HLA-matched relatives, matched unrelated volunteers, or unrelated cord blood. In this issue, 2 articles compare the outcome of unrelated cord blood and marrow transplantation in children.
Barker and colleagues (page 2957) report on a retrospective, single-center study of unrelated cord blood or marrow transplants in children with various diseases. While engraftment of granulocytes and platelets was slower with cord blood, survival was similar with either cord blood or marrow. Survival after cord blood transplantation in this series appears better than in larger series. Rocha and colleagues (page2962) report on a multicenter retrospective study of unrelated cord blood or marrow transplants for children with leukemia. Cord blood led to slower engraftment of granulocytes and platelets, and higher mortality, which occurred during the first 100 days. Relapse was similar after cord blood or marrow transplantation. Despite a higher degree of HLA incompatibility, cord blood was associated with less acute and chronic GVHD than was marrow. The median interval from the last leukemia remission induction to transplantation was 30 days shorter with cord blood than with marrow. Because patients with acute leukemia may deteriorate and die while waiting for transplantation, the survival gain from a shorter wait for a cord blood graft could partially offset the increased mortality after transplantation. The data from the 2 studies support the conclusion that transplantation of partially matched cord blood is effective treatment for children without a suitably matched unrelated marrow donor.
Transplantations in children have shown that HLA-identical sibling cord blood is associated with less GVHD and same survival compared to HLA-identical sibling marrow, suggesting that cord blood T-cell naı̈veté protects from severe GVHD and that HLA matching criteria for selection of unrelated cord blood may be relaxed. Almost all unrelated cord blood transplants, performed thanks to more than 65 000 units available globally, are mismatched for one or more HLA alleles. What constitutes an acceptable mismatched cord blood remains undefined. Slow engraftment and graft failures are the predominant hurdles that limit success of cord blood transplantation in children and even more in adults. HLA mismatching and low cell dose are risk factors for slow engraftment and graft loss. Therefore, new strategies for suppression of the host immune response across HLA disparity and, perhaps, stem cell expansion will be necessary for broadening the utilization of cord blood transplants.