It has been more than 10 years since the first unrelated-donor cord-blood transplantation for leukemia.1 In the intervening time, a number of major publications have confirmed the feasibility of using cord blood from mismatched unrelated donors for hematopoietic reconstitution following treatment with a myeloablative preparative regimen. The more important question to be addressed, however, regards utility: is unrelated-donor cord-blood transplantation effective in the treatment of acute myeloid leukemia (AML) in children? And the answer is.. .well, maybe.
In this issue of Blood, Michel et al (page 4290) review the outcomes of 95 children with AML following unrelated-donor cord-blood transplantation. Fewer than 10% of these children were HLA-matched with their donors; in fact, 44% of the patient-donor pairs had 2 or 3 antigens mismatched. Without T-cell depletion, this degree of histoincompatibility would be prohibitive when using marrow or peripheral blood as a source of stem cells, as a result of the high risk of severe and fatal graft-versus-host disease. The naive immune system of cord blood, however, appears to confer a substantial reduction in the risk of graft-versus-host disease. This, of course, begs the question of whether there is also a reduction in the graft-versus-leukemia effect.
Using cord blood cells, Michel et al report a 50% 2-year leukemia-free survival for patients who underwent transplantation in second remission. With intensive chemotherapy alone, children with AML who achieve a second remission following a long first remission have a prolonged survival, whereas those with a short first remission have a poor prognosis.2 In contrast, the good disease-free survival following unrelated-donor cord-blood transplantation was independent of the duration of first remission. Moreover, 2-year disease-free survival was 21% for children who underwent transplantation in relapse, a group with little chance of survival with chemotherapy alone. Thus, there appears to be a substantial graft-versus-leukemia effect with unrelated-donor cord-blood transplantation for patients with high-risk second remission and relapsed AML.
The question of benefit is less clear for the children who underwent transplantation in first remission. The 2-year leukemia-free survival following cord-blood transplantation was 59% for these children. However, prolonged remissions have been reported in 50% to 60% of children with AML using intensive chemotherapy regimens at diagnosis.3 Cytogenetics has been used to identify children with poor-risk AML who might benefit from transplantation in first remission,4 but Michel et al do not report the outcomes for this subgroup alone. Consequently, the role of unrelated donor cord blood transplantation for AML in first remission remains to be defined.