The use of umbilical cord blood as a source of marrow repopulating cells for the treatment of pediatric malignancies has been established. Given the general availability, the ease of procurement, and progenitor content, cord blood is an attractive alternative to bone marrow or growth factor mobilized peripheral blood cells as a source of transplantable hematopoietic tissue. However, there is a major potential limitation to the widespread use of cord blood as a source of hematopoietic stem cells for marrow replacement and gene therapy. There may be enough hematopoietic stem cells to reconstitute children, but the ability to engraft an adult might require ex vivo manipulations. We describe an in vitro system in which the growth of cord blood CD34+ cells is sustained and greatly expanded for more than 6 months by the simple combination of two hematopoietic growth factors. Progenitors and cells belonging to all hematopoietic lineages are continuously and increasingly generated (the number of colony-forming unit–granulocyte-macrophage [CFU-GM] present at the end of 6 months of culture are well over 2,000,000-fold the CFU-GM present at the beginning of the culture). Very primitive hematopoietic progenitors, including long-term culture-initiating cells (LTC-ICs) and blast cell colony-forming units, are also greatly expanded (after 20 weeks of liquid culture, LTC-IC number is over 200,000-fold the initial number). The extremely prolonged maintenance and the massive expansion of these progenitors, which share many similarities with murine long-term repopulating cells, suggest that extensive renewal and little differentiation take place. This system might prove useful in diverse clinical settings involving treatment of grown-up children and adults with transplantation of normal or genetically manipulated hematopoietic stem cells.

To assess the kinetics of lymphocyte subset recovery, 758 allografted patients were monitored by surface markers (CD3, CD4, CD8, CD56), with a 5-year follow-up. The donor was a matched sibling donor (MSD) (n ¼ 502) or an alternative donor (family mismatched or unrelated, AD) (n ¼ 256). The stem cell source was bone marrow for all patients. CD4 þ cell recovery was influenced-in univariate analysis-by three factors: donor type, patient age and GvHD. This was not the case for CD8 þ and CD56 þ cells. The median CD4 þ cell count on day þ 35 after HSCT was 86/ll. Patients achieving this CD4 þ cell count had significantly lower transplant-related mortality (TRM) compared to patients who did not achieve this CD4 þ cell count (20 vs 39%, P ¼ 0.00001), due to a lower risk of lethal infections (24 vs 47%, P ¼ 0.0003). In multivariate analysis MSD (RR 3.45, P ¼ 0.0001) and recipient age less than 16 years (RR 3.23, P ¼ 0.003) were significantly associated with a better CD4 þ cell recovery. CD4 þ counts on day þ 35 was predicted TRM (RR ¼ 1.97, P ¼ 0.0017) together with acute GvHD grade II-IV (RR 1.59, P ¼ 0.0097). No difference of TRM was observed for CD8 þ and CD56 þ cell counts.