Contribution of T-Cell Reconstitution during the Early Phase after Cord Blood Transplantation to Favorable Clinical Outcomes.

The immaturity of T cells in cord blood is well known in functional assays and phenotypic analyses. During the first several months after cord blood transplantation (CBT), the T cell compartment is recovered by peripheral expansion from those mature and naïve T cells in cord blood grafts and plays an important role in acute graft-versus-host disease (GVHD) and graft-versus-leukemia reaction. Recently, we have reported that adult patients with hematological malignancies receiving CBT from HLA-partially-mismatched unrelated donors (n=68) had a lower risk of severe acute GVHD (> grade II, 7% versus 26%) and transplant-related mortality (9% versus 29% at 1 year) and a higher probability of disease-free survival (74% versus 44% at 2 years) than HLA-matched unrelated bone marrow transplant (BMT) recipients (n=45) in our multivariate analysis (Takahashi et al., Blood, in press). We speculated that the immune reconstitution process over a period of several months after CBT might have contributed to these promising clinical results. Using four-color analysis with CD4, CD8, CD45RA, and CD62L, more than 90% of cord blood CD4⁺ and CD8⁺ T cells in the grafts belonged to the naïve fraction. Cytokine expression in cord blood T cells was also suppressed to 0.1% in CD4⁺ and to 0.9% in CD8⁺ with positive interferon-γ by intracellular staining, which were significantly lower than those in adult T cells (16.2% in CD4⁺ and 37.8% in CD8⁺). Circulating T cell counts normalized after 3 months for CD8⁺ and 4 months for CD4⁺ in our CBT recipients, both of which were significantly faster than in previously published studies, which were 9 months for CD8⁺ and 12 months for CD4⁺. After T cell recovery, peripheral blood T cells moved from the naïve to the central memory fraction immediately, and then moved to the effector memory fraction. A naïve subset of CD4⁺ T cells remained (median: 38 cells/μl on day 90, n=12) during the first 3 months, which was significantly higher than in the BMT control (median: 9 cells/μl on day 90, n=5, p=0.015), but showed a low level of CD8⁺ T cells (median: 14 cells/μl on day 90, n=12), almost the same as in BMT recipients (median: 13 cells/μl on day 90, n=5). Intracellular interferon-γ-producing T cells were detected at 3.4% (0.1–34.2%) in CD4⁺ and 32.3% (1.1–86.9%) in CD8⁺ at 1 month post-CBT (n=16), both of which were comparable to post-BMT. To investigate whether these T cells with memory phenotype are functional, we analyzed antigen-specific T-cell recovery using cytomegalovirus (CMV) as a specific antigen. CMV-responsive CD4⁺ T cells were detected within the first 4 months in all recipients with positive CMV antigenemia (n=13), but CD8⁺ T cells were detected only in 5 out of 13 cases, probably because of pre-emptive Gancyclovir administration in most antigenemia-positive patients. To conclude, naïve cord blood T cells rapidly increased in number and adopted a memory phenotype showing cytokine-production and antigen-recognition capacity in the early phase after CBT. These data suggest that mature T lymphocytes in cord blood have unique properties and contribute to the favorable clinical outcome of CBT.