Cytotoxic T Lymphocytes (CTL) Specific for Adenovirus and CMV Can Be Generated from Umbilical Cord Blood for Adoptive Immunotherapy

Umbilical cord blood (UCB) transplantation is a promising alternative source of hematopoietic stem cells for patients lacking HLA-matched donors. Nearly 60% of UCB transplants to date have been performed on minority individuals for whom an unrelated donor was not available; moreover, the naive phenotype of UCB cells may be responsible for the lower incidence and reduced severity of GvHD in these patients. The relatively low cell numbers and naivety of T lymphocyte populations in UCB grafts has, however, lead to delayed immune reconstitution and higher mortality due to infection. Reactivations of latent viruses such as cytomegalovirus (CMV) are particularly problematic, as is overt infection from adenovirus (Adv). Previous studies have shown that prophylactic adoptive immunotherapy with peripheral blood-derived CTL directed against CMV and Adv can effectively prevent the clinical manifestations of these viruses after hematopoietic stem cell transplant raising the possibility that a similar approach could be developed after UCB transplant. We hypothesized that virus-specific CTL could be generated from UCB for clinical use to restore anti-viral immunity and reduce viral infection post UCB transplant. Bi-virus specific CTL were generated from frozen UCB mononuclear cells using a clinical-grade recombinant adenovirus type5 vector pseudotyped with a type35 fiber carrying a transgene for CMVpp65 as a source of Adv and CMV antigens. UCB-derived dendritic cells were transduced with this Ad5f35pp65 vector as the initial source of antigen presenting cells to stimulate virus-specific CTL in the presence of IL-7, IL-12 and IL-15. This was followed by 2 rounds of weekly stimulation with autologous UCB-derived EBV-lymphoblastoid cell lines (LCL) transduced with the same vector in the presence of IL-15 and IL-2. UCB from donors of varied HLA types were selected. 40×10 6 UCB mononuclear cells (available in the 20% fraction of frozen UCB units) were thawed and used in the manufacturing process. After 3 rounds of stimulation, 9 CTL cultures contained a mean of 83% (range 64–94%) CD8+ve T-cells and 27% (range 12–40%) CD4+ve T-cells. Flow cytometric analysis of memory markers after 3 weeks expansion revealed a predominance of CD45RA− CD62L− T-cells (69±18%; range 25–93%) with a smaller population of CD45RA− CD62L+ T-cells (10±5%; range 1–23%). Evaluable UCB CTL lines showed specific cytolytic activity in 51 Cr release assays against targets loaded with CMV and Adv antigens. The observed cytotoxicity was specific because unloaded targets and MHC-mismatched targets were not killed. IFNγ ELISPOT assays on CTL lines demonstrated a mean of 209 (range 45–694) and 74 (range 0–188) spot forming cells/1×10 5 T-cells following incubation with CMV-pp65 and Adv-hexon/penton peptides respectively. No significant response to CMV-IE1 peptides was demonstrated. The expanded UCB CTL had a broad Vβ repertoire and were specific for multiple viral epitopes. In addition, the virus-specific T cells were shown to be expanded only from T-cells with a naive phenotype (CD45RA+/CCR7+). These results demonstrate that, despite the generally naive nature of UCB lymphocytes, bi-virus-specific responses can be expanded in vitro and could potentially be used clinically in UCBT patients who develop infectious complications prior to immune reconstitution.