Genetic Engineering and Significant Ex-Vivo Expansion of Cord Blood Natural Killer Cells: Implications for Post-Transplant Adoptive Cellular Immunotherapy

Natural killer (NK) cells are a critical component of both the innate and adaptive human immune response (Caligiuri et al, Blood 2008). Tumor target cell recognition by NK cells is a highly regulated and complex set of processes which are controlled by the balance between inhibitory and activating signals through the binding of a variety of ligands on tumor target cells by several distinct subtypes of NK cell receptors (Bryceson YT et al, Immun Rev, 2006). The major limitations of the use of NK cells in adoptive tumor cellular immunotherapy include lack of tumor recognition and activation and/or limited numbers of viable and functionally active NK cells (Shereck/Cairo et al, Pediatr Blood Cancer, 2007). To circumvent these limitations, methods to expand and/or activate peripheral blood NK cells have been developed. Over the past decade cord blood (CB) has been increasingly utilized as an alternative to peripheral blood for allogeneic stem cell transplantation (Cairo et al, Blood, 1997). We recently reported the successful expansion and functional activation of CB NK cells by ex-vivo cellular engineering with a cocktail of antibody and cytokines (Ayello/Cairo et al, BBMT, 2006). In addition, our group has had a major interest in the diagnosis, treatment and biology of childhood CD20+ B-NHL; and have identified subgroups of patients with a significantly poorer prognosis despite aggressive multiagent chemotherapy (Cairo et al, Blood, 2007). In this study we sought to to develop an adoptive cellular immunotherapy strategy to overcome chemotherapy drug resistant childhood B-NHL. Freshly isolated CB mononuclear cells (CBMC) were cultured with modified K562 cells expressing membrane bound IL15 and 4-1BB ligand (K562-mbIL15-41BBL; Imai et al, Blood, 2005). After irradiation with 100Gy, K562- mbIL15-41BBL cells were incubated in a 1:1 ratio with CBMC + 10 IU/mL rhIL-2 for 7–14 days. CD3 and CD56 expression was determined by flow cytometry at Days 0, 7 and 14. On Day 0, CBMC included a population of NK cells expressing CD56 of 3.9% ± 1.3% and CD3+ T cells of 48.3% ± 3.9%. After 7 days of culture with K562-mbIL15- 41BBL cells the percentage of CD56+/CD3− NK cells increased to 71.7% ± 3.9%, as compared to 9.7% ± 2.4% in cultures with media alone and 42.6% ± 5.9% in cultures with wild-type K562 cells (p<0.01). There was also a significant decrease in the percentage of T cells in cultures with the modified K562 cells compared to wild-type K562 and media alone (15.2% ± 2.2% vs 35.4% ± 4.4% vs 51.2% ± 7.1%, p<0.001). Overall, the percent of NK cells after 7 days of culture with K562-mbIL15-41BBL was 3374% ± 385% of the input cell number, i.e. an approximate 35-fold increase. This is significantly increased compared to culture with wild-type K562 (1771% ± 300%, p<0.05). On Day 14, there remained a significant difference in NK cell populations between CBMC incubated with modified K562 cells compared to wild-type K562 cells (62.0% ± 2.1% vs 27.9% ± 2.4%, p<0.001), and compared to media alone (5.5% ± 0.4%, p<0.001) but no further increase from Day 7. Expansion of NK cells using genetically modified K562 cells as a stimulus produced significantly higher numbers of NK cells than those previously observed using a cocktail of antibody and cytokines as a stimulus (Ayello/Cairo et al, BBMT, 2006): (71.7% ± 3.9% NK cells on Day 7 with modified K562 vs 33.9% ± 8.7% with AB/CY, p=0.0004). In summary, we have demonstrated CBMC can be stimulated by K562 cells expressing membrane bound IL15 and 4-1BB ligand (K562-mbIL15-41BBL) resulting in specific expansion of CB NK cells similar or higher than the expansion that can be obtained with peripheral blood. The method described here provides a means to promote CB NK-mediated cellular cytotoxicity for use in the post-transplant setting while minimizing the risk of graft-versus-host disease.