Patients’ Peripheral Blood as a Possible Source of Donor-Derived NK Cells for Ex Vivo Expansion and Genetic Modification for Post-Transplant Cellular Immunotherapy In Cord Blood Recipients

Abstract 2094

Haploidentical natural killer (NK) cells can induce and consolidate remission in patients with high-risk acute myeloid leukemia (AML) (Rubnitz et al. J Clin Onc 24: 371, 2010). Recently, significantly reduced relapse rates were observed in AML patients who received killer immunoglobulin-like receptor ligand-mismatched cord blood, suggesting effective alloreactivity of cord blood-derived NK cells (Willemze et al. Leukemia 23: 492, 2009). Cord blood transplantation (CBT) is an effective alternative source for allogeneic hematopoietic cell transplantation in both children and adults. However, its therapeutic efficacy for malignant diseases is limited by the lack of available donor effector cells, such as cytotoxic T lymphocytes, lymphokine-activated killer cells, NK-like T cells and NK cells, for treatment of hematological relapse and posttransplant lymphoproliferative disorder and/or for scheduled posttransplant cellular immunotherapy against refractory diseases. We previously reported a method that induces NK cells to proliferate and reliably allows their genetic modification in healthy individuals and leukemia patients in remission receiving maintenance chemotherapy (Imai et al. Blood 106: 376, 2005). To explore the possibility of using patients’ peripheral blood as a source for posttransplant NK cell therapy, we used our method to expand donor-derived NK cells from peripheral blood of CBT recipients early after engraftment. We also examined whether NK cells can be rendered cytotoxic against original leukemia blasts by transferring an antigen-specific artificial immunoreceptor gene. This study was approved by an institutional ethical committee. Patients received CBT for consolidation of hematological malignancy (n=7), neuroblastoma (n=1) or resolution of refractory EBV-associated hemophagocytic syndrome (n=1) with myeloablative (n=7) or reduced intensity conditioning (RIC) regimens (n=2). The patients were enrolled in the study after engraftment and peripheral blood was obtained after appropriate written consent was obtained. A chimerism study using short tandem repeat assays showed complete donor chimerism in all patients except one who received RIC-CBT. The peripheral blood was obtained at a median of 92 days post-CBT (range: 46–303 days) and subjected to ex vivo activation and expansion using a previously described protocol with slight modifications. Briefly, peripheral blood was coincubated with modified K562 cells expressing membrane-bound IL-15 and 4-1BB ligand (K562-mb15-41BBL) in the presence of low-dose IL-2 (10 U/mL). Most patients were on maintenance immunosuppressive therapy with calcineurin inhibitors with (n=3) or without (n=6) systemic corticosteroids. After 7 days of culture, a median 11.0-fold expansion (range: 5.3–28.9-fold) was observed in all but one patient who had been administered chemotherapy with Mylotarg for relapsed AML a few days before the blood sampling. The expansion rate in the first week was less efficient in CBT recipients than in healthy individuals (>20-fold), probably because of the immunosuppressants administered. However, an additional 2-week culture in the presence of high-dose IL-2 (1000 U/mL) yielded a median 206-fold expansion (range: 101–1381-fold in 21 days). The expanded NK cells exhibited upregulation of activating receptors including NKG2D, NCRp30 and NCRp44, and vigorous cytotoxicity against K562 cells (86.8–97.7% at an E/T ratio of 1:1). The NK cells were susceptible to retroviral genetic modification with the MSCV-IRES-GFP vector (median GFP-positive cells, 52.7%, n=10). Finally, peripheral NK cells from patients with acute lymphoblastic leukemia were expanded and transduced with the chimeric immunoreceptor gene anti-CD19-BB-ζ. The donor-derived NK cells expressed large amounts of anti-CD19 chimeric receptors on their surface and killed original leukemia blasts that were highly resistant to NK cell lysis (e.g. anti-CD19 vs. non-signaling receptor: 69% vs. 0% at an E/T ratio of 1:1). These results suggest that, in CBT recipients, ex vivo expansion and genetic modification of donor-derived NK cells from the patients’ peripheral blood is feasible. Because peripheral blood can be easily and repeatedly obtained, the method described here will allow multiple scheduled infusions. This preliminary study may lead to a novel strategy for posttransplant donor-NK cell therapy in CBT recipients.