Infusion of CD3/CD28 costimulated umbilical cord blood T cells at the time of single umbilical cord blood transplantation may enhance engraftment

Abstract

Limited cell numbers in umbilical cord blood (UCB) grafts present a major impediment to favorable outcomes in adult transplantation, largely related to delayed or failed engraftment. The advent of UCB transplantation (UCBT) using two grafts successfully circumvents this obstacle, despite the engraftment of only one unit. Preclinical models suggested that the addition of UCB T cells at the time of transplant can enhance engraftment. We tested whether ex vivo activation by CD3/CD28 costimulation and expansion of T cells from a single UCB graft would be safe and feasible in adults with advanced hematologic malignancies, with an overall objective of optimizing engraftment in single unit UCBT. In this phase 1 study, recipients of single UCB units were eligible if the unit was stored in two adequate fractions. Dose limiting toxicity was defined as grade 3 or grade 4 GVHD within 90 days of UCBT. Four patients underwent UCBT; all were treated at the first dose level (10(5) cells/kg). At the 10(5) cells/kg dose level two subjects experienced grade 3 intestinal GVHD, thus meeting stopping criteria. For three subjects, neutrophil engraftment was early (12, 17, and 20 days), while one subject experienced primary graft failure. We observed early donor T cell trafficking and found that expanded T cells produced supraphysiologic levels of cytokines relevant to engraftment and to lymphoid differentiation and function. Taken together, these preliminary data suggest rapid engraftment in recipients of a single UCBT combined with relatively low doses of activated T cells, though potentially complicated by severe GVHD.

Trial registration: ClinicalTrials.gov NCT00891592.

Conflict of interest statement

Conflict of interest: Nothing to report. Carl June and Bruce Levine. Can read: for work not related to manuscript, financial interest due to intellectual property and patents in the field of cell and gene therapy. Conflict of interest is managed in accordance with University of Pennsylvania policy and oversight.

© 2016 Wiley Periodicals, Inc.

Figures

Figure 1.. Early donor T cell trafficking (day +11) and hematopoietic recovery in a female recipient of UCBT and CD3/CD28 costimulated T cells

A: H&E stained sections of skin rash (20×) performed on day +11, demonstrating a sparse lymphocytic infiltrate in the superficial dermis and at the dermal-epidermal junction, associated with dyskeratotic keratinocytes. B: Fluorescence in situ hybridization (FISH) of skin (50×) detect XX (red-red) signals in stromal and epithelial cells and in some lymphocytes, of recipient (female) origin. A subset of lymphocytes contained an XY (red-green; dashed arrows) signal, compatible with male (donor) cells of origin in this sex-mismatched UCBT patient. C: H&E stained, 5× magnification and D: immunostain for CD3+; 40× magnification of bone marrow sections day +21 following UCBT, showing trilineage hematopoiesis with scattered T cells. E: H&E, 5× magnification and F: immunostain for CD79a; 20× magnification day +365 following UCBT highlights hematogones. Tandem complete white blood count 6.7 with a normal differential and absolute lymphocyte count of 2910 × 103/μL with normal absolute numbers of CD4+ and CD8+ cells, Hemoglobin 12.2 g/dL, Platelets 254 × 103/μL. Cytogenetic studies performed on bone marrow aspirate showed a normal 46,XY male (donor) karyotype in 30 metaphase cells. FISH studies were negative for anomalies of chromosomes 9 or 21, present in the ALL that preceded the diagnosis of therapy-related AML. Immunoglobulin heavy chain gene rearrangement by PCR showed a polyclonal distribution of peaks. Post-transplant chimerism studies detected 100% single donor DNA in peripheral blood (whole blood and CD3+ subset) and whole bone marrow on days +21, +28, and +365 following UCBT. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 2.. T cell polarization

A: Regulatory T cells are enriched in the T cell product and decline beyond day +10 following UCBT. B: Concentrations of cytokines were measured in supernatants from expanded cells (subjects 1, 3, and 5, top 3 rows; normalized per million cells) and in serum from UCBT recipients at pre-defined timepoints; baseline samples were drawn prior to day zero, pre-UCB, post-UCB, and post-aDLI were all drawn on day zero, prior to infusion of UCB MNCs, following UCB MNC infusion but prior to T cell infusion, and following T cell infusion respectively. C: Intracellular cytokine staining for IL-10 in CD4+ cells. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 3.. Antigen Responsiveness

A: Ex vivo alloreactivity generated by activated donor-derived UCB Tcells against autologous tumor lysate Autologous MNC (top row) or CD3/CD28 costimulated UCBT cells from the graft (bottom row) were incubated with PMA/Ionomycin (left) or with cell lysates prepared from banked tumor sample (bone marrow MNC at diagnosis) and PMA/Ionomycin (right). Degranulation was assessed by flow cytometry measurement of CD107a in CD8+ gated cells. B: Acquisition of EBV- and HOX-TLD specific cytotoxic Tcells. Batched serial samples from one subject (#2) who was EBV seropositive at baseline with an HLA-A2+ donor were analyzed for EBV-tetramer-and HOX-TLD-positive cells. HTLV-1-MHC tetramer served as a negative control.