1. ¹Gene Transfer and Somatic Cell Engineering Laboratory, Memorial Sloan Kettering Cancer Center, New York, NY
2. ²Medicine, Memorial Sloan Kettering Cancer Center, New York, NY
3. ³Xcyte Therapies, Inc., Seattle, WA

Abstract

In order to conduct a clinical trial of adoptive therapy with genetically modified T cells in patients with CLL, we are developing an ex vivo transduction and expansion protocol to generate biologically active tumor-reactive T cells. T cells are genetically modified to express the CD19-targeted 19-28z chimeric antigen receptor (CAR). The 19-28z CAR is derived from a single chain fragment length murine antibody specific to the human CD19 antigen expressed on CLL tumor cells, fused to the transmembrane and cytoplasmic domains of the human CD28 co-stimulatory receptor and the cytoplasmic domain of the CD3|[zeta]| chain of the human T cell receptor. We have previously shown that PHA-stimulated T cells subsequently expanded by co-culture on artificial antigen presenting cells (AAPCs) efficiently lyse CD19+ tumor cells in vitro and eradicate established CD19+ tumor cells in vivo in a SCID-Beige mouse tumor model (Brentjens et al. Nat. Med. 2003). However, as PHA is not suitable for clinical use, we are now investigating the feasibility of T cell transduction and expansion using Xcyte|[trade]| Dynabeads|[reg]|. In four experiments, we find that transduction efficiencies of healthy donor-derived T cells initially activated with Xcyte Dynabeads range from 58% to 80% 19-28z+ T cells as measured by flow cytometry analysis. The degree of 19-28z+ T cell expansion ranges from 2 to 3 log over 14 days of culture. Significantly, the biologic activity of the T cells activated with Xcyte Dynabeads was comparable to that of T cells activated with PHA/AAPCs based on standard 51Cr release assays targeting CD19+ Raji tumor cells (70 to 90% killing at 25:1 E:T ratio). To determine whether 19-28z+ T cells activated with Xcyte Dynabeads have cytotoxic activity in vivo, SCID-Beige mice with disseminated Raji cell lymphoma were treated with a single intravenous injection of either 2|[times]|107 19-28z transduced T cells (n=10) or 2|[times]|107 control Pz1 T cells transduced with an irrelevant CAR (n=4). After treatment, mice were routinely checked for tumor progression as determined by hind-limb paralysis. Mice with hind-limb paralysis were sacrificed. As expected, control mice treated with Pz1 transduced T cells all developed hind-limb paralysis 4-5 weeks after tumor cell injection. In contrast, mice treated with 19-28z+ T cells activated with Xcyte Dynabeads had either delayed progression of disease (n=3) or remain disease free (n=7) at >60 days. Based on our earlier studies, this result suggest that 19-28z+ T cells activated with Xcyte Dynabeads are as potent as 19-28z+ T cells activated with PHA/AAPCs. In order to confirm this result, mice will be observed for a longer period of time. Additional experiments including direct comparisons with PHA/AAPCs activated T cells and expansion of T cells derived from patients with CLL will be presented. These preliminary results suggest that an Xcyte Dynabeads-based approach should be suitable to transduce and expand T cells for our planned clinical trial.