T Cells Genetically Targeted to CD19 Eradicate B-ALL In a Novel Syngeneic Mouse Disease Model

Abstract 171

Adult precursor B cell acute lymphoblastic leukemia (B-ALL) is often incurable despite the induction of complete remissions, warranting the search for novel therapeutic modalities. Genetically targeted T cells have recently emerged as promising agents for directing targeted immune responses, as exemplified by the successful eradication of established CD19-positive tumors in immunodeficient mice treated with CD19-targeted human T cells (Brentjens et al, Nat Med, 2003). Based on these results, our program has developed a platform for the genetic modification of leukemia patients' T cells (Hollyman et al, J. Immunother, 2009) and opened two Phase I clinical trials to evaluate autologous, CD19-targeted T cells in patients with CLL (IRB# 06–138) and B-ALL (IRB# 09–114). The patients' T cells are retrovirally transduced with an anti-CD19 chimeric antigen receptor (CAR), termed 19–28z, which targets the T cells to normal and malignant B-lineage cells. To analyze and improve the function of the gene-modified T cells in adults with B-ALL, we have developed a mouse model of B-ALL in immunocompetent C57BL/6 mice. Starting from Eμ -myc mice, which express c-myc under the control of an immunoglobulin enhancer/promoter, we generated a spontaneous lymphoid tumor cell line, termed Eμ -ALL. Both its immunophenotype (CD5⁻CD19⁺CD43⁺HSA⁺BP1⁻IgM⁻) and immunoglobulin loci rearrangements are consistent with Eμ -ALL being derived from a precursor B cell. Intravenous injection of 1 × 10⁶ Eμ -ALL cells into adult C57BL/6 mice causes advanced morbidity in approximately 2 weeks, which is associated with bone marrow failure, as reflected by pancytopenia. In addition to the extensive leukemia involvement in the bone marrow there is also extensive infiltration in lymph nodes, spleen, and liver. We thus have a mouse tumor model that is molecularly, immunophenotypically, and clinically similar to precursor B-ALL. Significantly, the tumor model also mimics the poor survival despite inductions of remission in adults with B-ALL treated with chemotherapy. In our mouse model, treatment with intraperitoneal (IP) cyclophosphamide (100-200 mg/kg) prolongs survival by approximately two weeks, but mice nonetheless succumb to bone marrow failure from progressive leukemia. We have used this model to evaluate the function of anti-CD19 gene-modified T cells in syngeneic, immunocompetent mice. Having cloned a murine CD19-specific chimeric antigen receptor, termed m19-28z, we transduced C57BL/6 splenic T cells and infused them intravenously 24 hrs after treatment with IP cyclophosphamide (100-200 mg/kg) in Eμ -ALL-bearing mice. Cyclophosphamide alone is insufficient to induce long-term survival (n = 17). However, nearly all mice (n = 16) given cyclophosphamide and 10 × 10⁶ anti-CD19 gene-targeted T cells survived > 6 months and are still being monitored. Therefore, CD19-targeted T cells very effectively kill and/or contain the residual tumor cells remaining after cyclophosphamide. Furthermore, we detect long-term (> 6 months) B cell aplasia in some of the mice treated with anti-CD19 T cells, indicating that normal B-lineage cells are, as expected, eliminated by the adoptively transferred T cells. This in vivo tumor model of B-ALL will serve as an excellent system to evaluate the efficacy and toxicity of different anti-CD19 CARs, as well as different combination therapies.