A Phase I Clinical Trial of Treatment of B-Cell Malignancies with Autologous Anti-CD19-CAR-Transduced T Cells

Abstract 2865

T cells can be genetically modified to express chimeric antigen receptors (CARs) that specifically recognize the B-cell antigen CD19. Adoptive transfer of autologous T cells expressing anti-CD19 CARs is an attractive new approach for treating B-cell malignancies. We have constructed a CAR that consists of the variable regions of a mouse-anti-human-CD19 antibody coupled to the signaling domains of CD28 and CD3-zeta. We have treated 5 patients with 2 doses of 60 mg/kg of cyclophosphamide and 5 doses of 25 mg/m² of fludarabine followed by infusions of anti-CD19-CAR-transduced T cells and administration of high-dose IL-2. All of the patients received infusions of cells that produced cytokines in a CD19-specific manner. The percentage of the infused cells that expressed the anti-CD19 CAR as measured by flow cytometry ranged from 45% to 65%. The first patient enrolled on our trial has follicular lymphoma. He was treated twice. The patient obtained a partial remission (PR) from his first course of chemotherapy, 0.4×10⁹ anti-CD19-CAR-transduced T cells, and IL-2 (reported in Kochenderfer et al. Blood First Edition); however, he subsequently developed progressive disease, and 40 weeks after his first CAR-transduced T cell infusion he received a second course of chemotherapy followed by 2×10⁹ CAR-transduced T cells and IL-2. The second course of treatment resulted in an additional PR and was not associated with any toxicity that could be attributed to the CAR-transduced T cells. At last follow-up, a small amount residual disease detected only by positron emission tomography remained. In this first patient, the initial treatment course resulted in eradication of blood and bone marrow B-lineage cells for 39 weeks. In contrast to the prolonged eradication of B-lineage cells after the initial treatment course, the number of polyclonal blood B cells normalized 9 weeks after the second CAR-transduced T cell infusion. CAR-transduced T cells were present at a level of 0.1% of total peripheral blood mononuclear cells (PBMCs) one month after the first CAR-transduced T cell infusion. Despite the five-fold higher dose of CAR-transduced T cells administered with the second treatment, CAR-transduced T cells were not detected in the blood one month after the second CAR-transduced T cell infusion. The second patient treated on our protocol had follicular lymphoma and had received extensive prior therapy including autologous stem cell transplantation. After an initially uncomplicated course, this patient developed pneumonia caused by culture-proven influenza A virus and died 18 days after CAR-transduced T cell infusion. Quantitative PCR was used to measure the level of CAR-transduced cells in multiple tissues obtained from this patient at autopsy. CAR-transduced cells were widely distributed with the highest levels in the spleen and bone marrow. The third patient treated on our trial obtained a complete remission of advanced chronic lymphocytic leukemia (CLL) after treatment with chemotherapy, infusion of 2×10⁹ anti-CD19-CAR-transduced T cells, and IL-2. At the time of last follow-up, three months after treatment, adenopathy had resolved, CLL cells were not detected by flow cytometry analysis of the blood and bone marrow, and the number of normal polyclonal B cells in the blood was below normal levels. This patient had a period of fever and hypotension 7 days after cell infusion that was associated with an elevated serum interferon-gamma level of 1532 pg/mL. At the time of the hypotensive episode 7 days after cell infusion, anti-CD19-CAR-transduced cells made up 2.1% of PBMCs. The fourth patient treated on our study obtained a PR of splenic marginal zone lymphoma that continues 2 months after treatment with chemotherapy, 2×10⁹ CAR-transduced T cells, and IL-2. This patient did not have prolonged depletion of normal B cells after treatment, and he did not have any toxicity that could be attributed to the anti-CD19 CAR-transduced T cells. We recently treated a fifth patient who has CLL. Follow-up on this patient is too short to evaluate toxicity or response. In conclusion, we have shown that adoptive transfer of anti-CD19-CAR-transduced T cells with in vivo activity is feasible. The promising results obtained on this trial raise important questions for future research aimed at optimizing therapy with anti-CD19-CAR-transduced T cells.