Effect of Rituximab On the Activity of T Cells Expressing CD20-Specific Chimeric Antigen Receptors

Adoptive cellular therapy using autologous T cells that have been genetically modified to express a chimeric antigen receptor (CAR) has emerged as a promising therapy for lymphoma. Clinical trials for lymphoid malignancies to date have primarily targeted either the CD19 or CD20 antigens. While CD20 has a more established track record as an immunotherapy target, one potential drawback of targeting CD20 with CAR⁺ T cells is the theoretical possibility that residual levels of circulating anti-CD20 antibodies (Ab) from prior chemoimmunotherapy regimens could partially or completely block CAR-antigen interactions. This could negatively impact the efficacy of CD20-targeted CAR⁺ T cells. However, previous data from our group and others indicate that CD20 CAR⁺T cell function is only partially blocked by anti-CD20 Ab, and T cell function in the setting of anti-CD3 × anti-CD20 bispecific Ab is not blocked by rituximab (R) levels of up to 100 μg/ml. Collectively, these data suggest that a very low number of available CD20 binding sites may be sufficient to trigger CAR signaling and T cell activation.

We tested the effect of different levels of R on in vitro function of polyclonal T cells from healthy donors negatively selected by MACS, activated with anti-CD3/CD28 beads, and transduced with epHIV7 lentiviral vectors encoding 1^st or 3^rdgeneration (αCD20-ζ or αCD20-CD28–41BB-ζ) anti-CD20 CARs. T cells were re-stimulated 1 week after initial activation by co-culture with antigen presenting cells (APCs) that had been pre-incubated for 30 minutes with varying concentrations of R (ranging from 0 to 800 μg/ml). APCs were K562 cells transduced to express CD80 with or without CD20 (denoted “K80” and “K80-20”), or Ramos lymphoma cells. Proliferation, cytokine secretion, and cytotoxicity were then assessed as discussed below.

We first used flow cytometry to test whether varying concentrations of R blocked binding of the Leu16 Ab and, as expected, found a dose-dependent blockade of CD20 on each cell line, with 50 μg/ml and 200 μg/ml causing near-complete blockade of K80-20 and Ramos cells, respectively. However, despite this apparent blockade, proliferation was largely unimpaired in CFSE-labeled 1^st or 3^rd generation CAR⁺ T cells cultured with K80-20 or Ramos cells pre-incubated with R concentrations of up to 400 μg/ml. We concurrently measured cytokine secretion of these T cells using Luminex assays and found that IL-2 and IFN-γ secretion decreased with increasing R levels, but 50–85% of baseline levels were still achieved at R concentrations of up to 100 μg/ml. Cytotoxicity against K80-20 and Ramos target cells in standard ⁵¹Cr-release assays by 1^st and 3^rd generation CAR⁺ T cells was largely preserved at low R concentrations, and 50–75% of cytolytic activity was retained at 100 μg/ml. Nonspecific proliferation, cytokine secretion, and cytotoxicity were excluded in these experiments by using CAR⁺ T cells incubated with K80 cells lacking CD20 expression, or T cells transduced with an empty vector as negative controls. Mouse xenograft experiments are currently ongoing to test the effect of serum R levels on the in vivo anti-tumor efficacy of CD20-CAR T cells.

These in vitro results suggest that despite apparent blockade of the CD20 antigen, CAR⁺ T cells targeting CD20 retain significant activity in the presence of R concentrations of up to 100 ug/ml. Patients receiving 2–3 cycles of R-chemotherapy have serum R trough levels in the range of 30–70 μg/ml. We therefore predict that residual serum R levels will not present a significant impediment to CD20-targeted adoptive T cell therapy given after salvage R-chemotherapy.

Jensen:ZetaRx: Equity Ownership, Patents & Royalties. Maloney:Roche: Consultancy; Genentech: Consultancy. Off Label Use: Lentiviral vector encoding a CD20-specific chimeric antigen receptor, used to re-direct autologous T cells to recognize B cell lymphoma cells.