Combining a CD20 chimeric antigen receptor and an inducible caspase 9 suicide switch to improve the efficacy and safety of T cell adoptive immunotherapy for lymphoma

Abstract

Modification of T cells with chimeric antigen receptors (CAR) has emerged as a promising treatment modality for human malignancies. Integration of co-stimulatory domains into CARs can augment the activation and function of genetically targeted T cells against tumors. However, the potential for insertional mutagenesis and toxicities due to the infused cells have made development of safe methods for removing transferred cells an important consideration. We have genetically modified human T cells with a lentiviral vector to express a CD20-CAR containing both CD28 and CD137 co-stimulatory domains, a "suicide gene" relying on inducible activation of caspase 9 (iC9), and a truncated CD19 selectable marker. Rapid expansion (2000 fold) of the transduced T cells was achieved in 28 days after stimulation with artificial antigen presenting cells. Transduced T cells exhibited effective CD20-specific cytotoxic activity in vitro and in a mouse xenograft tumor model. Activation of the iC9 suicide switch resulted in efficient removal of transduced T cells both in vitro and in vivo. Our work demonstrates the feasibility and promise of this approach for treating CD20(+) malignancies in a safe and more efficient manner. A phase I clinical trial using this approach in patients with relapsed indolent B-NHL is planned.

Figure 1. Schematic diagram of the iC9-CD20CAR-Δ19 lentiviral vector.

The CD20-CAR encompasses a scFv derived from the 1F5 antibody linked to a human IgG1 Fc (hinge region, CH2 and CH3), a CD28 transmembrane domain (CD28 TM), a CD28 co-stimulatory domain, a 4-1BB co-stimulatory domain, and a CD3ζ intracellular domain. SP163, a translational control element; iC9, inducible caspase 9-based suicide gene; Δ19, a truncated CD19 shortening of the CD19 intracellular domain to only 19 amino acids and removal of all tyrosine residues; T2A, Thoseaasigna virus 2A peptide sequence; E2A, equine rhinitis A virus 2A peptide sequence.

Figure 2. Expression of CD20-CAR.

(A) Flow cytometric examination of surface expression of the CD20-CAR and CD19. Mock, non-transduced T cells. The CAR was detected with a PE-conjugated mouse anti-human IgG (Fc-specific) antibody. CD19 expression was detected with an APC conjugated mouse anti-human CD19 antibody. Cells were gated on total live cells 8 days after transduction. The insert value in the upper right quadrant indicates the percentage of CD20-CAR⁺ and CD19⁺ T cells. (B) Western blot analysis of CD20-CAR expression. Whole-cell lysates from transduced T cells (CAR T), T cells transduced with empty vector (vector T), and non-transduced T cells (mock T) were hybridized with mouse anti-human CD3ζ antibody under reducing conditions. The 16-kD band corresponds to the endogenous CD3ζ chain; the 77 kD band corresponds to the CD20-CAR.

Figure 3. Expansion of transduced T cells using the NIH3T3-based AAPCs.

(A) Rapid expansion of transduced T cells. Eight days after activation with CD3/28 beads, transduced T cells were cultured in plates coated with irradiated NIH3T3-20/80 (broken line) cells or NIH3T3-IV cells (solid line) in the presence of IL-2 (20 U/ml) and IL-15 (10 ng/ml on day 1 and 1 ng/ml subsequently). Cell counts were enumerated every 7 days by trypan blue exclusion. Absolute numbers of cells at various time points are depicted as mean ± SD of triplicate wells. Results are representative of three independent experiments. (B) Preferential enrichment of transduced T cells when co-cultured with 3T3-derived AAPCs. Eight days after activation with CD3/28 beads, transduced T cells were cultured in plates coated with irradiated NIH3T3-20/80 cells or NIH3T3-IV cells. The percentage of CD19⁺ cells was determined every 7 days by flow cytometirc analysis using a PE-conjugated mouse anti-human CD19 antibody. Results are shown as the mean percentage of CD19⁺ cells ± SD of triplicate wells at various time points and are representative of three independent experiments. (C) Central Memory Phenotype of expanded T cells (CD45RO⁺CD28⁺CD62L⁺). Flow cytometric analysis of T cells expanded after 2 re-stimulation cycles using NIH3T3 AAPCs was performed to determine the surface immunophenotype. The percent positivity (mean ± SD of triplicate cultures) for CD28, CD62L, CD45RO, and CD45RA was determined by corresponding antibody staining. Cells were gated on CD3⁺CD19⁺ T cells. Similar results were obtained with 4 independent experiments using cells cultured after either 2 or 3 re-stimulation cycles. Results were concordant in 5 experiments.

Figure 4. CD20-specific effector activity of transduced T cells in vitro.

(A) Transduced T cells exhibit CD20-specific cytolytic activity in vitro. iC9-CD20CAR-Δ19 T cells transduced with CD20-CAR (CD20-CAR T), or non-transduced (mock) T cells were co-cultured with EL4, Daudi, and Granta in various effector to target (E:T) ratios. Mean percent killing (± SD of triplicate culture) was determined by 4-hour standard ⁵¹C release assay. (B) Anti-tumor effect of the transduced T cells. Mock or iC9-CD20CAR-Δ19 transduced T cells were co-cultured with CD20⁺ Ramos cells in various Effector to Target ratios (E:T) for 48 hours. Cells were then stained with antibodies recognizing CD22 and CD3. Flow cytometric analysis was used to determine the presence of Ramos cells (CD22⁺CD3⁻) and T cells (CD22⁻CD3⁺). Similar results were obtained with three independent experiments. (C) Cytokine production. Expanded T cells secreted IL2, IFN-γ and TNF-α after co-culture with CD20-expressing Ramos target cells for 48 hours (mean ± SD of triplicate samples). There were negligible levels of cytokine production when the expanded T cells were cultured alone with no target cells. Results are representative of three independent experiments.

Figure 5. Anti-tumor activity of transduced T cells in vivo.

(A) Schema of in vivo study demonstrating the anti-CD20 anti-tumor activity of transduced T cells in a disseminated human B cell malignancy xenogeneic NOD/SICD model. Raji-FFLuc tumor cells were injected i.v. into NOD/SCID mice on day 0. Two doses (5×10⁶ million cells per dose) of either iC9-CD20CAR- Δ19 transduced T cells or non-transduced T cells were injected i.v. on day 2 and day 9 respectively. Whole mouse luciferase activity was measured at various time points. (B) Eradication of CD20⁺ Raji tumors in NOD-SCID mice by CD20-CAR T cells but not non-transduced T cells (Mock T) as measured by in vivo bioluminescent imaging. (C) Summary of the bioluminescence signal as a measurement of tumor growth by days after tumor cell infusion. Data represent mean + SD. (D) Kaplan-Meier survival curves of mice receiving iC9-CD20CAR-Δ19 T treatment (CD20-CAR T) or non-transduced mock T cell treatment (NT).

Figure 6. Transduced T cells are eradicated upon activation of the iC9 suicide gene.

(A) Effect of CID on transduced T cell viability in vitro. Transduced or non-transduced (Mock) T cells were cultured in media alone (-CID) or media containing 20 nM AP1903, a clinical prodrug of CID (+CID) for 0 to 72 hours. Number of live cells was determined by trypan blue exclusion. Results are shown as mean ± SD from triplicate cultures for each condition. (B) Efficiency of iC9-mediated deletion is dependent on the activation status of the transduced T cells. Transduced T cells at the end of the first re-stimulation cycle using NIH3T3-20/80 were re-stimulated with either NIH3T3 cells, NIH3T3-20, NIH3T3-20/80, or NIH3T3-IV in the presence of low dose IL-2 (50 U/ml) for 60 hours. Either media (-CID) or AP1903 (20 nM, +CID) were added at 24 hours. Relative percentage of live transduced T cells was determined by flow cytometric analysis using a PE-conjugated mouse anti-human CD19 antibody. Results are shown as mean ± SD from triplicate samples for each condition.

Figure 7. Effective removal of transduced T cells in vivo.

(A) Study schema of in vivo iC9 mediated T Cell Deletion. NOD/SCID mice engrafted i.v. with 5×10⁵ Raji cells, were given two doses (1×10⁷ each dose) of iC9-CD20CAR-Δ19 T cells on Day 2 and Day7. Irradiated NSO-IL-15 cells (1.5×10⁷) were also administered i.p. every other day (QOD) to provide a source of human IL-15 in the mouse host. Two doses of PBS or AP20187 were given i.p. on days 13 and 14. Flow cytometric analysis was used to determine the percentage of transferred T cells (CD19⁺CD22⁻) in bone marrow, peripheral blood, spleen and liver. (B) Activation of the iC9 gene effectively removed transferred iC9-CD20CAR-Δ19 T cells in vivo. Results from the control group are displayed in closed columns (-CID). Results from the AP20187 treatment group are displayed in the open columns (+CID).