Chemically Programmable and Switchable CAR-T Therapy

Abstract

Although macromolecules on cell surfaces are predominantly targeted and drugged with antibodies, they harbor pockets that are only accessible to small molecules and constitutes a rich subset of binding sites with immense potential diagnostic and therapeutic utility. Compared to antibodies, however, small molecules are disadvantaged by a less confined biodistribution, shorter circulatory half-life, and inability to communicate with the immune system. Presented herein is a method that endows small molecules with the ability to recruit and activate chimeric antigen receptor T cells (CAR-Ts). It is based on a CAR-T platform that uses a chemically programmed antibody fragment (cp-Fab) as on/off switch. In proof-of-concept studies, this cp-Fab/CAR-T system targeting folate binding proteins on the cell surface mediated potent and specific eradication of folate-receptor-expressing cancer cells in vitro and in vivo.

Keywords: CAR-Ts; antibodies; antitumor agents; cell-surface receptors; immunotherapy.

Figure 1.. Design and concept of cp-Fab/CAR-T system.

(a) Schematic illustration of the four tagged h38C2 Fabs with the reactive Lys (green) and the C- or N-terminally appended GCN4 peptide (NYHLENEVARLKKL; red). The linker sequence between the peptide and Fab is G4S (HCCT, LCCT, and LCNT) or LVGEAAAKEAAAKA (HCNT). WT h38C2 Fab without tag is shown on the right. (b) Schematic illustration of the CAR-T with scFv 52SR4 as anti-GCN4 peptide recognition domain fused to 4–1BB and CD3ζ cytoplasmic signalling domains. (c) The CAR-Ts only serve as functional effector cells in the presence of cp-Fab (right) but not unprogrammed Fab (middle) or no Fab (left), effectively rendering control of the CAR-T to a small molecule (light green) that targets a cancer cell surface receptor (light green). Chemical programming of WT h38C2 Fab includes the GCN4 peptide (red) synthetically fused to the small molecule (light green) via a modular PEG spacer (zigzag line).

Figure 2.. Structure of FOLR1-targeting compounds.

Structure of β-lactam-biotin-folate compound 1 and trifunctional β-lactam-GCN4-folate compounds 2, 3, 4, and 5.

Figure 3.. In vitro activity of cp-Fab/CAR-T.

Cytotoxicity of the FOLR1-targeting cp-Fab/CAR-T against IGROV-1 cells with titration of cp-Fabs based on tagged (a) or WT (b) h38C2 Fabs was tested at an E:T ratio of 10:1 and measured after 24 h incubation. Cytotoxicity mediated by the corresponding unprogrammed Fabs was also investigated. Shown are mean ± SD values from independent triplicates. The control experiments with untransduced T cells are shown in Supplementary Figure 4.

Figure 4.. CAR-T activation mediated by cp-Fabs.

The CAR-T was incubated with 20 nM of the indicated FOLR1-targeting cp-Fabs or the corresponding unprogrammed Fabs in the presence of IGROV-1 cells at an E:T ratio of 10:1 for 24 h. The percentage of activated T cells based on CD25 expression after incubation with cp-Fabs based on tagged (a) or WT (b) h38C2 Fabs was measured by flow cytometry. Cytokines released from the T cells in the presence of cp-Fabs based on tagged (c) or WT (d) h38C2 Fabs were measured by ELISA. Shown are mean ± SD values for independent triplicates. An unpaired two-tailed t-test was used to analyze significant differences (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

Figure 5.. In vivo activity of cp-Fab/CAR-T based on LCCT_1.

(a) Four cohorts of NSG mice (n = 5) were inoculated with 1 × 10⁶ IGROV-1/ffluc cells via i.p. injection. After 6 days, 2 × 10⁷ CAR-Ts and LCCT_1 (1 μg or 10 μg) or unprogrammed LCCT (10 μg) or vehicle alone (PBS) were administered by the same route. The mice received a total of one dose of CAR-T and 10 doses of LCCT_1 or controls daily. (b) Starting on day 6, all 20 mice were imaged (see Supplementary Figure 8) and their radiance was recorded (mean ± SD). Significant differences between cohorts treated with LCCT_1 and PBS were calculated using an unpaired two-tailed t-test (*, p < 0.05; **, p < 0.01; ***, p < 0.001). (c) Corresponding Kaplan-Meier survival curves with p-values using a log-rank (Mantel-Cox) test (**, p < 0.01). (d) The weight of all 20 mice was recorded on the indicated days (mean ± SD).

Figure 6.. In vivo activity of cp-Fab/CAR-T based on WT Fab_5.

Three cohorts of NSG mice (n = 5) were inoculated with 1 × 10⁶ IGROV-1/ffluc cells via i.p. injection. After 6 days, 2 × 10⁷ CAR-Ts and WT Fab_5 (10 μg), an equimolar amount of compound 6, or unprogrammed WT Fab (10 μg) were administered by the same route. The mice received a total of one dose of CAR-T and 10 doses of cp-Fab, compound, or Fab daily. (a) Bioluminescence images of all 15 mice were taken from day 6 (before treatment) to day 30 (after treatment) at the indicated time points. (b) Starting on day 6, all 15 mice were imaged and their radiance was recorded (mean ± SD). Significant differences between cohorts treated with WT Fab_5 and WT Fab or compound 6 and WT Fab were calculated using an unpaired two-tailed t-test (*, p < 0.05; **, p < 0.01; ***, p < 0.001). (c) Corresponding Kaplan-Meier survival curves with p-values using a log-rank (Mantel-Cox) test (*, p < 0.05; **, p < 0.01). (d) The weight of all 15 mice was recorded on the indicated days (mean ± SD).