CAR-Aptamers Enable Traceless Enrichment and Monitoring of CAR-Positive Cells and Overcome Tumor Immune Escape

Abstract

Chimeric antigen receptor (CAR)-positive cell therapy, specifically with anti-CD19 CAR-T (CAR19-T) cells, achieves a high complete response during tumor treatment for hematological malignancies. Large-scale production and application of CAR-T therapy can be achieved by developing efficient and low-cost enrichment methods for CAR-T cells, expansion monitoring in vivo, and overcoming tumor escape. Here, novel CAR-specific binding aptamers (CAR-ap) to traceless sort CAR-positive cells and obtain a high positive rate of CAR19-T cells is identified. Additionally, CAR-ap-enriched CAR19-T cells exhibit similar antitumor capacity as CAR-ab (anti-CAR antibody)-enriched CAR-T cells. Moreover, CAR-ap accurately monitors the expansion of CAR19-T cells in vivo and predicts the prognosis of CAR-T treatment. Essentially, a novel class of stable CAR-ap-based bispecific circular aptamers (CAR-bc-ap) is constructed by linking CAR-ap with a tumor surface antigen (TSA): protein tyrosine kinase 7 (PTK7) binding aptamer Sgc8. These CAR-bc-aps significantly enhance antitumor cytotoxicity with a loss of target antigens by retargeting CAR-T cells to the tumor in vitro and in vivo. Overall, novel CAR-aptamers are screened for traceless enrichment, monitoring of CAR-positive cells, and overcoming tumor cell immune escape. This provides a low-cost and high-throughput approach for CAR-positive cell-based immunotherapy.

Keywords: CAR-T cells; aptamers; monitor; retargeting; traceless enrichment.

Figure 1

Screening of CAR‐aptamers by SELEX. a) FMC63‐Fc and Fc protein expression by HEK‐293T cells were confirmed by SDS‐PAGE stained with coomassie brilliant blue. b) The binding performance of aptamers, 7s, 9s, 10s, 13s, 14s, and 18s to Jurkat cells and CAR19‐Jurkat cells. c) Fluorescence confocal microscopy of CAR19‐Jurkat cells stained with Hoechst (blue) and cocultured with ctrl‐ap, CAR‐ap9s, and CAR‐ap18s (red). Scale bar, 20 µm. d) Flow cytometry binding curve of CAR‐ap9s to CAR19‐Jurkat cells. K_d values were calculated by averaging the individual regression values of the independent experiments. e) Flow cytometry assay of CAR‐ab and CAR‐ap9s binding to human PBMC‐derived GFP⁺ CAR19‐T cells. Left, flow cytometry plots representing three replicates. Right, a statistical graph of the CAR‐positive rate, defined as the percentage of GFP⁺ T cells that were also positive for antibody or aptamer binding.

Scheme 1

a) The concept illustration of the CAR‐ap aptamer SELEX process. b) CAR‐ap aptamers used for traceless enrichment of CAR19‐T cells. c) CAR‐ap could specifically monitor CAR‐positive cells. d) CAR‐ap aptamers are used for retargeting CAR19‐T cells to CD19 antigen‐lost tumor cells.

Figure 2

Traceless sorting of CAR‐positive cells using a CAR‐ap‐based strategy. a) Schematic illustration of the traceless sorting of CAR‐positive Jurkat or T cells using CAR‐ap aptamers. Biotin‐labeled CAR‐ap preloaded onto Streptavidin‐labeled magnetic beads incubated with different cellular fractions. Unlabeled CAR‐negative cells that were not captured were removed using the washing fluid. Captured CAR‐positive cells were released by magnetic beads after adding the reversal agent to the eluting fluid. b) Flow cytometry assay to detect the CAR‐positive rate of the cell sample mixed by CAR19‐Jurkat cells and Jurkat cells before CAR‐ap sorting. c) Flow cytometry assay to detect the CAR‐positive rate of the cell sample in the washing fluid and eluting fluid after CAR‐ap sorting. d) Flow cytometry assay to detect the CAR‐positive rate of human PBMC‐derived CAR19‐T cell samples before and after CAR‐ap sorting. The plots represent three independent experiments within human PBMCs from three healthy donors. e) Fold change in the proportion of CAR‐positive cells by flow cytometry analysis before and after sorting.

Figure 3

Functional validation of CAR19‐T cells sorted with CAR‐ap9s in vitro and in vivo. a) In vitro antitumor cytotoxicity of CAR19‐T cells and CAR‐ap9s‐CAR19‐T cells to 697 cells at different E:T ratios. The CAR‐positive rate was adjusted to the same level before the experiment. b) Flow cytometry assay to detect CD25 and CD69 expression of CAR19‐T cells and CAR‐ap9s‐CAR19‐T cells during in vitro antitumor cytotoxicity. c) Flow cytometry assay to detect IFN‐𝛾, TNF‐ɑ, and IL‐2 cytokine expression of CAR19‐T cells and CAR‐ap9s‐CAR19‐T cells during in vitro antitumor cytotoxicity. d) Schematic illustration of the Nalm6 mouse model treated with CAR19‐T cells, CAR‐ap9s‐CAR19‐T cells, and CAR‐ab‐CAR19‐T cells. The CAR‐positive rate was adjusted to the same level in each group before the experiment. e) Flow cytometry assay to measure the tumor burden (the rate of Nalm6‐GFP cells) in bone marrow (BM), spleen (SP), and peripheral blood (PB) of mice in each group. Statistical analysis was performed using ANOVA.

Figure 4

Monitoring of CAR19‐T cell expansion using CAR‐ap9s in a Nalm6 mouse model. a) Schematic illustration of the detection of CAR19‐T cell expansion in a CAR19‐T‐treated leukemia mouse model. b, c) Flow cytometry assay of the CD4⁺ CAR⁺ T cell positive rate (b) and CD8⁺ CAR⁺ T cell positive rate (c) using CAR‐ap9s in the peripheral blood of mice on day 1, day 3, day 5, and day 7 after treatment with CAR19‐T cells in three groups with different treatment doses (Low dose, Middle dose, and High dose). d) Flow cytometry assay of the tumor cell positive rate in the bone marrow (BM) of mice on day 8 after treatment with CAR19‐T cells in the control group and three groups with different treatment doses. e) Kaplan−Meier survival curves of the control group and three groups with different treatment doses.

Figure 5

Construction and stability of the CAR‐ap‐based bispecific circular aptamer. a) Schematic illustrating the construction of bispecific circular aptamer CAR‐ap‐sgc8. b) Agarose gel electrophoresis of ssDNA CAR‐ap9s, ssDNA sgc8, and CAR‐ap9s‐sgc8. c) The stability of ssDNA CAR‐ap9s, ssDNA sgc8, and CAR‐ap9s‐sgc8 in FBS was determined by agarose gel electrophoresis at 37 °C for the different incubation times. d‐e) Flow cytometry assay to test the binding performance of CAR‐ap9s‐sgc8 to CEM cells and CD19‐KO BALL cells. f) Flow cytometry assay to measure the binding performance of CAR‐ab, ctrl‐ap, sgc8, CAR‐ap9s, and CAR‐ap9s‐sgc8 to CAR19‐T cells.

Figure 6

CAR‐ap9s‐sgc8 retargeted CAR19‐T cells to tumor cells with CD19 loss. a) Schematic illustration of CAR‐ap‐sgc8 mediated antitumor cytotoxicity of CAR19‐T cells to tumor cells with CD19 antigen loss. b) Fluorescence confocal microscopy of tumor cells stained with Hoechst (blue) co‐incubated with CFSE‐labeled CAR19‐T cells (green) in the presence of CAR‐apt9s(ap9s)‐sgc8 or ctrl‐ap. Left, CEM cells. Right, 697 cells. Scale bar, 20 µm. c) In vitro antitumor cytotoxicity of CAR19‐T cells to CEM, CD19‐KO‐697, and CD19‐KO‐Nalm6 cells at different E:T ratios in the presence or absence of CAR‐ap9s‐sgc8. d,e) Flow cytometry assay to detect active markers and cytokine expression of CAR19‐T cells during in vitro antitumor cytotoxicity in the presence of PBS, ssDNA CAR‐ap9s, ssDNA sgc8, and CAR‐ap9s‐sgc8. Statistical analysis was performed using ANOVA.

Figure 7

The potency of CAR‐ap9s‐sgc8‐mediated antitumor immunity of CAR19‐T cells in vivo. a) Schematic illustration of the therapeutic protocol used to investigate the function of CAR‐ap9s‐sgc8 mediating CAR‐T in CD19‐KO‐Nalm6 tumor xenografts. NSG mice (n = 4 per group) were inoculated in the flank with CD19‐KO‐Nalm6 tumor cells (1×10⁷) on day −14. After two weeks of tumor establishment, CAR19‐T cells (1×10⁷) were administered intravenously (i.v.) to tumor‐bearing NSG mice on day 0. Different groups of mice received different intratumoral (i.t.) treatments from day 1 to day 5 (50 µL of PBS for CTRL and CAR19‐T groups, bc‐ap9sRC‐sgc8RC (bc‐RC), or bc‐ap9s‐sgc8). b) Average tumor growth kinetics of mice receiving different treatments. Tumor volume (mm³) was monitored daily using the caliper method. Data are presented as means ±s.d. (n = 4). Statistical analyses were performed using two‐tailed paired Student's t‐tests (*p <0.05, **p<0.01, ***p <0.001, n.s., not significant). c) Ex vivo tumors in each group on day 12 after receiving various treatments as indicated. d) Body weight was monitored and recorded. Data are presented as means±s.d. (n = 4). e) Schematic illustration of the therapeutic protocol used to investigate the function of CAR‐ap9s‐sgc8 mediating CAR‐T in CCRF‐CEM tumor xenografts. NSG mice (n = 6 per group) were inoculated in the flank with CEM tumor cells (5×10⁶) on day −7. After allowing 1 week for tumor establishment, CAR19‐T cells (1×10⁷) were administered intravenously (i.v.) to tumor‐bearing NSG mice on days 0 and 3. Different groups of mice received different intratumoral (i.t.) treatments from day 1 to day 5 (50 µl of PBS for CTRL and CAR19‐T groups, bc‐RCap9s‐RCsgc8 (bc‐RC), or bc‐ap9s‐sgc8). f) Average tumor growth kinetics of mice under different treatments. Tumor volume (mm³) was monitored daily using the caliper method. Data are presented as means ±s.d. (n = 6). Statistical analyses were performed using a two‐tailed paired Student's t‐test (*p <0.05, **p<0.01, ***p <0.001, n.s., not significant). g) Ex vivo tumors in each group on day 11 after receiving indicated treatments. h) Body weight was monitored and recorded. Data are presented as means±s.d. (n = 6). i) H&E staining of tissue sections from CD19‐KO‐Nakm6 tumor xenografts after treatment. Scale bar: 100 µm.