Article Figures & Data
Figures
- Figure 1.
ICV-infused CAR T cells eradicate CNS lymphoma more efficiently than IV-infused CAR T cells. A, In our models of CNS lymphoma, NSG mice were injected intracranially (IC) with either EGFP+ffluc+ Daudi lymphoma cells or Jeko MCL cells. Following tumor engraftment, Tn/mem-derived CD19- or BAFF-R-CAR T cells were administered IV or ICV. For the CD19-CAR T-cell model, 3 × 106 cells and 1.3 × 106 cells were administered IV and ICV, respectively. For the BAFF-R-CAR T-cell model, equal numbers of cells (1 × 106) were administered for both IV and ICV conditions. B, For the CD19-CAR T-cell model, tumor burden, as measured in flux (photons/sec) by BLI, was evaluated once weekly. N = 5 mice per group in one experiment. Linear mixed models were used to compare logarithm-transformed flux of mice that received ICV versus IV CAR T cells over time; **, P < 0.01. C, For the BAFF-R-CAR T-cell model, Flux (photons/sec) was determined by measuring bioluminescence once per week. N = 5 mice per group in one experiment. Linear mixed models were used to compare logarithm transformed flux of mice that received ICV versus IV CAR T cells over time; **, P < 0.01. Survival of mice treated with ICV- versus IV-delivered BAFF-R-CAR T cells was analyzed by the log-rank test; **, P < 0.01. D and E, Blood was collected 105 days after CD19-CAR T-cell treatment and analyzed for expression of human CD45, CD8, CD4, and CAR (Erbitux) by flow cytometry. Data from a representative mouse for ICV- and IV-delivered CD19-CAR T cells are shown in D. Percentages (mean ±SD) of human T cells (hCD45+GFP−) and CD4+ and CD8+ cells for all mice are quantified in E. Gating was set based on isotype-matched monoclonal antibodies or streptavidin. Mean ± SDs from multiple mice/measurements are depicted.
- Figure 2.
CD19-CAR T cells administered by ICV infusion can completely eradicate both CNS and systemic lymphoma. A, NSG mice were injected SC with 3 × 106 EGFP+ ffluc+ Daudi lymphoma cells and 14 days later injected intracranially (IC) with 0.2 × 106 EGFP+ ffluc+ Daudi lymphoma cells. Five days following IC tumor inoculation, 2 × 106 mock-transduced T cells or CD19-CAR T cells were administered either ICV or IV. B, Flux (photons/sec) was determined by measuring bioluminescence once weekly. N = 5 mice per group. Linear mixed models were used to compare logarithm-transformed flux of ICV- and IV-treated mice over time; **, P < 0.01 and ***, P < 0.001. Survival of mice was analyzed by the log-rank test; **, P < 0.01. C–E, At the time of euthanasia (day 334 for the ICV group and between days 79 and 180 for the IV group), blood was collected and analyzed for expression of human CD45, CD8, CD4, CAR (EGFR), and central memory receptors by flow cytometry. Representative data from two separate experiments are presented. Percentages (mean ±SD) of human T cells in the blood (D) and T cells expressing CD62L, CD127, CD28, and CAR+ CD4 and CD8 T-cell subsets (E) are presented. ICV and IV were compared by the Student t test. ns, not significant; **, P < 0.01. F, Human T cells harvested from mouse spleens 334 days after ICV and 180 days after IV CAR T-cell infusion were restimulated with REM method containing OKT3, irradiated PBMC, and LCL for 14 days, and their T-cell receptor (TCR) repertoire was analyzed. Percentage of TCR Vβ expression in CD3-positive cells is depicted.
- Figure 3.
Antigen-primed CAR T cells possess better effector memory function and can resist tumor rechallenge. NSG mice were inoculated with SC injections of 3 × 106 Daudi lymphoma cells in the right flank 19 days prior to CAR T-cell treatment. On the day of CAR T-cell infusion, 2 × 106 CD19-CAR T cells were primed with irradiated Daudi cells at 10:1 (CAR:tumor) ratio for 1 hour prior to infusion by ICV (A) or IV (B) delivery. Mock T cells were infused into control mice. Flux (photons/sec) was determined by measuring bioluminescence by live imaging once per week. Note that 135 days after CAR T-cell infusion, surviving mice were rechallenged (dotted lines) with Daudi lymphoma cells by SC injection. Naïve mice were inoculated with Daudi lymphoma cells at rechallenge as controls. Mean ± SDs from 5 mice per group are presented. Linear mixed models were used to compare flux (logarithm transformed) among different groups over time, and log-rank test was used to compare survival functions among the groups. If a mouse died during imaging processes with no prior sign of tumor progression, then it was considered an anesthesia-related death, and the mouse was excluded from survival analysis. *, P < 0.05; **, P < 0.01; and ***, P < 0.001. C, Blood was collected retro-orbitally 133 days after CAR T-cell treatment and analyzed for expression of human CD45, CD3, and CAR (EGFR) by flow cytometry. No mice survived in the unprimed CD19-CAR IV group. Significance was determined by one-way ANOVA test. ***, P < 0.001.
- Figure 4.
ICV-infused CAR T cells exhibit similar trafficking but superior proliferation and persistence potential when compared with IV-delivered CAR T cells. A, CD19-CAR T cells were transduced with EGFP-ffluc and expanded in vitro. NSG mice were implanted SC with Daudi lymphoma in the right flank. Nineteen days after tumor engraftment, 2 × 106 EGFP+ffluc+ CAR T cells were administered ICV or IV, and CAR T-cell proliferation was determined by measuring bioluminescence by live imaging every other day (B–C). The same scale was used for each time point. D–E, Blood was collected at different time points after CAR T-cell infusion, and T-cell (human CD45+) and CAR+ T-cell (EGFR+) levels were detected by flow cytometry. Mean ±SDs from 5 mice per group are presented. Significance was determined with the Mann–Whitney test; *, P < 0.05 and ***, P < 0.001. Experiments were repeated twice, and representative data are presented.
- Figure 5.
CSF conditions CAR T cells for enhanced memory and reduced exhaustion. A, CD19-CAR T cells were cultured in RPMI or human CSF for 48 hours and subjected to RT-PCR array analysis. Fold changes of genes in CSF over RPMI of >1.3 are shown. N = 4 donors. B, CAR T cells cultured in CSF or RPMI for 48 hours were stimulated overnight with Daudi lymphoma cells. Cytokines released into the supernatant were measured with cytometric bead array using the Bio-Plex Human Cytokine 17-Plex Panel. Mean ± SDs from 6 replicates are presented. CSF and RPMI (logarithm transformation) were compared by the Student t test; ns, not significant; **, P < 0.01; and ***, P < 0.001. C, CAR T cells were activated, transduced, and expanded for 13 days in modified RPMI (60 mg/dL glucose; 2.8 mEq/L potassium) or in regular RPMI (200 mg/dL glucose; 5.3 mEq/L potassium). CAR T cells cultured in modified RPMI were subsequently cultured in regular RPMI (switched) for 14 days. Percentages (mean ± SD) of CD28-positive cells in CAR-gated populations from N = 4 different human donors are presented. Linear mixed models were used to compare the three conditions based on repeated measures from the same donor; *, P < 0.05; **, P < 0.01; and ***, P < 0.001. D, CAR T cells were cultured in modified or regular RPMI, stimulated with Daudi cells for 6 hours, and assessed for degranulation. Percentages (mean ± SD) of CD107+ cells in CAR-gated populations from N = 4 different human donors are presented.
- Figure 6.
ICV- and IV-delivered CAR T cells that persist in vivo have distinct gene signatures. T cells were harvested from the bone marrow of mice 68 days after ICV or IV CAR T-cell infusion, and the samples from 1 mouse in each group were subjected to single-cell RNA-seq (scRNA-seq) analysis. The subsequent analysis of human gene data was performed using “Seurat” package v3.0 and R scripts. t-SNE visualization plot of scRNA-seq data (A), and dot plots and feature plots in the identified cell clusters (B) were generated for specific T-cell populations. ICV-delivered cells were enriched in cluster 3, whereas IV-delivered cells were enriched in cluster 1. C, Cluster-specific markers were identified to generate a heatmap.
- Figure 7.
ICV-administered CD19-CAR T cells exhibit memory phenotype and function. A, T cells harvested from two representative mouse spleens (334 days after infusion for ICV mice and 180 days after infusion for IV mice) were stimulated with OKT3 and feeder cells, including PBMCs and LCL cells, for 2 cycles (14 days/cycle). B, Percentages of CAR T cells (CD3+EGFR+) and fold change following first and second REM stimulation are presented. C, CAR T cells were analyzed for expression of CD161, CD127, and CD62L following the second REM stimulation. D, Uptake of glucose analogue NBDG after 30 minutes was analyzed by flow cytometry. E, CD62L+ and CD62L− cell populations were assessed for the ability to uptake NBDG over time. Percentages (mean ± SD) of NBDG+ cells are presented. Linear mixed models were used to compare CD62+ and CD62L over time; **, P < 0.01. F, Mitochondria of T cells following the second REM stimulation were stained with 100 nmol/L MitoTracker (green) and imaged using an LSM880 confocal microscope. Images were taken at 63×; scale bars, 5 μm.
Additional Files
Supplementary Data
- Supplementary Figures - Supplementary Figures and Legends
Volume 9, Issue 1
