Fig 1.
Evaluation of ROR1 expression on tumor cells, pancreatic cancer, and healthy pancreas.
(a) Flow cytometry plots of samples stained with 4A5 mouse-anti-human ROR1 mAb (filled histograms) or matched isotype control (open histograms). Cell types tested were: EL4 murine T-cell lymphoma cell line, EL4 cells genetically modified to express ROR1 (EL4-ROR1), PBMC from patient with CLL diagnosis, non-t(1;19) B-ALL cell line NALM-6, t(1;19) B-ALL cell line Kasumi-2, and ovarian cancer cell lines A2780, EFO27, and OC314. (b) Fresh frozen sections of malignant and normal pancreas were stained with 4A5 mAb or isotype mAb, and developed using an anti-mouse secondary antibody conjugated with horse radish peroxidase. Tissues were visualized with diaminobenzidine and counterstained with hematoxylin. Slides were evaluated by the study pathologist to identify the tissue or cell-type stained and intensity of staining.
Fig 2.
Sustained numeric expansion of ROR1-specific CAR+ T cells upon clone#1 AaPC and cytokines.
(a) DNA plasmid vector maps for ROR1RCD28/pSBSO-SIM and ROR1RCD137/pSBSO-FRA. Abbreviations are IR/DR: Sleeping Beauty Inverted Repeat/Direct Repeat, phElF1α: Human Elongation Factor-1-α region hybrid promoter, ROR1RCD28: Human codon-optimized ROR1-specific scFv:Fc:CD28:CD3ζ CAR, ROR1RCD137: Human codon-optimized ROR1-specific scFv:Fc:CD137:CD3ζ CAR, SIM: “SIM” PCR tracking oligonucleotides, FRA: “FRA” PCR tracking oligonucleotides, BGHpA; bovine growth hormone polyadenylation sequence, Ori: minimal E.coli origin of replication, KanR: Bacterial selection gene encoding Kanamycin resistance, pKan: prokaryotic Kanamycin promoter. Digestion with BsrGI enzyme can distinguish the two plasmids, which have high degrees of similarity, and with PmlI enzyme can distinguish them from CD19RCD28 plasmid, which does not have PmlI site. The entire plasmid sequences were verified by Sanger-based sequencing techniques. (b) Parental K-562 (open histograms) and clone#1 AaPC (shaded histograms) were stained for CD19, CD86, CD137L, ROR1, and IL15 (membrane-bound IL-15; mIL15) where isotype (IgG) was used a negative control. (c) Expression CAR in ROR1RCD28 (middle) and ROR1RCD137 (right) T cells the day following electroporation (top panels) and after 28 days of co-culture on clone#1 AaPC (bottom panels) where “no DNA” T cells (left) were used as negative controls. T cells were marked by CD3 staining and CAR+ cells were detected with Fc-specific antibody. Quadrant frequencies are displayed in upper right corners. (d) Proliferation kinetics of total cells (circles) and CAR+ T cells (squares) on clone#1 AaPC over 28 days of co-culture. ROR1RCD28 displayed on the left and ROR1RCD137 shown on the right. Arrows represent addition of γ-irradiated clone#1 AaPC. Data are representative of 3 donors expanded in 3 independent experiments. Expansion data can be found in S1 Dataset, Fig 2 tab.
Fig 3.
Lymphocyte transcriptional profile and memory markers expressed on CAR+ T-cell surface.
After 29 days of expansion on clone#1 AaPC/IL-2/IL-21, ROR1RCD28 and ROR1RCD137 cells were (i) lysed for mRNA expression analysis using bar-codes or (ii) phenotyped for T-cell surface markers by flow cytometry. (a) RNA lysates were profiled for expression of a selected group of lymphocyte genes with non-enzymatic digital multiplex array of mRNA transcripts (NanoString) where transcription factors are shown on the left, genes associated with survival, co-stimulation, and trafficking are shown on the middle, and genes associated with effector function are shown on the right. Dashed line represents the limit-of-detection calculated by mean + 2xSD of negative controls. (b) Flow cytometry of ROR1RCD28 and ROR1RCD137 T cells showing co-staining for CD3 and CD56, CD4 and CD8, CD28 and CD27, CD62L and CCR7, CD45RO and CD62L, or CD95 and CD57 in cells gated for CAR expression based staining with Fc-specific antibody. ROR1RCD28+ T cells are shown in the top panels and ROR1RCD137 T cells are displayed in the bottom panels. One of 3 representative donors is displayed and quadrant frequencies are shown in the upper right corners. (c) Cumulative frequencies of cells staining positive for each memory marker within an extended memory phenotype panel. (d) Multi-parameter flow cytometry was used to determine frequencies of cells staining positive for combinations of CD45RA, CD27, CD28, and CCR7. For (c) and (d) ROR1RCD28 are in open shapes/bars and ROR1RCD137 are in closed shapes/bars and lines displayed in (c) are means (n = 3) and in (d) are mean ± SD (n = 3). Student’s two-tailed t-tests were used for statistical analysis between the two groups. *p<0.05 Expression and phenotype data can be found in S1 Dataset, Fig 3 tab.
Fig 4.
IFNγ production by ROR1-specific CAR+ T cells in response to ROR1+ targets.
At day 29 of co-culture with AaPC/IL2/IL-21, CAR+ T cells were co-cultured for 6 hours at 37°C with tumor targets or non-specific mitogenic stimuli (PMA/Iono) then analyzed for expression of IFNγ in CAR+ T cells (gated based on Fc expression). Brefeldin-A (GolgiPlug) was added to T cells to block IFNγ secretion. (a) Representative flow cytometry plots where ROR1RCD28 cultures are on the top and ROR1RCD137 cultures are on the bottom. Percentage of max (y-axes) normalized total cell numbers in each sample for consistency between samples so that IFNγ fluorescent intensity (x-axes) could be compared between conditions. (b) Cumulative mean fluorescence intensities (MFI) of IFNγ staining from co-cultures where mean ± SD (n = 3 donors) is shown. Student’s paired, 1-tailed t-test for statistical analysis between target co-culture and T cells alone (Mock; media only). *p<0.05 and **p<0.01 IFNγ data can be found in S1 Dataset, Fig 4 tab.
Fig 5.
Specific cytolysis of ROR1+ tumor cells by CAR+ T cells.
Four-hour chromium release assay was used to assess specific lysis by T cells at decreasing effector to target (E:T) ratios. (a) No DNA (CARneg), CD19RCD28+, CD19RCD137+, ROR1RCD28+, or ROR1RCD137+ T cells generated from healthy donor PBMC were challenged with EL4-ROR1+ cell line (closed squares), EL4 parental (ROR1neg) cell line (open circles), Kasumi-2 cell line (closed inverted triangles), or NALM-6 cell line (open triangles). (b) CAR+ T cells generated from healthy donor PBMC were challenged with allogeneic CLL cells (closed inverted triangles), healthy donor allogeneic B-cell LCL (open triangles). (c) No DNA (CARneg), ROR1RCD28+, or ROR1RCD137+ T cells generated from CLL patient PBMC were challenged with EL4-ROR1+ cell line (closed squares), EL4 parental (ROR1neg) cell line (open circles), autologous B cells (closed inverted triangles), or autologous T cells (open triangles). Data are mean ± SD of triplicate measurements in CRA and are representative of four donors from four independent experiments. Lysis data can be found in S1 Dataset, Fig 5 tab.
Fig 6.
In vivo tumor clearance by ROR1-specific CAR+ T cells.
NSG mice were intravenously (i.v.) injected with Kasumi-2-ffLuc-mKate cells and were treated with three i.v. doses of T cells to assess the ability of ROR1-specific T cells to manage disease. High-dose IL-2 was added intraperitoneally the day of injection and the following day. (a) Non-invasive BLI reported as flux was serially measured in untreated (black circles), ROR1RCD28+ T cell-treated (blue squares), and ROR1RCD137+ T cell-treated (red triangles) mice. Data are mean ± SEM (n = 5). Two-way ANOVA was used for statistical analysis. (b) Representative BLI images at day +23 post-tumor cell engraftment. (c) Overall survival of mice. *p<0.05, **p<0.01, ***p<0.001 BLI and survival data can be found in Supporting Information, Fig 6 tab.