Document resource
Background Chimeric antigen receptor (CAR) T cells induce response rates of up to 90% in lymphoid malignancies but have had limited efficacy in solid tumors. Our group and others have shown that independent modulation of positive or negative regulators of T cell function can improve the antitumor activity of CAR T cells against solid tumors. We hypothesize that simultaneous overexpression of T cell survival-related genes and downregulation of T cell dysfunction-associated genes, using cell state-specific regulatory switches, will improve CAR T cell persistence and anti-tumor effect in solid tumors.Methods We developed Synthetic Gene Expression Regulatory Switches (SynGERS) for tumor antigen-specific and cell state dependent bi-directional regulation of genes in CAR T cells. SynGERS contain a cell-state specific sensor unit acting on a minimal promoter that drives overexpression of genes associated with T cell survival (enhancer unit), and an RNA interference-based unit to repress T cell dysfunction-related genes (repressor unit). We developed a novel activation-specific sensor responding to nuclear receptor 4A2 (NR4A2) that outperformed previously described Nuclear Factor of Activated T cells (NFAT) sensor unit by 10-fold ( figure 1 a-c). Using this NR4A2-inducible promoter, we generated a library of 50 SynGERS and controls and tested them in a bicistronic vector constitutively co-expressing our clinically validated glypican-3 (GPC3)-specific CAR. We targeted GPC3 as it is expressed in several pediatric solid tumors but not in non-malignant tissues. We screened the library of SynGERS-CAR in vitro for optimal combinations of positive and negative regulator genes resulting in superior expansion, persistence and killing function in repeat tumor challenge assays against GPC3+ tumor cells. The top performing SynGERS were tested in vivo against GPC3+ cells in NSG MHC I/II DKO mice.Results Compared to T cells expressing the GPC3-CAR alone, eight SynGERS had increased cytotoxicity and/or proliferation in vitro. The top four performing SynGERS (eIL15-rTOX, eIL15-rBTG1, eIL15-rRASA2, and eIL15-rREGNASE1) had up to 1.43 fold increased tumor killing capacity and had up to 2.75 fold superior expansion at round four of the repeat antigen challenge with GPC3+ tumor cells (figure 1d). The top performing SynGERS also showed lower PD-1 expression than the CAR-alone control at the end of the repeat tumor challenge (figure 1e). In particular, the eIL15-rTOX SynGERS CAR T cells showed enhanced survival and anti-tumor effect than the CAR-alone control (figure 2).Conclusions SynGERS have the potential to induce potent anti-tumor function by CART cells with improved persistence and survival in solid tumors.Abstract 207 Figure 1Development of SynGERS. a) Schematic of SynGERS: The expression cassette consists of a 5’ sensor cell state specific sensor, acting on a minimal promoter (minP) followed by the enhancer (e) and repressor (r) units. An independent promoter (SFFV) drives the GPC3-CAR (GBBz, GPC3-specific second-generation CAR incorporating the 41BB endodomain). The enhancer unit (e) incorporates the Q8 tag for induction detection, which is translated into a separate peptide with the use of a T2A sequence. b) Activation dependent induction and repression of target genes. T cells were transduced with indicated SynGERS and pre- and post- activation gene expression induction (eGFP) and repression (MHC Class I) was measured by FACS. One representative of three independent experiments. c) Generation of NR4a2 sensors. Transduced cells were stimulated via TCR and FACS measured induction of GFP expression. Combined results from three independent experiments. d) screening of inducible SynGERS-CAR T cell library in repeat killing assay against Huh7-eGFP cells. Tumor cells were seeded one day before the co-culture in T cell medium. Next day, CAR T cells were added at a 1:1 effector to target ratio.% of killing was calculated as: 100* (mean eGFP index sample/maximum mean eGFP index). Once killing was complete, CAR T cells were analyzed and added to fresh tumor cells at the same ratio. CAR was detected by flow cytometry using AF647-labeled anti- mouse Fab2 antibody (Jackson). eGFP expression was detected with the IncuCyte system (Sortarious). PD-1 expression at round 3 of repeat killing assay against HepG2-eGFP cells. Tumor cells were seeded a day prior to the co-culture in T cell medium. The assay was performed following the same protocol described in d. Data are presented as mean of 2 donorsAbstract 207 Figure 2Functionality of SynGERS-CAR T cells against HepG2-eGFP-Luciferase cells in vivo. A-C, NSG MHC I/II DK mice received a single i.v. Dose of 5x106 SynGERS-CAR T cells on day 7 post i.p. Injection of 1x10 HepG2-eGFP-Luciferase cells.; non-transduced cells (NT), eQ8-rScramble, and elL15-rScramble served as controls (N= 4 or 5 per group as indicaded in the image). A, Serial bioluminescence images. B, Quantification of bioluminescence. C, Kaplan-Meier survival