LAUSR

T cells genetically engineered to express CARs specific for CD19 have shown breakthrough clinical successes in patients with B-cell lymphoid malignancies in recent years (Grupp SA et al., 2013; Kalos M et al., 2011; Maude SL et al., 2014). The technology holds great promise for other types of cancer, including solid tumors for which conventional cytoreductive therapies often fail (Kohler BA et al., 2015). However, the early clinical testing of CAR T cells against solid tumors has thus far benefited only a small fraction of patients (Ahmed N et al., 2015; Pule MA et al., 2008; Gilham DE et al., 2012), highlighting the need to explore novel antigens and to optimize antigen-specific CAR design. Glypican-3 (GPC3), a membrane-bound proteoglycan, is expressed in several solid tumors including hepatocellular carcinoma (HCC) (Gao H et al., 2014). GPC3 is an attractive target for immunotherapy since it is not expressed at detectable levels in non-malignant tissues including normal or cirrhotic liver (Gao H et al., 2014). Recent early phase clinical trials tested a GPC3-specific GC33 mAb and demonstrated that targeting GPC3 is safe, well tolerated and, depending on the density of GPC3 expression on tumor cells, can achieve significant antitumor responses in patients with advanced HCC (Zhu AX et al., 2013). Unlike clinically tested CD19-CARs, which contained either CD28 or 4-1BB co-stimulatory endodomains, the GPC3-CAR construct reported in these studies contained both CD28 and 4-1BB endodomains (Gao H et al., 2014). Li et al. evaluated a series of CAR constructs targeting GPC3, and GBBz was picked out as the best choice for its Th2 cytokine polarization profile (Li W et al., 2016). Recently, Zhao et al. compared several constructs targeting CD19. By using an in vivo ‘‘stress test’’, 1928z-41BBL provided the highest therapeutic efficacy, showing balanced tumoricidal function, increased T cell persistence, and decreased exhaustion (Zhao et al., 2015). In this study, for the first time we evaluated the anti-tumor properties of G3-28BBz or G328z-41BBL-modified T cells in a solid tumor—HCC. To perform the study the synthesized G3-28BBz or G328z-41BBL CAR gene was cloned into lentivirus vector. G328BBz CAR encoded CD3ζ with costimulatory domains derived from CD28 and 4-1BB, and G3-28z-41BBL CAR encoded CD3ζ with costimulatory domains derived from CD28 and a separate 4-1BB ligand spaced by internal ribosome entry site (IRES) sequence (Fig. 1A). Immobilized CD3 antibody-activated T cells were transduced with prepared lentivirus vector encoding the indicated CAR construct. Cell surface expression of CARs wasmeasured by flow cytometry (Fig. 1B and 1C). Both CARs were efficiently expressed on the cell surface of T cells without significant differences between the two constructs (Fig. 1C). The generated CAR-T cells contained >95% CD3-positive T cells, which comprises CD4and CD8-positive T-cell subsets, and has the similar CD4/CD8 ratio as in non-transduced T cells (Fig. 1D). To test the cytotoxicity of G3-28BBz and G3-28z-41BBL CAR T cells towards GPC3-positive targets, two HCC cell lines (HepG2 and HuH-7) and one immortal hepatic cell line (L-02) have been tested for cell surface expression of GPC3 by flow cytometry. Except for L-02 cell line, both HepG2 and HuH-7 HCC cells expressed GPC3, but the expression level of GPC3 on HepG2 was more than that of HuH-7 (Fig. 1E and 1F). For GPC3 CAR-T cells cytotoxicity test, LDH release assays have been performed at the indicated effector to target ratios. Results showed that GPC3-CAR T cells specifically killed GPC3 positive HepG2 and HuH-7 cells, but not the GPC3 negative L-02 cells. When comparing the cytolytic activity between G3-28BBz and G3-28z-41BBL CAR T cells, the latter showed more potent cytolytic activity, especially towards GPC3 high positive HepG2 cells at hour 4 (Fig. 1G and 1H). In addition, mock T cells did not kill any of the target cells and the killing activity of CAR-Tcells also can be inhibited by the addition of soluble recombinant human GPC3 protein, demonstrating that the cytolytic activity of T cells depends on the expression of CARs specifically targeting GPC3 (Fig. 1G–I). In the in vitro cytokine release assay, the ability of GPC3 CAR-T cells to secrete cytokines has been evaluated. HepG2 and HuH-7 effectively induced cytokine production of both two GPC3 CAR-T cells, but not L-02 cells and the groups containing mock T cells. Striking differences were noted in cytokine release between G3-28BBz and G3-28z-