Effective adoptive immunotherapy of triple-negative breast cancer by folate receptor-alpha redirected CAR T cells is influenced by surface antigen expression level

Abstract

Background: The poor prognosis and the limited efficacy of targeted therapy in patients with triple-negative breast cancer (TNBC) have raised the need for alternative therapies. Recent studies have demonstrated that folate receptor-alpha (FRα) may represent an ideal tumor-associated marker for immunotherapy for TNBC.

Methods: The aim of the present study was to apply a chimeric antigen receptor (CAR) approach for the targeting of FRα expressed on TNBC cells and evaluate the antitumor activity of CAR-engineered T cells in vitro and in vivo.

Results: We found that human T cells expressing a FRα-specific CAR were potent and specific killers of TNBC cells that express moderate levels of FRα in vitro and significantly inhibited tumor outgrowth following infusion into immunodeficient mice bearing an MDA-MB-231 tumor xenograft. However, the antitumor activity of the FRα CAR T cells was modest when compared to the same CAR T cells applied in an ovarian tumor xenograft model where FRα expression is more abundant. Notably, FRα CAR T cells induced superior tumor regression in vivo against MDA-MB-231 that was engineered for overexpression of FRα.

Conclusions: Taken together, our results show that FRα CAR T cells can mediate antitumor activity against established TNBC tumor, particularly when FRα is expressed at higher levels. These results have significant implications for the pre-selection of patients with high antigen expression levels when utilizing CAR-based adoptive T cell therapies of cancer in future clinical trials.

Keywords: Chimeric antigen receptor; Folate receptor-alpha; Immunotherapy; Triple-negative breast cancer.

Fig. 1

Construction and expression of folate receptor-alpha (FRα)-specific chimeric antigen receptor (CAR). a Schematic representation of MOv19-based FRα CAR constructs containing the CD27 co-stimulatory module in combination with the CD3ζ cytosolic domain. b Primary human CD3 T cells can efficiently express FRα-specific CAR. Expression was detected via PE-conjugated goat anti-mouse F(ab′)₂ fragment and measured by flow cytometry. c Compared to untransduced (UNT) T cells, transduced T cells consisted of CD4- and CD8-positive cells with both subsets expressing FRα CAR. FRα CAR expression was detected via biotin-labeled recombinant FRα protein staining followed by streptavidin-PE after transduction with lentivirus. Transduction efficiencies are indicated with the percentage of CAR expression in parentheses

Fig. 2

FRα may represent a new therapeutic target in TNBC. FRα surface expression in TNBC cell lines was measured by flow cytometry. FRα-specific mAb MOv18 was used to measure FRα expression on various ovarian cancer (OC) and breast cancer (BC) cell lines (open empty histogram), compared to a matched isotype antibody control (filled gray histogram)

Fig. 3

FRα CAR T cells secrete Th1 cytokines and up-regulate CD137 in response to FRα (+) tumor cells. a IFN-γ secretion of FRα CAR-transduced T cells after 20 h co-culture with the indicated tumor lines at a 1:1 ratio. Untransduced (UNT) T cells were used as a negative control. b Antigen-specific T cell activation was detected by the induction of CD137 expression. c FRα CAR T cells were stimulated with MDA-231 cells for 5 h in the presence of Golgi stop and analyzed by flow cytometry for intracellular cytokines IFN-γ, TNF-α, and IL-2. UNT T cells served as our negative control, whereas PMA and ionomycin-treated T cells served as positive controls

Fig. 4

Anti-tumor activity of FRα CAR T cells in vitro and in vivo. a FRα CAR T cells lysed FRα+ MDA-231 cells but exhibited decreased lysis of the FRα-C30 cells at the indicated effector/target (E/T) ratio for ~20 h. Untransduced (UNT) T cells served as our negative control. b NSG mice bearing established subcutaneous (s.c.) tumor were treated with intravenous (i.v.) injections of 1 × 10⁷ CAR+ T cells on days 40 and 46 post tumor inoculation. Tumor growth was assessed by caliper measurement [V = 1/2(length × width²)]. c Peripheral blood was collected 3 weeks after the first T cell infusion and quantified for the absolute number of human CD4+ and CD8+ T cells/μL of blood. Mean cell count ± SEM is shown with n = 5 for all groups

Fig. 5

FRα CAR T cells preferentially lyse tumor cells overexpressing FRα. a The MDA-231 cell line was engineered to overexpress FRα (MDA-231.FRα). b FRα CAR T cells T cells lyse MDA-231.FRα and SKOV3 more efficiently compared to MDA-231 cells. c MDA-231 cells were engineered to express ~90 % GFP (MDA-231.GFP). MDA-231.GFP cells were then mixed with MDA-231.FRα cells at a 1:1 ratio (~50 % GFP expression). The mixed cells were treated with either UNT or FRα CAR T cells at a 3:1 of E/T ratio for 24 h. UNT T cell treatment had no impact on the mixed tumor cell (~50 % GFP expression), whereas treatment with FRα CAR T cells increased the number of GFP (+)-engineered tumor cells (~85 % GFP expression)

Fig. 6

FRα CAR T cells induce rapid tumor regression of MDA-231.FRα in vivo. a-d NSG mice were inoculated with MDA-231 (a, b) or MDA.231.FRα tumor cells (c, d). Mice bearing established MDA-231.FRα or MDA-231 tumors received tail vein injections of 1 × 10⁷ CAR+ T cells on days 40 and 46, and tumor growth was monitored by caliper measurements (a, c) and BLI (b, d)