Skip to main content
NCBI home page
Molecular Therapy logoLink to Molecular Therapy
. 2016 Oct 4;24(11):1987–1999. doi: 10.1038/mt.2016.149

Tumor Regression and Delayed Onset Toxicity Following B7-H4 CAR T Cell Therapy

Jenessa B Smith 1,2, Evripidis Lanitis 1,3, Denarda Dangaj 2,3, Elizabeth Buza 4, Mathilde Poussin 1,2, Caitlin Stashwick 2,5, Nathalie Scholler 2,6, Daniel J Powell Jr 1,2,*
PMCID: PMC5154474  PMID: 27439899

Abstract

B7-H4 protein is frequently overexpressed in ovarian cancer. Here, we engineered T cells with novel B7-H4-specific chimeric antigen receptors (CARs) that recognized both human and murine B7-H4 to test the hypothesis that B7-H4 CAR T cell therapy can be applied safely in preclinical models. B7-H4 CAR T cells specifically secreted IFN-γ and lysed B7-H4(+) targets. In vivo, B7-H4 CAR T cells displayed antitumor reactivity against B7-H4(+) human ovarian tumor xenografts. Unexpectedly, B7-H4 CAR T cell treatment reproducibly showed delayed, lethal toxicity 6–8 weeks after therapy. Comprehensive assessment of murine B7-H4 protein distribution uncovered expression in ductal and mucosal epithelial cells in normal tissues. Postmortem analysis revealed the presence of widespread histologic lesions that correlated with B7-H4(+) expression, and were inconsistent with graft versus host disease. Lastly, expression patterns of B7-H4 protein in normal human tissue were comparable to distribution in mice, advancing our understanding of B7-H4. We conclude that B7-H4 CAR therapy mediates control of cancer outgrowth. However, long-term engraftment of B7-H4 CAR T cells mediates lethal, off-tumor toxicity that is likely due to wide expression of B7-H4 in healthy mouse organs. This model system provides a unique opportunity for preclinical evaluation of safety approaches that limit CAR-mediated toxicity after tumor destruction in vivo.

Introduction

Ovarian cancer (OC) is the fifth most common cause of cancer-related deaths among women in the USA. In 2015, ~21,000 women will receive a new diagnosis and 14,180 women will die of OC.1 Primary treatments include cytoreductive surgery and chemotherapy, however, many women relapse and ultimately die of their disease.2

Engineering T cells with chimeric antigen receptors (CARs) is a pioneering therapy afforded by expression of a synthetic receptor that engages potent T cell signaling domains through MHC-independent, B cell receptor-like recognition.3 CAR therapy has successfully targeted hematological malignancies.4,5,6 However, antigen-specific T cell therapies targeting solid, epithelial tumors have not been efficacious to date. Many factors contribute to the lack of success including heterogeneity of tumor-associated antigens, inefficient trafficking and penetration of solid tumor masses, and expression of the target antigen in vital healthy tissues.7

B7-H4 was identified in 2003 and is less characterized than its fellow B7 family members PD-L1 and PD-L2, which only share 25% amino acid homology with B7-H4.8 Notably, the murine B7-H4 homolog shares 87% amino acid homology with human B7-H4.9,10,11 B7-H4 messenger RNA and protein are highly overexpressed in various types of human malignancies, including ovarian cancer.12,13,14,15 In normal murine and human tissue, B7-H4 messenger RNA is widely expressed in nonlymphoid tissues,9,10,15,16,17 but evaluation of protein expression has been inconsistent across publications.18 In 2015, Leong et al. reported positive B7-H4 staining in the ductal epithelium of several normal tissues including the breast, pancreas, and kidney,19 notably distinct from the primarily lymphoid distribution of PD-L1 and PD-L2 under noninflammatory conditions.20 Additionally, use of an anti-B7-H4 antibody-drug conjugate did not result in toxicity in preclinical models, suggesting that B7-H4-targeted therapy could be applied safely. Reports describing B7-H4 protein expression in normal murine tissues are limited,16,21,22 warranting further examination.

Surface B7-H4 protein binds a currently unknown receptor(s) on activated T cells that results in inhibition of T cell effector function via cell cycle arrest,9 decreased proliferation,9,10,11 and reduced IL-2 production9,10,11 in vitro. In addition to B7-H4's extrinsic effects on the immune system, tumor cells expressing B7-H4 demonstrate augmented tumor cell proliferation,23,24 superior antiapoptotic ability,12 and increased tumor growth even in the absence of a competent immune system.12,23,24 Preclinical models of autoimmunity,16,22,25 infection,21,26,27 and cancer28,29 implicate B7-H4 as a negative immune regulator of T cells, neutrophils, and myeloid-derived suppressor cells. However, questions about the functional role of B7-H4 in the tumor microenvironment exist as recent studies suggest a positive or bystander role for B7-H4 in spontaneous tumor development.17,30 Still, the high level of surface expression on cancer cells in comparison to normal tissue makes B7-H4 an attractive target for therapy if treatment is shown to be safe in preclinical models.

We hypothesized that B7-H4-directed CAR T cell therapy could be a promising new therapeutic option in OC. However, concerns about expression in normal tissue may result in toxicity post treatment. Therefore, we comprehensively assessed expression of B7-H4 protein in murine tissue and developed a novel B7-H4-specific CAR T cell platform that cross-react with both human and murine B7-H4, allowing us to test both antitumor efficacy and safety of B7-H4 CAR T cells using in vivo xenograft tumor models.

Results

Generation of B7-H4 CARs

Dangaj et al. isolated and characterized four, novel anti-B7-H4 single chain variable fragments (scFvs) from a yeast display library (26, 56, 3#68, 3#54), two of which (3#68 and 3#54 scFvs) were able to rescue functional inhibition of HER-2 TCR-engineered T cells.14 We utilized these four scFv sequences to generate B7-H4-specific CAR constructs. Anti-B7-H4 scFv sequences were cloned into previously validated lentiviral vectors containing a human CD8α leader, CD8α hinge, a CD28 transmembrane domain, and CD28 and CD3ζ intracellular signaling domains.31 The B7-H4 constructs also contained a green fluorescence protein (GFP) reporter separated by a viral P2A ribosomal skipping site to assess transgene efficiency after transduction. CARs are referred to as 26, 56, 3#68, and 3#54-CD28Z (Figure 1a, top). A CAR specific for human CD1932 was used as a specificity control for antigen-independent activity in all experiments (Figure 1a, bottom). The MOV19 CAR, specific for human FRα,33 was utilized as a positive control for tumor-specific reactivity (Figure 1a, bottom).

Figure 1.

Figure 1

Open in a new tab

CAR T cells bearing different anti-B7-H4 scFv bind recombinant B7-H4 with varying relative ability. (a) Schematic of lentiviral B7-H4 chimeric antigen receptor (CAR) constructs. All constructs are second generation CARs that utilize the CD28 and CD3ζ intracellular domains. B7-H4 CARs contain a green fluorescence protein (GFP) reporter linked to the CAR transgene by a viral P2A ribosomal skipping peptide. CD19-CD28Z and MOV19-CD28Z do not contain the GFP reporter. (b) GFP reporter (y-axis) expression versus binding of biotinylated, recombinant human B7-H4 protein (rhB7-H4) (x-axis) 6 days after transduction of human T cells with the indicated CARs. Frequency and median fluorescent intensity (MFI) of binding to rhB7-H4 is shown in the upper right quadrant. Cells are gated by size and viability (7AAD−). (c) Binding of the indicated CAR T cell populations to recombinant proteins human FRα (left), human B7-H4 (middle), and mouse B7-H4 (right) 6 days post-transduction. Cells are gated on size, viability (7AAD−), and CAR transgene(+) (GFP+) populations. (b-c) Incubation with biotinylated protein was followed with streptavidin-allophycocyanin (APC) secondary reagent. UNT, untransduced; GFP, green fluorescent protein transduced (no CAR). T cell donor shown is representative of greater than five independent experiments. VH, variable heavy; L, linker; VL, variable light; CD28, CD28 intracellular domain; CD3ζ, CD3ζ intracellular domain.

B7-H4 CARs are expressed in primary human T cells

We first confirmed expression of the various B7-H4 CARs in primary human T cells. Lentiviral B7-H4 or control CAR constructs showed high transduction efficiency in both CD8+ and CD4+ T cells from primary human donors, as assessed by GFP expression 6 days post transduction (see Supplementary Figure S1a). Additionally, CAR expression on the surface of T cells was tested using idiotype-specific antibodies for CARs composed of either human (see Supplementary Figure S1b) or murine scFvs (see Supplementary Figure S1c). 3#68 B7-H4, 3#54 B7-H4, and control CARs CD19 and MOV19 were highly expressed on the surface of T cells. The 26 and 56 B7-H4 CARs demonstrated lower surface CAR expression, despite similar GFP reporter expression (see Supplementary Figure S1c). All B7-H4 and control CAR-transduced T cell populations maintained high levels of GFP reporter expression after 14 days of expansion (data not shown).

B7-H4 CAR T cells composed of different scFvs have distinct antigen-binding patterns

Next, we evaluated the capacity of the B7-H4 CAR-bearing T cells to bind B7-H4 by flow cytometry. The four B7-H4 CARs had a differential ability to bind recombinant, human B7H4 protein (rhB7-H4). This was indicated by unique shifts in median fluorescence intensity (Figure 1b). None of the B7-H4 CARs bound the control FRα protein (Figure 1c, left), while the FRα-specific, MOV19 CAR only bound its cognate antigen (see Supplementary Figure S1d, left panel). Interestingly, the B7-H4 CARs also bound recombinant, murine B7H4 protein (rmB7-H4) with a similar pattern seen with rhB7-H4 (Figure 1c, right).

B7-H4 CAR-bearing T cells specifically exhibit effector function against B7-H4 (+) tumor cell lines in vitro

We evaluated endogenous B7-H4 protein in different cancer cell lines utilizing two different anti-B7-H4 antibodies. We found low to moderate surface B7-H4 expression on the human breast cancer cell line (SKBR3), the human ovarian cancer cell line (OVCAR3), and the human Epstein Bar Virus (EBV)-immortalized B cell line EBV-B, which is consistent with previous reports14,34 (Figure 2a). Interestingly, the AMP841 clone detected a higher frequency and intensity shift of surface B7-H4 in comparison to the commercially available MIH43 clone (Figure 2a).

Figure 2.

Figure 2

Open in a new tab

B7-H4 CAR T cells specifically secrete IFN-γ and lyse tumor cell lines expressing endogenous B7-H4. (a) Expression of B7-H4 on the surface of human tumor cell lines in vitro. Anti-B7-H4 mAb staining (open histograms) overlaid with isotype control (gray, filled histograms) as assessed by flow cytometry using two different antihuman B7-H4 antibodies. Staining of MIH43-PE (top) and biotinylated AMP841 (bottom) are shown. (b,c) B7-H4 CAR T cells specifically recognize B7-H4(+) tumors. CAR T cells were cocultured with B7-H4(+) cell lines SKBR3, OVCAR3, and EBV-B or B7-H4- cell lines C30 and OVCAR5 at a 1:1 effector to target (E:T) ratio. IFN-γ in 72-hour supernatants was quantified using enzyme linked immunosorbent assay (ELISA). Different primary human T cell donors were utilized for each experiment. Representative T cell donors are shown. (d) Tumor cell lines transduced with luciferase (luc) were cultured with T cells at the indicated effector to target ratios. Residual luciferase signal was calculated after 24 hours. Data is normalized to lysis of the CD19-CD28Z control CAR-bearing T cells against the cell lines indicated, which are all CD19(−). 3#68 and 3#54 B7-H4 CARs lyse B7-H4(+) OVCAR3-luc, but do not lyse B7-H4(−) cell lines 624 and MDA231. MOV19-CD28Z CAR T cells, specific for human FRα protein, lyse FRα(+) MDA231 and OVCAR3, but do not lyse FRα(−) 624. Lytic capability of the 3#68 B7-H4 CAR is significantly higher at the 1:1 E:T in comparison with the 3#54 B7-H4 CAR (P < 0.0001 utilizing an unpaired t-test with Welch's correction). Representative T cell donor is shown. (b-d) Error bars represent mean ± standard deviation (SD) of triplicate wells.

Next, we interrogated the effector capabilities of the different B7-H4 CARs in vitro by coculturing the CAR-bearing T cells with tumor cell lines expressing various levels of endogenous B7-H4. As expected, the control CD19 CAR secreted interferon-γ (IFN-γ) against the CD19(+) EBV-B cell line, but not the CD19(−) cell lines C30 and OVCAR3 (Figure 2b). While all four B7-H4 CARs secreted IFN-γ in response to the highest B7-H4-expressing cell line, EBV-B, the 3#68 and 3#54 B7-H4 CAR T cells maintained high reactivity against moderate B7-H4 expression in OVCAR3 and did not recognize B7-H4(−) C30 (Figure 2b).

These data rationalized further preclinical assessment with the 3#68 and 3#54 B7-H4 CARs against a wider panel of tumor cell lines. Both B7-H4 CARs demonstrated dose-dependent IFN-γ secretion in response to B7-H4(+) cell lines (Figure 2c). Additionally, we observed a similar trend in secretion of IL-2, TNF-α, and MIP1α as evidenced by cytokine bead array profiling (data not shown).

Finally, we assessed the in vitro cytolytic function of B7-H4 CAR T cells. Positive control MOV19 CAR T cells specifically lysed human FRα(+) cells lines MDA231 and OVCAR3, but not the FRα(−) cell line 624 (ref. 35) (Figure 2d, left). Both 3#68 and 3#54 B7-H4 CAR T cells specifically killed B7-H4(+) OVCAR3 without reactivity against B7-H4(−) cell lines 624 and MDA231 (Figure 2d, middle and right panels). We did not observe evidence of nonspecific targeting of B7-H4(–) tumor cells such as cross reactivity to PD-L1, as evidenced by lack of IFN-γ secretion in response to B7-H4(−)PD-L1(+) ovarian tumor cells SKOV3 and A1847 (data not shown) or killing of the B7-H4(−)PD-L1(+) breast cancer cell line MDA231 (Figure 2d). Together, these data establish the feasibility of targeting B7-H4 with CAR T cells in vitro. We chose to continue preclinical investigation of the 3#68 B7-H4 CAR (thereafter referred to as B7-H4 CAR) because of its enhanced killing at the lower effector to target ratios.

B7-H4-28Z CAR T cells exhibit antitumor efficacy in vivo

Next, we investigated the antitumor capability of B7-H4 CAR therapy in vivo in an established tumor model. We treated mice bearing established subcutaneous OVCAR3 tumors with CD19 negative-control, MOV19 positive-control or B7-H4 CAR T cells. B7-H4 CAR T-cell-treated mice exhibited significantly reduced tumor growth compared with CD19 control CAR-treated animals, as measured by bioluminescence imaging (Figure 3a,b) and caliper measurement of tumor volume (Figure 3c). We also tested the antitumor activity of B7-H4 CAR T cells in an additional model, treating mice with therapy earlier in tumor development (see Supplementary Figure S2a). In conclusion, B7-H4 CAR T cells significantly limited tumor growth in two in vivo treatment models. Together, these results establish the potential efficacy of B7-H4 CAR T cell therapy in ovarian cancer.

Figure 3.

Figure 3

Open in a new tab

3#68-CD-28Z CAR T cells show rapid antitumor efficacy in vivo, followed by late-onset toxicity and death. Mice bearing established (> 200 mm3) OVCAR3 tumors were treated intraperitoneally with one dose of 4e6 CAR(+) T cells. CD19-CD28Z served as a negative control. The MOV19 CAR targeting FRα served as a positive control for antitumor efficacy since OVCAR3-luc cells express endogenous levels of human FRα and human B7-H4. The number of CAR+ T cells per cohort was determined and normalized via flow cytometric staining with an anti-heavy chain light chain IgG antibody. (a) 3#68 B7-H4 CAR T cell-treated mice show a significant reduction in tumor growth in a late treatment model. Luminescence signal is shown 3 days before, as well as 11 and 25 days after T cell administration. (b) Quantified luminescence signal shows reduced OVCAR3-luc tumor burden compared with the negative control CD19 CAR. (c) Caliper measurement over time shows a reduction in OVCAR3-luc tumor volume (y-axis) in the 3#68 CAR T cell-treated cohort. (d) Survival curve for the tumor bearing mice (shown in a-c) shows death of the 3#68 B7-H4 CAR-treated mice with a median survival of 47 days. (e) Survival curve of nontumor bearing mice shows death of the 3#68 B7-H4 CAR-treated mice with a median survival of 48 days. (d,e) Log-rank (Mantel-Cox) test was used to determine the statistical significance of the survival curves. (f,g) TruCount analysis shows the concentration (cell/ul blood, y-axis) of CD45(+)CD3(+) (left), CD45(+)CD3(+)CD8(+) (middle), and CD45(+)CD3(+)CD8(−) (right) T cells of the indicated tumor-bearing mice cohorts 28 days (f) and 42 days (g) post T cell administration. (b,c, f,g) Error bars represent mean ± standard error of the mean (SEM) of five mice per group. Data represent one of three independent experiments with different T cell donors. (b,c; f,g) Unpaired, two-tailed Student t-tests, without assuming a consistent standard deviation, were performed between cohorts for each time point. *P < 0.05; **P < 0.01.

B7-H4 CAR T-cell-treated mice show signs of toxicity after antitumor effects

Surprisingly, B7-H4 CAR-treated mice showed overt signs of toxicity at a late time point following T cell treatment. Mice were lethargic, hunched, dehydrated, and had ruffled fur, while control CAR-treated mice did not show evident signs of distress. We observed a statistically significant decrease in body weight in B7-H4 CAR T-cell-treated mice in comparison with control mice sacrificed 53 days post treatment (3#68-CD28Z cohort 21.7 ± 1.1 grams versus CD19-CD28Z cohort 25.7 ± 0.9 grams; P < 0.05).

B7-H4 CAR-treated cohorts in the established tumor model presented with signs of overt toxicity 45–48 days post T cell treatment (Figure 3d). Toxicity was not attributed to the antitumor response since MOV19 CAR therapy completely eliminated OVCAR3 tumors (Figure 3a–c) without evident toxicity (Figure 3d).

We also assessed CAR T cell mediated toxicity in nontumor bearing mice. Again, we observed apparent signs of toxicity in B7-H4 CAR-treated mice without distress in control CAR T-cell treated mice (Figure 3e). This occurred at a similar kinetic as in the B7-H4 CAR-treated tumor-bearing mice (Figure 3d), suggesting that recognition of human B7-H4 in the context of tumor xenografts was not necessary for the development of toxicity mediated by the B7-H4 CAR.

To evaluate if in vivo T cell expansion correlated with the development of toxicity, we analyzed peripheral blood human T cells over the course of therapy in the tumor bearing mice. Analysis revealed an increased presence of CD8+ and CD4+ T cells in both the MOV19 and B7-H4 CAR-treated groups at 28 days post T cell treatment in comparison to the CD19 CAR cohort, consistent with in vivo activation and proliferation of T cells (Figure 3f). However, B7-H4 CAR-treated animals showed a contraction of T cells 43 days post treatment (Figure 3g). B7-H4 CAR-treated mice in the early treatment model also showed signs of toxicity and died post T cell treatment (see Supplementary Figure S2b). Peripheral T cell counts also followed a similar trend as in the established tumor model (see Supplementary Figure S2c,d). Finally, the number of T cells in the peripheral blood of B7-H4 or MOV19 CAR-treated nontumor-bearing mice was not statistically different 28 or 42 days post T-cell injection (data not shown). These data suggested that a high peripheral T cell count is not necessary for the observed toxicity.

B7-H4 CAR T cells do not demonstrate an auto-active phenotype or express B7-H4

Certain CAR structures result in antigen-independent, constitutive T cell-proliferation and preferential outgrowth of CAR(+) T cells due to tonic CAR signaling.36 Unlike the “continuous CARs”,36 all control and 3#68 B7-H4 CAR T cell populations showed similar growth kinetics, even in the presence of recombinant human IL-2 (see Supplementary Figure S3b). T cell populations maintained in the absence of rhIL-2 did not survive (data not shown). Additionally, CAR T cell populations maintained a similar frequency of untransduced cells over time (see Supplementary Figure S3c), further suggesting that B7-H4 CARs do not have an autoactive phenotype. Lastly, we were unable to detect surface B7-H4 antigen on control or B7-H4 CAR T cell populations (see Supplementary Figure S3d), ruling out activation through antigen expressed on T cells.

B7-H4 protein is expressed in numerous murine tissues

We hypothesized that the toxicity after B7-H4 CAR administration was due to recognition of B7-H4 protein expressed in normal murine tissue. Available data has been inconsistent regarding the expression of B7-H4 on murine tissue.18 Several groups were unable to detect B7-H4 protein in immune cells in vitro,16,21,22 despite early studies that reported expression.10,11 Murine B7-H4 protein is detectable in pancreatic islet cells16,21,22 and in epithelial cells lining the lung,21 however, expression in other organs has not been reported. Therefore, we developed an immunohistochemistry (IHC) protocol utilizing a novel B7-H4 antibody (AMP6H3). We validated the antigen-specific and species-specific reactivity of the antibody by flow cytometry and IHC as described in the Supplementary Materials and Methods and show that AMP6H3 cross-reacts with human or murine B7-H4(+) cell lines (see Supplementary Figure S4).

We tested numerous different murine tissues to comprehensively determine the bio-distribution of B7-H4 protein. Murine B7-H4 was not detected in the bone marrow, spleen, heart, skeletal muscle, tongue, thyroid, mesentery, or urinary bladder (see Supplementary Figure S5). Some, rare B7-H4(+) cells were found in tissues of the gastrointestinal tract including the small intestine, colon, cecum (see Supplementary Figure S5), adrenal gland, and stomach (Figure 4). Additionally, neurons in the spinal cord and brain expressed B7-H4 in the cytoplasm (Figure 4 and Supplementary Figure S5).

Figure 4.

Figure 4

Open in a new tab

B7-H4 is expressed endogenously in murine tissue. Immunohistochemistry was used to evaluate the expression of B7-H4 in murine tissue. NSG mice from the full autopsy study bearing OVCAR3-luc tumors in the flank were treated with 4e6 green fluorescence protein (GFP)(+) T cells and sacrificed 77 days post-tumor injection. Tissues were harvested, formalin-fixed, paraffin-embedded, and stained with the anti-B7-H4 antibody 6H3, which recognizes both human and murine B7-H4 protein. (a) Representative photomicrographs are shown. Magnification is shown in the bottom left corner. Black boxes and arrows indicate zoomed in cutout. (b) Bar graph represents the expression score obtained from evaluation of the photomicrographs. Scoring: 0, undetectable; 1, rare positive; 2, mildly positive; 3, moderately positive; 4, strongly positive; and 5, very strongly positive. All data represent three mice per tissue.

Similar to previous reports, murine B7-H4 was detected in the islet cells of the pancreas16,21,22 and the epithelial lining of the lung.21 Additionally, we uncovered previously unrecognized expression in the esophagus, trachea, salivary gland, mammary, gallbladder, liver, kidney, uterus, and haired skin (Figure 4). Generally, B7-H4(+) staining was restricted to ductal and mucosal epithelial cells in each tissue, such as the bile duct in the liver, the biliary epithelium in the gallbladder, bronchiolar and tracheal mucosal epithelium in the lung, and the transitional epithelium in the kidneys. Notably, we detected B7-H4 expression in the ductal epithelium of the pancreas in addition to the previously reported staining in islet cells.16 The endometrial glands/epithelium and the myometrium in the uterus and oviduct also stained strongly for B7-H4. IHC staining utilizing an isotype control did not show staining of any tissues analyzed (see Supplementary Figure S5; data not shown). Lastly, as expected, human B7-H4 was strongly expressed in engrafted OVCAR3 tumor xenografts (Figure 4). Results are summarized in Table 1.

Table 1. Summary of B7-H4 expression in normal murine tissue.

graphic file with name mt2016149t1.jpg

Open in a new tab

B7-H4 CAR-treatment results in multiorgan lymphocyte infiltration

After determining widespread expression of murine B7-H4 in our animal model, we evaluated tissues from B7-H4 CAR treated mice. Comprehensive autopsy and pathology analyses were performed on hematoxylin and eosin-stained slides from tissues of OVCAR3-tumor bearing mice that were treated with B7-H4 CAR or control CAR T cells. One cohort of mice was sacrificed prior to showing signs of overt toxicity (37 days) and a second cohort was sacrificed once they showed signs of evident toxicity (55 days). The histologic lesions within all cohorts are summarized in the Supplementary Table S1.

We did not observe tissue damage or the presence of lymphocytes in many B7-H4(−) tissues, such as skeletal muscle, bone, heart, tongue, adrenal gland, or thyroid gland of B7-H4 CAR or control CAR-treated mice 37 days (Figure 5a) or 55 days (Figure 5b) post T cell administration (see Supplementary Table S1). Notably, tissues displaying membranous B7-H4(+) staining (Figure 4 and Table 1) exhibited similar histologic lesions at 37 and 53 days post B7-H4 CAR therapy. Lesions consisted of lymphocytic infiltration in the salivary gland, female reproductive tract, lung, liver (portal region), and pancreas (see Supplementary Table S1). At day 37, B7-H4 CAR-treated mice exhibited lymphocytic infiltration of the pulmonary interstitium, portal area of the liver, salivary gland, vaginal/cervical submucosa, and periductally within the pancreas as compared with control CAR mice, which had no significant lesions (Figure 5a). B7-H4 CAR T-cell-treated mice sacrificed 55 days post-treatment showed similar, but more severe, histologic lesions and exhibited more extensive tissue destruction (see Supplementary Table S1 and Figure 5b).

Figure 5.

Figure 5

Open in a new tab

3#68-CD28Z CAR-T cell treated mice exhibit histologic lesions consisting primarily of multiorgan lymphocytic infiltration as compared with control CAR-T cell treated mice at 37 and 53 days post-treatment. OVCAR3-tumor bearing mice were treated with one dose of 4e6 CAR(+) T cells. Mice were sacrificed before the 3#68 B7-H4 CAR-treated animals showed signs of overt toxicity (37 days post T cell treatment) or when showing evident signs of toxicity (55 days post T cell treatment). Tissue pathology was evaluated by hemotoxylin and eosin (H&E) staining. (a) At day 37, 3#68 B7-H4 CAR-treated mice (top) exhibited lymphocytic infiltration of the pulmonary interstitium, portal area of the liver, renal interstitium, and periductally within the pancreas as compared to control CD19 CAR-treated mice (bottom), which had no significant lesions. No lesions were observed in the heart, thyroid gland or adrenal gland in any cohort at either time point. (b) At day 53, 3#68 B7-H4 CAR-treated mice (top) exhibited similar, but more severe, lesions as compared with the early time point. No lesions were observed in the CD19 CAR control CAR-treated mouse (bottom). Symbols indicate the absence (−) or presence (+) of pathology. ++ indicates a more severe phenotype. Tissues and photomicrographs shown are representative of three mice per group. Magnification is shown in the bottom left corner.

We did note histologic lesions in tissues that stained B7-H4(−), including the urinary bladder, cecum, mesentery, peripheral nerves, and haired skin in 1/3 mice at both time points post B7-H4 CAR treatment (see Supplementary Table S1). Additionally, several tissues examined from mice sacrificed at day 55 contained lesions that were not observed within the early time point (37 days). These included lymphocyte exocytosis along with basal cell necrosis in the squamous portion of stomach (B7-H4(−)), lymphocytic infiltration and exocytosis in the oral cavity (B7-H4(−)), and interstitial lymphocytic infiltrates within the kidney (B7-H4(+)).

We detected human CD3(+) cells via IHC in the tumor of B7-H4 CAR T cell-treated mice but not CD19 CAR T cell-treated animals (see Supplementary Figure S6a). We also observed hCD3(+) cells surrounding B7-H4(+) areas in the pancreas (see Supplementary Figure S6b) and liver (see Supplementary Figure S6c) of B7-H4 CAR-treated animals but not CD19 CAR-treated mice (see Supplementary Figure S6d,e) . Additionally, anti-GFP IHC staining correlated with antihuman CD3 expression, confirming the presence of CAR(+) T cells in the lesions of 3#68 B7-H4 CAR T-cell-treated mice (data not shown). These data further support the hypothesis that B7-H4 CAR T cells mediated pathology through recognition of murine B7-H4(+) in healthy tissues in treated mice.

B7-H4 protein is expressed in normal human tissues

To assess if our new findings of widespread murine B7-H4 expression were also true in human tissues, we utilized our established anti-B7-H4 antibody (AMP6H3) and staining protocol to conduct immunohistochemistry on a microarray of human tissues containing 20 tissues and an ovarian carcinoma cell line (Supplementary Figure S7). Paralleling murine B7-H4 expression, human B7-H4 protein was not expressed in heart, muscle, cerebellum (head), spinal cord, thyroid gland, lymph node, and spleen. Additionally, human B7-H4 protein was undetectable in thymus, adipose, parathyroid gland, and placenta. Rare cells in the adrenal gland and in colonic crypts were B7-H4(+), similar to murine expression.

We detected B7-H4 protein in the mammary gland, kidney, and in pancreatic islet cells, consistent with published reports.19,37 We also detected previously uncharacterized B7-H4 expression in the esophagus, salivary gland, and liver. Positive staining was generally found in glandular epithelial cells of the salivary and breast, ductal epithelium of the liver, and in the tubular epithelium of the kidney. Although previously undescribed, ductal epithelial cells of the pancreas also stained B7-H4 positive. Notably, the ovarian cancer sample strongly expressed B7-H4. Our results are summarized in Table 2. Together our study reveals the bio-distribution of B7-H4 protein.

Table 2. Summary of B7-H4 expression in normal human tissue.

graphic file with name mt2016149t2.jpg

Open in a new tab

Discussion

Here, we establish the first report of CAR T cell therapy targeting B7-H4. We created four different, fully human B7-H4 CAR lentiviral constructs that showed cross reactivity with murine B7-H4 and had a differential ability to recognize B7-H4(+) targets in vitro. We pursued preclinical development of the 3#68 B7-H4 CAR (herein referred to as B7-H4 CAR) and saw reduction of ovarian tumor burden in a xenograft model. Unexpectedly, tumor and nontumor bearing 3#68 B7-H4 CAR T cell-treated mice reproducibly showed symptoms of lethal toxicity emerging 6–8 weeks after therapy. Examination of murine and human tissue revealed widespread expression of B7-H4 protein. Notably, human B7-H4 expression patterns and location is largely similar to murine B7-H4 protein expression, suggesting that preclinical studies targeting the B7-H4 homolog in mice may be a relevant model to interrogate questions addressing B7-H4's function and targetability in vivo.

In our study, postmortem analysis of B7-H4 CAR T-cell-treated mice revealed multiorgan lymphocytic infiltrate that primarily correlated with positive membranous B7-H4 tissue expression determined via immunohistochemistry, suggesting an on-target effect of B7-H4 CAR therapy. However, lesions were also evident in tissues that were B7-H4(−). Perivascular and perineural lymphocytic infiltrates seen within multiple soft tissues such as mesentery, broad ligament, and other adipose tissue may be a nonspecific reaction to an inflammatory state. Additionally, lesions in the squamous portion of the stomach and oral mucosa only occur at the later time point, possibly due to chronic, systemic activation of B7-H4 CAR by murine antigen causing graft-versus-host-disease (GVHD)-like pathology. Induction of xenogeneic GVHD phenotypes can be highly variable across cohorts of mice, but commonly effect skin,38,39,40 liver,38,39,40,41,42 gastrointestinal tract,38,39 and the lung.38,40,41,42 We observed only mild skin lesions in 1/3 mice at each time point and importantly, control mice did not show evidence of widespread GVHD at concomitant time points. Although similarities in target organs between GVHD and B7-H4 CAR treatment exist, we also observed unique lesions in B7-H4(+) tissues that are unaffected by GVHD-like pathology, highly suggesting a contribution of on-target CAR T cell recognition.

Compounds such as anti-B7-H4 scFvs,14 ADCC-mediating antibodies43 and antibody drug conjugates19 that cross react with murine B7-H4 have antitumor effects, but lack toxicity in mouse models. This may inherently be due to the differences in treatment modalities, as scFvs and antibodies are rapidly cleared, may not efficiently penetrate tissue, and lack the same potent effector functions as T cells44. This has been evident clinically in targeting ErbB-2, a protein expressed at low levels on pulmonary parenchyma. Monoclonal antibody therapy (Herceptin) has not shown adverse events, but treatment with a relatively high dose of a third generation, Herceptin scFv-based CAR with IL-2 and previous lymphodepletion resulted in death of a patient and discontinuation of the trial.45

Our data suggests that B7-H4 CAR therapy can delay tumor progression, but due to our findings of widespread B7-H4 expression in normal tissues, clinical application should be carefully considered. Notably, death following B7-H4 CAR T cell therapy occurred at a delayed time point, while still controlling tumor burden. This finding suggests that regulation of our B7-H4 CAR therapy could potentially be applied safely. Transient expression of CAR using inducible suicide genes46 or RNA electroporation47 could be utilized to avoid long term on-target, off-tumor toxicity to normal tissue. Additionally, the B7-H4-CD28Z CAR could be modified to remove the CD3ζ endodomain, thereby preventing activation in the absence of signal 1 for T cell activation.48 When expressed in tumor-specific T cells, this switch receptor/chimera could act as a decoy for the putative B7-H4R on T cells and provide costimulation in the presence of tumor-associated antigen and B7-H4 expressed on tumor cells.

In our current model, we determined the presence of pathologic lesions in mice at late time points post T-cell transfer to gain insight into the observed death following B7-H4 CAR therapy. However, this does not rule out initiation of lesions at earlier time points. Previous studies that observed toxicity in murine models of CAR therapy targeting fibroblast activating protein49 and NKG2D ligands50 report treatment-mediated pathology within days post T-cell transfer. The differences in kinetics seen in our study may have occurred for several reasons including differences in CAR T-cell dose, preconditioning regimes, distribution of target antigen in normal tissue, and murine strain variability.51 Additionally, several groups have observed safe therapeutic targeting of fibroblast activating protein in murine models,52,53 suggesting that differences in the therapy could allow for safe targeting by CAR T cells. Future studies with regulatable B7-H4 CARs could evaluate T-cell trafficking, kinetics, and impact on tissue utilizing withdrawal of B7-H4 CAR treatment prior to induction of lethal toxicity. Additionally, it is possible that differences in epitope, affinity, or avidity conferred by the untested 26, 56, and 3#54 B7-H4 CAR may allow for safe application of B7-H4 CAR therapy in vivo. Lastly, we did not observe clearance of the OVCAR3 tumor in our NSG xenograft experiments following one dose of B7-H4 CAR treatment. IHC analysis of the remaining tumor revealed heterogeneous areas of B7-H4(+) and B7-H4(−) expression in comparison with control CAR T-cell-treated cohorts (data not shown). Treatment with a second dose of CAR T cell therapy or use of combinatorial therapies may be necessary to eliminate the last cancer cell in heterogeneous solid tumors.7 If applied safely, the B7-H4 CAR therapy could be utilized in repeated doses or in combination with other targeted treatment modalities, such the anti-FRα CAR (MOV19), in order to improve upon the antitumor efficacy we observed in the B7-H4 CAR cohorts.

Evidence for B7-H4's function in the periphery has been limited, since studies from 2003–2011, were unable to detect protein expression on human tissue or did not report staining in murine tissue. After detecting B7-H4 protein in pancreatic islet cells, Wei et al. showed that transfer of pancreatic-specific T cells lead to increased severity of diabetes in a B7-H4-deficient mouse and, conversely, that overexpression of murine B7-H4 in pancreatic islets abrogated disease induction.16,22 This report highlights the potential for B7-H4 to act in peripheral T cell tolerance. While we have used our B7-H4 platform as a classic CAR approach to target antigen on tumor cells, it could also be applied in an autoimmunity setting. Transfer of regulatory T cells expressing a carcinoembryonic antigen-specific CAR lessens the severity of colitis in preclinical animal models.54 Since pancreatic islet cells express B7-H4, investigation of B7-H4 CAR-bearing regulatory T cells could have therapeutic potential in Type-1 diabetes.

Here, we report the first use of a B7-H4-specific CAR T cell therapy and our results indicate that B7-H4(+) cancer outgrowth can be controlled in vivo. Long-term persistence of 3#68 B7-H4 CAR T cells, however, mediates lethal on-target, off-tumor toxicity that is likely due to wide expression of B7-H4 in healthy mouse organs. Other models of CAR-mediated toxicity in mice tend to cause acute toxicity.49,50 Our B7-H4 CAR model of toxicity could provide a unique preclinical model to evaluate safety approaches designed to limit long-term CAR-mediated toxicity following antitumor activity in vivo. Additionally, these studies reveal the endogenous expression pattern of B7-H4, advancing our understanding of the molecule and fostering intelligent design of novel B7-H4-specific therapeutics for the clinic.

Materials and Methods

Construction of the B7-H4 CAR vectors. Fully human anti-B7-H4 scFv sequences14,53 were amplified via polymerase chain reaction from plasmids containing each soluble scFv sequence using the following primers:

26-Sense 5′-ACGCAGATCTGATATTGTGATGACTCAGACTCCAGC-3′ (BglII), Anti-sense 5′-ACGCGCTAGCTTGTTCGGATCCCTCGAATGAAGAG-3′ (NheI); 56-Sense 5′-ACGCAGATCTCAGCCTGTGCTGACTCAGTCCCACT-3′ (BglII), Anti-sense 5′-ACGC GCTAGCTTGTTCGGATCCCTCGAATGAAGAG-3′ (NheI); 3#68-Sense 5′-ACGCTGATCACAGCCTGGGCTGACTCAGCCACCCT (BclI), Anti-sense 5′-ACGCGCTAGCTTGTTCGGATCCCTCGAATGAAGAGACG-3′ (NheI); 54-Sense 5′-ACGCAGATCTCGGCCCGTGCTGACTCAGCCACCCT-3′ (BglII); Anti-sense 5′-ACGCGCTAGCTTGTTCGGATCCCTCGAATGAGGAGA-3′ (NheI). Amplified scFv sequences were digested with the underlined restriction enzymes and ligated into BamHI/NheI digested backbone of a third generation, self-inactivating pELNS lentiviral vector containing GFP.31

Recombinant lentiviral vector production. Replication-defective lentiviral vectors were produced as previously described.31 Briefly, HEK293T-cells were transfected with pELNS transfer plasmid and lentiviral packaging plasmids VSV glycoprotein expression plasmid (pVSV), pRSV.REV (Rev expression plasmid), and pMDLg/p.RRE (Gag/Pol expression plasmid). Twenty-four and 48-hour supernatants were collected and concentrated by ultracentrifugation (Beckman Coulter, Carlsbad, CA) at 26,000×g for 2 hours. Concentrated virus was stored at −80°C until use. Titer was determined by GFP reporter or idiotype expression 2–3 days post-transduction in HEK293T cells.

Human T cell transduction. Primary human T cells were purchased from the University of Pennsylvania Human Immunology Core. Specimens were obtained from healthy donors with written consent under an approved University Institutional Review Board protocol. Human CD4(+) and CD8(+) T cells were stimulated with anti-CD3/CD28 beads (Invitrogen, Carlsbad, CA) at a 3:1 ratio. Lentiviral vectors (multiplicity of infection (MOI) 5–10) were added 24 hours postactivation. T cell cultures were maintained in complete media (CM) composed of Rosewell Park Memorial Institute (RPMI) 1640-Glutamax, 10% heat-inactivated fetal bovine serum (Sigma, St. Louis, MO), 100 IU/ml penicillin (Gibco), and 100 IU/ml streptomycin (Gibco, Waltham, MA). 50-100 IU/ml recombinant, human IL-2 (Peprotech) was added to CM until 2 days before functional assays.

Cell lines. Established human cell lines OVCAR3, OVCAR5, SKBR3, MDA231 (HTB-26), MDA468 (HTB-132), and HEK293T were purchased from American Type Culture Collection (ATCC). The human ovarian cancer cell line C30 was provided by George Coukos (Ludwig Institute for Cancer Research, Lausanne, Switzerland). Human melanoma cell line 624 was provided by S.A. Rosenberg (National Cancer Institute /National Institute of Health). The EBV-transformed B cell line, EBV-B, was provided by Raj Somasundaram (Wistar Institute). All cell lines were cultured in CM at 37°C. B7-H4 expression was tested when cells were confluent. 624 MDA231, OVCAR3 were transduced with lentiviral GFP.2A.firefly luciferase to generate lines for cytotoxicity and xenograft experiments. All cells lines were routinely tested for mycoplasma contamination.

Flow cytometry. Mouse antihuman B7-H4 (AMP6H3) and the humanized version (AMP841) were kindly provided by Amplimmune, kidlington, UK. AMP6H3/AMP841 cross react with mouse B7-H4. Mouse antihuman B7-H4 (MIH43)-PE was purchased from AbD Serotec. The cell viability solutions 7AAD (BD Biosciences, San Jose, CA) or fixable viability dye eFluor506 (eBioscience) were used to remove dead cells from analysis. Recombinant human FRα (R&D Systems, Minneapolis, MN), human B7-H4,14 and mouse B7-H4 (R&D Systems) proteins were biotinylated as per the manufacturer's instructions (EZ-Link Sulfo-NHS-LC-Biotin, Thermo, Waltham, MA). Streptavidin APC (eBioscience) was utilized as a secondary reagent. Data was acquired on a BD FACS Canto II with Diva Software and analyzed using Flojo 7.6.5.

Cytokine release/cytotoxicity. Functional assays were set up once T cells were rested (size <300-350 fL, Multisizer Coulter Counter, Beckman Coulter). Viability and number of cells was enumerated using a hemacytometer. T-cell populations were normalized by addition of untransduced T cells to achieve the same frequency of CAR(+) cells in each population. Transduction efficiency was generally between 50–70%.

Cytokine release. 1e6 T cells were co-cultured with 1e6 tumor cell targets in 200 ul CM in triplicate wells. After 72 hours, IFN-γ production in cell-free supernatants was assessed via enzyme-linked immunosorbent assay (ELISA) (BioLegend, San Diego, CA).

Cytotoxicity. 0.2e6 luciferase (+) targets were plated in triplicate in 96-well plates. T cells were added at the indicated E:T ratios. Cocultures were incubated for 24 hour at 37°C in phenol-free CM. Residual luciferase activity was measured using the extended glow bioluminescent reporter gene assay (Applied Biosystems, Foster City, CA). Percent lysis = 100–((average signal from indicated CAR-treated wells)/(average signal from control CAR-treated wells)) × 100.

Xenograft model. NOD-scid IL2Rγnull (NSG) mice were obtained from the University of Pennsylvania Stem Cell and Xenograft Core. 6–12 week old mice were maintained under pathogen-free conditions under protocols approved by the University of Pennsylvania Institutional Animal Care and Use Committee. 4e6 OVCAR3 tumor cells were subcutaneously injected into the flank of NSG mice. After tumors were palpable (>200 mm2; ~50 days), mice were treated with 4e6 CAR-bearing T-cells. Tumor growth was assessed via bioluminescence imaging and caliper measurement. Tumor volume = (length × (width)2)/2, where length is the greatest longitudinal diameter and width is the greatest transverse diameter. Each cohort of mice was treated with the same number of CAR(+) T cells as determined by flow cytometric staining of T cell populations on Day 14 with a species-specific biotinylatyed antiheavy chain light chain (HC/LC) IgG antibody. CD19 and MOV19 CAR T cells were stained with rabbit antimurine HC/LC IgG (Jackson ImmunoResearch). 3#68 B7-H4 CAR T cells were stained with rabbit antihuman HC/LC IgG (Jackson ImmunoResearch, West Grove, PA).

Bioluminescence imaging. Bioluminescence imaging of luciferase(+) tumor cells was performed as previously described.31 Living Image Software (Xenogen, Cranbury, NJ) was used to quantify luciferase(+) tumor cells per animal and render a pseudocolor image representing light intensity (scale 1e6–1e8).

TruCount. T cells in the peripheral blood were enumerated using TruCount assay as per the manufacturer's instructions (BD Biosciences). T cell subsets were gated using the following antibodies for flow cytometry: mouse antihuman CD8a-APC (BioLegend), mouse antihuman CD3-PerCpCy5.5 (eBioscience), and mouse anti-human CD45-PE (BioLegend).

Comprehensive autopsy study. OVCAR3 tumor-bearing NSG mice were treated with one dose of 4e6 CAR(+) cells. Both B7-H4 and control CAR treated cohorts were sacrificed at 37 days and 55 days post CAR T cell injection. No clinical signs were observed in both the B7-H4 CAR treated mice at 37 days, however, nonspecific clinical signs of weight loss and ruffled fur were observed at day 55. No clinical signs were observed in the control CAR treated mice at 37 or 55 days. One GFP control mouse was sacrificed at day 55. A complete autopsy, including gross examination and whole body tissue collection, was performed. All tissues were fixed with 10% neutral-buffered formalin, processed, and evaluated histologically on hematoxylin and eosin-stained slides. Slides were analyzed for pathology in a blinded fashion by a board-certified anatomic veterinary pathologist.

Immunohistochemistry. Staining was performed on formalin-fixed paraffin-embedded (FFPE) tissue from the comprehensive autopsy experiments described above. A normal human tissue microarray previously created from tissues obtained in a deidentified manner from the surgical pathology archives at the University of Pennsylvania (Michael Feldman) under IRB approval was obtained. B7-H4 staining was performed with mouse antihuman/mouse AMP6H3 using the Bond Polymer Refine Detection System (Leica Microsystems AR9800, Wetzlar, Germany) on a Leica Bond-III instrument at the Human Anatomic Pathology Core at the University of Pennsylvania. Slides were dewaxed for 30 minutes at 72°C, and then subjected to heat-induced antigen retrieval for 20 minutes at 99°C with ER1 solution (Leica Microsystems AR9640). AMP6H3 primary antibody (1:1,500) was added for 15 minutes at room temperature followed by use of the bond polymer refine detection kit (Leica) as per the manufacturer's instructions. Slides were counterstained with hematoxylin (Leica). All slides were washed three times between each step with bond wash buffer or water. Slides were imaged using CellSens software (Olympus, Tokyo, Japan) on an Olympus BX53 microscope with an Olympus DP25 camera. Images were assembled in ImageScope version 12.1.0.5029 (Aperio) or PowerPoint (Microsoft). All B7-H4 staining evaluations were performed by a board-certified anatomic veterinary pathologist in the Comparative Pathology Core at the University of Pennsylvania School Of Veterinary Medicine. Three mice were evaluated per group.

Statistical analysis. Data are reported as mean ± standard error of the mean (SEM) unless otherwise noted. Unpaired, two-tailed Student t-tests and Kaplan–Meier estimates using the log-rank (Mantel-Cox) test were performed using GraphPad Prism 6 (GraphPad software). P-values <0.05 were considered statistically significant.

SUPPLEMENTARY MATERIAL Figure S1. B7-H4 CAR expression in primary human T cells. Figure S2. Anti-tumor efficacy and safety of the 3#68-CD-28Z CAR T cells in an early treatment model. Figure S3. In vitro growth kinetics and evaluation of CAR T cells bearing different B7-H4 or control CARs. Figure S4. Validation of anti-B7-H4 antibodies for immunohistochemistry. Figure S5. Assessment of murine B7-H4 expression. Figure S6. Presence of hCD3 lymphocytes and B7-H4 expression in murine tissue. Figure S7. Expression on B7-H4 protein in normal human tissue. Table S1. Summarization of histopathologic findings along with B7-H4 expression in B7-H4 CAR or control CAR-treated NSG mice from 37 and 53 days post treatment. Materials and Methods.

Acknowledgments

The authors acknowledge the following cores: Cancer Histology Core (Tianying Jiang), Anatomic Pathology Core (Li Ping Wang), Human Immunology Core, and Comparative Pathology Core (Amy Durham and Elizabeth Buza). The authors also thank Rachel Lynn for valuable suggestions and edits and Elaine Ho for validation of the hCD3 IHC protocol. This work was supported by the Department of Defense (DOD) grant OC110673 (N.S. and D.J.P), NIH RO1-CA168900 (D.J.P and N.S.), the Immunobiology of Normal and Neoplastic Lymphocytes T32 Training Grant (J.B.S), and NCI grant CA016520, which supported imaging at the Small Animal Imaging Facility Optical/Bioluminescence Core. J.B.S. and D.J.P. planned the experiments, arranged the figures, and wrote the manuscript. J.B.S. performed the studies and analyzed the data. D.D. and N.S. isolated the B7-H4 scFvs. E.L. and D.D. designed the constructs and performed initial studies. E.B. performed and analyzed pathology and IHC. M.P. performed the animal injections and tumor progression measurements. C.S. assisted in performing in vitro experiments. N.S. and D.J.P. have ownership interest in patents related to B7-H4 specific scFvs and CAR technology. The other authors do not disclose any potential conflicts of interest.

Supplementary Material

Supplementary Figures and Tables
Click here for additional data file. (136.6MB, doc)

References

  1. American Cancer Society (2015). Ovarian cancer. Learn about Cancer, vol. 2015. www.cancer.org. [Google Scholar]
  2. Coleman, RL, Monk, BJ, Sood, AK and Herzog, TJ (2013). Latest research and treatment of advanced-stage epithelial ovarian cancer. Nat Rev Clin Oncol 10: 211–224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Gross, G, Waks, T and Eshhar, Z (1989). Expression of immunoglobulin-T-cell receptor chimeric molecules as functional receptors with antibody-type specificity. Proc Natl Acad Sci USA 86: 10024–10028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Grupp, SA, Kalos, M, Barrett, D, Aplenc, R, Porter, DL, Rheingold, SR et al. (2013). Chimeric antigen receptor-modified T cells for acute lymphoid leukemia. N Engl J Med 368: 1509–1518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Porter, DL, Levine, BL, Kalos, M, Bagg, A and June, CH (2011). Chimeric antigen receptor-modified T cells in chronic lymphoid leukemia. N Engl J Med 365: 725–733. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Maude, SL, Frey, N, Shaw, PA, Aplenc, R, Barrett, DM, Bunin, NJ et al. (2014). Chimeric antigen receptor T cells for sustained remissions in leukemia. N Engl J Med 371: 1507–1517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Kakarla, S and Gottschalk, S (2014). CAR T cells for solid tumors: armed and ready to go? Cancer J 20: 151–155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hansen, JD, Du Pasquier, L, Lefranc, MP, Lopez, V, Benmansour, A and Boudinot, P (2009). The B7 family of immunoregulatory receptors: a comparative and evolutionary perspective. Mol Immunol 46: 457–472. [DOI] [PubMed] [Google Scholar]
  9. Sica, GL, Choi, IH, Zhu, G, Tamada, K, Wang, SD, Tamura, H et al. (2003). B7-H4, a molecule of the B7 family, negatively regulates T cell immunity. Immunity 18: 849–861. [DOI] [PubMed] [Google Scholar]
  10. Zang, X, Loke, P, Kim, J, Murphy, K, Waitz, R and Allison, JP (2003). B7x: a widely expressed B7 family member that inhibits T cell activation. Proc Natl Acad Sci USA 100: 10388–10392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Prasad, DV, Richards, S, Mai, XM and Dong, C (2003). B7S1, a novel B7 family member that negatively regulates T cell activation. Immunity 18: 863–873. [DOI] [PubMed] [Google Scholar]
  12. Salceda, S, Tang, T, Kmet, M, Munteanu, A, Ghosh, M, Macina, R et al. (2005). The immunomodulatory protein B7-H4 is overexpressed in breast and ovarian cancers and promotes epithelial cell transformation. Exp Cell Res 306: 128–141. [DOI] [PubMed] [Google Scholar]
  13. Tringler, B, Liu, W, Corral, L, Torkko, KC, Enomoto, T, Davidson, S et al. (2006). B7-H4 overexpression in ovarian tumors. Gynecol Oncol 100: 44–52. [DOI] [PubMed] [Google Scholar]
  14. Dangaj, D, Lanitis, E, Zhao, A, Joshi, S, Cheng, Y, Sandaltzopoulos, R et al. (2013). Novel recombinant human b7-h4 antibodies overcome tumoral immune escape to potentiate T-cell antitumor responses. Cancer Res 73: 4820–4829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Choi, IH, Zhu, G, Sica, GL, Strome, SE, Cheville, JC, Lau, JS et al. (2003). Genomic organization and expression analysis of B7-H4, an immune inhibitory molecule of the B7 family. J Immunol 171: 4650–4654. [DOI] [PubMed] [Google Scholar]
  16. Wei, J, Loke, P, Zang, X and Allison, JP (2011). Tissue-specific expression of B7x protects from CD4 T cell-mediated autoimmunity. J Exp Med 208: 1683–1694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Rahbar, R, Lin, A, Ghazarian, M, Yau, HL, Paramathas, S, Lang, PA et al. (2015). B7-H4 expression by nonhematopoietic cells in the tumor microenvironment promotes antitumor immunity. Cancer Immunol Res 3: 184–195. [DOI] [PubMed] [Google Scholar]
  18. Smith, JB, Stashwick, C and Powell, DJ Jr (2014). B7-H4 as a potential target for immunotherapy for gynecologic cancers: a closer look. Gynecol Oncol 134: 181–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Leong, SR, Liang, WC, Wu, Y, Crocker, L, Cheng, E, Sampath, D et al. (2015). An anti-B7-H4 antibody-drug conjugate for the treatment of breast cancer. Mol Pharm 12: 1717–1729. [DOI] [PubMed] [Google Scholar]
  20. Ritprajak, P and Azuma, M (2015). Intrinsic and extrinsic control of expression of the immunoregulatory molecule PD-L1 in epithelial cells and squamous cell carcinoma. Oral Oncol 51: 221–228. [DOI] [PubMed] [Google Scholar]
  21. Hofmeyer, KA, Scandiuzzi, L, Ghosh, K, Pirofski, LA and Zang, X (2012). Tissue-expressed B7x affects the immune response to and outcome of lethal pulmonary infection. J Immunol 189: 3054–3063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lee, JS, Scandiuzzi, L, Ray, A, Wei, J, Hofmeyer, KA, Abadi, YM et al. (2012). B7x in the periphery abrogates pancreas-specific damage mediated by self-reactive CD8 T cells. J Immunol 189: 4165–4174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Zhang, L, Wu, H, Lu, D, Li, G, Sun, C, Song, H, et al. (2013). The costimulatory molecule B7-H4 promote tumor progression and cell proliferation through translocating into nucleus. Oncogene 32: 5347–5358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Cheng, L, Jiang, J, Gao, R, Wei, S, Nan, F, Li, S et al. (2009). B7-H4 expression promotes tumorigenesis in ovarian cancer. Int J Gynecol Cancer 19: 1481–1486. [DOI] [PubMed] [Google Scholar]
  25. Azuma, T, Zhu, G, Xu, H, Rietz, AC, Drake, CG, Matteson, EL et al. (2009). Potential role of decoy B7-H4 in the pathogenesis of rheumatoid arthritis: a mouse model informed by clinical data. PLoS Med 6: e1000166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Suh, WK, Wang, S, Duncan, GS, Miyazaki, Y, Cates, E, Walker, T et al. (2006). Generation and characterization of B7-H4/B7S1/B7x-deficient mice. Mol Cell Biol 26: 6403–6411. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Zhu, G, Augustine, MM, Azuma, T, Luo, L, Yao, S, Anand, S et al. (2009). B7-H4-deficient mice display augmented neutrophil-mediated innate immunity. Blood 113: 1759–1767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Abadi, YM, Jeon, H, Ohaegbulam, KC, Scandiuzzi, L, Ghosh, K, Hofmeyer, KA et al. (2013). Host b7x promotes pulmonary metastasis of breast cancer. J Immunol 190: 3806–3814. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Leung, J and Suh, WK (2013). Host B7-H4 regulates antitumor T cell responses through inhibition of myeloid-derived suppressor cells in a 4T1 tumor transplantation model. J Immunol 190: 6651–6661. [DOI] [PubMed] [Google Scholar]
  30. Kreymborg, K, Haak, S, Murali, R, Wei, J, Waitz, R, Gasteiger, G et al. (2015). Ablation of B7-H3 but not B7-H4 results in highly increased tumor burden in a murine model of spontaneous prostate cancer. Cancer Immunol Res 3: 849–854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lanitis, E, Poussin, M, Hagemann, IS, Coukos, G, Sandaltzopoulos, R, Scholler, N et al. (2012). Redirected antitumor activity of primary human lymphocytes transduced with a fully human antimesothelin chimeric receptor. Mol Ther 20: 633–643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Milone, MC, Fish, JD, Carpenito, C, Carroll, RG, Binder, GK, Teachey, D et al. (2009). Chimeric receptors containing CD137 signal transduction domains mediate enhanced survival of T cells and increased antileukemic efficacy in vivo. Mol Ther 17: 1453–1464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Miotti, S, Canevari, S, Ménard, S, Mezzanzanica, D, Porro, G, Pupa, SM et al. (1987). Characterization of human ovarian carcinoma-associated antigens defined by novel monoclonal antibodies with tumor-restricted specificity. Int J Cancer 39: 297–303. [DOI] [PubMed] [Google Scholar]
  34. Quandt, D, Fiedler, E, Boettcher, D, Marsch, WCh and Seliger, B (2011). B7-h4 expression in human melanoma: its association with patients' survival and antitumor immune response. Clin Cancer Res 17: 3100–3111. [DOI] [PubMed] [Google Scholar]
  35. Song, DG, Ye, Q, Poussin, M, Harms, GM, Figini, M and Powell, DJ Jr (2012). CD27 costimulation augments the survival and antitumor activity of redirected human T cells in vivo. Blood 119: 696–706. [DOI] [PubMed] [Google Scholar]
  36. Frigault, MJ, Lee, J, Basil, MC, Carpenito, C, Motohashi, S, Scholler, J et al. (2015). Identification of chimeric antigen receptors that mediate constitutive or inducible proliferation of T cells. Cancer Immunol Res 3: 356–367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Cheung , SS, Ou, D, Metzger, DL, Meloche, M, Ao, Z, Ng, SS et al. (2014). B7-H4 expression in normal and diseased human islet β cells. Pancreas 43: 128–134. [DOI] [PubMed] [Google Scholar]
  38. van Rijn, RS, Simonetti, ER, Hagenbeek, A, Hogenes, MC, de Weger, RA, Canninga-van Dijk, MR et al. (2003). A new xenograft model for graft-versus-host disease by intravenous transfer of human peripheral blood mononuclear cells in RAG2-/- gammac-/- double-mutant mice. Blood 102: 2522–2531. [DOI] [PubMed] [Google Scholar]
  39. Beilhack, A, Schulz, S, Baker, J, Beilhack, GF, Wieland, CB, Herman, EI et al. (2005). In vivo analyses of early events in acute graft-versus-host disease reveal sequential infiltration of T-cell subsets. Blood 106: 1113–1122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Nervi, B, Rettig, MP, Ritchey, JK, Wang, HL, Bauer, G, Walker, J et al. (2007). Factors affecting human T cell engraftment, trafficking, and associated xenogeneic graft-vs-host disease in NOD/SCID beta2mnull mice. Exp Hematol 35: 1823–1838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Panoskaltsis-Mortari, A, Price, A, Hermanson, JR, Taras, E, Lees, C, Serody, JS et al. (2004). In vivo imaging of graft-versus-host-disease in mice. Blood 103: 3590–3598. [DOI] [PubMed] [Google Scholar]
  42. King, MA, Covassin, L, Brehm, MA, Racki, W, Pearson, T, Leif, J et al. (2009). Human peripheral blood leucocyte non-obese diabetic-severe combined immunodeficiency interleukin-2 receptor gamma chain gene mouse model of xenogeneic graft-versus-host-like disease and the role of host major histocompatibility complex. Clin Exp Immunol 157: 104–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Jeon, H, Vigdorovich, V, Garrett-Thomson, SC, Janakiram, M, Ramagopal, UA, Abadi, YM et al. (2014). Structure and cancer immunotherapy of the B7 family member B7x. Cell Rep 9: 1089–1098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Chames, P, Van Regenmortel, M, Weiss, E and Baty, D (2009). Therapeutic antibodies: successes, limitations and hopes for the future. Br J Pharmacol 157: 220–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Morgan, RA, Yang, JC, Kitano, M, Dudley, ME, Laurencot, CM and Rosenberg, SA (2010). Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2. Mol Ther 18: 843–851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Dotti, G, Gottschalk, S, Savoldo, B and Brenner, MK (2014). Design and development of therapies using chimeric antigen receptor-expressing T cells. Immunol Rev 257: 107–126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Zhao, Y, Moon, E, Carpenito, C, Paulos, CM, Liu, X, Brennan, AL et al. (2010). Multiple injections of electroporated autologous T cells expressing a chimeric antigen receptor mediate regression of human disseminated tumor. Cancer Res 70: 9053–9061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Lanitis, E, Poussin, M, Klattenhoff, AW, Song, D, Sandaltzopoulos, R, June, CH et al. (2013). Chimeric antigen receptor T Cells with dissociated signaling domains exhibit focused antitumor activity with reduced potential for toxicity in vivo. Cancer Immunol Res 1: 43–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tran, E, Chinnasamy, D, Yu, Z, Morgan, RA, Lee, CC, Restifo, NP et al. (2013). Immune targeting of fibroblast activation protein triggers recognition of multipotent bone marrow stromal cells and cachexia. J Exp Med 210: 1125–1135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. VanSeggelen, H, Hammill, JA, Dvorkin-Gheva, A, Tantalo, DG, Kwiecien, JM, Denisova, GF et al. (2015). T cells engineered with chimeric antigen receptors targeting NKG2D ligands display lethal toxicity in mice. Mol Ther 23: 1600–1610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Lynn, RC and Powell, DJ Jr (2015). Strain-dependent lethal toxicity in NKG2D ligand-targeted CAR T-cell therapy. Mol Ther 23: 1559–1561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wang, LC, Lo, A, Scholler, J, Sun, J, Majumdar, RS, Kapoor, V et al. (2014). Targeting fibroblast activation protein in tumor stroma with chimeric antigen receptor T cells can inhibit tumor growth and augment host immunity without severe toxicity. Cancer Immunol Res 2: 154–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Spear, P, Barber, A, Rynda-Apple, A and Sentman, CL (2013). NKG2D CAR T-cell therapy inhibits the growth of NKG2D ligand heterogeneous tumors. Immunol Cell Biol 91: 435–440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Blat, D, Zigmond, E, Alteber, Z, Waks, T and Eshhar, Z (2014). Suppression of murine colitis and its associated cancer by carcinoembryonic antigen-specific regulatory T cells. Mol Ther 22: 1018–1028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Dangaj, D and Scholler, N (2015). Isolation and validation of anti-B7-H4 scFvs from an ovarian cancer scFv yeast-display library. Methods Mol Biol 1319: 37–49. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figures and Tables
Click here for additional data file. (136.6MB, doc)

Articles from Molecular Therapy are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES