Tissue Factor as a Novel Target for Treatment of Breast Cancer

Tissue factor (TF) is frequently expressed in a variety of human breast cancer types, including triple-negative breast cancer. This article examines the usefulness of TF as a target in combating the most aggressive cancers, in which TF is frequently highly expressed.

CME Learning Objectives

1. Explain the process by which tissue factor (TF) initiates blood coagulation and is implicated in tumor progression.

2. Describe the proposed mechanisms of targeting TF in malignancy.

3. Discuss the applications of TF targets in developing new treatments for aggressive cancers including triple-negative breast cancer.

Abstract

Tissue factor (TF), a 47-kDa transmembrane glycoprotein that initiates blood coagulation when complexed with factor VIIa (FVIIa), is expressed in several tumor types. TF has been shown to play a role in cell signaling, inflammation, angiogenesis, as well as tumor growth and metastasis. Activation of the TF signaling pathway has been implicated in mediating the function of many tumor cell types and has led to TF as a potential target in the treatment of several malignancies. Formation of the TF-FVIIa complex in breast cancer cells has been shown to exert an antiapoptotic effect and play a key role in tumor growth and metastasis. Breast cancer growth is suppressed by inhibition of TF-mediated PAR2 signaling, and deficiency in PAR2 delays spontaneous breast cancer development in mice. TF is expressed in triple-negative breast cancer (TNBC), an aggressive type of breast cancer in which there is currently a paucity of available targets. Various methods of targeting TF have been investigated and include immunoconjugates or icons, anti-TF antibodies, TF pathway inhibitors, targeted photodynamic therapy, and microRNAs. These investigations may give way to promising clinical therapies for breast cancer, especially in TNBC, for which there are relatively few effective treatment options.

Implications for Practice:

Tissue factor (TF), a kDa transmembrane glycoprotein that binds with factor VII during blood coagulation, has been expressed in many tumor types. It plays a role in tumor growth and metastasis, which has made it a potential target for disease intervention. One malignancy in which TF is frequently expressed, and for which it is a potential therapeutic target, is breast cancer – especially triple-negative breast cancer (TNBC). TF is highly expressed in aggressive breast cancers, and TNBC is an aggressive breast cancer that carries a poor prognosis. To date, few treatment options have been available for TNBC. Various methods of targeting TF have been investigated, including using anti-TF antibodies, immunoconjugates or icons, targeted photodynamic therapy, TF pathway inhibitors, and microRNAs. Each has had some success in experimental trials and is described in detail. Targeting TF is likely to lead to useful clinical applications in breast cancer, especially TNBC and other malignancies.

Introduction

Tissue factor (TF) is a 47-kDa transmembrane glycoprotein that initiates blood coagulation when complexed with its cofactor, factor VIIa (FVIIa). The TF molecule consists of a 219-amino-acid extracellular domain, a 23-amino-acid transmembrane domain, and a 21-amino-acid cytoplasmic domain [1]. The extracellular domain of TF is required for procoagulant function [2]. This domain consists of two fibronectin type III-like domains that resemble several growth factor and cytokine receptors [3]. The cytoplasmic domain of TF, which is not required for procoagulant function, contains three serine residues that can be phosphorylated [4] and have been implicated in cell signaling [4, 5]. TF is essential for normal hemostasis and embryonic development [6, 7]. In addition, TF is expressed in a variety of tumor cell types and has been linked to the pathogenesis of cancer [6].

Upon vessel injury, TF expressed in fibroblasts is exposed to the bloodstream. Blood coagulation is initiated when TF binds to the serine protease, FVIIa [1]. Formation of the TF-FVIIa complex leads to activation of factor X and factor IX that, in turn, generates activated factor X (FXa) and activated factor IX (FIXa), respectively. Generation of FXa leads to the conversion of prothrombin to thrombin. Thrombin subsequently cleaves soluble fibrinogen to form a fibrin clot [1]. In addition, there is a circulating pool of TF that contributes to clot propagation [8]. An alternatively spliced TF (asTF) protein has also been identified, which appears to be active in promoting tumor growth and angiogenesis, but its role in blood coagulation is still unclear [9, 10].

Abnormalities in the coagulation cascade leading to a hypercoagulable state are a well-known complication of malignancy. Trousseau syndrome, a migratory thrombophlebitis, is a common manifestation of the increased coagulability seen in patients with cancer and often precedes the diagnosis of a malignancy [11]. Many aspects of cancer contribute to hypercoagulability, including TF expression in tumor cells and upregulation of TF in vascular endothelium and monocytes by inflammatory cytokines, interleukin-1, and tumor necrosis factor-α [6]. In addition, extrinsic compression of blood vessels by tumors, impaired clearance of activated coagulation factors in the setting of hepatic disease, treatments of the cancer including medications (cytotoxic agents and angiogenesis inhibitors), surgery, and patient immobility contribute to the hypercoagulability in malignancy [6].

TF is expressed in many tumor cell types, including breast cancer, gliomas, lung cancer, colorectal cancer, pancreatic cancer, prostate cancer, ovarian cancer, renal cell cancer, and hepatocellular cancer [12]. Tumor cells that are poorly differentiated appear, in general, to have increased TF expression compared with those that are more differentiated [6]. Tumors associated with high bloodborne TF levels are associated with a greater risk of thromboembolic disease [13] and with decreased overall survival [14]. Elevated circulating TF levels result from tumor shedding of TF-bearing microparticles (MPs) through leaky tumor blood vessels, tumor-induced upregulation of TF expression in monocytes and endothelial cells, and upregulation of endothelial cell TF expression by chemotherapeutic agents [6].

TF has also been implicated in tumor progression. Increased TF expression in tumor cells promotes their hematogenous metastasis [15, 16]. In addition, fibrin formation as a result of activation of blood coagulation and platelets shields tumor cells from NK cells and immune surveillance [6]. Finally, the TF-mediated signaling pathway has been implicated in tumor growth. Formation of either TF-FVIIa-FXa complex or TF-FVIIa complex on tumor cells results in cellular signaling, which involves phosphorylation of a p44/42 mitogen-activated protein kinase, Akt/PKB, mammalian target of rapamycin, and p70 S6K1 via activation of G-protein coupled protease-activated receptors (PARs) [17–21]. The functional consequences of activation of these pathways result in enhanced tumor cell migration, inhibition of apoptosis, and promotion of cellular growth [17–21].

Tissue Factor in Breast Cancer

Human breast cancer tissue and cell lines have been shown to express TF [12, 22–24]. TF is highly expressed on aggressive breast cancer cell lines, including MDA-MB-231 and HS578 [25, 26]. The MCF-7 breast cancer cell line, which has low endogenous TF expression, displays increased metastatic potential when transfected with cDNA encoding for human TF compared with the vector transfected control line in immunocompromised mice [27]. Tumor TF expression is also of prognostic significance because high expression is predictive of decreased overall survival in patients with breast cancer [14].

TF plays a role in breast cancer cell signaling and promotes tumor cell migration and inhibition of apoptosis [17, 20, 21]. The presence of TF on breast cancer tumor vasculature, but not on the vasculature of benign breast masses, has implicated TF in neoplastic angiogenesis [28]. The role of TF in breast tumor growth is further strengthened in studies with mouse models, which have shown that breast cancer growth is suppressed by inhibition of the TF-mediated PAR2 signaling and that deficiency in PAR2 delays spontaneous breast cancer development in mice [29]. In addition, in patients with primary breast cancer, shorter recurrence-free survival correlated with tumors expressing phosphorylated TF (pTF) alone as well as coexpression of pTF and PAR-2. These results further support the role of TF-PAR2 signaling in breast cancer progression [30].

Several human breast cancer cell lines with the triple-negative breast cancer (TNBC) phenotype express TF, including MDA-MB-231, MDA-MB-468, and HCC-1806 [31]. TNBC, which lacks expression of receptors for estrogen (ER), progesterone (PR) and Her-2/neu, is an aggressive type of breast cancer that often affects younger women and has a poor prognosis [32]. Preliminary data from our laboratory has also shown that TF is expressed in 80% (36/45) of TNBC patient biopsies (unpublished observations). Given the poor prognosis for TNBC and the paucity of targeted therapies for this type of breast cancer, targeting TF might be a promising therapeutic strategy.

Tumors associated with high bloodborne TF levels are associated with a greater risk of thromboembolic disease and with decreased overall survival. Elevated circulating TF levels result from tumor shedding of TF-bearing microparticles (MPs) through leaky tumor blood vessels, tumor-induced upregulation of TF expression in monocytes and endothelial cells, and upregulation of endothelial cell TF expression by chemotherapeutic agents.

Antibodies Against Tissue Factor

Formation of TF-FVIIa and TF-FVIIa-Xa complexes leads to activation of PAR2 and PAR1 signaling, which promote migration and proliferation of tumor cells. Anti-TF antibodies have been shown to inhibit metastases of human breast cancer cell lines in murine models [33, 34]. Two antibodies that block its coagulation function are the murine antihuman TF antibody TF8–5G9 and a humanized version of this antibody, CNTO 859 [33, 35]. CNTO-859 was shown to be highly effective in inhibiting metastatic spread of human MDA-MD-231 breast cancer cells to the lungs of SCID Beige mice, with a near complete reduction in lung metastases in the treated mice [35, 33].

Anti-TF antibodies that do not alter the procoagulant function of TF also result in inhibition of tumor progression [34]. Versteeg et al. [34] showed that the antitumor activity of the anti-TF antibody Mab-10H10, which does not inhibit coagulant activity but uncouples the interaction of TF to β1 integrin, results in decreased proliferation of MDA-MD-231 breast cancer cells. TF-dependent PAR2 signaling was also implicated in promoting tumor growth and shown to be inhibited by specific cleavage-blocking antibodies to PAR2. This effect was not seen with anti-PAR1 antibodies [34].

Although the approach involving antibodies against TF and TF-FVIIa complex appears promising, there may be associated risks. The most concerning risk is bleeding, which has been observed with some anticoagulants in patients with cancer [36]. Although a trial using TF pathway inhibitor, inactivated recombinant VIIa, in patients with acute lung injury was associated with a trend toward serious bleeding events [37], studies with anti-TF antibodies thus far have not shown a major incidence of bleeding. No serious bleeding events have been reported in clinical trials with SunolcH36 (also known as ALT-836), a monoclonal antibody against TF, in patients with coronary artery disease or acute lung injury [38, 39]. Currently, the risk of bleeding with anti-TF antibodies that inhibit coagulation is not yet known. Alternatively, anti-TF antibodies have been developed that specifically inhibit TF-FVIIa signaling without affecting the pro-coagulant function [34]. Taken together, these studies suggest that targeting either the procoagulant or signaling functions of TF might be useful therapeutic strategies in breast cancer.

Immunoconjugates (Icons)

Another therapeutic strategy that targets TF in tumors is immunotherapy using immunoconjugates (icons). Icons are chimeric molecules that involve a mutated factor VII (mfVII) targeting domain, which does not have coagulant function, and an Fc effector domain of IgG1 Fc (mfVII/Fc icon) [40]. Using one approach, the icon was encoded in a replication-incompetent adenoviral vector and then injected into the tumor cells, which then expressed the icon. In this study, Hu et al. [40] showed that the mutated factor VII domain of the icon binds with high affinity to TF and the Fc domain of the icon activates an immune cytolytic attack, which is mediated through natural killer cells as well as the complement pathway.

Icon therapy has been successful in mouse models of both melanoma and prostate cancer. In prostate cancer, icon injection into tumor cells resulted in inhibition of tumor growth and tumor regression in a SCID mouse model [40]. Another promising finding with regard to treatment of metastatic prostate tumors was that the mice that were injected with the icon showed regression of the tumor that was injected as well as the tumor that was not injected, suggesting that this therapy might be useful in disseminated tumors that might not be accessible for injection.

With regard to icon treatment safety, the prothrombin time (PT) was studied as a measurable parameter to assess for bleeding risk [40]. Bleeding risk was not readily detected in any of the examined organs, and the concentration of icon in the mice used to produce a tumor response was 1% of the minimum concentration required to prolong the PT [40]. The icon did not appear to bind to any of the normal organs in the mice [40].

More recently, investigators have used mutant factor VII (fVII) along with a photosensitizer to TF expressing tumor vasculature and tumors followed by laser irradiation. Hu et al. conjugated the photosensitizer verteporfin (VP) to mutant fVII for ligand-targeted photodynamic therapy [41, 42]. This type of targeted photodynamic therapy induced a greater inhibition of breast cancer cell growth than nontargeted photodynamic therapy and may be a promising approach to chemotherapy-resistant breast cancer cells [43].

Although FVII is synthesized by the liver and released into the blood, recent studies have demonstrated that FVII is also produced ectopically by several cancer cell lines, including breast cancer [44–46]. Curcumin, a dietary compound, has been shown to inhibit constitutive ectopic expression through inhibition of p300/CBP activity. Icon therapy might also target ectopic factor VII as another approach to blocking the TF-FVIIa-PAR2 signaling pathway [46].

Tissue Factor Pathway Inhibitors

The TF pathway in blood coagulation is inhibited by TF pathway inhibitors (TFPIs). TFPI is expressed in breast, pancreatic, and colon cancer cell lines and may function to prevent autocoagulation of vessels within the tumor [47]. Cells in tumor microvasculature endothelium produce TFPI-1; its main function is inhibition of TF-FVIIa-FXa complex. Inhibition of tumor growth, including both primary tumor as well as metastases, was demonstrated with the use of B16 melanoma mouse models treated with TFPI [47].

Other investigators have studied the effect of treating B16 mouse models as well as Lewis lung carcinoma mouse models with two small anticoagulants isolated from the hematophagous nematode A. caninum: rNAPc2 and rNAP5. Results from these studies showed that rNAPc2, which is directed against TF-FVIIa, inhibited tumor growth in both models, whereas rNAP5, which is directed against FXa, did not inhibit tumor growth. This finding suggested that the mechanism of tumor inhibition is specific to a pathway involving TF-FVIIa and not FXa. Furthermore, the function of TF-FVIIa in cancer angiogenesis and metastasis appears distinct from its function in coagulation. Other studies had shown a role for FXa in tumor metastases; however, the mechanism by which this takes place is thought to be by facilitating tumor cell seeding and not angiogenesis [47].

MicroRNA

One of the most recent therapeutic strategies targeting TF has used microRNA. In a recent study, microRNA-19 (miR-19) was shown to be instrumental in regulating TF expression in breast cancer cells at the posttranscriptional level [48]. In this study, breast cancer cell lines that were less aggressive, including MCF-7, T47D, and ZR-75–1, minimally expressed TF; however, in more aggressive breast cancer cell lines, including MDA-MB-231 and BT-20, TF was highly expressed. MicroRNAs are conserved molecules in organisms that regulate gene expression. MicroRNAs can bind to the 3′untranslated (UTR) region of the transcript and thus repress protein translation or make the mRNA unstable. In MCF-7 cells, translation of the 3′UTR region was found to be necessary for inhibition of TF expression. However, TF was expressed in the MCF-7 cells if the miR-19 binding site to 3′UTR was deleted. In addition, overexpression of miR-19 in MDA-MB-231 cells resulted in downregulation of TF expression. These findings indicate that altering microRNAs that regulate TF expression might be another novel strategy in breast cancer treatment.

Using TF as a target might be useful in combating the most aggressive cancers, in which TF is frequently highly expressed. TF expression is observed in both human breast cell lines and tissues. Treating TNBCs, which has been difficult given their high grades, large tumor burdens, and historical lack of available therapeutic agents, may be a particularly useful clinical application for targeting TF.

Conclusions

TF is frequently expressed in a variety of human breast cancer types. Using TF as a target might be useful in combating the most aggressive cancers, in which TF is frequently highly expressed. TF expression is observed in both human breast cell lines and tissues. Treating TNBCs, which has been difficult given their high grades, large tumor burdens, and historical lack of available therapeutic agents, may be a particularly useful clinical application for targeting TF. Several different methods have been devised to target the TF pathway using anti-TF antibodies, icons, TFPIs, targeted photodynamic therapy, and microRNAs. Each of these has had some success in recent experimental trials; however, potential side effects, such as bleeding and the specificity of targeted agents, remain to be fully clarified. A greater understanding of the mechanisms involved and the role of TF-mediated signaling in tumor growth and metastasis, as well as the TF pathway's overall interaction with other components of the immune response, will likely lead to the further development of TF as a novel target in breast cancer, as well as other malignancies in the future.

Author Contributions

Conception and design: Marion Cole, Michael Bromberg

Manuscript writing: Marion Cole, Michael Bromberg

Disclosures

The authors indicated no financial relationships.

Section Editors: Gabriel Hortobágyi: Antigen Express, Galena Biopharma, Novartis, Rockpointe (C/A), Novartis (RF), Taivex (OI), and is a founder and member of the board of directors for Citizen's Oncology Foundation; Kathleen Pritchard: Novartis, Roche, AstraZeneca, Pfizer, Abraxis, Boehringer-Ingelheim, GlaxoSmithKline, Sanofi, Ortho-Biotech, YM Biosciences, Amgen, Bristol-Myers Squibb, Bayer Schering Pharma (C/A, H).

Reviewer “A”: Johnson and Johnson (C/A)

Reviewer “B”: None

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board