Tissue factor (TF) expressed on sub-cellular membrane vesicles, so-called plasma microparticles (MPs), has recently emerged as a potential key player in intravascular coagulation activation in various disease states. In this report, we demonstrate significantly increased levels of TF-specific procoagulant activity (PCA) of plasma MPs in five patients presenting with overt disseminated intravascular coagulation (DIC) due to an underlying malignancy, including non-small-cell lung cancer (n = 1), melanoma (n = 1), prostate cancer (n = 2), and acute promyelocytic leukemia (n = 1). Clotting experiments on available tumor cell samples suggested that cancer cells were a potential source of circulating TF-positive MPs at least in three of the five patients. Furthermore, follow-up plasma samples from two surviving patients revealed that response of their malignancies to specific anti-cancer therapy was paralleled by resolution of overt DIC and a significant decline in MP-associated TF PCA. Levels of plasma TF antigen, as assessed by an enzyme-linked immunosorbent assay, were also increased at presentation albeit to a lesser extent compared to MP-associated TF PCA, likely due to insufficient solubilization of the phospholipid-incorporated full-length TF molecule by the detergent. In summary, our findings suggest that MP-associated TF PCA may play an important pathogenic role in the evolution of overt DIC in various types of malignancy.

SummaryTissue factor (TF) plays a critical role in tumour growth and metastasis, and its enhanced release into plasma in association with cellular microparticles (MPs) has recently been associated with pathological cancer progression. We have previously demonstrated significantly elevated levels of plasma TF antigen as well as systemic coagulation and platelet activation in patients with localised prostate cancer. In this prospective study, we used a highly sensitive one-stage clotting assay to measure preoperative TF-specific procoagulant activity (PCA) of plasma MPs in 68 consecutive patients with early-stage prostate cancer to further explore the relevance of circulating TF in this tumour entity. Automated calibrated thrombography was used to monitor thrombin generation in cell-free plasma samples in the absence of exogenous TF or phospholipids. Compared to healthy male controls (n=20), patients had significantly increased levels of both D-dimer and TF-specific PCA of plasma MPs (p<0.001). Furthermore, MP-associated TF PCA was higher in patients with (n=29) than in those without (n=39) laboratory evidence of an acute-phase reaction (p=0.004) and decreased to normal levels within one week after radical prostatectomy. Overall, we found a significant correlation between TF-specific PCA of plasma MPs and plasma D-dimer (p=0.002), suggesting that plasma MPs contributed to in-vivo coagulation activation in a TF-dependent manner. Thrombin generation in plasma was also significantly increased in patients compared to controls (p<0.01). Collectively, our findings suggest that TF-specific PCA of plasma MPs contributes to intravascular coagulation activation in patients with early-stage prostate cancer and may represent a potential link between hypercoagulability, inflammation, and disease progression.

Activation of coagulation and inflammation is a characteristic finding in patients with advanced malignancies, including prostate cancer. Tissue factor (TF), a molecule involved in hemostasis, thrombosis and pro-inflammatory signaling pathways, is over-expressed on tumor cells and cells of the tumor microenvironment (i.e. endothelial cells, fibroblasts and tissue macrophages). Moreover, the enhanced release of TF into plasma in association with sub-cellular membrane vesicles, so-called plasma microparticles (MPs), has recently been associated with key events in molecular oncogenesis and cancer progression. In this study, we measured TF-specific procoagulant activity (PCA) of plasma MPs in 58 consecutive patients with clinically localized prostate cancer (mean age, 64±5 years) to explore its potential as a prognostic marker in this tumor entity. MPs were isolated from pre-operative plasma samples by sequential high-speed centrifugation for 1 h at 16,100 × g. TF-specific PCA of plasma MPs was quantified using a highly sensitive one-stage clotting assay in the presence and absence of inhibitory TF monoclonal antibody and calibration of clotting times against serial dilutions (1:10–1:105) of lipidated recombinant human full-length TF (rhTF1–263), showing a linear correlation in a log-log plot with R2>0.99. The lower detection limit of this assay for rhTF1–263 (33 kDa) was <5 pg/ml (<150 fM), and the intra- and inter-assay coefficients of variation were 7.3% and 5.4%, respectively. Total numbers of TF-positive MPs were measured by single-color flow cytometry using PE-conjugated TF monoclonal antibody (HTF-1) and microspheres for size calibration (1 μm) and sample flow standardization. TF antigen was quantified in plasma by ELISA. Calibrated automated thrombography (CAT) was used to monitor thrombin generation in platelet-free plasma samples over a 2-h period without the addition of exogenous TF or phospholipids. Intravascular coagulation activation was assessed by measuring plasma D-dimer. All assay systems were validated using MPs spontaneously shed from prostate cancer cell lines (PC-3, LNCaP and DU145) or from whole blood monocytes after challenge with endotoxin. Based on plasma fibrinogen and C-reactive protein levels, patients were stratified into those with (n=26) and those without (n=32) laboratory evidence of an acute-phase reaction (APR). Compared to healthy male controls (n=20), patients had significantly increased levels of both D-dimer (0.46±0.19 vs. 0.21±0.05 mg/l) and TF-specific PCA of plasma MPs (563±301 vs. 292±74 U/ml) (P<0.001). Among patients, laboratory evidence of an APR was associated with a significant increase in MP-associated TF PCA (699±351 vs. 452±196 U/ml) (P=0.001). Overall, we found a significant correlation between MP-associated TF PCA and plasma D-Dimer (P=0.015), suggesting that plasma MPs contributed to in-vivo coagulation activation in a TF-dependent manner. CAT also revealed significantly increased thrombin generation in patient compared to control plasmas, as indicated by a shortening in lag phase (25±4 vs. 29±5 min) and an increase in both peak thrombin generation (184±76 vs. 127±71 nM) and the endogenous thrombin potential, defined as the area under the thrombin generation curve (3576±509 vs. 2980±562 nM*min) (P<0.01). Importantly, TF-specific PCA of plasma MPs correlated neither with absolute numbers of TF-positive MPs nor with plasma TF antigen, suggesting that a substantial and variable fraction of the total plasma TF pool circulated as an inactive variant. Interestingly, systemic levels of IL-8, an inflammatory cytokine involved in TF/FVIIa-dependent, PAR-2-mediated pro-migratory signaling pathways in tumor cells and shown to be of biological relevance in advanced, hormone-refractory prostate cancer, were elevated in patients compared to controls (9±11 vs. 4±6 pg/ml) (P<0.01). In summary, our findings suggest that TF-specific PCA of plasma MPs contributes to intravascular coagulation activation in patients with early-stage prostate cancer and may represent an important molecular link between hypercoagulability, inflammation and disease progression. The above-described assay for the quantification of MP-associated TF PCA could thus be of prognostic value in the risk stratification of patients with localized prostate cancer with respect to thromboembolic complications and/or tumor recurrence.