Tumors Are a Major Source of Tissue Factor-Positive Microparticles in a Mouse Model.

Abstract 3190

Poster Board III-127

Tissue factor (TF) is present in blood in the form of cell-membrane vesicles called microparticles (MPs). Elevated levels of TF-positive MPs may increase the risk of venous thromboembolism (VTE) in patients with cancer. In mice containing human tumors, levels of human TF antigen significantly correlate with both levels of thrombin-antithrombin (TAT) and tumor size. The objectives of this study were to measure levels of MP TF activity in the plasma of tumor-bearing mice and analyze the relative contribution of host versus tumor to this pool of MPs. Human pancreatic cells (Hpaf-II, 2×10⁶ cells per mouse) were injected subcutaneously into female SCID mice. Five weeks later, whole blood was collected and plasma was prepared. The median tumor weight was 1.03 ± 0.64 grams (n=9). MP TF activity was measured using our recent described method (Wang et al. JTH 2009:1092-1098). The species-specific antibodies 1H1 and HTF-1 were used to inhibit mouse TF activity in MPs derived from the host and human TF activity in MPs derived from the tumor, respectively. We also measured levels of human TF antigen and TAT in the plasma. In control mice injected with saline, levels of human TF antigen and human TF activity in MPs were undetectable, whereas both human TF antigen (318.6 ± 256.2 pg/mL, n=9) and human TF activity in MPs (1.08 ± 1.18 pg/mL, n=9) were dramatically increased in tumor-bearing mice. Consistent with previous reports, there was a significant correlation between levels of human TF antigen and tumor size (r=0.67, n=9, p=0.030). Importantly, we found that levels of human TF activity in MPs significantly correlated with levels of human TF antigen (r=0.883, n=9, p=0.002) and tumor size (r=0.667, n=9, p=0.029). Furthermore, tumor-bearing mice had significant higher levels of TAT (7.5 ± 1.8 ng/mL, n=9) than control mice (3.9 ± 0.7 ng/mL, n=3) (p=0.008). Four of the 9 mice had metastatic tumors in the pancreas. Therefore, based on metastasis and tumor size, the tumor-bearing mice were classified into two groups: early stage (no pancreatic metastasis or tumor size ≤ 1 gram, 0.70 ± 0.35 gram, n=4) and late stage (pancreatic metastasis or tumor size ≥1 gram, 1.30 ± 0.72 gram, n=5). As expected, mice in the late stage group have significant higher levels of human TF antigen (468.1 ± 228.0 vs 131.8 ± 148.3 pg/ml, p=0.039) and human TF activity in MPs (1.74 ± 1.24 pg/ml vs 0.25 ± 0.08 pg/mL, p=0.049) than those in the early stage group. Finally, mice in the late stage group have significantly higher levels of human TF activity in MPs derived from the tumor (1.95 ± 1.54 pg/mL) than mouse TF activity in MPs derived from the host (0.20 ± 0.26 pg/mL, p=0.0362, n=5). In conclusion, mice containing human pancreatic tumors have significant elevated levels of total MP TF activity and TAT. Levels of human TF activity in MPs correlate with levels of human TF antigen and activation of coagulation. Tumors are a major source of TF-positive MPs in a mouse model.