Human pancreatic tumors grown in mice release tissue factor‐positive microvesicles that increase venous clot size

Abstract

Background Pancreatic cancer patients have a high rate of venous thromboembolism. Human pancreatic tumors and cell lines express high levels of tissue factor (TF), and release TF-positive microvesicles (TF MVs). In pancreatic cancer patients, tumor-derived TF MVs are present in the blood, and increased levels are associated with venous thromboembolism and decreased survival. Previous studies have shown that mice with orthotopic human or murine pancreatic tumors have circulating tumor-derived TF MVs, an activated clotting system, and increased incidence and mean clot weight in an inferior vena cava stenosis model. These results suggest that TF MVs contribute to thrombosis. However, the specific role of tumor-derived TF MVs in venous thrombosis in mice has not been determined. Objectives To test the hypothesis that tumor-derived TF MVs enhance thrombosis in mice. Methods We determined the contribution of TF MVs derived from human pancreatic tumors grown orthotopically in nude mice to venous clot formation by using an anti-human TF mAb. We used an inferior vena cava stasis model of venous thrombosis. Results Tumor-bearing mice had significantly larger clots than control mice. Clots from tumor-bearing mice contained human TF, suggesting the incorporation of tumor-derived MVs. Importantly, administration of an anti-human TF mAb reduced clot size in tumor-bearing mice but did not affect clot size in control mice. Conclusions Our results indicate that TF MVs released from orthotopic pancreatic tumors increase venous thrombosis in mice. This new model may be useful for evaluating the roles of different factors in cancer-associated thrombosis.

Mouse models of cancer-associated thrombosis

“…In another study, we used the human pancreatic cell line BxPc-3 [14]. Similar to the studies with HPAF-II, we found that nude mice bearing orthotopic BxPc-3 tumors had increased levels of TAT complex [14, 28]. Importantly, administration of an anti-human TF monoclonal antibody abolished the increase in TAT complex [13].…”

“…The most popular are those involving the infrarenal vena cava (IVC). An advantage of the IVC is that high frequency ultrasonography can be used for non-invasive longitudinal monitoring of thrombus formation [14, 27, 28]. Clot formation is typically analyzed between 0–48 hours.…”

“…In a second study, we found that nude mice bearing orthotopic BxPC-3 tumors had an increased clot area at 3, 6, 9 and 24 hours after IVC stenosis but the incidence of clots in tumor-bearing mice was not significantly increased compared to control mice [14]. Due to the variability of the IVC stenosis model, we changed to the IVC stasis model to study the role of tumor-derived TF+ MV in cancer-associated thrombosis [28]. In addition, we generated a new line of BxPc-3 that expresses luciferase to allow non-invasive monitoring of tumor size.…”

“…In addition, we did not observe an increase in clot weight in nude mice bearing either BxPc-3 or AsPc-1 subcutaneous tumors compared to clots in control mice (Figure 1). In contrast, we observed that nude mice bearing orthotopic BxPc-3 tumors had significantly larger clots at 48 hours compared with clots in control mice [28]. Analysis of the clot area at different time points revealed an increase in the clot area in tumor-bearing mice at 24 and 48 hours but not at 3 and 6 hours compared with mice without tumors [28].…”

“…In contrast, we observed that nude mice bearing orthotopic BxPc-3 tumors had significantly larger clots at 48 hours compared with clots in control mice [28]. Analysis of the clot area at different time points revealed an increase in the clot area in tumor-bearing mice at 24 and 48 hours but not at 3 and 6 hours compared with mice without tumors [28]. We believe that at 3 and 6 hours clot formation is primarily driven by vessel wall TF exposed during the surgery.…”