Von Willebrand factor, ADAMTS13 levels and prediction of venous thromboembolism in patients with cancer

M. Pepin, A. Kleinjan, D. Hajage, H.R. Büller, M. DiNisio, P.W. Kamphuisen, I. Mahe, A. Stepanian

Research output: Contribution to journalMeeting AbstractAcademic


Background: Cancer patients are at high risk for venous thromboembolism (VTE). However, thromboprophylaxis in these patients is associated with an increased hemorrhagic risk. The Khorana score is a risk scoring model for prediction of VTE that includes clinical and laboratory parameters. It has been expanded by incorporating two biomarkers, soluble P-selectin (P-sel), and D-Dimer. Other laboratory markers like prothrombin fragment 1 + 2 (F1 + 2) have also been reported to predict VTE in cancer patients. von Willebrand factor (VWF) is a mutimeric protein that promotes platelet adhesion and aggregation in high shear stress conditions; it is also implicated in venous thrombosis. ADAMTS13 (a desintegrin-like and métalloprot éase thrombospondin type 1 motif) specifically cleaves VWF multimers and subsequently regulates its activity; previous studies displayed decreased levels of ADAMTS13 in cancer patients when compared to patients without cancer. Aims: (i) to search for an association between VWF and ADAMTS13 levels and VTE in cancer patients; (ii) to compare Khorana score and expanded Khorana score before and after addition of VWF and ADAMTS13 levels. Methods: multicenter case-control study. Subjects were ambulatory patients receiving chemotherapy for stage III or IV cancer, prospectively followed for 6 months for the development of VTE. Each case (patient who developed VTE) was matched with seven controls (patients who did not develop VTE) for age, sex, type of cancer and stage. The Khorana score and the expanded score were calculated. ADAMTS13 activity was measured using a fluorescence resonance energy transfer assay; VWF, ADAMTS13 antigen, P-sel, D-Dimer and F1 + 2 levels were measured using ELISA methods. Univariate and multivariate analysis were performed to compare the levels between cases and controls. Logistic models were used to test whether VWF and ADAMTS13 levels would add significant prognostic information to Khorana and expanded Khorana score. Results are expressed as odds ratios (OR) and their 95% CIs. Results: We studied 20 cases and 140 controls. VWF levels were significantly higher in cases when compared to controls (326 ± 185% vs. 242 ± 158%; P = 0.02), whereas ADAMTS13 levels were not significantly different (activity 87 ± 18.9% vs. 81.5 ± 18.9; antigen 579 ± 108 ng/mL vs. 564 ± 137 ng/mL). VWF and ADAMTS13 levels were not correlated (r = -0.06 Pearson). VWF levels significantly increased with increasing stage of cancer (P = 0.03) whereas ADAMTS13 activity levels only tended to decrease (P = 0.06). The addition of VWF levels significantly improved the Khorana score (OR = 3.5 [1.2-10.2]). ADAMTS13 activity levels significantly improved the Khorana and the expanded Khorana scores (OR = 5 [1.4-23.5]) and 5.8 [1.4-28.2], respectively). Summary/Conclusions: Prospective studies are needed to confirm our results and determine if VWF and ADAMTS13 can be used as predictive markers of VTE in stage III/IV cancer patients. Our results also highlight a role of VWF and ADAMTS13 in the pathophysiology of VTE in cancer.
Original languageEnglish
Pages (from-to)985-986
Number of pages2
JournalJournal of Thrombosis and Haemostasis
Publication statusPublished - Jul-2013


  • von Willebrand factor
  • antigen
  • D dimer
  • marker
  • protein
  • prothrombin
  • thrombospondin
  • PADGEM protein
  • biological marker
  • prediction
  • venous thromboembolism
  • patient
  • human
  • neoplasm
  • society
  • thrombosis
  • hemostasis
  • cancer patient
  • risk
  • laboratory
  • thrombocyte adhesion
  • statistical model
  • multivariate analysis
  • pathophysiology
  • shear stress
  • assay
  • fluorescence resonance energy transfer
  • chemotherapy
  • case control study
  • vein thrombosis
  • parameters
  • prospective study
  • model
  • enzyme linked immunosorbent assay

Cite this