Shedding Light on Hemostasis in Patients With Inflammatory Bowel Diseases

Journal Pre-proof Shedding Light on Hemostasis in Patients With Inflammatory Bowel Diseases Jeremy Lagrange, PhD, Patrick Lacolley, MD, Denis Wahl, Laurent Peyrin-Biroulet, MD, Véronique Regnault, PhD

PII: DOI: Reference:

S1542-3565(20)30056-2 https://doi.org/10.1016/j.cgh.2019.12.043 YJCGH 56961

To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 31 December 2019 Please cite this article as: Lagrange J, Lacolley P, Wahl D, Peyrin-Biroulet L, Regnault V, Shedding Light on Hemostasis in Patients With Inflammatory Bowel Diseases, Clinical Gastroenterology and Hepatology (2020), doi: https://doi.org/10.1016/j.cgh.2019.12.043. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 by the AGA Institute

Title: Shedding Light on Hemostasis in Patients With Inflammatory Bowel Diseases

Jeremy Lagrange, PhD1,2,3, Patrick Lacolley, MD1,2,4, Denis Wahl1,2,5, Laurent PeyrinBiroulet, MD2,6,7, Véronique Regnault, PhD1,2,4 1

INSERM U1116, Faculté de Médecine, Vandœuvre-lès-Nancy, France Université de Lorraine, Nancy, France 3 Center for Thrombosis and Hemostasis, University Medical Center Mainz, Germany 4 CHRU Nancy, Vandœuvre-lès-Nancy, France 5 Division of Vascular Medicine, CHRU Nancy, Vandœuvre-lès-Nancy, France 6 INSERM U1256, Faculté de Médecine, Vandœuvre-lès-Nancy, France 7 Department of Gastroenterology, CHRU Nancy, Vandœuvre-lès-Nancy, France 2

Correspondance: Jeremy Lagrange PhD, INSERM 1116 DCAC 9 avenue de la forêt de Haye 54 505 Vandœuvre-lès-Nancy, France [email protected] Abstract : 158 words Word count (references included) : 5990 Key words: inflammatory bowel disease; thrombosis; platelet; fibrinolysis Conflicts of interest: The authors state that they have no conflict of interest. Authors contribution: J.L; original draft preparation, P.L, D.W and L.P.B; critical review of the manuscript, VR; critical review and editing.

1

Abstract: Patients with inflammatory bowel diseases (IBD) have an increased risk of thrombosis, possibly due to changes in blood cells and molecules involved in hemostasis. They have increased platelet counts and reactivity as well as increased platelet-derived large extracellular vesicles. Coagulation is continuously activated in patients with IBD, based on measured markers of thrombin generation, and the anticoagulant functions of endothelial cells are damaged. Furthermore, fibrinogen is increased and fibrin clots are denser. However, pathogenesis of thrombosis in patients with IBD appears to differ from that of patients without IBD. Patients with IBD also take drugs that might contribute to risk of thrombosis, complicating the picture. We review the features of homeostasis that are altered in patients with IBD and possible mechanisms of this relationship.

Introduction

2

Venous thromboembolism (VTE) affects between 0.5 and 2 persons per 1000 in the general population and is a major cause of death in hospitalized patients.1 Independent risk factors for VTE include age, sex, genetic factors (such as mutated factor V Leiden and prothrombin G20210A, or deficiency in antithrombin, or protein C or S), surgery or trauma. Inflammatory bowel disease (IBD), which includes Crohn’s disease (CD) and ulcerative colitis (UC), is also an independent risk factor for first and recurrent VTE and large population-based studies have shown that the risks of both pulmonary embolism (PE) and VTE are several-fold higher in patients with IBD compared with the general population.2 Meta-analyses based on large populations have revealed that the risk for VTE was increased more than two times in IBD patients.3,4 VTE is a life threatening complication of IBD and hospitalized patients display higher risk of VTE5. Indeed, 80% of VTE occurred in patients with active IBD and hospitalized IBD patients with VTE had increased risk of mortality compared to hospitalized IBD patients without VTE (OR = 2.5; 95% CI: 1.83-3.43) and even in comparison to non-IBD patients with VTE (OR= 2.1; 95% CI: 1.6-2.9)6. IBD leads to an increased risk of recurrent VTE compared to patients with no IBD. After a first VTE event the probability of recurrence 5 years after discontinuation of anticoagulation therapy was higher among patients with IBD than patients without IBD.7 Disease activity in IBD is also an independent risk factor for VTE. Venous events are not the only thrombotic complications in IBD patients: arterial thrombotic events (myocardial infarction, splanchnic ischemia, stroke) are also increased in IBD patients and in particular in hospitalized patients.8 Active disease amplifies the arterial thrombotic risk in IBD as well as for VTE. In addition, the VTE risk directly related to the disease, recent reports suggest that newly developed drugs could also participate in the increased risk of VTE9,10. Contrary to the general population a large group of IBD patients develop VTE for no identifiable reason11. Herein we provide an extensive review of data collected in IBD patients about hemostasis alterations that we will separated between (1) platelets, (2) coagulation, and (3) fibrinolysis, in order to summarize the relevant parameters implicated in increased VTE risk. Platelets Quantitative, qualitative and functional platelet alterations Increased platelet count in IBD was first reported in 196812 and described since in many other studies (Figure 1)13–18 and was regarded as reactive thrombocytosis. Platelet count was considered as a marker to distinguish IBD from infective diarrhea.19 In the dextran sulfate sodium (DSS)-induced mouse model of colitis, platelet count and platelet turnover were increased.20 Both circulating activated platelets and platelets aggregated with neutrophils, monocytes, or lymphocytes were elevated following DSS treatment. Mean platelet volume (MPV) is also affected by reactive thrombocytosis. MPV decrease is associated with an increased thrombopoiesis. There are two possible explanations for this MPV decrease in IBD: elevated inflammation activates megakaryopoiesis leading to an accelerated production and decreased half-life of platelets21, or, larger platelets that are known to be more active could infiltrate sites of inflammation leading to a lower MPV of blood platelets. MPV is related both to inflammatory and thrombotic disorders.22 MPV was often reported as being decreased and associated with disease activity in IBD.23–25 However, a sex difference might exist with females being less affected by MPV variations.17 Interestingly MPV was found to be a good predictor in CD patients for sustained response to anti-tumor necrosis factor (TNF)-α therapy (Infliximab).26 Concerning platelet functions, activation and aggregation were found to be increased in IBD with an effect of diseased activity.27 3

Others found an increased aggregation response even in inactive-phases.18 Markers of platelet activation such as platelet factor 4 (PF4) were found to be increased and even correlated with activity of the disease.27,28 It was shown that without any stimulation, platelets from IBD patients presented spontaneous aggregation and hypersensitivity.29 More recently proteaseactivated receptors (PAR)-1 and -4 that mediate thrombin signaling were found to increase platelet responsiveness in whole blood platelet aggregometry in CD.30 Another factor that could influence platelet reactivity is von Willebrand Factor (VWF). Endothelial cells produce 80% of total VWF, the remaining 20% being synthesized by megakaryocytes and stored in platelets. VWF is important for thrombus formation via binding and transport of factor VIII (FVIII) and by mediating platelet activation and adhesion via the platelet receptor glycoprotein (GP) Ibα.31 IBD are associated with increased circulating levels of VWF and a decrease in the level and activity of a disintegrin and metalloprotease with thrombospondin repeats, member 13 (ADAMTS13)32, which cleaves large VWF multimers to smaller, lessactive multimers, thereby determining VWF hemostatic and proinflammatory potential. VWF plasma levels are also increased with disease activity in IBD.33 The complex formed between VWF and ADAMTS13 was increased in active and in inactive disease states.32 In UC patients an increased VWF in active disease was considered as a marker of endothelial dysfunction triggered by systemic inflammation.34 Importantly a lower incidence of IBD in patients with inherited bleeding disorders, such as hemophilia A (FVIII deficiency) and B (FIX deficiency), and von Willebrand’s disease has been reported.35 Finally, in mice lacking ADAMTS13 DSSinduced colitis was worsened while administration of ADAMTS13 decreased colitis severity indicating a direct role of VWF in IBD pathogenesis.36 Platelet contribution to inflammation as well as platelet-derived large extracellular vesicles (P-LEV) in IBD are described in the supplement. In summary, IBD patients present clear increased platelet count and decreased MPV with higher reactivity which could participate in the increased thrombotic risk. Inflammation increases platelet responsiveness through CD40 and PARs pathways and P-LEV having an important procoagulant activity are highly increased. Moreover, given the involvement of platelets in coagulation, inflammation and immune responses, antiplatelet drugs could positively affect the clinical course of IBD by reducing both the procoagulant and inflammatory states. Interestingly, less frequent rectal bleeding was reported in IBD patients taking antiplatelet therapy which highlights the possible role of platelets as modulators independent of their hemostasis function.37 In the general population platelet count was found to be associated with arterial thrombosis but not with venous thrombosis38 while in IBD patients platelet count was associated with the overall risk of thrombosis.39 Coagulation Inhibition of the coagulation cascade has been taken into account for a long time by physicians dealing with IBD. The use of low-molecular weight heparin (LMWH) is part of the common prophylaxis administered to all patients hospitalized following disease flare. The benefit of LMWH is still discussed but several studies reported a use in active but not in moderately active UC. 40,41 A summary of the primary and secondary hemostasis is visible in the supplement. In vivo coagulation activation and thrombin formation In vivo markers of thrombin generation, prothrombin fragments 1 and 2 (F1+2) generated during the conversion of prothrombin to thrombin and thrombin-antithrombin complexes (TAT) have been measured in some publications (Table 1). All publications reported an 4

elevation of TAT and F1+2 both in active CD and UC while changes were not clear in inactive diseases.15–17,42–44 This implies that there is upregulation of the coagulation system and thus increased thrombin formation during the active disease state, though this is less clear in periods of disease inactivity. Prothrombin concentration was indifferent in IBD patients45,46. Concerning the measurements of antithrombin (AT); they showed high variations and were contradictory depending on the study.13,15,46,47 The thrombin-generating capacity assessed in vitro by CAT was not significantly different compared to controls but significantly higher after addition of thrombomodulin (TM), to evaluate inhibition through the APC pathway, and adjustment to CRP levels.45 A second report described also a markedly increased thrombin generation only in acute-phase IBD.48 In pediatric patients, thrombin generation was increased but with a prolonged lag time and correlated with disease activity.49 Therefore, it remains unclear whether the thrombin-generating capacity is increased among IBD patients. Other diseases such as antiphospholipid antibodies and common genetic mutations of coagulation factors (FV Leiden, prothrombin G20210A) affecting thrombin formation and risk of VTE are discussed in the supplement. All together, the current findings on thrombin generation changes are not conclusive to explain the increased risk of VTE and arterial thrombosis in IBD patients despite the increased in vivo activation of coagulation highlighted by increased in vivo markers. Extrinsic pathway In IBD where the thrombotic risk is increased, modifications of an important number of coagulation factors have been reported.13,14,47 The literature for the coagulation factor changes are presented in the supplementary table. TF, the trigger for the extrinsic pathway of coagulation, is present in the sub-endothelial layers of the vascular wall (Figure 2). Alterations of the anticoagulant properties of the vascular endothelium are discussed in the supplement. In blood, the so-called blood-borne TF can be expressed by activated monocytes and by LEV.50 In IBD, TF activity related to monocytes as well as LEV exposing TF were found to be increased. 51,52 However, LEV exposing TF were not associated with in vivo coagulation activation (indirectly accessed by plasma levels of D-dimer, a fibrinolysis product). TF plasma activity was correlated to the level of TAT, which is another marker for in vivo coagulation activation, but not with inflammatory markers or clinical activity in IBD.53 Interestingly TF was found to be detrimental in experimental colitis.54,55 The use of an antibody against TF in the DSS model of colitis or genetic ablation of TF with 1% remaining TF activity to avoid embryonic lethality has been reported to limit the immune cell influx into the colon and to lower inflammation and organ damage. Regarding coagulation FVII which forms a complex with TF to activate FX, contradictory results have been published. Yazici et al did not find a significant difference for FVII activity in IBD patients while Hudson et al found FVII activity significant increased.13,56 In both studies a difference appeared between CD and UC patients with a higher procoagulant activity of FVII in CD patients. FX was found to be increased in active UC and increased FV was described for both CD and UC.14,15,47 Activated FX (FXa) in concert with FVa can convert prothrombin to thrombin, the key enzyme of the coagulation cascade. Activated FX was not correlated to clinical activity of IBD.53 Intrinsic pathway Again the variation of circulating FXII concentration varies in IBD.15,16 FXI is one of the markedly increased coagulation factors in active and inactive disease (Figure 2 and supplementary table).15,16 FVIII procoagulant activity was reported to be increased in most 5

publications.13–15,45,47 The effect of this coagulation factor on the occurrence of deep-vein thrombosis is well documented and could have a pivotal role in the thrombotic risk associated with IBD.57 However elevated FVIII was not found to be an independent predictor of the risk of recurrent VTE.7 TF was not the only coagulation factor capable of limiting DSS-induced colitis in mice. Indeed FIX deficient mice presented a decreased colonic neutrophil infiltration.58 Moreover, the intestinal epithelium was described to be able to synthesize FIX. Local synthesis of coagulation factors could increase PARs activation and induce detrimental cellular effects. Intriguingly, mice lacking FXII were not protected against DSS-induced colitis.59 This dichotomy in the intrinsic pathway may be the result of thrombin-driven FXI activation which was described to trigger and amplify vascular inflammation independently of FXII.60 Indeed, mice lacking FXI where protected against angiotensin II-induced vascular inflammation while mice lacking FXII display normal vascular inflammation. Further studies should focus on the role of ectopic synthesis of coagulation factors triggered by active IBD. In summary, no strong definitive data have been published on most coagulation factors and most of the conclusions were based on a limited number of patients. Large-scale studies with higher numbers of patients and systematic measurement of coagulation factors in relation with age and disease activity will be necessary to conclude on the possible causal effect of some coagulation factors in increasing the thrombotic risk in IBD. The relevance of thrombin generation was established in the prediction of a first VTE event but until now no longitudinal studies have been conducted to evaluate the usefulness of this tool to stratify thrombotic risk in IBD patients.61 Moreover, some coagulation factors (TF, FXa, thrombin) are now well described to have direct or indirect cellular effects through the PARs and could play a role on microvascular alterations present in IBD patients.62 Finally, other autoimmune diseases, such as the antiphospholipid syndrome, could also have an increased prevalence in IBD and have a role in the pathogenesis of thrombosis. Very few data are available concerning the direct relation between coagulation proteins and VTE risk in IBD patients. Protein S deficiency is associated with a very high VTE risk while other factors as FVIII were not statistically different39 Fibrin clot formation and fibrinolysis alteration Fibrin clot formation Discrepant results have been published regarding the circulating concentrations of fibrinogen, some studies reporting increased fibrinogen in CD and UC13,16,46,49 and others only in CD13,56 or UC.17 These differences might result from the fact that fibrinogen concentration is well known to be correlated with inflammation which is highly variable depending on the disease activity state. In order to stabilize the fibrin clot, FXIII forms cross-links with fibrin polymers. FXIII was decreased in active IBD patients and correlated with disease activity.44,63,64 Decreased circulating levels of FXIII may result from an increased consumption due to the formation of microthrombi. As for in vivo markers of fibrin formation, results from laboratory tests for plasma clotting (APTT: activated partial thromboplastin time, PT: prothrombin time) were not consistent (Table 1). Some reported no changes in APTT or PT in CD or UC patients while others presented slightly or significant delayed clotting times.13,15–17,46,47 No study has presented faster clotting times. Fibrinolysis 6

Fibrinolysis is the pathway counteracting coagulation and degrading the fibrin clot. The key molecule of fibrinolysis is plasmin. Its inactive zymogen (plasminogen) is activated mainly by tissue plasminogen activator (t-PA) or urokinase plasminogen activator (uPA). Plasmin in return can be inhibited by plasminogen activator inhibitor-1 and 2 (PAI-1, -2) via inhibition of tPA and uPA. Hypofibrinolysis and alterations in thrombus resolution constitute a risk factor for VTE.65 As for changes in coagulation, a reduction in the fibrinolytic capacity of IBD patients is not clear but some findings suggest an imbalance of fibrinolysis (Table 2). Plasminogen is unchanged14–16,46 while variations for tPA are inconsistent since some studies show an increase16,44 and others a decrease for both antigen and activity46,66 or no variations at all.67,68 PAI-1 concentrations were measured in several studies and discrepant findings have been reported. However, PAI-1 is most likely increased in IBD.16,44,46,67–70 Recently Kaiko et al. showed that PAI-1 expression was highly elevated in IBD patients with active disease who did not respond to anti-TNF therapy.71 PAI-1 exacerbated mucosal damage by blocking tPA and inhibition of PAI-1 reduced both mucosal damage and inflammation. PAI-1 can interact with specific matrix components such as vitronectin or low-density lipoprotein-1 and uPA, with its receptor, play a role in cell migration.72 PAI-1 inhibition was found to decrease vascular smooth muscle cell migration and neointima formation.73 These cellular effects may have a role in IBD but have not yet been studied. Very recently PAI-1 was discovered to be able to directly bind FXIa on the surface of endothelial cells, thereby inducing its clearance and degradation.74 This new mechanism emphasizing a crosstalk between coagulation, fibrinolysis and endothelial function may be impaired in IBD patients as they exhibit alterations in all these processes. In line with the PAI-1 increase suggesting a decrease of fibrinolysis, thrombin-activatable fibrinolysis inhibitor (TAFI) was found to be decreased in IBD patients.69 Focusing on clot lysis is a more integrative way to evaluate fibrinolysis compared with separate measurements of fibrinolysis-related molecules. Indeed, IBD patients display a denser fibrin network with prolonged clot lysis time (Figure 3).75 Moreover, in IBD patients with a history of thromboembolic events the area under the fibrinolytic curve, which reflects the overall coagulation/fibrinolysis profile, was increased.75 Fibrinolysis alteration is very likely related to the thrombotic risk associated with IBD and could be used as a parameter for risk assessment. D-dimer results from fibrin degradation and are fibrinolysis markers. Most studies in IBD patients presented increased D-dimers13,15,16,44,67,70,76,77 but some did not find any differences.14,34,47 (Table 2). However, in all publications focusing on D-dimers and diseased activity a positive correlation was reported which is also in favor of a continuous activation of the coagulation during disease flares. In summary, fibrinogen is increased in active IBD and clot formation is affected with an elevated consumption of FXIII while the coagulation times are not significantly impaired. However, fibrinolysis is decreased as shown with PAI-1 elevation. The balance between coagulation activation and fibrinolysis conterregulation is dysregulated towards a prothrombotic phenotype. In the study of Andrade et al, as for FVIII, fibrinogen concentration was not found to be significantly associated with the occurrence of VTE in IBD patients. Concerning fibrinolysis itself no data are available. Conclusion In IBD patients VTE may be considered unprovoked in the absence of recent surgery with general/spinal/epidural anesthesia for longer than 30 min, hospital admission and confinement to bed (only bathroom privileges) for acute illness for at least three days, or active malignancy (i.e., no curative treatment received, ongoing curative treatment, or recurrence/progression 7

despite curative treatment), and patients with an age < 65. In this situation minor clinical risk factors should be investigated together with laboratory workup, including proteins C and S and antithrombin deficiencies, factor V Leiden and 20210 prothrombin gene mutations as well as assays for antiphospholipid antibodies. Unprovoked proximal deep vein thrombosis or pulmonary embolism, recurrent VTE and major thrombophilia such as antithrombin deficiencies or antiphospholipid syndrome may require long term anticoagulant treatment. On the fundamental side, until now clear changes in the coagulation system which were reported by several groups for active CD and UC include increased TAT, F1+2 and D-dimer for the in vivo markers of coagulation and fibrinolysis and elevated circulating FV, FVIII, VWF, fibrinogen and PAI-1. However, coagulation changes are hardly related to the procoagulant state of IBD patients since clotting times are unchanged or even delayed despite an important increase in fibrinogen concentration. Substantial increases in terms of platelet reactivity and the clear increased platelet count in IBD patients argues strongly for a crucial role of platelets in the increased thrombotic risk. P-LEV are also good candidates as contributors to this increased risk since they can act at different vascular beds far from the ones where platelets are aggregated. Fibrinolysis was also identified as strongly altered in IBD. Additionally, it is clear that inconsistent data have been reported in different studies performed on the hemostasis system in IBD. Some of the discrepancies could originate from pre-analytical abnormalities in the preparation of blood and plasma samples. Therefore, further studies should follow the guidelines from the Scientific and Standardization Committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH)78–80. Finally, the endothelial barrier was impaired with decreased surface expression of anticoagulant proteins. So far, no extensive review has been completed on the effect of IBD treatments in active and inactive diseases and on the thrombotic risk. A recent meta-analysis on the VTE risk in IBD after steroid or anti-TNF-α medications reported a higher risk after steroid treatment but a protection with anti-TNF-α.81 Important open questions are summarized below. This review provides a framework for better understanding of how IBD affects hemostasis and in particular platelet and fibrinolytic function. In the upcoming years, studies should be designed in a way that integrate all blood cells and components in IBD risk stratification since the use of individual plasma components or platelet activation are not sufficient to give a complete overview of the blood-dependent procoagulant state.

Open research questions Could antiplatelet therapy improve disease activity? Is there an indication for anticoagulant treatment in the active phase of IBD? How do cellular effects of coagulation factors (tissue factor, activated factor X, thrombin) affect IBD pathogenesis? Are other autoimmune diseases augmented in IBD and worsening the risk of thrombosis? Search strategy and selection criteria: Each protein and cell of hemostasis was searched in PubMed with the terms inflammatory bowel disease, Crohn’s disease or ulcerative colitis and also with the abbreviation of each term. Only papers published in English were reviewed. The final reference list was generated on the basis of relevance to the broad scope of this Review. Acknowledgements: The authors thank Professor S.N. Thornton for revision of the English. 8

References 1. Heit, J. A. et al. The epidemiology of venous thromboembolism in the community. Thromb. Haemost. 86, 452–463 (2001). 2. Murthy, S. K. & Nguyen, G. C. Venous thromboembolism in inflammatory bowel disease: an epidemiological review. Am. J. Gastroenterol. 106, 713–718 (2011). 3. Nguyen, G. C. et al. Consensus statements on the risk, prevention, and treatment of venous thromboembolism in inflammatory bowel disease: Canadian Association of Gastroenterology. Gastroenterology 146, 835-848.e6 (2014). 4. Fumery, M. et al. Thromboembolic events and cardiovascular mortality in inflammatory bowel diseases: a meta-analysis of observational studies. J Crohns Colitis 8, 469–479 (2014). 5. Zezos, P., Kouklakis, G. & Saibil, F. Inflammatory bowel disease and thromboembolism. World J Gastroenterol 20, 13863–13878 (2014). 6. Solem, C. A., Loftus, E. V., Tremaine, W. J. & Sandborn, W. J. Venous thromboembolism in inflammatory bowel disease. Am. J. Gastroenterol. 99, 97–101 (2004). 7. Novacek, G. et al. Inflammatory bowel disease is a risk factor for recurrent venous thromboembolism. Gastroenterology 139, 779–787, 787.e1 (2010). 8. Kirchgesner, J. et al. Increased risk of acute arterial events in young patients and severely active IBD: a nationwide French cohort study. Gut 67, 1261–1268 (2018). 9. Desai, R. J., Pawar, A., Weinblatt, M. E. & Kim, S. C. Comparative Risk of Venous Thromboembolism in Rheumatoid Arthritis Patients Receiving Tofacitinib Versus Those Receiving Tumor Necrosis Factor Inhibitors: An Observational Cohort Study. Arthritis & Rheumatology (Hoboken, N.J.) 71, 892–900 (2019). 10. Sandborn, W. J. et al. Venous thromboembolic events in the tofacitinib ulcerative colitis clinical development programme. Aliment. Pharmacol. Ther. 50, 1068–1076 (2019). 11. Jackson, L. M. et al. Thrombosis in inflammatory bowel disease: clinical setting, procoagulant profile and factor V Leiden. QJM 90, 183–188 (1997). 12. Morowitz, D. A., Allen, L. W. & Kirsner, J. B. Thrombocytosis in chronic inflammatory bowel disease. Ann. Intern. Med. 68, 1013–1021 (1968). 13. Yazici, A. et al. Thrombophilic Risk Factors in Patients With Inflammatory Bowel Disease. Gastroenterology Res 3, 112–119 (2010). 14. Dolapcioglu, C. et al. Coagulation parameters in inflammatory bowel disease. Int J Clin Exp Med 7, 1442–1448 (2014). 15. Kume, K., Yamasaki, M., Tashiro, M., Yoshikawa, I. & Otsuki, M. Activations of coagulation and fibrinolysis secondary to bowel inflammation in patients with ulcerative colitis. Intern. Med. 46, 1323–1329 (2007). 16. Alkim, H., Ayaz, S., Alkim, C., Ulker, A. & Sahin, B. Continuous active state of coagulation system in patients with nonthrombotic inflammatory bowel disease. Clin. Appl. Thromb. Hemost. 17, 600–604 (2011). 17. Shen, J. et al. Biomarkers of altered coagulation and fibrinolysis as measures of disease activity in active inflammatory bowel disease: A gender-stratified, cohort analysis. Thrombosis Research 123, 604–611 (2009). 18. Andoh, A. et al. Increased aggregation response of platelets in patients with inflammatory bowel disease. J. Gastroenterol. 41, 47–54 (2006). 19. Harries, A. D., Beeching, N. J., Rogerson, S. J. & Nye, F. J. The platelet count as a simple measure to distinguish inflammatory bowel disease from infective diarrhoea. Journal of Infection 22, 247–250 (1991). 20. Yan, S. LS. et al. Platelet abnormalities during colonic inflammation. Inflamm Bowel Dis 19, 1245–1253 (2013). 9

21. Talstad, I., Rootwelt, K. & Gjone, E. Thrombocytosis in ulcerative colitis and Crohn’s disease. Scand. J. Gastroenterol. 8, 135–138 (1973). 22. Aliosmanoglu, I. et al. Can mean platelet volume be a new risk factor in portal venous thrombosis? Clin. Appl. Thromb. Hemost. 19, 433–436 (2013). 23. Liu, Z. et al. Prevention of experimental colitis in SCID mice reconstituted with CD45RBhigh CD4+ T cells by blocking the CD40-CD154 interactions. J. Immunol. 164, 6005–6014 (2000). 24. Järemo, P. & Sandberg-Gertzen, H. Platelet density and size in inflammatory bowel disease. Thromb. Haemost. 75, 560–561 (1996). 25. Kapsoritakis, A. N. et al. Mean platelet volume: a useful marker of inflammatory bowel disease activity. Am. J. Gastroenterol. 96, 776–781 (2001). 26. Sobolewska, A., Włodarczyk, M., Stec-Michalska, K., Fichna, J. & WiśniewskaJarosińska, M. Mean Platelet Volume in Crohn’s Disease Patients Predicts Sustained Response to a 52-Week Infliximab Therapy: A Pilot Study. Dig Dis Sci 61, 542–549 (2016). 27. Collins, C. E., Cahill, M. R., Newland, A. C. & Rampton, D. S. Platelets circulate in an activated state in inflammatory bowel disease. Gastroenterology 106, 840–845 (1994). 28. Ye, L. et al. Serum platelet factor 4 is a reliable activity parameter in adult patients with inflammatory bowel disease. Medicine (Baltimore) 96, (2017). 29. Webberley, M. J., Hart, M. T. & Melikian, V. Thromboembolism in inflammatory bowel disease: role of platelets. Gut 34, 247–251 (1993). 30. Schmid, W. et al. Increased responsiveness to thrombin through protease-activated receptors (PAR)-1 and -4 in active Crohn’s disease. J Crohns Colitis 8, 495–503 (2014). 31. Broos, K., Feys, H. B., De Meyer, S. F., Vanhoorelbeke, K. & Deckmyn, H. Platelets at work in primary hemostasis. Blood Rev. 25, 155–167 (2011). 32. Cibor, D. et al. Levels and activities of von Willebrand factor and metalloproteinase with thrombospondin type-1 motif, number 13 in inflammatory bowel diseases. World J. Gastroenterol. 23, 4796–4805 (2017). 33. Stevens, T. R. et al. Circulating von Willebrand factor in inflammatory bowel disease. Gut 33, 502–506 (1992). 34. Zezos, P. et al. Elevated plasma von Willebrand factor levels in patients with active ulcerative colitis reflect endothelial perturbation due to systemic inflammation. World J. Gastroenterol. 11, 7639–7645 (2005). 35. Thompson, N. P., Wakefield, A. J. & Pounder, R. E. Inherited disorders of coagulation appear to protect against inflammatory bowel disease. Gastroenterology 108, 1011–1015 (1995). 36. Zitomersky, N. L. et al. ADAMTS13 Deficiency Worsens Colitis and Exogenous ADAMTS13 Administration Decreases Colitis Severity in Mice. TH Open 1, e11–e23 (2017). 37. Vinod, J. et al. The effect of antiplatelet therapy in patients with inflammatory bowel disease. J. Clin. Gastroenterol. 46, 527–529 (2012). 38. Warny, M., Helby, J., Birgens, H. S., Bojesen, S. E. & Nordestgaard, B. G. Arterial and venous thrombosis by high platelet count and high hematocrit: 108 521 individuals from the Copenhagen General Population Study. J. Thromb. Haemost. 17, 1898–1911 (2019). 39. Andrade, A. R. et al. Risk of thrombosis and mortality in inflammatory bowel disease. Clin Transl Gastroenterol 9, (2018). 40. de Bièvre, M. A. et al. Randomized, placebo-controlled trial of low molecular weight heparin in active ulcerative colitis. Inflamm. Bowel Dis. 13, 753–758 (2007).

10

41. Chande, N., McDonald, J. W., Macdonald, J. K. & Wang, J. J. Unfractionated or lowmolecular weight heparin for induction of remission in ulcerative colitis. Cochrane Database Syst Rev CD006774 (2010) doi:10.1002/14651858.CD006774.pub3. 42. Chamouard, P. et al. Prothrombin fragment 1 + 2 and thrombin-antithrombin III complex as markers of activation of blood coagulation in inflammatory bowel diseases. Eur J Gastroenterol Hepatol 7, 1183–1188 (1995). 43. Smith, C. J., Haire, W. D., Kaufman, S. S. & Mack, D. R. Determination of prothrombin activation fragments in young patients with inflammatory bowel disease. Am. J. Gastroenterol. 91, 1221–1225 (1996). 44. Hayat, M., Ariëns, R. A. S., Moayyedi, P., Grant, P. J. & O’Mahony, S. Coagulation factor XIII and markers of thrombin generation and fibrinolysis in patients with inflammatory bowel disease. Eur J Gastroenterol Hepatol 14, 249–256 (2002). 45. Saibeni, S. et al. Increased thrombin generation in inflammatory bowel diseases. Thromb. Res. 125, 278–282 (2010). 46. de Jong, E., Porte, R. J., Knot, E. A., Verheijen, J. H. & Dees, J. Disturbed fibrinolysis in patients with inflammatory bowel disease. A study in blood plasma, colon mucosa, and faeces. Gut 30, 188–194 (1989). 47. Lam, A., Borda, I. T., Inwood, M. J. & Thomson, S. Coagulation studies in ulcerative colitis and Crohn’s disease. Gastroenterology 68, 245–251 (1975). 48. Saladino, V. et al. Increased Thrombin Generation in Inflammatory Bowel Diseases. Gastrointestinal Endoscopy 67, AB319 (2008). 49. Bernhard, H. et al. Calibrated automated thrombin generation in paediatric patients with inflammatory bowel disease. Hamostaseologie 29 Suppl 1, S90-93 (2009). 50. Østerud, B. & Bjørklid, E. Blood-Borne Tissue Factor (Including Microparticles). (Landes Bioscience, 2013). 51. Edwards, R. L. et al. Activation of blood coagulation in Crohn’s disease. Increased plasma fibrinopeptide A levels and enhanced generation of monocyte tissue factor activity. Gastroenterology 92, 329–337 (1987). 52. Palkovits, J. et al. Tissue factor exposing microparticles in inflammatory bowel disease. J Crohns Colitis 7, 222–229 (2013). 53. Undas, A. et al. Activated factor XI and tissue factor in inflammatory bowel disease. Inflamm Bowel Dis 16, 1447–1448 (2010). 54. Anthoni, C. et al. Tissue factor: a mediator of inflammatory cell recruitment, tissue injury, and thrombus formation in experimental colitis. J. Exp. Med. 204, 1595–1601 (2007). 55. Queiroz, K. C. S. et al. Tissue factor-dependent chemokine production aggravates experimental colitis. Mol. Med. 17, 1119–1126 (2011). 56. Hudson, M. et al. Thrombotic vascular risk factors in inflammatory bowel disease. Gut 38, 733–737 (1996). 57. Koster, T., Blann, A. D., Briët, E., Vandenbroucke, J. P. & Rosendaal, F. R. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 345, 152–155 (1995). 58. Khandagale, A. et al. Coagulation factor 9-deficient mice are protected against dextran sulfate sodium-induced colitis. Biology Open 7, bio034140 (2018). 59. Wang, B. et al. The Plasma Kallikrein–Kininogen Pathway Is Critical in the Pathogenesis of Colitis in Mice. Front Immunol 9, (2018). 60. Kossmann, S. et al. Platelet-localized FXI promotes a vascular coagulationinflammatory circuit in arterial hypertension. Sci Transl Med 9, (2017). 61. van Hylckama Vlieg, A. et al. The risk of a first and a recurrent venous thrombosis associated with an elevated D-dimer level and an elevated thrombin potential: results of the THE-VTE study. J. Thromb. Haemost. 13, 1642–1652 (2015). 11

62. Lagrange, J. & Wenzel, P. The regulatory role of coagulation factors in vascular function. Front Biosci (Landmark Ed) 24, 494–513 (2019). 63. Chamouard, P. et al. Significance of diminished factor XIII in Crohn’s disease. Am. J. Gastroenterol. 93, 610–614 (1998). 64. Seitz, R. et al. Ulcerative colitis and Crohn’s disease: factor XIII, inflammation and haemostasis. Digestion 55, 361–367 (1994). 65. Lisman, T., Groot, P. G. de, Meijers, J. C. M. & Rosendaal, F. R. Reduced plasma fibrinolytic potential is a risk factor for venous thrombosis. Blood 105, 1102–1105 (2005). 66. Gris, J. C. et al. Impaired fibrinolytic capacity in patients with inflammatory bowel disease. Thromb. Haemost. 63, 472–475 (1990). 67. Weber, P., Husemann, S., Vielhaber, H., Zimmer, K. P. & Nowak-Göttl, U. Coagulation and fibrinolysis in children, adolescents, and young adults with inflammatory bowel disease. J. Pediatr. Gastroenterol. Nutr. 28, 418–422 (1999). 68. Souto, J. C. et al. Prothrombotic state and signs of endothelial lesion in plasma of patients with inflammatory bowel disease. Dig. Dis. Sci. 40, 1883–1889 (1995). 69. Koutroubakis, I. E. et al. Plasma thrombin-activatable fibrinolysis inhibitor and plasminogen activator inhibitor-1 levels in inflammatory bowel disease. Eur J Gastroenterol Hepatol 20, 912–916 (2008). 70. Vrij, A. A., Rijken, J., Wersch, J. W. J. van & Stockbrügger, R. W. Coagulation and Fibrinolysis in Inflammatory Bowel Disease and in Giant Cell Arteritis. PHT 33, 75–83 (2003). 71. Kaiko, G. E. et al. PAI-1 augments mucosal damage in colitis. Sci Transl Med 11, (2019). 72. Czekay, R.-P. et al. PAI-1: An Integrator of Cell Signaling and Migration. Int J Cell Biol 2011, (2011). 73. Ji, Y. et al. Pharmacological Targeting of Plasminogen Activator Inhibitor-1 Decreases Vascular Smooth Muscle Cell Migration and Neointima Formation. Arterioscler Thromb Vasc Biol 36, 2167–2175 (2016). 74. Puy, C. et al. Endothelial PAI-1 (Plasminogen Activator Inhibitor-1) Blocks the Intrinsic Pathway of Coagulation, Inducing the Clearance and Degradation of FXIa (Activated Factor XI). Arterioscler. Thromb. Vasc. Biol. 39, 1390–1401 (2019). 75. Bollen, L. et al. The Occurrence of Thrombosis in Inflammatory Bowel Disease Is Reflected in the Clot Lysis Profile. Inflamm. Bowel Dis. 21, 2540–2548 (2015). 76. Kjeldsen, J., Lassen, J. F., Brandslund, I. & Schaffalitzky de Muckadell, O. B. Markers of coagulation and fibrinolysis as measures of disease activity in inflammatory bowel disease. Scand. J. Gastroenterol. 33, 637–643 (1998). 77. Zhang, J. et al. D-Dimer levels are correlated with disease activity in Crohn’s patients. Oncotarget 8, 63971–63977 (2017). 78. DARGAUD, Y., WOLBERG, A. S., GRAY, E., NEGRIER, C. & HEMKER, H. C. Proposal for standardized preanalytical and analytical conditions for measuring thrombin generation in hemophilia: communication from the SSC of the ISTH. J Thromb Haemost 15, 1704–1707 (2017). 79. Lacroix, R. et al. Standardization of pre-analytical variables in plasma microparticle determination: results of the International Society on Thrombosis and Haemostasis SSC Collaborative workshop. J Thromb Haemost (2013) doi:10.1111/jth.12207. 80. Cattaneo, M. et al. Recommendations for the standardization of light transmission aggregometry: a consensus of the working party from the platelet physiology subcommittee of SSC/ISTH. J Thromb Haemost 11, 1183–1189 (2013). 81. Sarlos, P. et al. Steroid but not Biological Therapy Elevates the risk of Venous Thromboembolic Events in Inflammatory Bowel Disease: A Meta-Analysis. J Crohns Colitis 12, 489–498 (2018). 12

82. Deutschmann, A. et al. Increased procoagulant function of microparticles in pediatric inflammatory bowel disease: role in increased thrombin generation. J. Pediatr. Gastroenterol. Nutr. 56, 401–407 (2013).

13

Table 1: Clotting times, thrombin generation and markers of in vivo thrombin generation in IBD. Markers Activated partial thromboplastin time (APTT) Prothrombin time (PT) Calibrated automated thrombinography (CAT) Thrombinantithrombin complex (TAT)

Population IBD patients UC patients IBD patients

IBD pediatric patients IBD patients IBD patients UC patients

CD and UC patients Fragment 1+2 (F1+2)

UC patients IBD Young patients IBD pediatric patients

Observed modification

Ref.

= =

13

↗ in active CD and UC females

17

= with UC activity = ↗ PT ↗ PT = Endogenous thrombin potential ↗ in active and inactive IBD compared to controls = Endogenous thrombin potential ↗ in active IBD ↗ in active compared to inactive ↗ in active and ischemic UC = in actives UC (high variability) Highly ↗ in active and inactive IBD ↗ in active IBD ↗ in active IBD and with a tendency in inactive IBD ↗ both in CD and UC patients ↗ in active compared to inactive IBD ↗ in active and ischemic UC ↗ in active but not in inactive IBD =

15

correlated with the disease activity

49

46

13,47 16 46 17

82

45 42 70 15 34 16 42 44 76 70 15 43 67

Table 2: Fibrinolysis markers. 14

Marker

Population

Observed modification ↗ in CD and UC ↗ in active but not inactive IBD ↗ in CD (and with activity) but not in UC. ↗ in active IBD ↗ in active and ischemic UC ↗ in active CD patients ↗ in active disease

Ref.

↗

49

= in CD increased in UC = = in active and inactive IBD = with UC activity ↗ in active and inactive IBD ↗ in active = in inactive (antigen) ↘ antigen ↘ activity = antigen = antigen

14

IBD patients

↘ antigen

68

IBD patients

↘ but not associated with activity ↘ antigen ↗ in active and inactive IBD

69

= in active and inactive (antigen)

44

↗ with activity (antigen) ↗ in active compared to inactive ↗ ↗ in active compared to inactive

69

IBD patients Fibrinogen

Plasminogen

UC patients CD patients IBD young patients IBD pediatric patients IBD patients UC patients

Tissue plasminogen activator (tPA)

IBD patients

Young IBD patients Urokinase (uPA) TAFI

plasminogen activator inhibitor-1 (PAI-1)

16,46,47 32 13,56 17 15,34 43 67

46 16 15 16 44 66 46 68 67

68 16

IBD patients

Young IBD patients

IBD patients D-Dimer Young IBD patients CD patients UC patients

(antigen) = fibrinogen degradation products ↗ fibrinogen degradation products ↗ D-dimer in CD (and with activity) ↗ D-dimer in active disease and ↗ fibrinogen degradation products = D-dimers ↗ in active compared to inactive ↗ in active = inactive IBD ↗ in active IBD ↗ in active CD ↗ in active and ischemic UC = in actives UC (high variability)

70 46 67 47 76 13 16 14 70 44 67 77 15 34

15

Figure 1: Platelet alterations in IBD patients. IBD patients display increased platelet count related to reactive thrombocytosis. Mean platelet volume (MPV) is decreased and plateletderived large extracellular vesicles (P-LEV) are increased and both are associated with the disease activity. P-LEV have high procoagulant activities and in vivo prothrombin activation is increased in active IBD. Increase in platelet reactivity results from (i) increased circulating concentrations of von Willebrand factor (VWF) and fibrinogen that activate platelet receptors (GPIIb/IIIa and GPIb-V-IX); (ii) activation of thrombin receptor proteases activated receptor (PAR)-1 and -4; (iii) autoactivation through the CD40 ligand (CD40L) released following PAR-1 activation and cleaved from the platelet surface; ie soluble CD40 (sCD40), can in turn activate platelets via CD40. Platelet activation leads to increased released of dense and α granule contents including fibrinogen, VWF, platelet factor 4 (PF4), factor (F) V, XI, or plasminogen activator inhibitor-1 (PA-1). VWF-cleaving protease (ADAMTS13) is increased and may limit VWF functions. Figure 2. Changes in pro- and anticoagulant reactions in IBD patients. (A) IBD leads to increased large extracellular vesicles (LEV) bearing tissue factor (TF). TF can also come from subendothelial layers in case of a vascular breach. Endothelium alteration impairs its anticoagulant properties: tissue factor pathway inhibitor (TFPI) activity is decreased. Prothrombin conversion to thrombin activation is increased due to elevated levels of procoagulant factors (FVII, FV, FX). (B) Thrombin activation amplifies its own formation which is exacerbated by increased levels of procoagulant factors (FXI, FXI, FVIII) leading to increased thrombin generation as shown by elevated concentrations in prothrombin fragment 1+2 (F1+2) and thrombin-antithrombin complex (TAT). Shedding of thrombomodulin (TM) and endothelial protein C receptor (EPCR) are increased. The thrombin-induced anticoagulant pathway, the conversion of protein C (PC) to activated PC (APC), is also altered reducing the downregulation of the coagulation cascade. Black curved arrows: “conversion to”; green arrows: “activation of” red T-bar: “inhibition of”. Little black arrows next to the factors or complexes indicate if it is predominately increased or decreased in IBD patients in the literature. Figure 3. Fibrinolysis modifications in IBD patients. Fibrinogen concentration increases with inflammation. Thrombin leads to fibrin fiber polymerization. Thrombin activates also FXIII which crosslinks fibrin in order to stabilize the clot. Fibrinolysis is decreased due to increased plasminogen activator inhibitor 1 (PAI-1) that inhibits tissue plasminogen activator (tPA) and urokinase-type plasminogen activator and, plasmin activation inhibition in IBD. The fibrin network is denser and clot lysis time delayed. Fibrin degradation products (FDP and D-Dimer) hallmark continuous activation of fibrinolysis. TAFI: thrombin-activatable fibrinolysis inhibitor.

16

Primary and secondary hemostasis The formation of a blood clot resulting from the combined activation of platelets (primary hemostasis) and the coagulation cascade (secondary hemostasis) should limit blood loss following vascular injury.1 In several diseases such as atherothrombosis platelets are important mediators and adhere quickly after the plaque rupture to the exposed subendothelium molecules such as collagen and von Willebrand Factor (VWF). Coagulation is a dynamic process that contributes to secondary hemostasis. The coagulation system is classically divided in two cascades, the extrinsic and intrinsic pathways that converge in factor X (FX) activation. The extrinsic pathway starts with the release of tissue factor (TF) in case of vascular injury. In the presence of calcium, exposed TF make a complex with FVII leading to its activation. This complex, called extrinsic Xase can activate FX to FXa. FXa will then make a complex with FVa called “prothrombinase complex”. FXa will cleave prothrombin leading to the formation thrombin. The intrinsic pathway is initiated by FXII that activates FXI. This pathway may be more important for thrombosis than for physiological hemostasis.2 FXI can activate FIX which will form a complex with FVIIIa that activates FX leading to prothrombin activation. Thrombin formation is the central step of the coagulation process since thrombin triggers the polymerization of fibrinogen into fibrin fibers and activates the inducible anticoagulant systems such as the activated protein C pathway (APC). In addition, thrombin is able to autoamplify its formation (for example through activation of FV, VIII and XI). Fibrinogen cleavage by thrombin triggers its polymerization and fibrin clot formation. Platelet-derived large extracellular vesicles Platelet-derived large extracellular vesicles (P-LEV), the most abundant LEV in the blood, can expose TF and procoagulant phospholipids on their membrane and are mediators of hemostasis, inflammation, angiogenesis and can participate in cardiovascular diseases. P-LEV are commonly increased in a wide range of diseases characterized by inflammatory processes such as diabetes mellitus and rheumatoid arthritis.3,4 P-LEV procoagulant activity is 50- to 100-fold higher than the procoagulant activity of activated platelets.5 This increased procoagulant activity mainly results from negatively charged phospholipid exposure (in particular phosphatidylserine). In IBD patients P-LEV were found to be correlated with disease activity (Figure 1).6,7 In pediatric patients, LEV were increased but not related to thrombin generation variation assessed by calibrated automated thrombography (CAT).8 IBD are characterized by the formation of microthrombi in the intestinal tissue while a lifethreatening thrombus appears in larger vessels. An emerging concept is to consider P-LEV as therapeutic targets.9,10 P-LEV are decreased following treatments with acetylsalicylic acid or platelet receptor inhibitors but also drugs targeting diseases displaying low grade chronic inflammation.11 No studies have been performed on the importance of P-LEV in IBD patients with venous or arterial thrombosis. One aspect that makes P-LEV relevant for the thrombotic risk in IBD is their ability to promote coagulation at distinct sites from those where platelets are aggregated. Platelets and inflammatory pathways Platelets are not only involved in hemostasis, they are now well recognized as key players for inflammation.12 They can contribute to the inflammatory response through a cross-talk with immune and vascular cells independently of physiological hemostasis. Activation of platelets

leads to the release of inflammatory and procoagulant molecules and expression of inflammatory receptors. The implication of platelets in inflammatory responses may be considered as part of the missing link between hemostasis and inflammation.13 CD40 and its ligand, the soluble CD40 ligand (sCD40L or CD154), have a central role in immune cell responses and regulate oxidative stress which can affect IBD onset.14 It is expressed by activated platelets, as well as T and B lymphocytes15. In hemostasis, CD40L enhances thrombin generation and thrombosis formation in association with inflammation.16 SCD40L is able to activate platelets under high shear stress and to stabilize arterial thrombi.17 This function is possible through activation of the GPIIb/IIIa receptor which is the main platelet receptor for fibrinogen. Another implication of sCD40L in hemostasis is its binding to endothelial cells that induces tissue factor (TF) upregulation and a procoagulant state.18 Platelet-derived CD40L, which is increased in IBD19, can be used as a marker of in vivo platelet activation (Figure 1). SCD40L is increased in CD and UC patients with active disease compared to patients with inactive disease.20 Moreover, plasma levels of sCD40L are associated with platelet count. In CD, inhibition of TNF-α decreased the CD40/CD40L axis in the mucosal microcirculation.21 In mouse models of colitis blocking the CD40 axis was beneficial22,23. In DSS-treated mice platelet expression of an activation epitope of GPIIb/IIIa was increased.24 The activation of this receptor in concert with the macrophage-1 antigen (Mac-1 or integrin αMβ2) triggered a strong interaction between platelets and leukocytes and their subsequent adhesion to the vessel wall.25 An important number of clinical trials have been conducted using GPIIb/IIIa inhibitors but none in IBD patients because of a high bleeding risk.26 Since other integrin inhibitors, such as α4β7 inhibitors, are used in IBD to limit direct lymphocyte trafficking to the intestinal tissue, targeting integrin GPIIb/IIIa (or αvβ3) might represent a potential therapeutic option to limit leukocyte trafficking and to decrease thrombotic risk. Antiphospholipid antibodies Antiphospholipid antibodies (aPL) are a heterogeneous group of autoantibodies directed at plasma proteins such as β2-glycoprotein I (β2GPI) and prothrombin, and in some cases protein C or protein S. These antibodies are associated with an increased risk of arterial and venous thrombosis, and are associated with obstetric manifestations such as recurrent early spontaneous abortions or late fetal losses. These antibodies binding prothrombin, protein C, protein S or phosphatidylserine, can respectively lead to more thrombin activation, less activated protein C and a decreased platelet count. Several studies have tested aPL in IBD patients and found a higher frequency of anticoardiolipin antibodies and antiphosphatidylserine/prothrombin than controls but it has not been possible to confirm their prognostic value in patients with IBD27,28 (Supplementary Figure). Of note some individual case reports of an association of definite antiphospholipid syndrome and IBD have been recorded. Common genetic mutations Common genetic mutations predisposing to thrombosis (FV Leiden, prothrombin G20210A) were not associated with a further increased thrombotic risk in IBD patients compared to patients displaying these mutations in the general population29,30. Moreover FV Leiden and prothrombin G20210A where not associated with IBD prevalence and activity31,32. However, FV Leiden was associated with the history of VTE in IBD patients33,34.

Vascular endothelium: anticoagulation dysfunction The endothelium is a barrier between blood components and vascular tissue. It has numerous functions, one of them being its anticoagulant properties. The main anticoagulant functions consist in the synthesis of TFPI and proteins enabling PC activation such as TM and endothelial protein C receptor (EPCR). Under physiological conditions TFPI is present at the surface of endothelial cells allowing local inhibition of the extrinsic pathway. Endothelial activation/dysfunction triggered by inflammation leads to TFPI release. In IBD patients circulating TFPI is increased and correlates with biochemical markers of inflammation and disease activity (Table 2 and Figure 2).35 TFPI as well as TF were both found to be increased in colon tissue from UC patients36. Intriguingly TFPI was recently found to have limited anticoagulant properties in children with IBD.37 Indeed, addition of TFPI to plasma samples of IBD patients only delayed slightly thrombin generation assessed by CAT compared to controls. This limited effect could explain the fact that even with both TF and TFPI being increased, IBD patients display an increased activation of the coagulation system. Soluble TM was elevated in plasma of IBD patients while endothelial cells displayed reduced TM expression.38,39 Finally EPCR which increases the activation rate of protein C presented similar variations as TM.39 Similar decrease in TM and EPCR on the surface of the microvasculature in colonic mucosa were reported by Scaldaferri et al.40 Protein C was unchanged in IBD41–44, even though one study reported increased PC levels in active CD.45 Protein S, the cofactor of APC, was often described to be decreased in IBD 43,44 but again contradictory reports found no significant changes.41,42,45 Low protein S is not a general marker for VTE risk.46 However, it must be noted that protein S is also a cofactor of TFPI which is not only able to inhibit TF but also FXa. Therefore, alteration of both TFPI or PS may be deleterious for the downregulation of thrombin formation. Human intestinal microvascular endothelial cells incubated with TNF-α and DSS treated mice exhibited reduced capacity to activate protein C. Mice with very low levels of protein C display spontaneous colitis, highlighting the tight relation existing between coagulation and inflammation.39 Transitory acquired protein C and protein S deficiency were observed in a patient with IBD experiencing cerebral arterial thrombosis47 and a few other studies anticipated that impairment of the protein C pathway may contribute to the development of VTE.44,48 Protein C is well known for its protective effect occurring through PARs and similar beneficial effects were described in epithelial cells in IBD.49 IBD patients display impaired endothelial anticoagulant function. In addition, alterations of the anticoagulant protein C pathway, especially in microvessels, could favor microthrombosis which in turn participate in further vascular alterations by limiting the beneficial APC cellular effects. Supplementary table: Changes in pro and anticoagulant factors in IBD.

Factor Tissue factor (TF) FVII

Population IBD patients UC patients IBD patients UC patients

Observed modifications Activity correlates with TAT ↗TF in colon tissue = procoagulant FVII ↗ procoagulant FVII ↗ in ischemic UC

Ref. 50 36 41 51 45

UC patients

↗ in active UC

45 52

Prothrombin

IBD patients

= =

Thrombin

UC patients

↗ in active UC ↗ procoagulant FV ↗ procoagulant FV

45

= procoagulant FV ↗ in active and ischemic UC ↗ in active and ischemic UC = in active and inactive IBD = in active and inactive IBD ↗ in active and ischemic UC ↗ in active and inactive IBD ↗ in active and ischemic UC ↗ in CD but not in UC = Tendency of ↗ in CD and UC, ↗ in active CD versus remission ↗ procoagulant FVIII ↗ in IBD ↗ in active and ischemic UC ↗ in CD and UC ↗ in active but not inactive IBD ↗ in CD (and with activity) but not in UC. ↗ in active IBD ↗ in active and ischemic UC ↗ in active CD patients ↗ in active disease

43

FX

FV

FXII FXI FIX

FVIII

IBD patients UC patients UC patients IBD patients IBD patients UC patients IBD patients UC patients

IBD patients

UC patients

IBD patients Fibrinogen

UC patients CD patients IBD young patients IBD pediatric patients IBD patients

FXIII CD patients IBD patients Tissue factor pathway inhibitor (TFPI)

Antithrombin (AT) Protein S (PS)

UC patients IBD pediatric patients IBD patients Young IBD patients UC patients IBD patients

53

54 42

45 45 43 43 45 43 45 54 43 41 42 52 45 43,53,54 55 41,51 56 45,57 58 59 60

↗

High ↘ in many active IBD ↘ FXIII in active and inactive IBD but = activity FXIIIA correlated with CD activity correlates with activity ↘ anticoagulant TFPI activity ↗ in colon tissue

61 62

63 35 37 36 60

↘ activity with IBD ↘ in CD and UC patients

= = = Increased in active UC = PS ↘ with IBD

54 41 53 59 45 41,42 44

Protein C (PC)

UC patients IBD patients UC patients

Thrombomodulin IBD patients (TM) Young IBD patients Endothelial protein C IBD patients receptor (EPCR)

↘ PS in active IBD = in active and ischemic UC = ↗ in active UC ↗ in plasma ↘ on endothelial cells ↗ in plasma ↘ on mucosal vessels ↗ in plasma and with activity ↗ in plasma ↘ on mucosal vessels

43 45 41–44 45 38 39 59

39

Supplementary Figure: IBD patients display increased frequency of antiphospholipid antibodies (aPL) which affect hemostasis. APL can be directed against β2-glycoprotein I (β2GPI) and prothrombin but also protein C (PC) and protein S (PS). Inhibition of β2GPI activates endothelial cells and platelets and in combination with anti-prothrombin antibodies limits activated coagulation factor V (FVa) inactivation (FVai) leading to increased thrombin generation. Inhibition of PC or PS decreases thrombin inhibition and subsequently increases thrombin generation.

References 1. Broos, K., Feys, H. B., De Meyer, S. F., Vanhoorelbeke, K. & Deckmyn, H. Platelets at work in primary hemostasis. Blood Rev. 25, 155–167 (2011). 2. Nickel, K. F., Long, A. T., Fuchs, T. A., Butler, L. M. & Renné, T. Factor XII as a Therapeutic Target in Thromboembolic and Inflammatory Diseases. Arterioscler. Thromb. Vasc. Biol. 37, 13–20 (2017). 3. Li, S. et al. Cell-Derived Microparticles in Patients with Type 2 Diabetes Mellitus: a Systematic Review and Meta-Analysis. Cell. Physiol. Biochem. Int. J. Exp. Cell. Physiol. Biochem. Pharmacol. 39, 2439–2450 (2016). 4. Boilard, E. et al. Platelets amplify inflammation in arthritis via collagen-dependent microparticle production. Science 327, 580–583 (2010). 5. Sinauridze, E. I. et al. Platelet microparticle membranes have 50- to 100-fold higher specific procoagulant activity than activated platelets. Thromb. Haemost. 97, 425–434 (2007). 6. Andoh, A. et al. Elevated circulating platelet-derived microparticles in patients with active inflammatory bowel disease. Am. J. Gastroenterol. 100, 2042–2048 (2005). 7. Leonetti, D. et al. Circulating microparticles from Crohn’s disease patients cause endothelial and vascular dysfunctions. PloS One 8, e73088 (2013). 8. Deutschmann, A. et al. Increased procoagulant function of microparticles in pediatric inflammatory bowel disease: role in increased thrombin generation. J. Pediatr. Gastroenterol. Nutr. 56, 401–407 (2013). 9. Lopez, E., Srivastava, A. K., Pati, S., Holcomb, J. B. & Wade, C. E. Platelet-Derived Microvesicles: A Potential Therapy for Trauma-Induced Coagulopathy. Shock Augusta Ga 49, 243–248 (2018). 10. Chen, Y., Li, G. & Liu, M.-L. Microvesicles as Emerging Biomarkers and Therapeutic Targets in Cardiometabolic Diseases. Genomics Proteomics Bioinformatics 16, 50–62 (2018). 11. Rosińska, J., Łukasik, M. & Kozubski, W. The Impact of Vascular Disease Treatment on Platelet-Derived Microvesicles. Cardiovasc. Drugs Ther. 31, 627–644 (2017). 12. Stokes, K. Y. & Granger, D. N. Platelets: a critical link between inflammation and microvascular dysfunction. J. Physiol. 590, 1023–1034 (2012). 13. Voudoukis, E., Karmiris, K. & Koutroubakis, I. E. Multipotent role of platelets in inflammatory bowel diseases: A clinical approach. World J. Gastroenterol. WJG 20, 3180–3190 (2014). 14. Senhaji, N., Kojok, K., Darif, Y., Fadainia, C. & Zaid, Y. The Contribution of CD40/CD40L Axis in Inflammatory Bowel Disease: An Update. Front. Immunol. 6, (2015). 15. Henn, V. et al. CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 391, 591–594 (1998). 16. Gavins, F. N. E., Li, G., Russell, J., Perretti, M. & Granger, D. N. Microvascular thrombosis and CD40/CD40L signaling. J. Thromb. Haemost. JTH 9, 574–581 (2011). 17. André, P. et al. CD40L stabilizes arterial thrombi by a beta3 integrin--dependent mechanism. Nat. Med. 8, 247–252 (2002).

18. Slupsky, J. R. et al. Activated platelets induce tissue factor expression on human umbilical vein endothelial cells by ligation of CD40. Thromb. Haemost. 80, 1008–1014 (1998). 19. Ludwiczek, O., Kaser, A. & Tilg, H. Plasma levels of soluble CD40 ligand are elevated in inflammatory bowel diseases. Int. J. Colorectal Dis. 18, 142–147 (2003). 20. Koutroubakis, I. et al. Association between enhanced soluble CD40 ligand and prothrombotic state in inflammatory bowel disease. Eur. J. Gastroenterol. Hepatol. 16, 1147–1152 (2004). 21. Danese, S. et al. TNF-alpha blockade down-regulates the CD40/CD40L pathway in the mucosal microcirculation: a novel anti-inflammatory mechanism of infliximab in Crohn’s disease. J. Immunol. Baltim. Md 1950 176, 2617–2624 (2006). 22. De Jong, Y. P. et al. Chronic murine colitis is dependent on the CD154/CD40 pathway and can be attenuated by anti-CD154 administration. Gastroenterology 119, 715–723 (2000). 23. Liu, Z. et al. Prevention of experimental colitis in SCID mice reconstituted with CD45RBhigh CD4+ T cells by blocking the CD40-CD154 interactions. J. Immunol. Baltim. Md 1950 164, 6005–6014 (2000). 24. Yan, S. LS. et al. Platelet abnormalities during colonic inflammation. Inflamm. Bowel Dis. 19, 1245–1253 (2013). 25. Zarbock, A., Polanowska-Grabowska, R. K. & Ley, K. Platelet-neutrophil-interactions: linking hemostasis and inflammation. Blood Rev. 21, 99–111 (2007). 26. Ley, K., Rivera-Nieves, J., Sandborn, W. J. & Shattil, S. Integrin-based Therapeutics: Biological Basis, Clinical Use and New Drugs. Nat. Rev. Drug Discov. 15, 173–183 (2016). 27. Chamouard, P. et al. Prothrombin fragment 1 + 2 and thrombin-antithrombin III complex as markers of activation of blood coagulation in inflammatory bowel diseases. Eur. J. Gastroenterol. Hepatol. 7, 1183–1188 (1995). 28. Sipeki, N. et al. Prevalence, significance and predictive value of antiphospholipid antibodies in Crohn’s disease. World J. Gastroenterol. 21, 6952–6964 (2015). 29. Vecchi, M. et al. Inflammatory bowel diseases are not associated with major hereditary conditions predisposing to thrombosis. Dig. Dis. Sci. 45, 1465–1469 (2000). 30. Turri, D. et al. Factor V Leiden and prothrombin gene mutation in inflammatory bowel disease in a Mediterranean area. Dig. Liver Dis. Off. J. Ital. Soc. Gastroenterol. Ital. Assoc. Study Liver 33, 559–562 (2001). 31. Over-Hamzaoglu, H. et al. Prothrombotic gene mutations and Crohn’s disease; is there any association? Hepatogastroenterology. 52, 1467–1469 (2005). 32. Haslam, N., Standen, G. R. & Probert, C. S. An investigation of the association of the prothrombin G20210A gene mutation and inflammatory bowel disease: Factor II and IBD. Inflamm. Bowel Dis. 7, 133–135 (2001). 33. Oldenburg, B., Van Tuyl, B. A. C., van der Griend, R., Fijnheer, R. & van Berge Henegouwen, G. P. Risk factors for thromboembolic complications in inflammatory bowel disease: the role of hyperhomocysteinaemia. Dig. Dis. Sci. 50, 235–240 (2005). 34. Zhong, M., Dong, X. W., Zheng, Q., Tong, J. L. & Ran, Z. H. Factor V Leiden and thrombosis in patients with inflammatory bowel disease (IBD): a meta-analysis. Thromb. Res. 128, 403–409 (2011). 35. Cibor, D., Szczeklik, K., Mach, T. & Owczarek, D. Levels of tissue factor pathway inhibitor in patients with inflammatory bowel disease. Pol. Arch. Intern. Med. 129, 253– 258 (2019).

36. He, H.-L., Zhang, J.-B. & Li, Q. Clinical significance of expression of tissue factor and tissue factor pathway inhibitor in ulcerative colitis. World J. Gastroenterol. 20, 7461 (2014). 37. Schlagenhauf, A. et al. Children with Inflammatory Bowel Disease Exhibit Insensitivity to Tissue Factor Pathway Inhibitor. Blood 132, 2504–2504 (2018). 38. Boehme, M. W. et al. Elevated serum levels and reduced immunohistochemical expression of thrombomodulin in active ulcerative colitis. Gastroenterology 113, 107–117 (1997). 39. Faioni, E. M. et al. Expression of endothelial protein C receptor and thrombomodulin in the intestinal tissue of patients with inflammatory bowel disease. Crit. Care Med. 32, S266-270 (2004). 40. Scaldaferri, F. et al. Crucial role of the protein C pathway in governing microvascular inflammation in inflammatory bowel disease. J. Clin. Invest. 117, 1951–1960 (2007). 41. Yazici, A. et al. Thrombophilic Risk Factors in Patients With Inflammatory Bowel Disease. Gastroenterol. Res. 3, 112–119 (2010). 42. Dolapcioglu, C. et al. Coagulation parameters in inflammatory bowel disease. Int. J. Clin. Exp. Med. 7, 1442–1448 (2014). 43. Alkim, H., Ayaz, S., Alkim, C., Ulker, A. & Sahin, B. Continuous active state of coagulation system in patients with nonthrombotic inflammatory bowel disease. Clin. Appl. Thromb. Off. J. Int. Acad. Clin. Appl. Thromb. 17, 600–604 (2011). 44. Aadland, E., Odegaard, O. R., Røseth, A. & Try, K. Free protein S deficiency in patients with chronic inflammatory bowel disease. Scand. J. Gastroenterol. 27, 957–960 (1992). 45. Kume, K., Yamasaki, M., Tashiro, M., Yoshikawa, I. & Otsuki, M. Activations of coagulation and fibrinolysis secondary to bowel inflammation in patients with ulcerative colitis. Intern. Med. Tokyo Jpn. 46, 1323–1329 (2007). 46. Pintao, M. C. et al. Protein S levels and the risk of venous thrombosis: results from the MEGA case-control study. Blood 122, 3210–3219 (2013). 47. Jorens, P. G. et al. Acquired protein C and S deficiency, inflammatory bowel disease and cerebral arterial thrombosis. Blut 61, 307–310 (1990). 48. Aadland, E., Odegaard, O. R., Røseth, A. & Try, K. Free protein S deficiency in patients with Crohn’s disease. Scand. J. Gastroenterol. 29, 333–335 (1994). 49. Vetrano, S. et al. Unexpected role of anticoagulant protein C in controlling epithelial barrier integrity and intestinal inflammation. Proc. Natl. Acad. Sci. U. S. A. 108, 19830– 19835 (2011). 50. Undas, A. et al. Activated factor XI and tissue factor in inflammatory bowel disease. Inflamm. Bowel Dis. 16, 1447–1448 (2010). 51. Hudson, M. et al. Thrombotic vascular risk factors in inflammatory bowel disease. Gut 38, 733–737 (1996). 52. Saibeni, S. et al. Increased thrombin generation in inflammatory bowel diseases. Thromb. Res. 125, 278–282 (2010). 53. de Jong, E., Porte, R. J., Knot, E. A., Verheijen, J. H. & Dees, J. Disturbed fibrinolysis in patients with inflammatory bowel disease. A study in blood plasma, colon mucosa, and faeces. Gut 30, 188–194 (1989). 54. Lam, A., Borda, I. T., Inwood, M. J. & Thomson, S. Coagulation studies in ulcerative colitis and Crohn’s disease. Gastroenterology 68, 245–251 (1975). 55. Cibor, D. et al. Levels and activities of von Willebrand factor and metalloproteinase with thrombospondin type-1 motif, number 13 in inflammatory bowel diseases. World J. Gastroenterol. 23, 4796–4805 (2017).

56. Shen, J. et al. Biomarkers of altered coagulation and fibrinolysis as measures of disease activity in active inflammatory bowel disease: A gender-stratified, cohort analysis. Thromb. Res. 123, 604–611 (2009). 57. Zezos, P. et al. Elevated plasma von Willebrand factor levels in patients with active ulcerative colitis reflect endothelial perturbation due to systemic inflammation. World J. Gastroenterol. 11, 7639–7645 (2005). 58. Smith, C. J., Haire, W. D., Kaufman, S. S. & Mack, D. R. Determination of prothrombin activation fragments in young patients with inflammatory bowel disease. Am. J. Gastroenterol. 91, 1221–1225 (1996). 59. Weber, P., Husemann, S., Vielhaber, H., Zimmer, K. P. & Nowak-Göttl, U. Coagulation and fibrinolysis in children, adolescents, and young adults with inflammatory bowel disease. J. Pediatr. Gastroenterol. Nutr. 28, 418–422 (1999). 60. Bernhard, H. et al. Calibrated automated thrombin generation in paediatric patients with inflammatory bowel disease. Hamostaseologie 29 Suppl 1, S90-93 (2009). 61. Seitz, R. et al. Ulcerative colitis and Crohn’s disease: factor XIII, inflammation and haemostasis. Digestion 55, 361–367 (1994). 62. Hayat, M., Ariëns, R. A. S., Moayyedi, P., Grant, P. J. & O’Mahony, S. Coagulation factor XIII and markers of thrombin generation and fibrinolysis in patients with inflammatory bowel disease. Eur. J. Gastroenterol. Hepatol. 14, 249–256 (2002). 63. Chamouard, P. et al. Significance of diminished factor XIII in Crohn’s disease. Am. J. Gastroenterol. 93, 610–614 (1998).