Prediction of bleeding risk in patients taking vitamin K antagonists using thrombin generation testing

Thrombin generation, bleeding and hemostasis in humans: Protocol for a scoping review of the literature

INTRODUCTION

Hemostasis and bleeding are difficult to measure. Thrombin generation assays (TGAs) can measure both procoagulant and anticoagulant contributions to coagulation. TGAs might prove useful for the study of bleeding disorders. There has been much progress in TGA methodology over the past two decades, but its clinical significance is uncertain. We will undertake a scoping review of the literature to synthesize available information on the application of TGAs towards the study of bleeding and hemostasis, TGA methodologies being used and to summarize available literature on associations between TGA parameters, bleeding and hemostatic outcomes.

METHODS AND ANALYSIS

MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) will be searched in collaboration with an information specialist. Title/abstract and full-text screening will be carried out independently and in duplicate; eligible study types will include randomized controlled trials, non-randomized studies, systematic reviews, and case series reporting TGA results and bleeding/hemostatic outcomes among humans. Mapping the information identified will be carried out with results presented using qualitative data analytical techniques.

ETHICS AND DISSEMINATION

This scoping review will use published, publicly available information. Research ethics approval will not be required. We will disseminate our findings using conference presentations, peer-reviewed publications, social media, and engagement with knowledge users. This review will outline knowledge gaps concerning TGAs, better delineate its applicability as a clinically relevant assay for bleeding. and seek to identify ongoing barriers to its widespread adoption in clinical research, and eventually, in the clinical setting.

TRAIL REGULATIONS

Registration ID with Open Science Framework: osf.io/zp4ge.

Pharmacokinetic Interactions Between Abiraterone, Apalutamide, Darolutamide or Enzalutamide and Antithrombotic Drugs: Prediction of Clinical Events and Review of Pharmacological Information

PURPOSE

Abiraterone, apalutamide, darolutamide and enzalutamide are second-generation hormone therapies used for advanced prostate cancer; the majority of patients receiving these treatments are elderly, poly-medicated patients. Since their first market authorizations, their pharmacokinetic (PK) characteristics are increasingly well known. A potential risk of drug-drug interaction (DDI), especially with cardiovascular drugs, needs to be considered. In the case of antithrombotics, treatment imbalance can lead to severe consequences.

OBJECTIVES

To describe PK profiles of hormone therapies and antithrombotics and to predict DDIs and potentially related clinical events.

METHODS

PK profiles (CYP450 and P-gp substrate, inducer or inhibitor) are described by cross-referencing data sources (summary of product characteristics, European public assessment reports, PubMed database, Micromedex^®, etc.); a description of the potential interactions with anti-cancer drugs for each DDI and related clinical events is provided. We discuss management recommendations, including those set out in international guidelines.

RESULTS

Antithrombotics are mainly metabolized by CYP 2C9, 2C19 or 3A4. For abiraterone (CYP 2C8, 2D6 inhibitor) and darolutamide (CYP 3A4 inducer), no interaction was identified with antithrombotics. For apalutamide (CYP 2C9, 2C19, 3A4 and P-gp inducer) and enzalutamide (CYP 2C9, 2C19, 3A4 inducer and P-gp inhibitor), several PK interactions were identified with antithrombotics, which could lead to various clinical events (haemorrhage or thromboembolism).

CONCLUSION

Numerous interactions are expected between enzalutamide or apalutamide and antithrombotics, for which management should be deployed on a case-by-case basis. PK and pharmaco-epidemiological studies could shed light on whether or not there are clinically significant events related to DDIs with antithrombotics.

Thrombin generation assays to personalize treatment in bleeding and thrombotic diseases

Treatment of bleeding and thrombotic disorders is highly standardized and based on evidence-based medicine guidelines. These evidence-based treatment schemes are well accepted but may lead to either insufficient treatment or over-dosing, because the individuals' hemostatic properties are not taken into account. This can potentially introduce bleeding or thrombotic complications in individual patients. With the incorporation of pharmacokinetic (PK) and pharmacodynamic (PK-PD) parameters, based on global assays such as thrombin generation assays (TGAs), a more personalized approach can be applied to treat either bleeding or thrombotic disorders. In this review, we will discuss the recent literature about the technical aspects of TGAs and the relation to diagnosis and management of bleeding and thrombotic disorders. In patients with bleeding disorders, such as hemophilia A or factor VII deficiency, TGAs can be used to identify patients with a more severe bleeding phenotype and also in the management with non-replacement therapy and/or bypassing therapy. These assays have also a role in patients with venous thrombo-embolism, but the usage of TGAs in patients with arterial thrombosis is less clear. However, there is a potential role for TGAs in the monitoring of (long-term) antithrombotic therapy, for example with the use of direct oral anticoagulants. Finally this review will discuss controversies, limitations and knowledge gaps in relation to the introduction of TGAs to personalize medicine in daily medical practice.

Assessing the individual roles of FII, FV, and FX activity in the thrombin generation process

Thrombin generation (TG) is known as a physiological approach to assess the hemostatic function. Although it correlates well with thrombosis and bleeding, in the current setup it is not sensitive to the effects of fluctuations in single coagulation factors. We optimized the calibrated automated thrombinography (CAT) method to quantify FII, FV and FX activity within the coagulation system. The CAT assay was fine-tuned for the assessment of FII, FV and FX by diluting the samples in FII-, FV-, or FX-deficient plasma, respectively, and measuring TG. Plasma FII levels correlated linearly with the ETP up to a plasma concentration of 100% FII. FV and FX levels correlated linearly with the peak height up to a plasma level of 2.5% FV and 10% FX, respectively. Sensitized CAT protocols were designed by adding a fixed volume of a pre-diluted patient sample to FII, FV, and FX deficient plasma in TG experiments. This approach makes the TG measurement dependent on the activity of the respective coagulation factor. The ETP or peak height were quantified as readouts for the coagulation factor activity. The intra- and inter-assay variation coefficients varied from 5.0 to 8.6%, and from 3.5 to 5.9%, respectively. Reference values were determined in 120 healthy subjects and the assays were clinically validated in 60 patients undergoing coronary artery bypass grafting (CABG). The sensitized CAT assays revealed that the contribution of FII, FV, and FX to the TG process was reduced after CABG surgery, leading to reduced prothrombin conversion and subsequently, lower TG.

Major bleeding during oral anticoagulant therapy associated with factor V activation by factor Xa

OBJECTIVE

Plasma thrombin generation (TG) provides important information on coagulation status; however, current TG output parameters do not predict major bleeding of patients on anticoagulants. We recently reported that factor V (FV) activation by factor X (FX)a contributes importantly to the initiation phase of TG. Here we investigated how this pathway varies in the normal population and whether FXa-mediated activation of FV is associated with major bleeding in patients on anticoagulant therapy.

APPROACH

We employed TIX-5, a specific inhibitor of FV activation by FXa, to estimate the contribution of FXa-mediated FV activation to tissue factor (TF)-initiated TG.

RESULTS

We show that the contribution of this pathway to plasma TG varies considerably in the normal population, as measured by the time needed to form the first traces of thrombin (TG lag time; mean prolongation by TIX-5 40%, range 0%-116%). Comparing patients on vitamin K antagonists (VKA) of the BLEED study (263 patients with and 538 patients without major bleeding), showed a marked prolongation in the median TG lag time in the presence of TIX-5 in cases (12.83 versus 11.00 minutes, P = 0.0030), while the TG lag time without TIX-5 only showed a minor although significant difference (5.83 vs. 5.67 minutes, P = 0.0198). The TIX-5 sensitivity (lag time + TIX-5/lag time + vehicle) in the upper quartile was associated with a 1.62-fold (95% confidence interval 1.04-2.52) increased risk of major bleeding compared to the lowest quartile.

CONCLUSION

A greater dependence on FXa-mediated activation of FV of TG is associated with increased risk of major bleeding during VKA therapy.

Potential value of the calibrated automated thrombogram in patients after a cerebral venous sinus thrombosis; an exploratory study

Background

Cerebral venous sinus thrombosis (CVST) is a relatively rare, but potentially lethal condition. In approximately 15% of the patients, the cause of CVST remains unclear. Conventional clotting tests such as prothrombin time and activated partial thromboplastin time are not sensitive enough to detect prothrombotic conditions nor mild haemostatic abnormalities. The calibrated automated thrombogram (CAT) is a physiological function test that might be able to detect minor aberrations in haemostasis. Therefore, we aimed to detect the presence of a prothrombotic state in patients who endured idiopathic CVST with the CAT assay.

Methods

Five adult patients with an idiopathic, radiologically proven CVST that had been admitted during the past 3 years were included in this study. The control group consisted of five age/gender matched healthy volunteers. Exclusion criteria were known haematological disorders, malignancy (current/past) or hormonal and anticoagulant therapy recipients. We obtained venous blood samples from all participants following cessation of anticoagulation. Using the CAT assay, we determined lag time, normalized endogenous thrombin potential (ETP), ETP reduction and normalized peak height. In addition, prothrombin concentrations were determined.

Results

We found no significant differences in lag time (4.7 min [4.5–4.9] vs 5.3 min [3.7–5.7], p = 0.691), normalized ETP (142% [124–148] vs 124% [88–138], p = 0.222), ETP reduction (29% [26–35] vs 28% [24–58], p > 0.999), and normalized peak height (155% [153–175] vs 137 [94–154], p = 0.056) between patients and their age/gender matched controls. In addition, prothrombin concentrations did not significantly differ between patients and controls (120% [105–132] vs 127% [87–139], p > 0.999).

Conclusion

Reasons for absent overt hypercoagulability within this study population may be the small patient sample, long time since the event (e.g. 3 years) and avoidance of acquired risk factors like oral contraception. Given the fact that CVST is a serious condition with a more than negligible risk of venous thrombosis event recurrence, exclusion of clinically relevant hypercoagulability remains a challenging topic to further study at the acute and later time points, particularly in patients with idiopathic CVST.

Thrombin Generation as a Method to Identify the Risk of Bleeding in High Clinical-Risk Patients Using Dual Antiplatelet Therapy

Background: Patients using dual antiplatelet therapy after percutaneous coronary intervention are at risk for bleeding. It is currently unknown whether thrombin generation can be used to identify patients receiving dual antiplatelet therapy with increased bleeding risk. Objectives: To investigate whether thrombin generation measurement in plasma provides additional insight into the assessment of bleeding risk for high clinical-risk patients using dual antiplatelet therapy. Methods: Coagulation factors and thrombin generation in platelet-poor plasma were measured in 93 high clinical-risk frail patients using dual antiplatelet therapy after percutaneous coronary intervention. During 12-month follow-up, clinically relevant bleedings were reported. Thrombin generation at 1 and 6 months after percutaneous coronary intervention was compared between patients with and without bleeding events. Results: One month after percutaneous coronary intervention, the parameters of thrombin generation, endogenous thrombin potential, peak height, and velocity index were significantly lower in patients with bleeding in the following months compared to patients without bleeding. At 6 months follow-up, endogenous thrombin potential, peak height, and velocity index were still (significantly) decreased in the bleeding group as compared to non-bleeders. Thrombin generation in the patients' plasma was strongly dependent on factor II, V, and VIII activity and fibrinogen. Conclusion: High clinical-risk patients using dual antiplatelet therapy with clinically relevant bleeding during follow-up show reduced and delayed thrombin generation in platelet-poor plasma, possibly due to variation in coagulation factors. Thus, impaired thrombin-generating potential may be a "second hit" on top of dual antiplatelet therapy, increasing the bleeding risk in high clinical-risk patients. Thrombin generation has the potential to improve the identification of patients using dual antiplatelet therapy at increased risk of bleeding.

Plasma levels do not predict thrombin generation in patients taking direct oral anticoagulants

BACKGROUND

The antithrombotic effect of direct oral anticoagulants (DOAC) in specific clinical scenarios is difficult to assess.

OBJECTIVE

This study aimed to evaluate the effect of DOAC on thrombin generation (TG) in relation to their plasma level.

METHODS

Eighty patients newly started on anticoagulation were included, 20 patients for each DOAC-apixaban, edoxaban, rivaroxaban, and dabigatran. Plasma was sampled before DOAC (baseline), at plasma peak time, 6 and 12 hours after starting DOAC for quantification of drug levels and TG.

RESULTS

The baseline TG before DOAC intake showed inter-individual variability. All DOACs significantly prolonged lag time (LT) and time to peak (TTP), but did not change endogenous thrombin potential (ETP). Anti-Xa inhibitors but not dabigatran reduced thrombin peak, but the effect of apixaban at plasma peak was less pronounced (factor 1.6). LT and TTP prolongation of dabigatran was lower compared to anti-Xa inhibitors. All DOACs showed a nonlinear dose-response relationship, with the greatest antithrombotic effect at lower DOAC plasma levels. The inhibition of TG parameters between baseline and peak was parallel between individual patients but the coefficient of variation of TG was lower compared to drug levels.

CONCLUSION

The antithrombotic effect at DOAC peak plasma level measured by TG depends on the patient-specific baseline TG level and the drug-specific inhibition by the particular DOAC. Although peak plasma levels have a high variability, the variation of TG is lower compared to drug levels. Therefore, TG assays may be superior to plasma levels in the assessment of the intensity of anticoagulation.

Recommendations for the measurement of thrombin generation: Communication from the ISTH SSC Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibodies

Thrombin generation (TG) assay is an overall assay to assess the functionality of the hemostatic system and may be a useful tool in diagnosing patients with hyper- and hypocoagulability. Lack of standardization in performing the assays contributes largely to poor correlation between assays and study results. The current lack of standardization remains a major issue in the setting of TG, as illustrated in a recent survey of the ISTH/SSC indicating differences in pre-, analytical, and post-analytical factors among users. These factors may considerably affect the between-laboratory reproducibility of results. Based on the results of the survey and a current review of the literature, along with insights and strong consensus of key investigators in the field, we present guidance for measurement of TG in a clinical setting. Recommendations on blood drawing, handling, processing, and sample storage; reagent concentration and source; analytical conditions on dilution of samples and temperature; calibration and replicate testing; calculation and interpretation of results; and reference values are addressed to help in reducing interlaboratory variation. These recommendations aim at harmonization between methods and laboratories to support the application of TG in patient diagnosis and management.

Delayed Thrombin Generation Is Associated with Minor Bleedings in Venous Thromboembolism Patients on Rivaroxaban: Usefulness of Calibrated Automated Thrombography

Bleeding is the most feared and difficult to predict adverse event of anticoagulation. We sought to investigate whether calibrated automated thrombography (CAT) parameters are associated with minor bleeding (MB) in anticoagulated patients following venous thromboembolism (VTE). Enrolled were 132 patients on rivaroxaban, 145 on vitamin K antagonists (VKA) and 31 controls who stopped anticoagulation. Prior to the next dose of the anticoagulant, we measured CAT parameters, along with rivaroxaban concentration and INR. During a median follow-up of 10 months, we recorded minor and major bleedings. On rivaroxaban, 27 (20.5%) patients with MB had longer time to start thrombin generation, lower peak thrombin generation and lower endogenous thrombin potential compared with subjects without MB (all p < 0.001). All CAT parameters, except for peak thrombin generation (p = 0.049), were similar in VKA patients with (n = 25, 17.2%) vs. without MBs. By logistic regression, time to start thrombin generation (p = 0.007) and unprovoked VTE (p = 0.041) independently predicted MBs on rivaroxaban. Major bleedings were more frequent in patients with MBs (17.3% vs. 1.8%, p < 0.001). Abnormal CAT parameters characterize VTE patients prone to MBs on rivaroxaban, but not on VKA. Time to start thrombin generation measured about 24 h since the last rivaroxaban dose might help predict MBs.

Inhibition of triggering receptor expressed on myeloid cells-1 impairs thrombin generation

BACKGROUND

New evidence has shown the link between inflammation and thrombosis. Triggering receptor expressed on myeloid cells-1 (TREM-1) is an immunoreceptor expressed mostly on neutrophils and monocytes/macrophages. TREM-1 acts as an amplifier of the inflammatory response, and its pharmacological inhibition displays protective effects in various models of inflammatory disorders, in particular by dampening coagulation abnormalities and thrombocytopenia observed during acute inflammation.

OBJECTIVES

We aimed to decipher the role of TREM-1 in fostering thrombin generation.

METHODS

We measured thrombin generation (TG) by the use of calibrated automated thrombography with whole blood, and isolated primary human neutrophils and monocytes upon stimulation with lipopolysaccharide (LPS). Tissue factor (TF) expression was measured by flow cytometry and its activity by ELISA. Phosphatidylserine (PtdSer) exposure was determined by flow cytometry. A dodecapeptide (LR12) was used as a specific inhibitor of TREM-1.

RESULTS

LPS increased TG, TF expression, and activity, as well as the exposure of PtdSer on the surface of monocytes. LR12 dampened TF activity through the decrease of PtdSer exposure, leading to a reduction of thrombin generation.

CONCLUSIONS

TREM-1 inhibition decreases thrombin generation and could be an interesting target for the development of new inhibitors of leukocyte-associated thrombotic activity.

Heparins: A Shift of Paradigm

Heparins inhibit the thrombin forming capacity of plasma, i. e., the endogenous thrombin potential (ETP), by their anti-thrombin (aIIa) activity, the anti-factor Xa (aXa) activity is of minimal importance. This holds for both unfractionated heparin (UFH) and low molecular weight heparin (LMWH) at aXa/aIIa ratios < 25. Clinical experience and epidemiological evidence show a direct relationship between the ETP and the risk of thrombosis and bleeding. Consequently, the therapeutic potency of a heparin is determined by its aIIa activity, i.e., the concentration of a domain in which 12 sugar flank the high affinity antithrombin-binding pentasaccharide (HA5) at one side. The response of individual plasmas to a fixed dose of any heparin is highly variable. This suggests that individualization of heparin dosage, on basis of the ETP, might reduce bleeding or re-thrombosis. There exist simple laboratory methods for both the ETP and the concentration of the active domain. These methods can be used both for unequivocally characterization of a heparin preparation and for controlling heparin therapy and allow arbitrary units relative to a standard to be abandoned. These tests are as robust as any hematological routine test but not yet routinely available, which severely encumbers progress in the field.

Rivaroxaban pharmacodynamics in healthy volunteers evaluated with thrombin generation and the active protein C system: Modeling and assessing interindividual variability

BACKGROUND

Rivaroxaban is a direct factor Xa inhibitor with substantial inter-individual pharmacokinetic (PK) variability. Pharmacodynamic (PD) variability, especially assessed with thrombin generation (TG), has been less documented.

OBJECTIVES

(i) To assess TG parameter time profiles in healthy volunteers, with TG being studied under different conditions and (ii) to model the relationship between rivaroxaban concentrations and TG parameters and subsequently estimate interindividual variability.

METHODS

Sixty healthy male volunteers (DRIVING-NCT01627665) received a single 40-mg rivaroxaban dose. Blood sampling was performed at baseline and 10 predefined time points over 24 h. The TG was investigated with the fully automated ST-Genesia system (Stago), using two tissue-factor (TF) concentrations, in the absence (-), or presence (+) of thrombomodulin (TM) for the lowest one. The PD models were built to characterize the relationships between plasma rivaroxaban concentrations and endogenous thrombin potential (ETP) or peak height induced by the lowest TF concentration.

RESULTS

Thrombin generation parameter time profiles with the lowest TF concentration showed a good sensitivity to rivaroxaban, especially +TM (active protein C negative feedback). The relationship between rivaroxaban concentrations and TG parameters was modeled with a sigmoidal relation. Mean rivaroxaban concentrations halving the baseline value of ETP and peak height (-TM) (C50 ) were of 284 and 33.2 ng/mL, respectively: +TM, C50 declined to 19.4 and 13.8 ng/mL, reflecting a powerful inhibitory effect. The estimated C50 population coefficients of variation were of 12.2% (-TM) and 31.3% (+TM) with the peak height models, 34.8% (+TM) with the ETP model.

CONCLUSIONS

This low-rivaroxaban to moderate-rivaroxaban PD variability in healthy volunteers contrasts with the substantial PK variability and deserves to be studied in different patient settings.

Calibrated Automated Thrombinography (CAT), a Tool to Identify Patients at Risk of Bleeding during Anticoagulant Therapy: A Systematic Review

Background  Bleeding is a feared adverse event during anticoagulant treatment. In patients on vitamin K antagonists, most bleedings occur with the international normalized ratio (INR) in the therapeutic range. Currently, identification of high-risk patients via laboratory methods is not reliable. In this systematic review, we assessed the ability of calibrated automated thrombin generation (CAT-TG) to predict bleeding in patients on anticoagulant treatment. Methods  A systematic search was executed in three databases: Medline, Embase, and Cochrane. Results  Seven studies were included; two were of good methodological quality. One study showed that patients on warfarin with INRs in range (2-3) admitted for hemorrhage ( n  = 28), had lower CAT-TG levels (endogenous thrombin potential [ETP]: 333 ± 89 nM/min) than patients on warfarin admitted for other reasons (ETP: 436 ± 207 nM/min; p  < 0.001). Another study found no difference in ETP or peak levels between bleeding and nonbleeding patients in PPP or PRP. When measured in whole blood, both levels were significantly lower in patients with bleeding compared with nonbleeding patients (median [interquartile range, IQR] ETP: 182.5 [157.2-2,847 nM/min] vs. median [IQR] ETP: 256.2 [194.9-344.2 nM/min]; p  < 0.001) and median [IQR] peak: 23.9 [19.6-41.8 nM] vs. median [IQR] peak: 39.1 [24.9-53.2 nM]; p  < 0.05). From the remaining studies, four suggested that CAT-TG is more sensitive in detecting hemostatic abnormalities than INR and one article found ETP and INR to be equally useful. However, insufficient data were provided to validate these conclusions. Conclusion  Studies investigating the direct association between decreased CAT-TG values and hemorrhagic events are rare. Therefore, the clinical consequences of low CAT-TG values remain to be further investigated.

Thrombin generation in low plasma volumes

Accurate thrombin generation determination by calibrated automated thrombinography can be sustained when reducing the plasma and reagent volumes up to half, but not for higher reductions or plasma dilutions.

Continuous thrombin generation in whole blood: New applications for assessing activators and inhibitors of coagulation

Hemostatic tests have been utilized to clarify the blood coagulation potential. The novel thrombin generation (TG) assay of this study provides explicit information and is the most physiologically-relevant hemostatic test ex vivo. We describe how this assay allows for TG under a number of relevant circumstances. First, whole blood (WB) from healthy individuals was analyzed ± 5 pM tissue factor (TF) and ± contact pathway inhibition. Without an exogenous initiator TG was decreased and delayed, but addition of 5 pM TF shortened the lag phase and increased peak thrombin. Additional experiments included fresh WB from a trauma patient analyzed for endogenous activity and TG from healthy donors subjected to TG antagonists which prolonged the lag phase whereas TG agonists consistently shortened the lag phase in a dose dependent manner. Lastly, platelet-poor plasma was reconstituted with packed red blood cells and TG was monitored in the presence and absence of both TF as an activator and PCPS as a phospholipid surface. Our data illustrate the potential that this continuous TG assay has in the evaluation of disorders relevant to blood coagulation and in the monitoring of treatments administered in response to these disorders.