Risk of recurrent venous thromboembolism in patients with cancer: an individual patient data meta-analysis and development of a prediction model.

Background About 7% of patients with cancer-associated venous thromboembolism (CAT) develop a recurrence during anticoagulant treatment. Identification of high-risk patients may help guide treatment decisions. Aim To identify clinical predictors and develop a prediction model for on-treatment recurrent CAT. Methods For this individual patient data (IPD) meta-analysis, we used data from four randomized controlled trials evaluating low-molecular-weight heparin (LMWH) or direct oral anticoagulants (DOACs) for CAT (Hokusai VTE Cancer, SELECT-D, CLOT, and CATCH). The primary outcome was adjudicated on-treatment recurrent CAT during 6-month follow-up. A clinical prediction model was developed using multivariable logistic regression analysis with backward selection. This model was validated using internal-external cross validation. Performance was assessed by the c-statistic and a calibration plot. Results After excluding patients using vitamin K antagonists, the combined dataset comprised 2,245 patients with cancer and acute CAT who were treated with edoxaban (23%), rivaroxaban (9%), dalteparin (47%), or tinzaparin (20%). Recurrent on-treatment CAT during 6-month follow-up occurred in 150 (6.7%) patients. Predictors included in the final model were age (restricted cubic spline), breast cancer (OR 0.42; 95%-CI 0.20-0.87), metastatic disease (OR 1.44; 95%-CI 1.01-2.05), treatment with DOAC (OR 0.66; 95%-CI 0.44-0.98), and deep vein thrombosis only as index event (OR 1.72; 95%-CI 1.31-2.27). The c-statistic of the model was 0.63 (95%-CI 0.54-0.72) after internal-external cross validation. Calibration varied across studies. Conclusions The prediction model for recurrent CAT included five clinical predictors and has only modest discrimination. Prediction of recurrent CAT at the initiation of anticoagulation remains challenging.

Direct factor Xa inhibitors and the risk of cancer and cancer mortality: A Danish population-based cohort study.

BACKGROUND: Preclinical animal studies have suggested that myeloid cell-synthesized coagulation factor X dampens antitumor immunity and that rivaroxaban, a direct factor Xa inhibitor, can be used to promote tumor immunity. This study was aimed at assessing whether patients with atrial fibrillation taking direct factor Xa inhibitors have lower risk of cancer and cancer-related mortality than patients taking the direct thrombin inhibitor dabigatran. METHODS AND FINDINGS: This nationwide population-based cohort study in Denmark included adult patients with atrial fibrillation and without a history of cancer, who started taking a factor Xa inhibitor or dabigatran between 2011 and 2015. Data on medical history, outcomes, and drug use were acquired through Danish healthcare registries. The primary outcome was any cancer. Secondary outcomes were cancer-related mortality and all-cause mortality. Outcome events were assessed during 5 years of follow-up in an intention-to-treat analysis. The propensity score-based inverse probability of treatment weighting was used to compute cumulative incidence and subdistribution hazard ratios (SHRs) and corresponding 95% confidence intervals (CIs), with death as a competing event. Propensity scores were estimated using logistic regression and including in the model sex, age group at index date, comorbidities, and use of comedications. A total of 11,742 patients with atrial fibrillation starting a factor Xa inhibitor and 11,970 patients starting dabigatran were included. Mean age was 75.2 years (standard deviation [SD] 11.2) in the factor Xa cohort and 71.7 years (SD 11.1) in the dabigatran cohort. On the basis of the propensity score-weighted models, after 5 years of follow-up, no substantial difference in the cumulative incidence of cancer was observed between the factor Xa inhibitor (2,157/23,711; 9.11%, 95% CI [8.61%,9.63%]) and dabigatran (2,294/23,715; 9.68%, 95% CI [9.14%,10.25%]) groups (SHR 0.94, 95% CI [0.89,1.00], P value 0.0357). We observed no difference in cancer-related mortality (factor Xa inhibitors cohort 1,028/23,711; 4.33%, 95% CI [4.02%,4.68%]. Dabigatran cohort 1,001/23,715; 4.22%, 95% CI [3.83%,4.66%]; SHR 1.03, 95% CI [0.94,1.12]), but all-cause mortality was higher in the factor Xa inhibitor cohort (factor Xa inhibitors cohort 7,416/23,711; 31.31%, 95% CI [30.37%,32.29%]. Dabigatran cohort 6,531/23,715; 27.56%, 95% CI [26.69%,28.45%]; HR 1.17, 95% CI [1.13,1.21]). The main limitations of the study were the possibility of residual confounding and the short follow-up period. CONCLUSIONS: In this population based cohort study, factor Xa inhibitor use was not associated with an overall lower incidence of cancer or cancer-related mortality when compared to dabigatran. We did observe an increase in all-cause mortality in the factor Xa inhibitor cohort.

Targeted exome analysis in patients with rare bleeding disorders: data from the Rare Bleeding Disorders in the Netherlands study.

Background: Rare coagulation factor deficiencies and disorders of fibrinolysis (defined as rare bleeding disorders [RBDs]) present with a heterogeneous bleeding phenotype, and bleeding severity is difficult to predict. Objectives: Describe underlying rare genetic variants in the Dutch RBD population and investigate the relationship between genotype, laboratory phenotype, and clinical phenotype. Methods: The Rare Bleeding Disorders in the Netherlands is a cross-sectional, nationwide study conducted between October 1, 2017, and November 30, 2019. Bleeding scores and blood samples were collected during a single study visit. Coagulation factor levels were measured centrally, and targeted exome analysis was performed on 156 genes involved in thrombosis and hemostasis. Pathogenicity was assigned according to the Association for Clinical Genetic Science guidelines. Results: Rare genetic variants specific to the diagnosed RBD were found in 132 of 156 patients (85%). Of the 214 rare genetic variants identified, 57% (n = 123) were clearly pathogenic, 19% (n = 40) were likely pathogenic, and 24% (n = 51) were variants of unknown significance. No explanatory genetic variants were found in patients with plasminogen activator inhibitor type 1 deficiency or hyperfibrinolysis. A correlation existed between factor activity levels and the presence of a genetic variant in the corresponding gene in patients with rare coagulation factor deficiencies and alpha-2-antiplasmin deficiency. Co-occurrence of multiple genetic variants was present in a quarter of patients, but effect on phenotype remains unclear. Conclusion: Targeted exome analysis may offer advantages over single-gene analysis, emphasized by a number of combined deficiencies in this study. Further studies are required to determine the role of co-occurring hemostasis gene variants on the bleeding phenotype in RBDs.

Risk assessment tools for bleeding in patients with unprovoked venous thromboembolism: an analysis of the PLATO-VTE study.

BACKGROUND: Guidelines suggest indefinite anticoagulation after unprovoked venous thromboembolism (VTE) unless the bleeding risk is high, yet there is no consistent guidance on assessing bleeding risk. OBJECTIVES: This study aimed to evaluate the performance of 5 bleeding risk tools (RIETE, VTE-BLEED, CHAP, VTE-PREDICT, and ABC-Bleeding). METHODS: PLATO-VTE, a prospective cohort study, included patients aged ≥40 years with a first unprovoked VTE. Risk estimates were calculated at VTE diagnosis and after 3 months of treatment. Primary outcome was clinically relevant bleeding, as per International Society on Thrombosis and Haemostasis criteria, during 24-month follow-up. Discrimination was assessed by the area under the receiver operating characteristic curve (AUROC). Patients were classified as having a "high risk" and "non-high risk" of bleeding according to predefined thresholds; bleeding risk in both groups was compared by hazard ratios (HRs). RESULTS: Of 514 patients, 38 (7.4%) had an on-treatment bleeding. AUROCs were 0.58 (95% CI, 0.48-0.68) for ABC-Bleeding, 0.56 (95% CI, 0.46-0.66) for RIETE, 0.53 (95% CI, 0.43-0.64) for CHAP, 0.50 (95% CI, 0.41-0.59) for VTE-BLEED, and 0.50 (95% CI, 0.40-0.60) for VTE-PREDICT. The proportion of high-risk patients ranged from 1.4% with RIETE to 36.9% with VTE-BLEED. The bleeding incidence in the high-risk groups ranged from 0% with RIETE to 13.0% with ABC-Bleeding, and in the non-high-risk groups, it varied from 7.7% with ABC-Bleeding to 9.6% with RIETE. HRs ranged from 0.93 (95% CI, 0.46-1.9) for VTE-BLEED to 1.67 (95% CI, 0.86-3.2) for ABC-Bleeding. Recalibration at 3-month follow-up did not alter the results. CONCLUSION: In this cohort, discrimination of currently available bleeding risk tools was poor. These data do not support their use in patients with unprovoked VTE.

The effect of current antithrombotic therapy on mortality in nursing home residents with COVID-19: a multicentre retrospective cohort study.

BACKGROUND: The first wave of COVID led to an alarmingly high mortality rate among nursing home residents (NHRs). In hospitalised patients, the use of anticoagulants may be associated with a favourable prognosis. However, it is unknown whether the use of antithrombotic medication also protected NHRs from COVID-19-related mortality. OBJECTIVES: To investigate the effect of current antithrombotic therapy in NHRs with COVID-19 on 30-day all-cause mortality during the first COVID-19 wave. METHODS: We performed a retrospective cohort study linking electronic health records and pharmacy data in NHRs with COVID-19. A propensity score was used to match NHRs with current use of therapeutic dose anticoagulants to NHRs not using anticoagulant medication. The primary outcome was 30-day all-cause mortality, which was evaluated using a logistic regression model. In a secondary analysis, multivariable logistic regression was performed in the complete study group to compare NHRs with current use of therapeutic dose anticoagulants and those with current use of antiplatelet therapy to those without such medication. RESULTS: We included 3521 NHRs with COVID-19 based on a positive RT-PCR for SARS-CoV-2 or with a well-defined clinical suspicion of COVID-19. In the matched propensity score analysis, NHRs with current use of therapeutic dose anticoagulants had a significantly lower all-cause mortality (OR = 0.73; 95% CI: 0.58-0.92) compared to NHRs who did not use therapeutic anticoagulants. In the secondary analysis, current use of therapeutic dose anticoagulants (OR: 0.62; 95% CI: 0.48-0.82) and current use of antiplatelet therapy (OR 0.80; 95% CI: 0.64-0.99) were both associated with decreased mortality. CONCLUSIONS: During the first COVID-19 wave, therapeutic anticoagulation and antiplatelet use were associated with a reduced risk of all-cause mortality in NHRs. Whether these potentially protective effects are maintained in vaccinated patients or patients with other COVID-19 variants, remains unknown.

The Incidence of Pulmonary Embolism in Hospitalized Non-ICU Patients with COVID-19 during the First Wave: A Multicenter Retrospective Cohort Study in the Netherlands.

INTRODUCTION: During the first COVID-19 outbreak in 2020 in the Netherlands, the incidence of pulmonary embolism (PE) appeared to be high in COVID-19 patients admitted to the intensive care unit (ICU). This study was performed to evaluate the incidence of PE during hospital stay in COVID-19 patients not admitted to the ICU. METHODS: Data were retrospectively collected from 8 hospitals in the Netherlands. Patients admitted between February 27, 2020, and July 31, 2020, were included. Data extracted comprised clinical characteristics, medication use, first onset of COVID-19-related symptoms, admission date due to COVID-19, and date of PE diagnosis. Only polymerase chain reaction (PCR)-positive patients were included. All PEs were diagnosed with computed tomography pulmonary angiography (CTPA). RESULTS: Data from 1,852 patients who were admitted to the hospital ward were collected. Forty patients (2.2%) were diagnosed with PE within 28 days following hospital admission. The median time to PE since admission was 4.5 days (IQR 0.0-9.0). In all 40 patients, PE was diagnosed within the first 2 weeks after hospital admission and for 22 (55%) patients within 2 weeks after onset of symptoms. Patient characteristics, pre-existing comorbidities, anticoagulant use, and laboratory parameters at admission were not related to the development of PE. CONCLUSION: In this retrospective multicenter cohort study of 1,852 COVID-19 patients only admitted to the non-ICU wards, the incidence of CTPA-confirmed PE was 2.2% during the first 4 weeks after onset of symptoms and occurred exclusively within 2 weeks after hospital admission.

Unmet needs and barriers in venous thromboembolism education and awareness among people living with cancer: a global survey.

BACKGROUND: Venous thromboembolism (VTE) is a major preventable cause of morbidity, disability, and mortality in subjects with cancer. A global appraisal of cancer-associated VTE education and awareness is not available. OBJECTIVES: To evaluate VTE-related education, awareness, and unmet needs from the perspective of people living with cancer using a quantitative and qualitative approach. METHODS: This cross-sectional study used data from an online-based survey covering multidimensional domains of cancer-associated VTE. Data are presented descriptively. Potential differences across participant subgroups were explored. RESULTS: Among 2262 patients with cancer from 42 countries worldwide, 55.3% received no VTE education throughout their cancer journey, and an additional 8.2% received education at the time of VTE diagnosis only, leading to 63.5% receiving no or inappropriately delayed education. When education was delivered, only 67.8% received instructions to seek medical attention in case of VTE suspicion, and 36.9% reported scarce understanding. One-third of participants (32.4%) felt psychologically distressed when becoming aware of the potential risks and implications connected with cancer-associated VTE. Most responders (78.8%) deemed VTE awareness highly relevant, but almost half expressed concerns about the quality of education received. While overall consistent, findings in selected survey domains appeared to numerically differ across age group, ethnicity, continent of residence, educational level, metastatic status, and VTE history. CONCLUSION: This study involving a large and diverse population of individuals living with cancer identifies important unmet needs in VTE-related education, awareness, and support across healthcare systems globally. These findings unveil multilevel opportunities to expedite patient-centered care in cancer-associated VTE prevention and management.

Tailored anticoagulant treatment after a first venous thromboembolism: protocol of the Leiden Thrombosis Recurrence Risk Prevention (L-TRRiP) study - cohort-based randomised controlled trial.

INTRODUCTION: Patients with a first venous thromboembolism (VTE) are at risk of recurrence. Recurrent VTE (rVTE) can be prevented by extended anticoagulant therapy, but this comes at the cost of an increased risk of bleeding. It is still uncertain whether patients with an intermediate recurrence risk or with a high recurrence and high bleeding risk will benefit from extended anticoagulant treatment, and whether a strategy where anticoagulant duration is tailored on the predicted risks of rVTE and bleeding can improve outcomes. The aim of the Leiden Thrombosis Recurrence Risk Prevention (L-TRRiP) study is to evaluate the outcomes of tailored duration of long-term anticoagulant treatment based on individualised assessment of rVTE and major bleeding risks. METHODS AND ANALYSIS: The L-TRRiP study is a multicentre, open-label, cohort-based, randomised controlled trial, including patients with a first VTE. We classify the risk of rVTE and major bleeding using the L-TRRiP and VTE-BLEED scores, respectively. After 3 months of anticoagulant therapy, patients with a low rVTE risk will discontinue anticoagulant treatment, patients with a high rVTE and low bleeding risk will continue anticoagulant treatment, whereas all other patients will be randomised to continue or discontinue anticoagulant treatment. All patients will be followed up for at least 2 years. Inclusion will continue until the randomised group consists of 608 patients; we estimate to include 1600 patients in total. The primary outcome is the combined incidence of rVTE and major bleeding in the randomised group after 2 years of follow-up. Secondary outcomes include the incidence of rVTE and major bleeding, functional outcomes, quality of life and cost-effectiveness in all patients. ETHICS AND DISSEMINATION: The protocol was approved by the Medical Research Ethics Committee Leiden-Den Haag-Delft. Results are expected in 2028 and will be disseminated through peer-reviewed journals and during (inter)national conferences. TRIAL REGISTRATION NUMBER: NCT06087952.

Elevated levels of citrullinated fibrinogen in patients with cancer.

Neutrophil released peptidyl arginine deiminase 4 (PAD4) converts arginine residues on plasma proteins into citrulline. Here, we developed an assay to quantify citrullinated fibrinogen. We employed a biotin-conjugated phenylglyoxal (biotin-phenylglyoxal (PG)) compound that selectively labels citrulline. Patient samples were derived from a multicenter prospective cohort study that aimed to identify cancer patients at high risk for venous thromboembolism (VTE). Our data show that cancer patients have higher (median 2-fold increased) citrullinated fibrinogen levels when compared to normal human plasma and a cohort of healthy donors. Our results show that citrullination of fibrinogen is a common posttranslational modification in patients with cancer.

RNA-sequencing to discover genes and signaling pathways associated with venous thromboembolism in glioblastoma patients: A case-control study.

BACKGROUND: Glioblastoma patients are at high risk of developing venous thromboembolism (VTE). Tumor-intrinsic features are considered to play a role, but the underlying pathophysiological mechanisms remain incompletely understood. OBJECTIVES: To identify tumor-expressed genes and signaling pathways that associate with glioblastoma-related VTE by using next generation RNA-sequencing (RNA-Seq). METHODS: The tumor gene expression profile of 23 glioblastoma patients with VTE and 23 glioblastoma patients without VTE was compared using an unpaired analysis. Ingenuity Pathway Analysis (IPA) core analysis was performed on the top 50 differentially expressed genes to explore associated functions and pathways. Based on full RNA-Seq data, molecular glioblastoma subtypes were determined by performing cluster analysis. RESULTS: Of the 19,327 genes, 1246 (6.4 %) were differentially expressed between glioblastoma patients with and without VTE (unadjusted P < 0.05). The most highly overexpressed gene was GLI1, a classical target gene in the Sonic Hedgehog (Shh) signaling pathway (log2 fold change: 3.7; unadjusted P < 0.0001, adjusted P = 0.219). In line, Shh signaling was among the top canonical pathways and processes associated with VTE. The proportion of patients with the proneural/neural glioblastoma subtype was higher among those with VTE than controls. CONCLUSION: Shh signaling may be involved in the development of glioblastoma-related VTE.

Impact of venous thromboembolism on the mortality in patients with cancer: a population-based cohort study.

Background: Despite recent improvements in the treatment of cancer, little is known about the long-term survival in patients with cancer and venous thromboembolism. We aimed to examine the five-year mortality of venous thromboembolism in cancer patients in a large population-based cohort study. Methods: Using Danish healthcare registries from 1995 to 2020, we obtained data on cancer patients with venous thromboembolism and comparison cohorts of cancer patients without venous thromboembolism, matched in terms of cancer type, age, sex, and year of cancer diagnosis, and adjusted for level of comorbidity and frailty using the Charlson Comorbidity Index Score and Hospital Frailty Risk Score, marital status, use of selected medications, and recent surgery (<90 days). Findings: During the study period, 886,536 patients were diagnosed with cancer. Of 1882 cancer patients diagnosed at the time of their venous thromboembolism, 44.4% (835/1882) had distant metastases. In this cohort, the one- and five-year mortality cumulative incidences were 68% (1284/1882) and 84% (1578/1882), respectively, in contrast to 38% (2135/5549) and 67% (3653/5549) in the comparison cohort. The mortality rate ratio was 4.34 (95% confidence interval [CI], 3.95-4.78) for the first year of follow-up and 3.44 (95% CI 3.17-3.73) for the five-year follow-up period. Of the 23,366 patients diagnosed with venous thromboembolism after cancer diagnosis, 18% (4183/23,366) had distant metastases at the time of cancer diagnosis. The cumulative incidence of death at one year was 45% (10,465/23,366; mortality rate ratio 3.48, 95% CI 3.37-3.60) and at five years 69% (15,669/23,366; mortality rate ratio 2.57, 95% CI 2.50-2.63). Interpretation: Despite improved cancer treatment, venous thromboembolism in cancer patients is strongly associated with a poor prognosis. Funding: The study was supported by grants from the Independent Research Fund Denmark (record no. 3101-00102B) and the Karen Elise Jensen Foundation.

Performance of the 4-Level Pulmonary Embolism Clinical Probability Score (4PEPS) in the diagnostic management of pulmonary embolism: An external validation study.

BACKGROUND: The recently published 4-level Pulmonary Embolism Clinical Probability Score (4PEPS) integrates different aspects from currently available diagnostic strategies to further reduce imaging testing in patients with clinically suspected pulmonary embolism (PE). AIM: To externally validate the performance of 4PEPS in an independent cohort. METHODS: In this post-hoc analysis of the prospective diagnostic management YEARS study, the primary outcome measures were discrimination, calibration, efficiency (proportion of imaging tests potentially avoided), and failure rate (venous thromboembolism (VTE) diagnosis at baseline or follow-up in patients with a negative 4PEPS algorithm). Multiple imputation was used for missing 4PEPS items. Based on 4PEPS, PE was considered ruled out in patients with a very low clinical pre-test probability (CPTP) without D-dimer testing, in patients with a low CPTP and D-dimer <1000 µg/L, and in patients with a moderate CPP and D-dimer below the age-adjusted threshold. RESULTS: Of the 3465 patients, 474 (14 %) were diagnosed with VTE at baseline or during 3-month follow-up. Discriminatory performance of the 4PEPS items was good (area under ROC-curve, 0.82; 95%CI, 0.80-0.84) as was calibration. Based on 4PEPS, PE could be considered ruled out without imaging in 58 % (95%CI 57-60) of patients (efficiency), for an overall failure rate of 1.3 % (95%CI 0.86-1.9). CONCLUSION: In this retrospective external validation, 4PEPS appeared to safely rule out PE with a high efficiency. Nevertheless, although not exceeding the failure rate margin by ISTH standards, the observed failure rate in our analysis appeared to be higher than in the original 4PEPS derivation and validation study. This highlights the importance of a prospective outcome study.

Polygenic risk scores for prediction of cancer-associated venous thromboembolism in the UK Biobank cohort study.

BACKGROUND: Guidelines recommend thromboprophylaxis for patients with cancer at high risk of venous thromboembolism (VTE). Polygenic risk scores may improve VTE prediction but have not yet been evaluated in patients with cancer. OBJECTIVES: We assessed the performance of the 5-, 37-, 297-, extended 297- (additionally including factor V Leiden and prothrombin G20210A), and 100-single-nucleotide polymorphism (SNP) scores in predicting cancer-associated VTE in the UK Biobank, a population-based, prospective cohort study. METHODS: The primary outcome was VTE during 12 months after cancer diagnosis. Cancer and VTE diagnosis were based on ICD-10 codes. Discrimination was evaluated by c-indices and subdistribution hazard ratios in the upper vs 3 lower quartiles of the scores in a competing risk model. As a comparison, the c-index was calculated for the Khorana cancer type risk classification. RESULTS: Of 36 150 patients with cancer (median age, 66 years; 48.7% females), 1018 (2.8%) developed VTE. C-indices at 12 months ranged from 0.56 (95% CI, 0.54-0.58) for the 5-SNP to 0.60 (95% CI, 0.58-0.62) for the extended 297-SNP scores. The subdistribution hazard ratios ranged from 1.36 (95% CI, 1.19-1.56) for the 5-SNP to 1.90 (95% CI, 1.68-2.16) for the extended 297-SNP scores and were consistent after adjusting for cancer type. For the Khorana cancer type classification, the c-index was 0.60 (95% CI, 0.58-0.61), which increased to 0.65 (95% CI, 0.63-0.67, +0.05; 95% CI, 0.04-0.07) when combined with the extended 297-SNP score. CONCLUSION: These findings demonstrate that polygenic VTE risk scores can identify patients with cancer with a 1.9-fold higher VTE risk independent of cancer type. Combined clinical-genetic scores to improve cancer-associated VTE prediction should be evaluated further.

Diagnostic management of acute pulmonary embolism: a prediction model based on a patient data meta-analysis.

AIMS: Risk stratification is used for decisions regarding need for imaging in patients with clinically suspected acute pulmonary embolism (PE). The aim was to develop a clinical prediction model that provides an individualized, accurate probability estimate for the presence of acute PE in patients with suspected disease based on readily available clinical items and D-dimer concentrations. METHODS AND RESULTS: An individual patient data meta-analysis was performed based on sixteen cross-sectional or prospective studies with data from 28 305 adult patients with clinically suspected PE from various clinical settings, including primary care, emergency care, hospitalized and nursing home patients. A multilevel logistic regression model was built and validated including ten a priori defined objective candidate predictors to predict objectively confirmed PE at baseline or venous thromboembolism (VTE) during follow-up of 30 to 90 days. Multiple imputation was used for missing data. Backward elimination was performed with a P-value <0.10. Discrimination (c-statistic with 95% confidence intervals [CI] and prediction intervals [PI]) and calibration (outcome:expected [O:E] ratio and calibration plot) were evaluated based on internal-external cross-validation. The accuracy of the model was subsequently compared with algorithms based on the Wells score and D-dimer testing. The final model included age (in years), sex, previous VTE, recent surgery or immobilization, haemoptysis, cancer, clinical signs of deep vein thrombosis, inpatient status, D-dimer (in µg/L), and an interaction term between age and D-dimer. The pooled c-statistic was 0.87 (95% CI, 0.85-0.89; 95% PI, 0.77-0.93) and overall calibration was very good (pooled O:E ratio, 0.99; 95% CI, 0.87-1.14; 95% PI, 0.55-1.79). The model slightly overestimated VTE probability in the lower range of estimated probabilities. Discrimination of the current model in the validation data sets was better than that of the Wells score combined with a D-dimer threshold based on age (c-statistic 0.73; 95% CI, 0.70-0.75) or structured clinical pretest probability (c-statistic 0.79; 95% CI, 0.76-0.81). CONCLUSION: The present model provides an absolute, individualized probability of PE presence in a broad population of patients with suspected PE, with very good discrimination and calibration. Its clinical utility needs to be evaluated in a prospective management or impact study. REGISTRATION: PROSPERO ID 89366.

Covariate-specific ROC curve analysis can accommodate differences between covariate subgroups in the evaluation of diagnostic accuracy.

OBJECTIVES: We present an illustrative application of methods that account for covariates in receiver operating characteristic (ROC) curve analysis, using individual patient data on D-dimer testing for excluding pulmonary embolism. STUDY DESIGN AND SETTING: Bayesian nonparametric covariate-specific ROC curves were constructed to examine the performance/positivity thresholds in covariate subgroups. Standard ROC curves were constructed. Three scenarios were outlined based on comparison between subgroups and standard ROC curve conclusion: (1) identical distribution/identical performance, (2) different distribution/identical performance, and (3) different distribution/different performance. Scenarios were illustrated using clinical covariates. Covariate-adjusted ROC curves were also constructed. RESULTS: Age groups had prominent differences in D-dimer concentration, paired with differences in performance (Scenario 3). Different positivity thresholds were required to achieve the same level of sensitivity. D-dimer had identical performance, but different distributions for YEARS algorithm items (Scenario 2), and similar distributions for sex (Scenario 1). For the later covariates, comparable positivity thresholds achieved the same sensitivity. All covariate-adjusted models had AUCs comparable to the standard approach. CONCLUSION: Subgroup differences in performance and distribution of results can indicate that the conventional ROC curve is not a fair representation of test performance. Estimating conditional ROC curves can improve the ability to select thresholds with greater applicability.

Contact system and intrinsic pathway activation in patients with advanced pancreatic cancer: a prospective cohort study.

BACKGROUND: Despite high risk of venous thromboembolism (VTE) in patients with pancreatic cancer, there are little data on contact system activation in these patients. OBJECTIVES: To quantify contact system and intrinsic pathway activation and subsequent VTE risk in patients with pancreatic cancer. METHODS: Patients with advanced pancreatic cancer were compared with controls. Blood was drawn at baseline and patients were followed for 6 months. Complexes of proteases with their natural inhibitors, C1-esterase inhibitor (C1-INH), antithrombin (AT), or alpha-1 antitrypsin (α1at), were measured for complexes containing kallikrein (PKa:C1-INH), factor (F)XIIa (FXIIa:C1-INH), and FXIa (FXIa:C1-INH, FXIa:AT, FXIa:α1at). The association of cancer with complex levels was assessed in a linear regression model, adjusted for age, sex, and body mass index. In a competing risk regression model, we assessed associations between complex levels and VTE. RESULTS: One hundred nine patients with pancreatic cancer and 22 controls were included. The mean age was 66 years (SD, 8.4) in the cancer cohort and 52 years (SD, 10.1) in controls. In the cancer cohort, 18 (16.7%) patients developed VTE during follow-up. In the multivariable regression model, pancreatic cancer was associated with increased complexes of PKa:C1-INH (P < .001), FXIa:C1-INH (P < .001), and FXIa:AT (P < .001). High FXIa:α1at (subdistribution hazard ratio, 1.48 per log increase; 95% CI, 1.02-2.16) and FXIa:AT (subdistribution hazard ratio, 2.78 highest vs lower quartiles; 95% CI, 1.10-7.00) were associated with VTE. CONCLUSION: Complexes of proteases with their natural inhibitors were elevated in patients with cancer. These data suggest that the contact system and intrinsic pathway activation are increased in patients with pancreatic cancer.

Accuracy of physicians' intuitive risk estimation in the diagnostic management of pulmonary embolism: an individual patient data meta-analysis.

BACKGROUND: In patients clinically suspected of having pulmonary embolism (PE), physicians often rely on intuitive estimation ("gestalt") of PE presence. Although shown to be predictive, gestalt is criticized for its assumed variation across physicians and lack of standardization. OBJECTIVES: To assess the diagnostic accuracy of gestalt in the diagnosis of PE and gain insight into its possible variation. METHODS: We performed an individual patient data meta-analysis including patients suspected of having PE. The primary outcome was diagnostic accuracy of gestalt for the diagnosis of PE, quantified as risk ratio (RR) between gestalt and PE based on 2-stage random-effect log-binomial meta-analysis regression as well as gestalts' sensitivity and specificity. The variability of these measures was explored across different health care settings, publication period, PE prevalence, patient subgroups (sex, heart failure, chronic lung disease, and items of the Wells score other than gestalt), and age. RESULTS: We analyzed 20 770 patients suspected of having PE from 16 original studies. The prevalence of PE in patients with and without a positive gestalt was 28.8% vs 9.1%, respectively. The overall RR was 3.02 (95% CI, 2.35-3.87), and the overall sensitivity and specificity were 74% (95% CI, 68%-79%) and 61% (95% CI, 53%-68%), respectively. Although variation was observed across individual studies (I2, 90.63%), the diagnostic accuracy was consistent across all subgroups and health care settings. CONCLUSION: A positive gestalt was associated with a 3-fold increased risk of PE in suspected patients. Although variation was observed across studies, the RR of gestalt was similar across prespecified subgroups and health care settings, exemplifying its diagnostic value for all patients suspected of having PE.

Effect of the P-glycoprotein inhibitor tamoxifen on edoxaban plasma levels in women with breast cancer.

BACKGROUND: Concomitant use of P-glycoprotein inhibitors can reduce clearance of edoxaban and increase its plasma concentration. Caution is advised with simultaneous use of edoxaban and the frequently used P-glycoprotein inhibitor tamoxifen. However, pharmacokinetic data are lacking. OBJECTIVES: This study aimed to assess the effect of tamoxifen on edoxaban clearance. METHODS: This was a prospective, self-controlled, pharmacokinetic study in breast cancer participants starting tamoxifen. Edoxaban was given at a dose of 60 mg once daily for 4 consecutive days, first without tamoxifen and later with concomitant tamoxifen in steady-state. On day 4 of both edoxaban sequences, serial blood samples were taken. A population pharmacokinetic model was developed using nonlinear mixed effects modelling in which the effect of tamoxifen on edoxaban clearance was assessed. Additionally, mean area under the curves (AUC) were estimated. Geometric least square means (GLM) ratios were calculated and no interaction was concluded if the 90 % CI was within the 80-125 % no-effect boundaries. RESULTS: Twenty-four women with breast cancer scheduled for tamoxifen were included. The median age was 56 years (IQR 51-63). The average edoxaban clearance was 32.0 L/h (95 % CI, 11.1-35.0 L/h). There was no effect of tamoxifen on edoxaban clearance, with a fraction of 100 % (95 % CI 92-108) compared to clearance without tamoxifen. The mean AUCs were 1923 ng*h/ml (SD 695) without tamoxifen and 1947 ng*h/ml (SD 595) with tamoxifen (GLM-ratio 100.4; 90 % CI 98.6-102.2). CONCLUSIONS: Concomitant use of the P-glycoprotein inhibitor tamoxifen does not lead to reduced clearance of edoxaban in patients with breast cancer.

Tissue factor activity of small and large extracellular vesicles in different diseases.

Background: Circulating procoagulant extracellular vesicles (EVs) are increased in diseases, such as cancer, sepsis, and COVID-19. EV tissue factor (TF) activity is associated with disseminated intravascular coagulation in sepsis and venous thrombosis in patients with pancreatic cancer and COVID-19. EVs are commonly isolated by centrifugation at ∼20,000 g. Objectives: In this study, we analyzed the TF activity of 2 EV populations enriched for large and small EVs in patients with either sepsis, pancreatic cancer, or COVID-19. Methods: EVs were isolated from plasma by sequential centrifugation at 20,000 g (large EVs, LEVs) and then 100,000 g (small EVs, SEVs). We analyzed EVs from plasma prepared from whole blood samples from healthy individuals with or without lipopolysaccharide (LPS) stimulation as well as EVs from plasma samples from patients with either sepsis, pancreatic cancer, or COVID-19. TF-dependent (EV-TF activity) and TF-independent factor Xa (FXa) generation of the EVs was measured. Results: LPS increased EV-TF activity in LEVs but not SEVs. Similarly, in 2 patients with sepsis who had EV-TF activity above the background of the assay we observed EV-TF activity in LEVs but not SEVs. Patients with pancreatic cancer or COVID-19 had circulating EV-TF activity in both LEVs and SEVs. Conclusion: We recommend that EVs are isolated from plasma from patients by centrifugation at 100,000 g rather than 20,000 g to obtain a more accurate measure of levels of circulating EV-TF activity.