Advertisement

Incidence, risk factors, and evolution of venous thromboembolic events in patients diagnosed with pancreatic carcinoma and treated with chemotherapy on an outpatient basis

Open AccessPublished:August 02, 2022DOI:https://doi.org/10.1016/j.ejim.2022.07.020

      Highlights

      • Pancreatic carcinoma was associated with an incidence of thrombosis as high as 22%.
      • Venous thromboembolic events were associated with worse survival in this population.
      • Symptomatic events correlated with higher mortality than asymptomatic events.
      • Visceral vein thrombosis negatively affected survival in these patients.
      • Khorana score did not discriminate between patients with or without thrombosis.

      Abstract

      Background

      Pancreatic carcinoma is one of the tumors associated with a higher risk for thromboembolic events, with incidence rates ranging from 5% to 41% in previous retrospective series.

      Patients and methods

      We conducted a retrospective study in eleven Spanish hospitals that included 666 patients diagnosed with pancreatic carcinoma (any stage) between 2008 and 2011 and treated with chemotherapy. The main objective was to evaluate the incidence of venous thromboembolic events (VTE) in this population, as well as potential risk factors for thrombosis. The impact of VTE on mortality was also assessed.

      Results

      With a median follow-up of 9.3 months, the incidence of VTE was 22.1%; 52% were diagnosed incidentally. Our study was unable to confirm the ability of the Khorana score to discriminate between patients in the intermediate or high risk category for thrombosis. The presence of VTE proved to be an independent prognostic factor associated with increased risk of death (HR 2.39, 95% CI 1.96–2.92). Symptomatic events correlated with higher mortality than asymptomatic events (HR 1.72; 95% CI, 1.21–2.45; p = 0.002), but incidental VTE, including visceral vein thrombosis (VVT), negatively affected survival compared to patients without VTE. Subjects who developed VTE within the first 3 months of diagnosis of pancreatic carcinoma had lower survival rates than those with VTE after 3 months (HR 1.92, 95% CI 1.30–2.84; p<0.001).

      Conclusions

      Pancreatic carcinoma is associated with a high incidence of VTE, which, when present, correlates with worse survival, even when thrombosis is incidental. Early onset VTE has a particularly negative impact.

      Keywords

      1. Introduction

      Venous thromboembolic disease is considered one of the leading causes of morbi-mortality [
      • Sørensen H.T.
      • Mellemkjaer L.
      • Olsen J.H.
      • Baron J.A.
      Prognosis of cancers associated with venous thromboembolism.
      ,
      • Khorana A.A.
      • Francis C.W.
      • Culakova E.
      • Kuderer N.M.
      • Lyman G.H.
      Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy.
      ]. in oncological patients. Cancer multiplies the risk of thromboembolic events 6- or 7-fold compared to the general population without cancer [
      • Heit J.A.
      • Silverstein M.D.
      • Mohr D.N.
      • Petterson T.M.
      • O’Fallon W.M.
      • Melton L.J.
      Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case-control study.
      ,
      • Blom J.W.
      • Doggen C.J.
      • Osanto S.
      • Rosendaal F.R.
      Malignancies, prothrombotic mutations, and the risk of venous thrombosis.
      ], and this risk is greater in certain tumor sites and in people receiving chemotherapy. The incidence of VTE in individuals with pancreatic cancer is high, although it varies widely in the literature [
      • Oh S.Y.
      • Kim J.H.
      • Lee K.W.
      • Bang S.M.
      • Hwang J.H.
      • Oh D.
      • Lee J.S.
      Venous thromboembolism in patients with pancreatic adenocarcinoma: lower incidence in Asian ethnicity.
      ,
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ,
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Mandalà M.
      • Reni M.
      • Cascinu S.
      • Barni S.
      • Floriani I.
      • Cereda S.
      • et al.
      Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
      ,
      • Khorana A.A.
      • Dalal M.
      • Lin J.
      • Connolly G.C.
      Incidence and predictors of venous thromboembolism (VTE) among ambulatory high-risk cancerpatients undergoing chemotherapy in the United States.
      ,
      • Kondo S.
      • Sasaki M.
      • Hosoi H.
      • Sakamoto Y.
      • Morizane C.
      • Ueno H.
      • Okusaka T.
      Incidence and risk factors for venous thromboembolism in patients with pretreated advanced pancreatic carcinoma.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Ouaissi M.
      • Frasconi C.
      • Mege D.
      • Panicot-Dubois L.
      • Boiron L.
      • Dahan L.
      • et al.
      Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
      ,
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • van Es N.
      • Franke V.F.
      • Middeldorp S.
      • Wilmink J.W.
      • Büller H.R
      The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Hanna-Sawires R.G.
      • Groen J.V.
      • Hamming A.
      • Tollenaar R.A.E.M.
      • Mesker W.E.
      • et al.
      Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ] due, in part, to the differences in the populations studied. In fact, several publications report incidence rates exceeding 30% [
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ]. Data regarding the impact of these events on survival are contradictory [
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Ouaissi M.
      • Frasconi C.
      • Mege D.
      • Panicot-Dubois L.
      • Boiron L.
      • Dahan L.
      • et al.
      Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ].
      Chemotherapy has been identified as an independent risk factor for VTE and most chemotherapy treatments are administered on an outpatient basis. A risk model to identify individuals at high risk of developing symptomatic VTE among those receiving chemotherapy on an outpatient basis was validated a decade ago by Khorana et al. [
      • Khorana A.A.
      • Kuderer N.M.
      • Culakova E.
      • Lyman G.H.
      • Francis C.W.
      Development and validation of a predictive model for chemotherapy-associated thrombosis.
      ]. The Khorana score stratifies patients into three risk categories depending on the primary tumor site, prechemotherapy blood cell count (leukocytes, platelets, and hemoglobin), and body mass index (BMI). One of the advantages of this model is that it can be easily calculated in routine clinical practice. Furthermore, it has proven to be predictive of mortality and progression-free survival in cancer patients [
      • Kuderer N.M.
      • Poniewierski M.S.
      • Culakova E.
      • Lyman G.H.
      • Khorana A.A.
      • Pabinger I.
      • et al.
      Predictors of venous thromboembolism and early mortality in lung cancer: results from a global prospective study (CANTARISK).
      ]. However, several observational studies conducted in different tumors [
      • Mansfield A.S.
      • Tafur A.J.
      • Wang C.E.
      • Kourelis T.V.
      • Wysokinska E.M.
      • Yang P.
      Predictors of active cancer thromboembolic outcomes: validation of the Khorana score among patients with lung cancer.
      ,
      • Ramos J.D.
      • Casey M.F.
      • Bamias A.
      • De Giorgi U.
      • Bellmunt J.
      • Harshman L.C.
      • et al.
      Retrospective international study of cancers of the urothelium (RISC) investigators. The Khorana score in predicting venous thromboembolism for patients with metastatic urothelial carcinoma and variant histology treated with chemotherapy.
      ,
      • Lim S.H.
      • Woo S.Y.
      • Kim S.
      • Ko Y.H.
      • Kim W.S.
      • Kim S.J.
      Cross-sectional study of patients with diffuse large B-cell lymphoma: assessing the effect of host status, tumor burden, and inflammatory activity on venous thromboembolism.
      ,
      • Chaudhury A.
      • Balakrishnan A.
      • Thai C.
      • Holmstrom B.
      • Nanjappa S.
      • Ma Z.
      • et al.
      Validation of the Khorana score in a large cohort of cancer patients with venous thromboembolism.
      ,
      • Tully C.M.
      • Apolo A.B.
      • Zabor E.C.
      • et al.
      The high incidence of vascular thromboembolic events in patients with metastatic or unresectable urothelial cancer treated with platinum chemotherapy agents.
      ], including pancreatic carcinoma [
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • van Es N.
      • Franke V.F.
      • Middeldorp S.
      • Wilmink J.W.
      • Büller H.R
      The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ], have failed to confirm the predictive value of the Khorana score to assess the risk of thrombosis.
      Although most VTE occur within the first few months of tumor diagnosis, they can appear at any time in the course of the disease. It has recently been suggested that early VTE prognosticates worse survival, not only in pancreatic cancer [
      • Mandalà M.
      • Reni M.
      • Cascinu S.
      • Barni S.
      • Floriani I.
      • Cereda S.
      • et al.
      Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ], but in other tumors as well [
      • Kourelis T.V.
      • Wysokinska E.M.
      • Wang Y.
      • Yang P.
      • Mansfield A.S.
      • Tafur A.J
      Early venous thromboembolic events are associated with worse prognosis in patients with lung cancer.
      ].
      Therefore, we conducted a retrospective study to establish the incidence of VTE in subjects diagnosed with pancreatic carcinoma and undergoing outpatient chemotherapy at 11 hospitals belonging to the Thrombosis and Cancer Section of the Spanish Society of Medical Oncology (SEOM, for its acronym in Spanish). Simultaneously, we examined the relationship between the Khorana score and the occurrence of thrombosis in this population, as well as the frequency and evolution of incidental events in our series, and the possible impact of early thrombosis on patient survival.

      2. Material and method

      2.1 Patients

      Consecutive adult patients diagnosed with histologically and/or cytologically confirmed exocrine pancreatic carcinoma (all stages) between January 1, 2008 and December 31, 2011 receiving outpatient chemotherapy at eleven hospitals belonging to the Thrombosis and Cancer Section of SEOM were included. The last follow-up visit was in July 2016. Cases of pancreatic neuroendocrine tumors were excluded. The inclusion of patients who were receiving anticoagulant treatment at diagnosis of pancreatic carcinoma for other reasons was allowed.
      Medical records and radiological test reports (computed tomography, magnetic resonance imaging, and ultrasound) of all individuals admitted were retrospectively reviewed. Laboratory data were collected from the blood tests immediately prior to initiating chemotherapy to calculate the Khorana score.
      The primary end point was to evaluate the incidence of VTE in our population. To this end, we included any VTE, either symptomatic or incidental, that was identified following diagnosis of pancreatic carcinoma or in the three months immediately preceding. Diagnoses of pulmonary embolism (PE), deep vein thrombosis (DVT), and all locations of visceral venous thrombosis (VVT) were included. This “extended” definition of VTE is based on the rationale that incidental thromboses and VVT would behave similarly to symptomatic events, in addition to having analogous pathophysiology. Moreover, this “extended” definition has been used in similar, recently published studies [
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Hanna-Sawires R.G.
      • Groen J.V.
      • Hamming A.
      • Tollenaar R.A.E.M.
      • Mesker W.E.
      • et al.
      Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
      ]. Catheter-associated thromboses were analyzed as DVT.
      Arterial thromboembolic events were not recorded in our series.

      2.2 Statistical analysis

      Continuous variables are presented as medians and interquartile ranges and categorical variables as absolute and relative frequencies. Statistical comparisons were performed using the χ2 test (or Fisher's exact test when necessary) for categorical variables and the Mann–Whitney–Wilcoxon test for continuous variables.
      Time to VTE was defined as the time from histological diagnosis of pancreatic cancer to VTE diagnosis. Early thrombosis was defined as occurring within the first three months after diagnosis of the pancreatic tumor.
      A univariate logistic regression model was used to explore the risk factors associated with VTE; once identified, a multivariate model was constructed to globally explain the variables involved in development of VTE. Factors with p ≤ 0.25 by univariate analysis were further analyzed in a multivariate analysis.
      Overall survival (OS) was defined as the time from diagnosis of pancreatic carcinoma to all-cause mortality. Patients who presented VTE before diagnosis of the tumor were not included in the survival analysis. The risk of the presence of VTE and other variables potentially implicated in survival was assessed through a Cox proportional hazards model in which VTE was analyzed as a time-dependent variable. In the case of stratified analyses, median survival was computed using Kaplan-Meier curves, together with hazard ratios (HR) associated with each subgroup compared to the subgroup without a diagnosis of VTE.
      Statistical analyses were performed using the statistical software package R v 4.1.1. Statistical significance was established when the p value was 0.05.

      3. Results

      The study sample consisted of 666 patients who met the inclusion criteria and were included in the analyses. Table 1 reflects the clinical characteristics of the study population.
      Table 1Study population characteristics.
      Patients with VTE
      NOverall, N = 6661No, N = 519 (78%)
      Median [IQR]; n (%).
      Yes, N = 147 (22%)
      Median [IQR]; n (%).
      p-value
      Pearson's Chi-squared test; Wilcoxon rank sum test; Fisher's exact test.
      Sex6660.003
      Male384 (57.7%)315 (60.7%)69 (46.9%)
      Female282 (42.3%)204 (39.3%)78 (53.1%)
      Age66665 [57–72]65 [58–72]64 [55–72]0.10
      BMI6450.033
      Underweight (<18.5)40 (6.2%)29 (5.8%)11 (7.6%)
      Normal (18.5–24.9)381 (59.0%)298 (59.5%)83 (57.6%)
      Overweight (25–29.9)161 (25.0%)133 (26.5%)28 (19.5%)
      Obese (>30)63 (9.8%)41 (8.2%)22 (15.3%)
      Tumor site649<0.001
      Head398 (61.3%)327 (64.7%)71 (49.3%)
      Body86 (13.3%)55 (10.9%)31 (21.5%)
      Tail84 (12.9%)55 (10.9%)29 (20.2%)
      Overlapping81 (12.5%)68 (13.5%)13 (9.0%)
      Stage6650.004
      Localized76 (11.4%)67 (12.9%)9 (6.1%)
      Locally advanced198 (29.8%)163 (31.5%)35 (23.8%)
      Metastatic391 (58.8%)288 (55.6%)103 (70.1%)
      ECOG650>0.9
      ECOG 0–1500 (76.9%)389 (77.0%)111 (76.6%)
      ECOG ≥ 2150 (23.1%)116 (23.0%)34 (23.4%)
      Previous VTE6660.3
      No644 (96.7%)504 (97.1%)140 (95.2%)
      Yes22 (3.3%)15 (2.9%)7 (4.8%)
      Catheter665<0.001
      No526 (79.1%)425 (82.0%)101 (68.7%)
      Yes139 (20.9%)93 (18.0%)46 (31.3%)
      Gemcitabine5730.614
      No42 (7.3%)36 (7.6%)

      6 (6.1%)
      Yes531 (92.7%)439 (92.4%)92 (93.9%)
      Capecitabine5720.671
      No44737275
      Yes12510223
      Erlotinib5730.505
      No42935069
      Yes15412529
      Oxaliplatin5730.039
      No36831454
      Yes20516144
      5FU5730.047
      No47840375
      Yes957223
      Irinotecan5730.088
      No50842682
      Yes654916
      Other chemotherapy5730.369
      No50141883
      Yes725515
      Follow up6669.3 [4.7–17.0]9.2 [4.6–17.4]9.4 [4.9–16.3]0.8
      1 Median [IQR]; n (%).
      2 Pearson's Chi-squared test; Wilcoxon rank sum test; Fisher's exact test.
      With a median follow-up of 9.3 months (range 0.4–72.0 months), 147 patients developed VTE, representing 22.1% of the total population. In 20, more than one VTE was diagnosed simultaneously (by and large, PE and DVT). The type of VTE is displayed in Table 2. Fifty-two percent of the VTE (77 of the 147 cases) were incidental findings during the extension study following diagnosis of the tumor (35 subjects) or in radiological examinations during follow-up. In 20 individuals, the VTE was diagnosed in the 3 months prior to the diagnosis of pancreatic carcinoma (16 in the month previous). The diagnosis of VTE led to the diagnosis of pancreatic cancer in 15 participants. Sixty-two patients were receiving anticoagulant treatment at diagnosis of pancreatic carcinoma (37 acenocoumarol, 12 LMWH at therapeutic doses, 9 LMWH at prophylactic doses and 4 unknown), 10 of whom developed VTE. Table 3
      Table 2VTE characteristics.
      VTE characteristics
      N = 1471
      Type
      VVT (portal, SMV …)49 (33.3%)
      DVT45 (30.6%)
      PE25 (17.0%)
      Simultaneous events20 (13.6%)
      Another type of VTE8 (5.4%)
      Detection of VTE
      Asymptomatic / incidental77 (52.4%)
      Symptomatic70 (47.6%)
      1n (%).
      Abbreviations: VVT: visceral venous thrombosis; SMV: superior mesenteric vein;.
      DVT: deep vein thrombosis; PE: pulmonary thrombosis; VTE: venous thrombotic event.
      Table 3Comparative of Khorana score and VTE.
      Khorana Score
      Intermediate risk, N = 372 (63%)
      n (%).
      High risk, N = 220 (37%)
      n (%).
      p-value
      Pearson's Chi-squared test.
      VTE0.3
      No314 (84.4%)179 (81.4%)
      Yes58 (15.6%)41 (18.6%)
      1 n (%).
      2 Pearson's Chi-squared test.
      At the time of VTE diagnosis, 70% of the sample presented distant metastases and 24%, locally advanced disease. Only 9 cases occurred in patients with localized cancer, 4 of whom following complete resection of the tumor, with a median of 2.9 months (range 0.2–4.1 months) since surgery. Two thirds of all VTE occurred in the first 6 months subsequent to the diagnosis of pancreatic cancer. Median time to VTE appearance was 2.93 months (IQR 0.30–8.38). Once those individuals who had presented the event in the 3 months prior to cancer diagnosis were excluded (n = 20), the cumulative incidence of VTE at 3, 6, and 12 months was 1.6% (95% CI 0.8%−2.9%), 8.2% (95% CI 6.0%−10.7%), and 15.6% (95% CI 12.4%−19.1%), respectively.
      The Khorana score for 30 participants was unknown, given that their body mass index (BMI) had not been recorded in their medical history. Excluding the 44 subject who presented VTE prior to the start of chemotherapy, 372 (63%) were included in the intermediate risk category and 220, in the high risk category (3 points: 153 patients; 4 points: 48 subjects; 5 points: 19 participants). The incidence of VTE in the Khorana score-based intermediate and high risk categories was 15.6% (58/372) and 18.6% (41/220), respectively, though these differences were not significant (p = 0.3). Moreover, no significant differences were detected when comparing VTE incidence across individuals with a Khorana score of 2–3 versus 4–5 (16.8% vs 16.4%, p >0.9).
      The results of the univariate and multivariate analyses to evaluate the prognostic factors associated with a greater risk of developing VTE are illustrated in Table 4. In the multivariate analysis, being female and the presence of a central catheter were significantly associated with a greater risk of developing VTE.Thirteen of the 147 patients (8.8%) developed recurrence of the VTE during follow up (4 PE, 2 DVT, 2 VVT, 3 other events, and 2 exhibited several simultaneous events). Twelve of the 13 recurrences occurred in the first 6 months following diagnosis of the first thrombotic event. The antithrombotic treatment at the time of recurrence was not recorded.
      Table 4Risk factors for VTE. Univariate and multivariate analysis.
      UnivariateMultivariate
      OR
      OR=Odds Ratio, CI=Confidence Interval.
      95% CI
      OR=Odds Ratio, CI=Confidence Interval.
      p-valueOR
      OR=Odds Ratio, CI=Confidence Interval.
      95% CI
      OR=Odds Ratio, CI=Confidence Interval.
      p-value
      Sex
      Male
      Female1.751.21, 2.530.0032.09(1.30, 3.39)0.003
      Age (years)
      <= 65 years
      > 65 years0.880.61, 1.270.5
      BMI
      < 25
      ≥ 251.000.67, 1.47> 0.9
      ECOG
      ECOG 0–1
      ECOG ≥ 21.030.66, 1.58>0.9
      Stage
      Localized
      Locally advanced1.600.76, 3.700.2
      Metastatic2.661.34, 5.900.009
      Tumor site
      Head
      Body2.601.55, 4.31<0.0011,690.85, 3.230,122
      Tail2.431.44, 4.06<0.0011,750.86, 3.410,109
      Overlapping0.880.44, 1.630.40,60.24, 1.350,252
      Catheter
      No
      Yes2.081.37, 3.14<0.0011,81.03, 3.110,036
      Previous VTE
      No
      Yes1.680.63–4.070.3
      Gemcitabine
      No
      Yes1.260.55, 3.400.6
      Capecitabine
      No
      Yes1.120.66, 1.850.7
      Erlotinib
      No
      Yes1.180.72, 1.880.5
      Oxaliplatin
      No
      Yes1.591.02 2.470.041,560.93, 2.610,087
      5FU
      No
      Yes1.710.99, 2.880.047
      Irinotecan
      No
      Yes1.700.90, 3.070.09
      Other chemotherapy
      No
      Yes1.330.68, 2.410.4
      Khorana
      Intermediate risk
      High risk1.240.79, 1.920.3
      1 OR=Odds Ratio, CI=Confidence Interval.
      Median survival since diagnosis of pancreatic cancer was 5.2 months (95% CI 4.3–6.0 months) for subjects diagnosed with VTE as opposed to 11.9 months (95% CI 10.5–13.5 months) for those who did not exhibit the event (p < 0.001) (Fig. 1a). To assess the impact VTE had on OS, Cox proportional risks models were performed using VTE as the time-dependent covariable. The multivariate analysis (Table 5) identified VTE as an independent poor prognostic factor that doubled the risk of death in cases with VTE versus those without (HR 1.94, 95% CI 1.57–2.40; p <0.001). Similarly, poor functional status (ECOG Performance Status ≥ 2), male sex, older age and locally advanced or metastatic disease at diagnosis of pancreatic carcinoma were associated with worse overall survival. At the time of data analysis, 128 of the VTE patients had died. Most of the deaths were secondary to oncological disease, although 8 of them were a consequence of VTE and in 3 cases the cause of death was unknown.
      Fig 1
      Fig. 1Fig. 1A: Kaplan-Meier curve comparing survival from time of cancer diagnosis to demise with and without VTE. B: Kaplan-Meier curve comparing survival from time of cancer diagnosis to demise among patients without VTE, those with VVT, DVT and PE. C: Kaplan-Meier curve comparing survival from time of cancer diagnosis to demise among patients without VTE, those with incidental VTE, and those with symptomatic VTE. D: Kaplan-Meier curves comparing survival from time of cancer diagnosis to demise among between “Early VTE” (< 3 months) and “Non-early VTE”.
      Table 5Risk factors for death. Univariate and multivariate analysis.
      UnivariateMultivariate
      HR
      HR=Hazard Ratio, CI=Confidence Interval.
      95% CI
      HR=Hazard Ratio, CI=Confidence Interval.
      p-valueHR
      HR=Hazard Ratio, CI=Confidence Interval.
      95% CI
      HR=Hazard Ratio, CI=Confidence Interval.
      p-value
      VTE2.391.96, 2.92<0.0011.941.57, 2.40<0.001
      Sex
      Male
      Female0.900.76, 1.070.20.810.68, 0.970.022
      Age (years)
      ≤ 65 years
      > 65 years1.261.06, 1.480.0081.211.01, 1.440.036
      BMI
      < 25
      ≥ 251.040.87, 1.240.7
      ECOG
      ECOG 0–1
      ECOG ≥ 22.331.91, 2.84<0.0012.061.67, 2.53<0.001
      Stage
      Localized
      Locally advanced2.251.60, 3.18<0.0011.911.32, 2.76<0.001
      Metastatic4.923.54, 6.84<0.0013.912.74, 5.59<0.001
      Tumor site
      Head
      Body1.481.15, 1.910.002
      Tail1.551.20, 2.01<0.001
      Overlapping1.230.95, 1.580.12
      Khorana
      Intermediate risk
      High risk1.030.86, 1.230.8
      1 HR=Hazard Ratio, CI=Confidence Interval.

      3.1 Stratified analysis

      3.1.1 Analysis of visceral thromboses vs. PE/ DVT

      Forty-nine participants (7.4% of total population) developed only VVT. Median survival was 5.9 months (95% CI 4.3–12.2 months) for individuals with VVT, significantly lower than those without VTE (HR 1.57; 95% CI 1.14–2.17; p = 0.006). As Fig. 1b shows, for patients with PE (HR 4.03; 95% CI 2.90–5.59; p <0.001) or DVT (HR 2.38; 95% CI 1.70–3.32) median overall survival was also inferior to that of patients without VTE. Prognosis was significantly worse for subjects who exhibited PE relative to those who only had VVT (HR 2.56; 95% CI, 1.64–3.98, p < 0.001) or DVT (HR 1.69; 95% CI 1.08–2.65; p = 0.021). A non-significant trend for longer survival was observed for patients with VVT compared to those with DVT (HR 0.66; 95% CI 0.42–1.03: p = 0.07).

      3.1.2 Analysis of incidental versus symptomatic events

      Of the 25 individuals who presented PE, 15 (60%) such events were incidental findings in asymptomatic patients. In contrast, 90% of the DVT were symptomatic. All the VVT were diagnosed incidentally. As Fig. 1c illustrates, the risk of death was significantly greater for both individuals with incidental (HR 1.90; 95% CI 1.47–2.47; p < 0.001) and symptomatic (HR 3.28; 95% CI 2.51–4.30; p < 0.001) VTE relative to those without VTE. Median OS rates were 5.8 months (95% CI 4.0–10.2 months) and 5.1 months (95% CI 4.1–6.1 months) for subjects with incidental and symptomatic VTE, respectively. In contrast, the risk of death was significantly higher for those with symptomatic VTE versus incidental events (HR 1.72, 95% CI 1.21–2.45; p = 0.002).

      3.1.3 Analysis of early events

      Participants with a history of VTE prior to their cancer diagnosis were excluded from this analysis. There were 54 cases of VTE in the 3 months after tumor diagnosis. Early appearance of thrombosis was associated with worse prognosis (Fig. 1d), with a median survival of 5.9 months (95% CI 4.8–9.5) for individuals with early thrombosis compared to 10.8 months for the rest (HR 1.52, 95% CI 1.12–2.08; p <0.001). These differences were also observed (HR 1.92 95% CI 1.30–2.84; p <0.001) when we compared OS of subjects who presented VTE in the first 3 months with those who developed VTE more than 3 months after cancer diagnosis (median 14.1 months; 95% CI 12.7–17.4).

      4. Discussion

      Studies published to date report VTE incidence rates in patients with pancreatic cancer that vary between 5% and 41% [
      • Oh S.Y.
      • Kim J.H.
      • Lee K.W.
      • Bang S.M.
      • Hwang J.H.
      • Oh D.
      • Lee J.S.
      Venous thromboembolism in patients with pancreatic adenocarcinoma: lower incidence in Asian ethnicity.
      ,
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ,
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Mandalà M.
      • Reni M.
      • Cascinu S.
      • Barni S.
      • Floriani I.
      • Cereda S.
      • et al.
      Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
      ,
      • Khorana A.A.
      • Dalal M.
      • Lin J.
      • Connolly G.C.
      Incidence and predictors of venous thromboembolism (VTE) among ambulatory high-risk cancerpatients undergoing chemotherapy in the United States.
      ,
      • Kondo S.
      • Sasaki M.
      • Hosoi H.
      • Sakamoto Y.
      • Morizane C.
      • Ueno H.
      • Okusaka T.
      Incidence and risk factors for venous thromboembolism in patients with pretreated advanced pancreatic carcinoma.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Ouaissi M.
      • Frasconi C.
      • Mege D.
      • Panicot-Dubois L.
      • Boiron L.
      • Dahan L.
      • et al.
      Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
      ,
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • van Es N.
      • Franke V.F.
      • Middeldorp S.
      • Wilmink J.W.
      • Büller H.R
      The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Hanna-Sawires R.G.
      • Groen J.V.
      • Hamming A.
      • Tollenaar R.A.E.M.
      • Mesker W.E.
      • et al.
      Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ] and even higher if autopsy series are analyzed. This great variability is due, in part, to racial differences, given the well-known lower incidence of VTE in Asian populations [
      • Oh S.Y.
      • Kim J.H.
      • Lee K.W.
      • Bang S.M.
      • Hwang J.H.
      • Oh D.
      • Lee J.S.
      Venous thromboembolism in patients with pancreatic adenocarcinoma: lower incidence in Asian ethnicity.
      ,
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ,
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ], and partially, to the inclusion of heterogeneous patient groups in the series (some studies included only cases of locally advanced or metastatic disease [
      • Mandalà M.
      • Reni M.
      • Cascinu S.
      • Barni S.
      • Floriani I.
      • Cereda S.
      • et al.
      Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
      ,
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ]; others, such as ours, also admitted individuals with localized disease [
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Ouaissi M.
      • Frasconi C.
      • Mege D.
      • Panicot-Dubois L.
      • Boiron L.
      • Dahan L.
      • et al.
      Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
      ,
      • van Es N.
      • Franke V.F.
      • Middeldorp S.
      • Wilmink J.W.
      • Büller H.R
      The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ,
      • Hanna-Sawires R.G.
      • Groen J.V.
      • Hamming A.
      • Tollenaar R.A.E.M.
      • Mesker W.E.
      • et al.
      Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
      ]). In contrast, some studies include only cases treated with chemotherapy [
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ,
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Mandalà M.
      • Reni M.
      • Cascinu S.
      • Barni S.
      • Floriani I.
      • Cereda S.
      • et al.
      Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
      ,
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ], while others admit patients with or without specific cancer treatment. Finally, when comparing incidence rates across studies, the definition of VTE in each must be evaluated, bearing in mind whether or not VVT, incidental events, arterial thromboses, etc. are analyzed. Our series, one of the most extensive published to date, examines individuals with pancreatic cancer treated with outpatient chemotherapy in daily clinical practice and confirms the elevated incidence of VTE in this population (22.1%), similar to reports in recent studies [
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Hanna-Sawires R.G.
      • Groen J.V.
      • Hamming A.
      • Tollenaar R.A.E.M.
      • Mesker W.E.
      • et al.
      Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
      ]. The main objective of the study was to analyze the incidence of VTE in this population, as well as to evaluate the factors associated with a higher risk of thrombosis and its impact on patient survival.
      Our study was unable to confirm the Khorana score's capacity to discriminate between intermediate or high risk. The incidence of VTE was 18.6% and 15.6% in high risk and intermediate risk, respectively, very much in line with the findings of the recently published BACAP-VTE study [
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ]. These findings can be explained by the fact that incidental events, very common in our series, were not assessed in the study by Khorana et al. One study conducted in patients with pancreatic carcinoma, presented as an abstract [
      • Price L.H.
      • Nguyen M.B.
      • Picozzi V.J.
      • Kozarek R.A.
      Portal vein thrombosis in pancreatic cancer: natural history, risk factors and implications for patient management.
      ], revealed that the Khorana score was useful only in predicting symptomatic VTE, but that its value to prognosticate incidental events and VVT was less clear. Furthermore, we must point out that fewer than 2% of the participants in the cohorts to develop and validate the Khorana score had a diagnosis of pancreatic carcinoma. Several studies similar to ours have likewise failed to demonstrate the merit of the Khorana score to predict risk of thrombosis in patients with pancreatic cancer [
      • Muñoz Martín A.J.
      • García Alfonso P.
      • Rupérez Blanco A.B.
      • Pérez Ramírez S.
      • Blanco Codesido M.
      • Martín Jiménez M.
      Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • van Es N.
      • Franke V.F.
      • Middeldorp S.
      • Wilmink J.W.
      • Büller H.R
      The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ]. Nevertheless, in the publication by Barrau et al. [
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ], individuals with pancreatic carcinoma and a Khorana score ≥ 4 had twice the risk of VTE than those scoring 2–3 (HR=1.96,95% CI 1.05–3.65; p = 0.03), differences that were not detected in our study. Recently, van Es et al. have published a meta-analysis of individual data from participants in seven randomized clinical trials of thromboprophylaxis [
      • van Es N.
      • Ventresca M.
      • Di Nisio M.
      • Zhou Q.
      • Noble S.
      • et al.
      IPDMA heparin use in cancer patients research group. The Khorana score for prediction of venous thromboembolism in cancer patients: an individual patient data meta-analysis.
      ]. In this meta-analysis, a Khorana score indicating high risk correlated significantly with the appearance of VTE in subjects with pancreatic cancer (odds ratio [OR], 2.2; 95% CI, 1.02–4.9), but not in the other solid tumors examined. The recent publication by Godinho et al. [
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ] that evaluated 165 cases of pancreatic cancer suggests that the ONKOTEV score is significantly better suited to predict VTE than the Khorana score, albeit in their analysis only 66% of the study populations received chemotherapy. The ONKOTEV score looks at four risk variables: Khorana score ≥ 3, prior history of VTE, metastatic disease, and compression of vascular structures by the tumor [
      • Cella C.A.
      • Di Minno G.
      • Carlomagno C.
      • Arcopinto M.
      • Cerbone A.M.
      • Matano E.
      • et al.
      Preventing venous thromboembolism in ambulatory cancer patients: the ONKOTEV study.
      ]. Unfortunately, we did not contemplate this last variable in our series; hence, we were unable to assess the predictive value of ONKOTEV in our patients.
      Regarding the use of chemotherapy, we must point out that the cytostatics received by the patient throughout their oncological disease were collected, but we do not know which ones the patient was receiving at the time of thrombosis. In the univariate analysis, patients who received oxaliplatin or 5FU had a higher risk of thrombosis, which is possibly related to the more frequent use of central catheters in these patients (41% and 72% of patients who received oxaliplatin and 5FU respectively, compared to 10% of those who do not receive either drug). The fact that catheter-associated thromboses were analyzed as DVT in our series may represent a limitation in the interpretation of the results.
      The studies that have probed the impact of thrombosis on survival in these patients yield dissimilar results; whereas some studies find a negative impact on survival [
      • Mandalà M.
      • Reni M.
      • Cascinu S.
      • Barni S.
      • Floriani I.
      • Cereda S.
      • et al.
      Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Epstein A.S.
      • Soff G.A.
      • Capanu M.
      • Crosbie C.
      • Shah M.A.
      • Kelsen D.P.
      • et al.
      Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
      ,
      • Ouaissi M.
      • Frasconi C.
      • Mege D.
      • Panicot-Dubois L.
      • Boiron L.
      • Dahan L.
      • et al.
      Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ], others evince no significant differences [
      • Oh S.Y.
      • Kim J.H.
      • Lee K.W.
      • Bang S.M.
      • Hwang J.H.
      • Oh D.
      • Lee J.S.
      Venous thromboembolism in patients with pancreatic adenocarcinoma: lower incidence in Asian ethnicity.
      ,
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Godinho J.
      • Casa-Nova M.
      • Moreira-Pinto J.
      • Simões P.
      • Paralta Branco F.
      • Leal-Costa L.
      • et al.
      ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Hanna-Sawires R.G.
      • Groen J.V.
      • Hamming A.
      • Tollenaar R.A.E.M.
      • Mesker W.E.
      • et al.
      Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ]. Our results illustrate that the presence of VTE is associated with shorter OS in individuals with pancreatic cancer. The fact that VTE was associated with poor prognosis, even after adjusting for stage, could be related to a higher tumor burden in patients who developed thrombosis, compared to patients of the same stage who did not. VTE may identify a more aggressive disease, or cancer phenotype, which could be associated with a higher risk of death.
      Consistent with other studies [
      • Ouaissi M.
      • Frasconi C.
      • Mege D.
      • Panicot-Dubois L.
      • Boiron L.
      • Dahan L.
      • et al.
      Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • van Es N.
      • Franke V.F.
      • Middeldorp S.
      • Wilmink J.W.
      • Büller H.R
      The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ], more than half of the VTE in our series were diagnosed incidentally on routine radiological scans and these asymptomatic events were associated with a better prognosis than symptomatic events, despite having a lower survival rate than subjects without VTE (HR 1.90; 95% CI 1.47–2.47). In a population of pancreatic cancer, Lee et al. [
      • Lee J.C.
      • Ro Y.S.
      • Cho J.
      • Park Y.
      • Lee J.H.
      • Hwang J.H.
      • et al.
      Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
      ] found that symptomatic VTE correlated with a worse prognosis than incidental ones (HR 1.87; 95% CI 1.26–2.78; p = 0.0002) and that PE/ DVT events were also linked with a worse prognosis than VVT (HR 1.53; 95% CI 1.02–2.20; p = 0.022), consistent with our findings. Nevertheless, other studies in pancreatic cancer have not found differences in survival between symptomatic and incidental events [
      • Ishigaki K.
      • Nakai Y.
      • Isayama H.
      • Saito K.
      • Hamada T.
      • Takahara N.
      • et al.
      Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
      ,
      • Menapace L.A.
      • Peterson D.R.
      • Berry A.
      • Sousou T.
      • Khorana A.A.
      Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
      ,
      • Kruger S.
      • Haas M.
      • Burkl C.
      • Goehring P.
      • Kleespies A.
      • Roeder F.
      • et al.
      Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
      ,
      • Berger A.K.
      • Singh H.M.
      • Werft W.
      • Muckenhuber A.
      • Sprick M.R.
      • et al.
      High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
      ,
      • Frere C.
      • Bournet B.
      • Gourgou S.
      • Fraisse J.
      • Canivet C.
      • et al.
      BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
      ]. Further, a high percentage of the thromboses that occur in individuals with pancreatic carcinoma are located at the spleno-portal axis and most of these events are diagnosed incidentally. Management of these VVT and their impact on mortality in the context of pancreatic cancer have yet to be ascertained, although a recent publication [
      • Afzal A.
      • Suhong L.
      • Gage B.F.
      • Schoen M.W.
      • Carson K.
      • et al.
      Splanchnic vein thrombosis predicts worse survival in patients with advanced pancreatic cancer.
      ] demonstrates that the presence of splanchnic thrombosis in patients with advanced pancreatic cancer negatively affects survival, in keeping with findings previously published by Søgaard et al. [
      • Søgaard K.K.
      • Farkas D.K.
      • Pedersen L.
      • Sørensen H.T.
      Splanchnic venous thrombosis is a marker of cancer and a prognostic factor for cancer survival.
      ] Treating these VVT in daily clinical practice entails a complex decision that must factor in the risk of bleeding in relation to the benefit of anticoagulant treatment, as stated in the publication by Mier-Hicks et al. [
      • Mier-Hicks A.
      • Raj M.
      • Do R.K.
      • Yu K.H.
      • Lowery M.A.
      • Varghese A.
      • et al.
      Incidence, management, and implications of visceral thrombosis in pancreatic ductal adenocarcinoma.
      ] in which 8% of the subjects with pancreatic cancer and VVT included in the study, exhibited bleeding complications secondary to anticoagulation, despite the fact that almost one fourth of the patients did not receive anticoagulant therapy. There again, the publication by Afzal et al. [
      • Afzal A.
      • Suhong L.
      • Gage B.F.
      • Schoen M.W.
      • Carson K.
      • et al.
      Splanchnic vein thrombosis predicts worse survival in patients with advanced pancreatic cancer.
      ] revealed that, despite the negative effect of VVT on survival, anticoagulant therapy did not significantly improve prognosis and, in contrast, almost tripled the risk of bleeding.
      It has been well established that VTE in pancreatic cancer is a time-dependent event, which has led recent studies to examine whether the timing of the VTE impacts survival. While the definition of early thrombosis varies from one study to another [
      • Kim J.S.
      • Kang E.J.
      • Kim D.S.
      • Choi Y.J.
      • Lee S.Y.
      • et al.
      Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
      ,
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ,
      • Chen J.S.
      • Hung C.Y.
      • Chang H.
      • Liu C.T.
      • Chen Y.Y.
      • et al.
      Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
      ], all reveal that early thrombosis following diagnosis is predictive of worse survival. The recent work by Barrau et al. [
      • Barrau M.
      • Maoui K.
      • Le Roy B.
      • Roblin X.
      • Mismetti P.
      • et al.
      Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
      ] found that individuals who developed VTE in the first 3 months following inclusion in the study had a median OS that was significantly lower than the rest (8 months vs. 14 months; HR 2.62; p = 0.002), comparable to our study.
      This work has certain limitations, the main ones having to do with its retrospective analyses. The Khorana score, one of the main variables to be analyzed, could not be ascertained in 30 of the patients included as some data were not available in the clinical history. The antithrombotic treatment administered following the first VTE was not recorded. This information would have been relevant to assess observed recurrences, as well as if they occurred in properly anticoagulated subjects, which is important considering that some authors suggest reducing the dose of anticoagulant treatment (>90% of recurrences occurred in the first 6 months after VTE). Likewise, we are unaware of how the VVT (all of which were incidental) were managed and the treatment of which may or may not have affected OS. On the other hand, the fact that this is a multicenter study has enabled us to analyze a large number of patients, although we have also found that the frequency of VVT has been considerably lower in some centers than in others, leading us to think that the interobserver variability of the radiologists across the different centers may have influenced the results, probably underestimating the true incidence of VTE.
      The high incidence of VTE observed in our study warrants an evaluation of the possibility of thromboprophylaxis in this subgroup of patients, especially in the first 6 months after the diagnosis of pancreatic carcinoma, when up to two thirds of thrombotic events occur. Four randomized clinical trials have been published in recent years that examined the role of LMWH thromboprophylaxis in cancer patients undergoing outpatient chemotherapy. Two of these studies were conducted in a heterogeneous population of cancer patients, including patients with pancreatic cancer (PROTECHT and SAVE ONCO) [
      • Agnelli G.
      • Gussoni G.
      • Bianchini C.
      • Verso M.
      • Mandalà M.
      • Cavanna L.
      • et al.
      PROTECHT investigators. Nadroparin for the prevention of thromboembolic events in ambulatory patients with metastatic or locally advanced solid cancer receiving chemotherapy: a randomised, placebo-controlled, double-blind study.
      ,
      • Agnelli G.
      • George D.J.
      • Kakkar A.K.
      • Fisher W.
      • Lassen M.R.
      • Mismetti P.
      • et al.
      SAVE-ONCO investigators. Semuloparin for thromboprophylaxis in patients receiving chemotherapy for cancer.
      ], and another two were performed specifically in pancreatic carcinoma (FRAGEM and CONKO 004) [
      • Maraveyas A.
      • Waters J.
      • Roy R.
      • Fyfe D.
      • Propper D.
      • Lofts F.
      • et al.
      Gemcitabine versus gemcitabine plus dalteparin thromboprophylaxis in pancreatic cancer.
      ,
      • Pelzer U.
      • Opitz B.
      • Deutschinoff G.
      • Stauch M.
      • Reitzig P.C.
      • Hahnfeld S.
      • et al.
      Efficacy of prophylactic low-molecular weight heparin for ambulatory patients with advanced pancreatic cancer: outcomes from the CONKO-004 Trial.
      ]. All these studies confirm a significant reduction of the incidence of VTE, albeit without affecting OS. More recently, two clinical trials (AVERT and CASSINI) [
      • Carrier M.
      • Abou-Nassar K.
      • Mallick R.
      • Tagalakis V.
      • Shivakumar S.
      • Schattner A.
      • et al.
      AVERT investigators. Apixaban to prevent venous thromboembolism in patients with cancer.
      ,
      • Khorana A.A.
      • Soff G.A.
      • Kakkar A.K.
      • Vadhan-Raj S.
      • Riess H.
      • Wun T.
      • et al.
      CASSINI investigators. Rivaroxaban for thromboprophylaxis in high-risk ambulatory patients with cancer.
      ] have appraised the possible benefit of new direct oral anticoagulants (DOACs) in thromboprophylaxis in oncology patients and with Khorana scores ≥2. In the AVERT trial [
      • Carrier M.
      • Abou-Nassar K.
      • Mallick R.
      • Tagalakis V.
      • Shivakumar S.
      • Schattner A.
      • et al.
      AVERT investigators. Apixaban to prevent venous thromboembolism in patients with cancer.
      ], in which 14% of the study population had pancreatic cancer, apixaban significantly decreased the incidence of VTE during the study's follow-up period. Rivaroxaban failed to demonstrate a significant reduction of the incidence of VTE or death due to VTE during the 180-day period (primary end point) in the CASSINI study [
      • Pelzer U.
      • Opitz B.
      • Deutschinoff G.
      • Stauch M.
      • Reitzig P.C.
      • Hahnfeld S.
      • et al.
      Efficacy of prophylactic low-molecular weight heparin for ambulatory patients with advanced pancreatic cancer: outcomes from the CONKO-004 Trial.
      ], despite being associated with a lower incidence of VTE during the intervention period. The results of a pre-specified sub-analysis of patients with pancreatic cancer (n = 273, 32.6% of the total) included in the CASSINI study has recently been published, yielding results that were consistent with findings in the overall population [
      • Vadhan-Raj S.
      • McNamara M.G.
      • Venerito M.
      • Riess H.
      • O'Reilly E.M.
      • Overman M.J.
      • et al.
      Rivaroxaban thromboprophylaxis in ambulatory patients with pancreatic cancer: results from a pre-specified subgroup analysis of the randomized CASSINI study.
      ]. The basic concern regarding the use of thromboprophylaxis in this population is the risk of bleeding. Clinical guidelines coincide in that thromboprophylaxis should not be routinely administered to cancer patients who are initiating chemotherapy, but rather suggest that its use be contemplated in individuals with an intermediate-high risk (Khorana score ≥ 2), as long as there is no significant risk of bleeding [
      • Key N.S.
      • Khorana A.A.
      • Kuderer N.M.
      • Bohlke K.
      • Lee A.Y.
      • Arcelus J.I.
      • et al.
      Venous thromboembolism prophylaxis and treatment in patients with cancer: ASCO clinical practice guideline update.
      ,

      National Comprehensive Cancer Network (NCCN) Guideline. Cancer-Associated Venous Thromboembolic Disease. (2022) Available at: http://www.nccn.org/professionals/physician_gls/pdf/vte.pdf. Accessed 15 Mar 2022.

      ,
      • Farge D.
      • Frere C.
      • Connors J.M.
      • Ay C.
      • Khorana A.A.
      • et al.
      2019 international clinical practice guidelines for the treatment and prophylaxis of venous thromboembolism in patients with cancer.
      ]. Guidelines deem LMWH, apixaban, or rivaroxaban to be suitable in this context.
      In conclusion, VTE is a common complication in individuals with pancreatic cancer that affects survival, even when the events are incidental, which justifies probing the use of thromboprophylaxis in this population.

      Ethical statements

      The study was approved by the Ethics Committee of Hospital General Universitario Gregorio Marañón. Informed consent was not requested as it is not required for this type of study.

      Funding

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Acknowledgments

      SEOM for promoting this study. Priscilla Chase for editing and translating the manuscript.

      References

        • Sørensen H.T.
        • Mellemkjaer L.
        • Olsen J.H.
        • Baron J.A.
        Prognosis of cancers associated with venous thromboembolism.
        N Engl J Med. 2000; 343: 1846-1850https://doi.org/10.1056/NEJM200012213432504
        • Khorana A.A.
        • Francis C.W.
        • Culakova E.
        • Kuderer N.M.
        • Lyman G.H.
        Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy.
        J Thromb Haemost. 2007; 5: 632-634https://doi.org/10.1111/j.1538-7836.2007.02374.x
        • Heit J.A.
        • Silverstein M.D.
        • Mohr D.N.
        • Petterson T.M.
        • O’Fallon W.M.
        • Melton L.J.
        Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case-control study.
        Arch Intern Med. 2000 Mar 27; 160: 809-815https://doi.org/10.1001/archinte.160.6.809
        • Blom J.W.
        • Doggen C.J.
        • Osanto S.
        • Rosendaal F.R.
        Malignancies, prothrombotic mutations, and the risk of venous thrombosis.
        JAMA. 2005; 293: 715-722https://doi.org/10.1001/jama.293.6.715
        • Oh S.Y.
        • Kim J.H.
        • Lee K.W.
        • Bang S.M.
        • Hwang J.H.
        • Oh D.
        • Lee J.S.
        Venous thromboembolism in patients with pancreatic adenocarcinoma: lower incidence in Asian ethnicity.
        Thromb Res. 2008; 122: 485-490https://doi.org/10.1016/j.thromres.2007.12.015
        • Lee J.C.
        • Ro Y.S.
        • Cho J.
        • Park Y.
        • Lee J.H.
        • Hwang J.H.
        • et al.
        Characteristics of venous thromboembolism in pancreatic adenocarcinoma in East Asian ethnics: a large population-based observational study.
        Medicine (Baltimore). 2016; 95: e3472https://doi.org/10.1097/MD.0000000000003472
        • Ishigaki K.
        • Nakai Y.
        • Isayama H.
        • Saito K.
        • Hamada T.
        • Takahara N.
        • et al.
        Thromboembolisms in advanced pancreatic cancer: a retrospective analysis of 475 patients.
        Pancreas. 2017; 46: 1069-1075https://doi.org/10.1097/MPA.0000000000000889
        • Mandalà M.
        • Reni M.
        • Cascinu S.
        • Barni S.
        • Floriani I.
        • Cereda S.
        • et al.
        Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients.
        Ann Oncol. 2007; 18: 1660-1665https://doi.org/10.1093/annonc/mdm284
        • Khorana A.A.
        • Dalal M.
        • Lin J.
        • Connolly G.C.
        Incidence and predictors of venous thromboembolism (VTE) among ambulatory high-risk cancerpatients undergoing chemotherapy in the United States.
        Cancer. 2013; 119: 648-655https://doi.org/10.1002/cncr.27772
        • Kondo S.
        • Sasaki M.
        • Hosoi H.
        • Sakamoto Y.
        • Morizane C.
        • Ueno H.
        • Okusaka T.
        Incidence and risk factors for venous thromboembolism in patients with pretreated advanced pancreatic carcinoma.
        Oncotarget. 2018 Mar 30; 9: 16883-16890https://doi.org/10.18632/oncotarget.24721
        • Menapace L.A.
        • Peterson D.R.
        • Berry A.
        • Sousou T.
        • Khorana A.A.
        Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer.
        Thromb Haemost. 2011; 106: 371-378https://doi.org/10.1160/TH10-12-0789
        • Epstein A.S.
        • Soff G.A.
        • Capanu M.
        • Crosbie C.
        • Shah M.A.
        • Kelsen D.P.
        • et al.
        Analysis of incidence and clinical outcomes in patients with thromboembolic events and invasive exocrine pancreatic cancer.
        Cancer. 2012; 118: 3053-3061https://doi.org/10.1002/cncr.26600
        • Ouaissi M.
        • Frasconi C.
        • Mege D.
        • Panicot-Dubois L.
        • Boiron L.
        • Dahan L.
        • et al.
        Impact of venous thromboembolism on the natural history of pancreatic adenocarcinoma.
        Hepatobiliary Pancreat Dis Int. 2015; 14: 436-442https://doi.org/10.1016/s1499-3872(15)60397-6
        • Muñoz Martín A.J.
        • García Alfonso P.
        • Rupérez Blanco A.B.
        • Pérez Ramírez S.
        • Blanco Codesido M.
        • Martín Jiménez M.
        Incidence of venous thromboembolism (VTE) in ambulatory pancreatic cancer patients receiving chemotherapy and analysis of Khorana's predictive model.
        Clin Transl Oncol. 2014; 16: 927-930https://doi.org/10.1007/s12094-014-1165-y
        • Kruger S.
        • Haas M.
        • Burkl C.
        • Goehring P.
        • Kleespies A.
        • Roeder F.
        • et al.
        Incidence, outcome and risk stratification tools for venous thromboembolism in advanced pancreatic cancer - A retrospective cohort study.
        Thromb Res. 2017; 157: 9-15https://doi.org/10.1016/j.thromres.2017.06.021
        • van Es N.
        • Franke V.F.
        • Middeldorp S.
        • Wilmink J.W.
        • Büller H.R
        The Khorana score for the prediction of venous thromboembolism in patients with pancreatic cancer.
        Thromb Res. 2017; 150: 30-32https://doi.org/10.1016/j.thromres.2016.12.013
        • Berger A.K.
        • Singh H.M.
        • Werft W.
        • Muckenhuber A.
        • Sprick M.R.
        • et al.
        High prevalence of incidental and symptomatic venous thromboembolic events in patients with advanced pancreatic cancer under palliative chemotherapy: a retrospective cohort study.
        Pancreatology. 2017; 17: 629-634https://doi.org/10.1016/j.pan.2017.04.012
        • Kim J.S.
        • Kang E.J.
        • Kim D.S.
        • Choi Y.J.
        • Lee S.Y.
        • et al.
        Early venous thromboembolism at the beginning of palliative chemotherapy is a poor prognostic factor in patients with metastatic pancreatic cancer: a retrospective study.
        BMC Cancer. 2018; 18: 1260https://doi.org/10.1186/s12885-018-5154-3
        • Frere C.
        • Bournet B.
        • Gourgou S.
        • Fraisse J.
        • Canivet C.
        • et al.
        BACAP consortium. Incidence of venous thromboembolism in patients with newly diagnosed pancreatic cancer and factors associated With outcomes.
        Gastroenterology. 2020; 158 (e4): 1346-1358https://doi.org/10.1053/j.gastro.2019.12.009
        • Godinho J.
        • Casa-Nova M.
        • Moreira-Pinto J.
        • Simões P.
        • Paralta Branco F.
        • Leal-Costa L.
        • et al.
        ONKOTEV score as a predictive tool for thromboembolic events in pancreatic cancer-A retrospective analysis.
        Oncologist. 2020; 25: e284-e290https://doi.org/10.1634/theoncologist.2019-0510
        • Barrau M.
        • Maoui K.
        • Le Roy B.
        • Roblin X.
        • Mismetti P.
        • et al.
        Early venous thromboembolism is a strong prognostic factor in patients with advanced pancreatic ductal adenocarcinoma.
        J Cancer Res Clin Oncol. 2021; 147: 3447-3454https://doi.org/10.1007/s00432-021-03590-x
        • Hanna-Sawires R.G.
        • Groen J.V.
        • Hamming A.
        • Tollenaar R.A.E.M.
        • Mesker W.E.
        • et al.
        Incidence, timing and risk factors of venous thromboembolic events in patients with pancreatic cancer.
        Thromb Res. 2021; 207: 134-139https://doi.org/10.1016/j.thromres.2021.08.002
        • Chen J.S.
        • Hung C.Y.
        • Chang H.
        • Liu C.T.
        • Chen Y.Y.
        • et al.
        Venous thromboembolism in asian patients with pancreatic cancer following palliative chemotherapy: low incidence but a negative prognosticator for those with early onset.
        Cancers (Basel). 2018; 10: 501https://doi.org/10.3390/cancers10120501
        • Khorana A.A.
        • Kuderer N.M.
        • Culakova E.
        • Lyman G.H.
        • Francis C.W.
        Development and validation of a predictive model for chemotherapy-associated thrombosis.
        Blood. 2008; 111: 4902-4907https://doi.org/10.1182/blood-2007-10-116327
        • Kuderer N.M.
        • Poniewierski M.S.
        • Culakova E.
        • Lyman G.H.
        • Khorana A.A.
        • Pabinger I.
        • et al.
        Predictors of venous thromboembolism and early mortality in lung cancer: results from a global prospective study (CANTARISK).
        Oncologist. 2018; 23: 247-255https://doi.org/10.1634/theoncologist.2017-0205
        • Mansfield A.S.
        • Tafur A.J.
        • Wang C.E.
        • Kourelis T.V.
        • Wysokinska E.M.
        • Yang P.
        Predictors of active cancer thromboembolic outcomes: validation of the Khorana score among patients with lung cancer.
        J Thromb Haemost. 2016; 14: 1773-1778https://doi.org/10.1111/jth.13378
        • Ramos J.D.
        • Casey M.F.
        • Bamias A.
        • De Giorgi U.
        • Bellmunt J.
        • Harshman L.C.
        • et al.
        Retrospective international study of cancers of the urothelium (RISC) investigators. The Khorana score in predicting venous thromboembolism for patients with metastatic urothelial carcinoma and variant histology treated with chemotherapy.
        Clin Appl Thromb Hemost. 2017; 23: 755-760https://doi.org/10.1177/1076029616668405
        • Lim S.H.
        • Woo S.Y.
        • Kim S.
        • Ko Y.H.
        • Kim W.S.
        • Kim S.J.
        Cross-sectional study of patients with diffuse large B-cell lymphoma: assessing the effect of host status, tumor burden, and inflammatory activity on venous thromboembolism.
        Cancer Res Treat. 2016; 48: 312-321https://doi.org/10.4143/crt.2014.266
        • Chaudhury A.
        • Balakrishnan A.
        • Thai C.
        • Holmstrom B.
        • Nanjappa S.
        • Ma Z.
        • et al.
        Validation of the Khorana score in a large cohort of cancer patients with venous thromboembolism.
        Blood. 2016; 128: 879https://doi.org/10.1182/blood.V128.22.879.879
        • Tully C.M.
        • Apolo A.B.
        • Zabor E.C.
        • et al.
        The high incidence of vascular thromboembolic events in patients with metastatic or unresectable urothelial cancer treated with platinum chemotherapy agents.
        Cancer. 2016; 122: 712-721https://doi.org/10.1002/cncr.29801
        • Kourelis T.V.
        • Wysokinska E.M.
        • Wang Y.
        • Yang P.
        • Mansfield A.S.
        • Tafur A.J
        Early venous thromboembolic events are associated with worse prognosis in patients with lung cancer.
        Lung Cancer. 2014; 86: 358-362https://doi.org/10.1016/j.lungcan.2014.10.003
        • Price L.H.
        • Nguyen M.B.
        • Picozzi V.J.
        • Kozarek R.A.
        Portal vein thrombosis in pancreatic cancer: natural history, risk factors and implications for patient management.
        in: Proceedings of the 7th Gastrointestinal Cancer Symposium. 22-24 Jan 2010 (Orlando, Florida (Abstract 143))
        • van Es N.
        • Ventresca M.
        • Di Nisio M.
        • Zhou Q.
        • Noble S.
        • et al.
        IPDMA heparin use in cancer patients research group. The Khorana score for prediction of venous thromboembolism in cancer patients: an individual patient data meta-analysis.
        J Thromb Haemost. 2020; 18: 1940-1951https://doi.org/10.1111/jth.14824
        • Cella C.A.
        • Di Minno G.
        • Carlomagno C.
        • Arcopinto M.
        • Cerbone A.M.
        • Matano E.
        • et al.
        Preventing venous thromboembolism in ambulatory cancer patients: the ONKOTEV study.
        Oncologist. 2017; 22: 601-608https://doi.org/10.1634/theoncologist.2016-0246
        • Afzal A.
        • Suhong L.
        • Gage B.F.
        • Schoen M.W.
        • Carson K.
        • et al.
        Splanchnic vein thrombosis predicts worse survival in patients with advanced pancreatic cancer.
        Thromb Res. 2020; 185: 125-131https://doi.org/10.1016/j.thromres.2019.11.023
        • Søgaard K.K.
        • Farkas D.K.
        • Pedersen L.
        • Sørensen H.T.
        Splanchnic venous thrombosis is a marker of cancer and a prognostic factor for cancer survival.
        Blood. 2015; 126: 957-963https://doi.org/10.1182/blood-2015-03-631119
        • Mier-Hicks A.
        • Raj M.
        • Do R.K.
        • Yu K.H.
        • Lowery M.A.
        • Varghese A.
        • et al.
        Incidence, management, and implications of visceral thrombosis in pancreatic ductal adenocarcinoma.
        Clin Color Cancer. 2018; : 121-128https://doi.org/10.1016/j.clcc.2018.01.008
        • Agnelli G.
        • Gussoni G.
        • Bianchini C.
        • Verso M.
        • Mandalà M.
        • Cavanna L.
        • et al.
        PROTECHT investigators. Nadroparin for the prevention of thromboembolic events in ambulatory patients with metastatic or locally advanced solid cancer receiving chemotherapy: a randomised, placebo-controlled, double-blind study.
        Lancet Oncol. 2009; 10: 943-949https://doi.org/10.1016/S1470-2045(09)70232-3
        • Agnelli G.
        • George D.J.
        • Kakkar A.K.
        • Fisher W.
        • Lassen M.R.
        • Mismetti P.
        • et al.
        SAVE-ONCO investigators. Semuloparin for thromboprophylaxis in patients receiving chemotherapy for cancer.
        N Engl J Med. 2012; 366: 601-609https://doi.org/10.1056/NEJMoa1108898
        • Maraveyas A.
        • Waters J.
        • Roy R.
        • Fyfe D.
        • Propper D.
        • Lofts F.
        • et al.
        Gemcitabine versus gemcitabine plus dalteparin thromboprophylaxis in pancreatic cancer.
        Eur J Cancer. 2012; 48: 1283-1292https://doi.org/10.1016/j.ejca.2011.10.017
        • Pelzer U.
        • Opitz B.
        • Deutschinoff G.
        • Stauch M.
        • Reitzig P.C.
        • Hahnfeld S.
        • et al.
        Efficacy of prophylactic low-molecular weight heparin for ambulatory patients with advanced pancreatic cancer: outcomes from the CONKO-004 Trial.
        J Clin Oncol. 2015; 33: 2028-2034https://doi.org/10.1200/JCO.2014.55.1481
        • Carrier M.
        • Abou-Nassar K.
        • Mallick R.
        • Tagalakis V.
        • Shivakumar S.
        • Schattner A.
        • et al.
        AVERT investigators. Apixaban to prevent venous thromboembolism in patients with cancer.
        N Engl J Med. 2019; 380: 711-719https://doi.org/10.1056/NEJMoa1814468
        • Khorana A.A.
        • Soff G.A.
        • Kakkar A.K.
        • Vadhan-Raj S.
        • Riess H.
        • Wun T.
        • et al.
        CASSINI investigators. Rivaroxaban for thromboprophylaxis in high-risk ambulatory patients with cancer.
        N Engl J Med. 2019; 380: 720-728https://doi.org/10.1056/NEJMoa1814630
        • Vadhan-Raj S.
        • McNamara M.G.
        • Venerito M.
        • Riess H.
        • O'Reilly E.M.
        • Overman M.J.
        • et al.
        Rivaroxaban thromboprophylaxis in ambulatory patients with pancreatic cancer: results from a pre-specified subgroup analysis of the randomized CASSINI study.
        Cancer Med. 2020; 9: 6196-6204https://doi.org/10.1002/cam4.3269
        • Key N.S.
        • Khorana A.A.
        • Kuderer N.M.
        • Bohlke K.
        • Lee A.Y.
        • Arcelus J.I.
        • et al.
        Venous thromboembolism prophylaxis and treatment in patients with cancer: ASCO clinical practice guideline update.
        J Clin Oncol. 2020; 38: 496-520https://doi.org/10.1200/JCO.19.01461
      1. National Comprehensive Cancer Network (NCCN) Guideline. Cancer-Associated Venous Thromboembolic Disease. (2022) Available at: http://www.nccn.org/professionals/physician_gls/pdf/vte.pdf. Accessed 15 Mar 2022.

        • Farge D.
        • Frere C.
        • Connors J.M.
        • Ay C.
        • Khorana A.A.
        • et al.
        2019 international clinical practice guidelines for the treatment and prophylaxis of venous thromboembolism in patients with cancer.
        Lancet Oncol. 2019; 20: e566-e581https://doi.org/10.1016/S1470-2045(19)30336-5