Clinical and Translational Oncology (2021) 23:697–708 https://doi.org/10.1007/s12094-020-02477-6 SPECIAL ARTICLE Consensus statement of the Spanish Society of Internal Medicine and the Spanish Society of Medical Oncology on secondary thromboprophylaxis in patients with cancer T. Quintanar1 · C. Font2 · E. Gallardo3 · R. Barba4 · B. Obispo5 · C. Díaz‑Pedroche6 Received: 1 August 2020 / Accepted: 10 August 2020 / Published online: 3 September 2020 © The Author(s) 2020 Abstract Up to 20% of cancer patients will develop some manifestation of venous thromboembolic disease (VTD) during their clinical course. VTD greatly impacts morbidity, mortality, quality of life and pharmaceutical expenditure. In addition, both thrombotic relapse and major haemorrhages derived from VTD treatment are more likely in oncological patients. To make the decision to establish secondary thromboprophylaxis as an indefinite treatment in these patients, it is important to review all the risk factors involved, whether related to the disease, the patient or the prior thrombotic event. The objectives of this consensus of the Spanish Society of Internal Medicine (Sociedad Española de Medicina Interna—SEMI) and the Spanish Society of Medical Oncology (Sociedad Española de Oncología Médica—SEOM) are to establish recommendations that help assess the risk of recurrence of VTD and haemorrhagic risk in patients with cancer, as well as to analyse the evidence that exists on the currently available drugs, which will allow the establishment of a protocol for shared decision-making with the informed patient. Keywords Anticoagulants · Haemorrhage · Low-molecular-weight heparin · Neoplasms · Vitamin K · Venous thrombosis Introduction after diagnosis [2] with a major impact on morbidity, mor- tality, quality of life, and pharmaceutical expenditure. Both The advances in antineoplastic treatment of most solid recurrence of and major haemorrhages from VTD treat- tumours have resulted in an increase in overall survival in ment are more frequent in cancer patients [3]. This has led patients with advanced disease [1]. Managing complica- researchers to explore other anticoagulant treatment options tions of cancer treatment, therefore, has value in improving in this scenario. Low-molecular-weight heparin (LMWH) survival and quality of life. Up to 20% of cancer patients has been established as the main therapeutic tool based on will develop some manifestation of venous thromboembolic its greater efficacy observed against vitamin K antagonists disease (VTD) during their clinical course, making it one (VKA) in the CLOT study and other studies during the first of the main cardiovascular complications from the first year 3–6 months of treatment [4, 5]. 1 * T. Quintanar Department of Medical Oncology, Hospital General teresaqv22@yahoo.es Universitario de Elche y Vega Baja, Elche, Alicante, Spain 2 C. Font Department of Internal Medicine, Hospital Clinic, Barcelona, CFONT@clinic.cat Spain 3 E. Gallardo Department of Medical Oncology, Parc Taulí Hospital egallardo@tauli.cat Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain R. Barba 4 raquel.barba@hospitalreyjuancarlos.es Department of Internal Medicine, Hospital Universitario Rey Juan Carlos, Móstoles, Madrid, Spain B. Obispo 5 berta.obispo@gmail.com Department of Medical Oncology, Hospital Universitario Infanta Leonor, Madrid, Spain C. Díaz‑Pedroche 6 cdiazpedroche@gmail.com Department of Internal Medicine, Hospital Universitario Doce de Octubre, Madrid, Spain 13 Vol.:(0123456789) 698 Clinical and Translational Oncology (2021) 23:697–708 In recent years, several clinical trials have evaluated Factors influencing decision‑making the efficacy and safety of direct-acting oral anticoagu- lants (DOAC) in cancer patients, which has allowed an Factors related to the disease expansion of the therapeutic repertoire. In the clinical practice guidelines of the main scientific societies [6–8], Table 1 describes the considerations for and against the DOAC is not recommended as a therapeutic option for maintenance of long-term secondary thromboprophylaxis patients with gastrointestinal or genitourinary tumours, as related to tumour pathology. these tumours are associated with an increased haemor- rhagic risk [6, 8]. In select patients, dicoumarinic agents Location of the primary tumour and vascular compression may be a therapeutic option for long-term anticoagulant treatment, particularly for those who are not receiving The presence of cancer increases the risk of VTD four to chemotherapy or other agents that can interact with them sevenfold with respect to the general population, such that pharmacologically, and also only providing that adequate approximately 20% of the episodes of thrombosis observed control of the international normalized ratio (values of in the community will occur in cancer patients. Not all approximately 2–3) can be proven. However, there is little tumours have the same thrombogenic potential [11]. Pan- evidence for optimal therapy after 6 months of treatment, creatic cancer, central nervous system tumours, and gas- which requires individualized decision-making for each troesophageal cancer have the highest annual incidence patient [9, 10]. of thrombosis (6–14%). At an intermediate level are renal The objectives of this consensus of the Spanish Society cancer, lung cancer, ovarian cancer, and haematological of Internal Medicine (Sociedad Española de Medicina tumours (3–4%), while cancers such as breast, prostate, and Interna—SEMI) and the Spanish Society of Medical colon have a low thromboembolic risk (1–3%). Oncology (Sociedad Española de Oncología Médica— The aetiopathogenesis of VTD in cancer patients is com- SEOM) are to review the available evidence and establish plex and multifactorial. The state of hypercoagulability, recommendations, when possible, to help assess the risk endothelial damage caused by treatments, and circulatory of recurrence of VTD and the haemorrhagic risk in cancer stasis are contributory factors. Circulatory stasis may be patients. To do this, the rates of comorbidities, potential caused by the limitation of the patient’s mobility, as well pharmacological interactions, and other factors must be as by the extrinsic vascular compression exerted by the taken into account to decide the best possible therapeu- tumour itself or its metastases (e.g., tumours with bulky tic management. Establishing a standardized evaluation lymphadenopathies). protocol should facilitate shared decision-making with the informed patient and continuous re-evaluation according Histology and stage of the disease to the oncological evolution of the disease and the com- plications that have occurred. In some neoplasms, certain histological subtypes are asso- ciated with greater thrombotic risk, for example, in ovarian cancer and lung cancer, although this has not been seen Table 1  Cancer factor For Against considerations for and against maintaining long-term High thrombogenic risk: Low thrombogenic risk: secondary thromboprophylaxis Pancreatic cancer Breast cancer CNS tumours Prostate cancer Gastroesophageal cancer Histology of adenocarcinoma Other histologies Tumours with high degree of differentiation (G3) Locally advanced or metastatic disease Localized disease Presence of molecular alterations in KRAS, ALK, and ROS Absence of molecular alterations Recent surgery Hormone therapy in breast cancer Treatment with cyclin inhibitors in breast cancer Cisplatin chemotherapy treatment Antiangiogenic therapy CNS central nervous system 13 Clinical and Translational Oncology (2021) 23:697–708 699 in the different histological subtypes of breast and colon In April 2019, an increased risk of thrombosis was tumours. Patients with adenocarcinoma of the lung have a reported in patients carrying ROS1 translocation, similar to higher risk of thrombosis than those with squamous-cell that seen in carriers of ALK rearrangements [17]. carcinoma [12]. In addition, an increased risk of throm- bosis has been observed in patients with mucin-producing Treatment of the disease adenocarcinomas such as those of the pancreas and gas- trointestinal system. Surgery The risk of thrombosis is increased in all patients The published results of the Vienna-CATS study have who undergo surgery, both due to the prothrombotic state shown that the degree of tumour differentiation is associ- associated with cancer and the high morbidity and complex- ated with the risk of thrombosis: Patients with high-grade ity of surgery in these patients. In patients with cancer, the tumours (G3) have twice the risk of developing a throm- risks of deep vein thrombosis (DVT) and pulmonary throm- botic event than those with low-grade tumours (G1 or G2) boembolism (PTE) are up to 2 and 3 times higher, respec- [13]. tively, than they are in patients not requiring surgery. The risk of thrombosis also increases with the tumour Over the years, the risk of thrombosis associated with stage. Locally advanced and metastatic disease is associated surgery has decreased due to the improvement of surgical with a higher risk of thrombosis than localized disease. The techniques, the rapid mobilization of patients after surgery risk of VTD in patients with cancer without distant involve- and the improvement of prophylaxis and perioperative care. ment is up to 4 times higher than in the general population Surgery performed on the pelvis and abdomen is associated and can increase to up to 58 times with respect to the gen- with an increased risk of thrombosis. eral population if the patient has disseminated disease [14]. Patients with solid tumours and disseminated disease have Hormone therapy and cyclin inhibitors Hormone therapy 20 times the risk of those with tumour pathology but without for breast cancer is associated with an increased risk of metastasis. Finally, the risk of recurrence of thrombosis is thrombosis, especially in patients treated with tamoxifen higher in patients with advanced disease than in patients compared to those treated with aromatase inhibitors. with localized disease (5 versus 2–3 times) and in patients Recently, cyclin inhibitors (palbociclib, abemaciclib with active disease. and ribociclib) have been incorporated into the treatment of metastatic breast cancer positive for hormone receptors and negative for human epidermal growth factor receptor Molecular alterations (KRAS, ALK, ROS1) 2 (HER2), and they are administered in combination with hormone therapy. Among their side effects, an increased risk KRAS mutation is a predictive biomarker commonly used of thrombosis (up to 5%) has been described, and there is in colorectal cancer to identify patients who will not benefit debate about their pathophysiological mechanisms [18]. from treatment with drugs against epidermal growth factor receptor (EGFR). Regarding the risk of thrombosis, there Chemotherapy Cancer patients undergoing chemotherapy is evidence that the expression of tissue factors on the sur- are 6–7 times more likely to develop a thrombotic event face of colon tumour cells is activated by KRAS oncogene than those not receiving this treatment. Khorana et al. dem- mutations and inactivation of the TP53 tumour suppressor onstrated that the thrombosis rate was higher in a cohort gene. Given that tissue factor is a procoagulant associated of cancer patients 12 months after starting chemotherapy with systemic hypercoagulability, a multicentre study was (12.6%) than in the control cohort (1.4%) [19]. designed to test whether patients with KRAS mutations had In addition, an increased risk of thrombosis has been a higher risk of thrombosis. Results showed that patients associated with different types of chemotherapy. Cisplatin with mutated KRAS colorectal cancer had 3 times the risk is one of the most widely used drugs for malignant tumours, of developing thrombotic events as those with unmutated usually in combination with other drugs, and published KRAS, making it the first study that showed the implication evidence links it with an increase in venous and arterial of a genetic mutation on the risk of developing thrombosis thrombosis [20]. The combination of cisplatin with other [15]. chemotherapies carries up to twice the risk of thrombotic In 3–5% of patients with lung adenocarcinoma, ALK complications in gastroesophageal tumours as with regimens gene rearrangements are detected. There are now three with platinum salts, such as oxaliplatin (7% with cisplatin drugs directed against this alteration (crizotinib, alectinib versus 1% with oxaliplatin). Carboplatin, another commonly and ceritinib). In 2017, a study showed that patients with used platinum salt, is considered to have an intermediate ALK rearrangements had a three to fivefold higher risk of thrombotic risk (5.5 vs. 7.0% with cisplatin). thrombosis than patients with non-small-cell lung cancer Other immunosuppressive and chemotherapeutic (NSCLC) without ALK rearrangements [16]. drugs, such as thalidomide, lenalidomide, gemcitabine, 13 700 Clinical and Translational Oncology (2021) 23:697–708 l-asparaginase, anthracyclines, fluoropyrimidines, and iri- associated with an increased risk of recurrence of thrombo- notecan, have also been associated with an increased risk of sis in patients with cancer [30], and it is a variable included thrombosis [21]. in the Ottawa thrombosis recurrence predictive model along with other oncological variables and previous thromboem- Angiogenesis inhibitors The pharmacological inhibition bolic episodes [31]. However, the Ottawa model has not of the physiological functions of the vascular endothe- been validated consistently in other cohorts of patients, and lial growth factor (VEGF)–VEGF receptor (VEGFR) its practical utility is limited [31]. In fact, among patients axis affects the maintenance and protection of the normal with "unprovoked" thrombosis, the risk of VTD is higher endothelium, which may contribute to a greater predisposi- in men. Some data suggest that there is a greater risk of tion to arterial and venous thromboembolic complications. thrombosis in African-American cancer patients than those However, the published data do not show a clear causal rela- of Asian origin [32, 33], although there are no data on the tionship between the use of antiangiogenic monoclonal anti- risk of recurrence or long-term bleeding according to race bodies (such as bevacizumab, aflibercept and ramucirumab) or ethnicity [9, 34]. Both sedentary lifestyle and smoking and the increased risk of VTD when drug administration is are associated with increased thrombotic risk, while alcohol adjusted for duration by units of time. An increase in the risk consumption can increase the risk of haemorrhagic compli- of VTD was not observed with the use of oral antiangio- cations. In an analysis of the RIETE registry, immobility genic tyrosine kinase inhibitors. However, antiangiogenic was also independently associated with increased haemor- therapy significantly increases the incidence and risk of rhagic risk [29]. In this sense, it is of interest to take into arterial thrombosis [22]. In addition, elderly patients are at account other aspects of profession and lifestyle, such as an increased risk of stroke. participating in sports, that may help promote healthy life- style habits and provide an opportunity to improve throm- Factors related to the patient botic or haemorrhagic risk. Regarding the presence of comorbidities, the coexistence VTD is a pathology of multiple causes, which include risk of infection, arterial thromboembolism, kidney disease, lung factors associated with hypercoagulability both acquired and disease, trauma, or arthrodegenerative, neurological or psy- genetic, and specific to each patient, that overlap the risk chiatric diseases that lead to immobility can increase the of thrombosis from cancer and its treatment. Thrombotic risk of thrombosis. Similarly, myeloproliferative syndromes, risk factors can also be categorized as transient (and poten- nocturnal paroxysmal haemoglobinuria and haematological tially modifiable) or persistent. Likewise, thrombotic events diseases that present with anaemia, leukocytosis, or throm- themselves can be categorized as provoked or unprovoked, bocytosis can also increase the risk (Table 2). Among auto- depending on the risk factors related to VTD [23–26]. The immune inflammatory diseases that increase the risk of various factors that may have contributed to the development thrombosis are inflammatory bowel disease, antiphospho- of the initial thrombotic event should be considered to esti- lipid syndrome and vasculitis with endothelial involvement mate with greater precision the eventual risk of recurrence (Behçet disease). The presence of previous venous disease, of thrombosis. Similarly, in the evaluation of the patient, or a central venous catheter or other intravascular devices personal characteristics that could increase the haemorrhagic also increase the risk, along with other mechanical factors risk during anticoagulant treatment should be considered that can increase venous stasis, such as vascular malforma- [27, 28]. The balance between thrombotic and haemorrhagic tions or obesity. risk factors allows clinicians to individualize the benefit-risk Comorbidities that increase the haemorrhagic risk include ratio associated with the decision whether to maintain the a history of recent major haemorrhage and diseases associ- treatment of secondary thromboprophylaxis in each patient, ated with digestive bleeding (peptic ulcer, intestinal angiod- and whether to adjust its intensity to a full or a medium- or ysplasia or colonic polyposis) or bleeding in other locations, long-term prophylactic anticoagulant dose (Table 2). particularly if the cause is not corrected. Renal and hepatic Regarding demographic characteristics, younger age insufficiency, portal hypertension and the presence of anae- (< 65 years) has been associated with an increased risk of mia and platelet disease are also associated with increased recurrence in cancer patients [3, 29], while advanced age haemorrhagic risk. (≥ 65 years) carries both a thrombotic and a haemorrhagic In the presence of such comorbidities, the use of other risk in the general population on anticoagulants for VTD medications that may reduce thrombotic risk should always [27, 28]. There are other, more common haemorrhagic risk be considered, such as erythropoiesis-stimulating agents, variables, although these are not exclusive to the geriatric thalidomide, and hormonal therapy in young premenopau- population, such as the risk of falls, immobility, renal fail- sal patients for contraceptive purposes or as adjuvant treat- ure, and the concomitant use of antiplatelet therapy due to ments in survivors of hormone-sensitive neoplasms. In con- the previous cardiovascular disease. Female sex has been trast, antiplatelet drugs increase the haemorrhagic risk. In 13 Clinical and Translational Oncology (2021) 23:697–708 701 Table 2  Patient-related factors Factors favouring the recurrence of thrombosis Factors favouring haemorrhagic risk favouring recurrence of thrombosis and bleeding during Demographic factors anticoagulant treatment Age < 65 years old Age > 65 years, in particular > 75 years Female sex African–American vs. Asian race Pregnancy and postpartum Lifestyle Sedentary Immobility Immobility Reduced functionality Smoking Risk of falling High-risk professions and sports Alcohol abuse Comorbidities Obesity Previous bleeding, particularly if the Autoimmune diseases cause is not corrected Inflammatory bowel Severe renal insufficiency or uraemia Vasculitis (e.g., Behçet) Portal hypertension Antiphospholipid syndrome Hepatic failure Myeloproliferative syndromes Previous stroke Monoclonal gammopathy Diabetes mellitus Nocturnal paroxysmal haemoglobinuria Anaemia Anaemia or transfusions Thrombopenia Thrombocytosis Leukocytosis Concomitant pharmacological treatments Erythropoietin Antiplatelet agents Hormonal contraceptives Poor VKA-INR control Tamoxifen Corticotherapy Genetic factors Congenital thrombophilia Haemophilia Deficiency of: Von Willebrand disease Protein C Rendu–Osler–Weber disease Protein S Antithrombin Factor V Leiden Prothrombin G20210A Personal or family history of VTD Mechanical factors Venous catheter Other endovascular devices Vascular compression Varicose veins, varicectomy Vena cava agenesis VKA vitamin K antagonist, VTD venous thromboembolic disease, INR international normalized ratio general, the possibility of pharmacological interactions that routine clinical practice is not recommended in most patients may increase or decrease the effect of anticoagulant therapy with VTD. If such genetic factors are present, they must be should be carefully reviewed, particularly if long-term use interpreted individually as to whether they are thrombotic risk of VKA or DOAC is planned. factors. In any case, the personal and family history of first- Finally, the possibility of genetic thrombotic risk factors or degree VTD should be taken into account in any decision- congenital thrombophilia should be taken into account, such as making [35, 36]. deficiency of natural anticoagulants (very infrequent with an estimated prevalence < 1% in the general population) including antithrombin, protein C, and protein S. The presence of factor V Leiden and prothrombin G20210A (PTG20210A) is more prevalent in the general population, although their screening in 13 702 Clinical and Translational Oncology (2021) 23:697–708 Factors related to prior thrombotic events to patients with thrombosis in other areas of the body, with the first month into anticoagulation presenting the highest To decide whether anticoagulation should be indefinite, the risk of bleeding [40]. thrombotic event must be evaluated taking into consideration There are no clinically significant differences in the risk its location, severity, and sequelae (Table 3). of major haemorrhage or recurrence depending on whether the thromboembolic episode is incidental or symptomatic [40, 41], and therefore, it should not determine the dura- Location of the prior thrombotic event tion of treatment. However, the number of affected vessels (vascular territory at risk), the chronicity of the lesions, their The location of an initial thrombotic event influences the relationship with surgery and the haemorrhagic risks are decision about duration of anticoagulant treatment. In the main factors that must be taken into account to prolong patients with pulmonary embolism, mortality from the anticoagulation. In distal or upper-extremity thrombosis in recurrence of the thrombotic event is higher than in those cancer patients, their relationship with the catheter guides with DVT [37, 38], and therefore, in these patients, it is decisions. A total of 3–4% of patients with cancer and particularly advisable to maintain anticoagulation if the permanent central venous catheters will develop catheter- cancer is active. Along the same lines, proximal venous related thrombosis within one year [42]. The guidelines do thrombosis of the lower extremities is associated with an not recommend removing the central catheter unless it is increased risk of pulmonary embolism, and the duration of faulty, contaminated and/or the clinical symptoms worsen treatment should be longer than in distal or upper-extremity despite anticoagulation [7, 43]. In addition, it is advisable thrombosis, in which the risk of pulmonary embolism is to maintain anticoagulation, while the catheter is in place. If lower; therefore, a shorter duration of treatment should be the catheter is removed, the risk of recurrence of thrombosis considered. However, distal thrombosis in patients with can- is low, and therefore, it is not necessary to maintain antico- cer followed in the long term has a worse prognosis than in agulation [44]. Finally, in venous thrombosis of the central patients without cancer, with a higher haemorrhagic risk nervous system, indefinite treatment is usually recommended and rethrombosis risk, and so treatment decisions should be unless the initial episode is related to the administration of individualized [39]. drugs (e.g., chemotherapy and/or hormonal treatment) that Thrombosis of the splanchnic territory is increasingly can be suspended [45]. common in cancer patients, although it is not suspected in up to 59% of cases [40]. Although patients with cancer and The severity of the prior thrombotic event splanchnic thrombosis have more comorbidities that increase the risk of haemorrhage (e.g., platelet disease, impaired liver The mortality rate, the relapse rate, and the risk of major function, gastrointestinal disease or gastrointestinal cancer), haemorrhage in patients with cancer are similar to those in more than 50% of patients treatment is prolonged beyond of non-oncological patients with symptomatic pulmonary 6 months with LMWH doses similar to those administered embolism or incidental pulmonary embolism [46]. However, Table 3  Factors related to thrombotic events for and against initiating secondary thromboprophylaxis For indefinite treatment For withdrawal Thrombotic event Pulmonary embolism Distal deep vein thrombosis Proximal vein thrombosis Venous thrombosis in upper extremities with catheter removal Venous thrombosis in upper extremities with catheter maintenance Isolated, chronic, or postsurgical splenic thrombosis Acute unprovoked symptomatic multivessel splenic thrombosis CNS thrombosis Symptoms and/or tumour burden No influence Subsegmental asymptomatic incidental pulmonary embolism VTD without predisposing factors (false positive) Post-surgical VTD Aftermath Chronic VTD with or without pulmonary hypertension. In patients with per- Absence of residual thrombosis sistent dyspnoea, high central tumour burden, high-intermediate-risk PTE, high-risk PTE, or acute pulmonary hypertension. Evaluate lung scintigraphy or echocardiogram at 3–6 months VTD venous thromboembolic disease; CNS central nervous system; PTE pulmonary thromboembolism 13 Clinical and Translational Oncology (2021) 23:697–708 703 in patients with untreated incidental pulmonary embolism, D-dimer level has been correlated with the risk of the risks of recurrence and death are higher than in antico- recurrence in patients with DVT; therefore, a low D-dimer agulated patients, without significant differences between might suggest anticoagulation should be stopped, although patients with central and subsegmental embolism [47]. there are no conclusive data to support this practice [52]. Therefore, in most guidelines, anticoagulation is recom- In cancer patients, D-dimer may be elevated, even without mended in patients with cancer and isolated incidental sub- DVT. In a study of 167 patients with active cancer and a segmental pulmonary embolism even if it is asymptomatic, 12% incidence of VTD, D-dimer gradually increased in but whether the diagnosis is correct should be evaluated by VTD patients and remained constant in patients without having images reviewed by an expert radiologist and a com- VTD and an increased risk of recurrence in case of dupli- plete study carried out using venous Doppler ultrasound of cation (HR 2.78; 95% CI 1.69–4.58; p < 0.0001) [53]. In the lower extremities [48]. another study with 114 patients with cancer and VTD and an incidence of relapse of 9%, the elevation of D-dimer Sequelae of the prior thrombotic episode and high-sensitivity C-reactive protein was predictors of relapse of the thromboembolic episode (D-dimer: HR In patients with cancer, the sequelae of a prior thrombotic 5.81; 95% CI 1.1–31.7; C-reactive protein: HR 9.82; 95% event should also be evaluated. Thus, chronic thromboem- CI 19–52), and therefore, they may be useful to identify bolic pulmonary hypertension is an indication of indefinite susceptible patients following the interruption of antico- anticoagulation, and cancer is a risk factor for its develop- agulation [54]. ment. The role of residual thrombosis (RT) in the lower Therefore, given the limited evidence for parameters such extremities in the prediction of recurrence is not clear. In as RT and D-dimer in prediction of VTD in cancer patients, patients with cancer, the absence of thrombi in the re-eval- it is not possible at this time to recommend an anticoagula- uation of the episode at 6 months means that recurrence tion strategy. is unlikely [49], and therefore, the decision to withdraw anticoagulation is reinforced in patients with medium–high haemorrhagic risk, but the presence of RT does not increase the risk of recurrence. Options for secondary thromboprophylaxis Here, the evidence for each of the therapeutic options is The contribution of previous diagnostic reviewed. tests to decision‑making The use of RT as a predictor of recurrence after DVT has Vitamin K antagonists not been validated due to the use of different diagnostic techniques and testing times and the absence of a standard Few clinical trials have compared the efficacy of VKA definition. In a meta-analysis of 2527 patients with DVT versus LMWH in the long-term treatment of patients with without cancer, the predictive power of RT was weak (haz- cancer and VTD. In addition, most have a low or moder- ard ratio [HR]: 1.32; 95% confidence interval [CI 1.06–1.65) ate degree of evidence, and patient follow-up is usually no and absent when ultrasound was performed beyond 3 months longer than 12 months. In a Cochrane review published (HR 1.19, 95% CI 0.87–1.61). RT was observed in 55% of in 2018 [55], it was concluded that during the first year, patients, and relapse was observed in 16% [50]. In a study the risk of recurrence was higher in patients treated with of 347 patients with cancer and DVT, 70% of whom had RT, VKA than in those treated with LMWH (HR 0.58; 95% CI participants were randomized to maintain or not maintain 0.43–0.77). Regarding safety variables, the differences in the anticoagulation for more than 6 months, but no significant risk of major or minor haemorrhage did not reach statisti- differences were observed in relapses (HR 1.37, 95% CI cal significance, as was also the case with all-cause mortal- 0.7–2.5; p = 0.311) [49]. The absence of RT was a power- ity. None of the studies included quality-of-life variables. ful predictor of nonrecurrence of DVT, as there was a 3% Therefore, in terms of efficacy, the review concluded that recurrence rate after stopping anticoagulation (HR 6.0, 95% VKA is less effective than LMWH in the long-term treat- CI 1.7–21.2, p = 0.005). However, the possibility of bias in ment of VTD, although they are similar in terms of safety. that study means that the results remain inconclusive. In However, a recent study based on the RIETE registry has 153 patients evaluated (46% RT) in a post hoc analysis of a shown that patients treated with LMWH who switch to VKA cohort study in patients with cancer and DVT, RT showed after 6 months of treatment have a recurrence or bleeding no predictive power of relapse (odds ratio [OR]: 2.30; 95% rate similar to that of patients who continue with LMWH, so CI 0.73–7.22) [51]. this treatment should be considered, especially in patients in 13 704 Clinical and Translational Oncology (2021) 23:697–708 which LMWH or DOAC are contraindicated (e.g., patients to 7.9% in the group treated with dalteparin (HR 0.63; with creatinine clearance below 30 ml/min) [56]. p < 0.001 for non-inferiority), with no increased risk of gas- trointestinal haemorrhage in the first 6 months (1.9% with LMWH at full doses apixaban vs. 1.7% with dalteparin; HR 1.05) [64]. There are no results on the use of apixaban beyond 6 months, but one LMWH is the treatment of choice during the first 6 months ongoing study is comparing a full dose (5 mg/12 h) with following a thromboembolic event. Beyond this period, lower doses, simulating other studies in patients without evidence is scarce. The DALTECAN trial demonstrated, in cancer [65]. 100 patients with cancer and VTD with a follow-up longer In summary, most of the guidelines consider treatment than 6 months [57], that dalteparin maintains its efficacy with DOAC, specifically treatment with rivaroxaban and over time in terms of decreasing recurrence, with a stabi- edoxaban, to be a reasonable alternative to HPBM except in lization of the haemorrhage rate, which was higher in the patients with gastrointestinal or urological tumours or with first month (3.6%) but was gradually reduced (0.7% in the mucosal involvement and high haemorrhagic risk [66–68]. It seventh month). Similar results have been obtained with tin- is important to discuss with patients the different therapeutic zaparin in treatment up to 12 months [58]. options available, recognizing that for many, oral treatment is considered an advantage only if it is more effective than LMWH at prophylactic doses subcutaneous treatment [69]. There are no studies to support the use of reduced doses of Antiplatelet agents LMWH in cancer patients in whom long-term treatment is maintained [59]. It is recommended to administer the dose Currently, there is no evidence that antiplatelet agents can according to the actual, rather than the ideal, patient weight. be useful in long-term secondary thromboprophylaxis in patients with cancer and prior VTD. Studies that demon- DOAC strate its efficacy have been performed in patients with unprovoked VTD, but excluding those with neoplasia [70]. Two clinical trials have been reported in patients with neo- plasia and VTD with a follow-up beyond 12 months. In the HOKUSAI trial, the noninferiority of edoxaban compared Shared decision‑making with the informed to LMWH was demonstrated when the combined objective patient of recurrence and haemorrhage was evaluated [60]. There were no differences in the relapse rate or mortality rate, but Precision or personalized medicine undoubtedly requires the incidence of bleeding was higher in patients treated with a proactive role for the patient in shared decision-making DOAC. In a post hoc analysis, the greatest haemorrhagic (SDM) [71, 72]. Assessing the benefits and drawbacks of risk was observed in patients with gastrointestinal or uro- each therapeutic option with the patient should complement logical neoplasms. In the SELECT-D trial [61], patients who the individualization of treatment and therapeutic adjust- received rivaroxaban had a lower recurrence rate and a clini- ments based on the increasing availability of biomarkers and cally relevant higher haemorrhage rate than those treated the predictive models of response or toxicity (Fig. 1). The with dalteparin, especially among patients with gastrointes- role of the patient in SDM is particularly important in situ- tinal or urological tumours, although there were no differ- ations where the available evidence is limited and when ences in mortality rate between treatment groups. discussing treatments that can have a high impact on the In addition to these studies, several meta-analyses have quality of life or the lifestyle of the patient [73, 74]. been performed, and the results show that although the risk The patient must be able to participate in SDM, taking of recurrence is reduced, the risk of bleeding is slightly into consideration the advantages and disadvantages from higher [62]. It is possible that a higher net benefit is obtained several viewpoints [75, 76]: (1) whether or not to continue in patients treated with DOAC if those with a low risk of anticoagulationt; (2) establishing the expected duration of haemorrhage are selected. anticoagulation and the timelines to review and adjust if In the ADAM-VTD study [63], the risk of haemorrhage necessary; (3) assessing the possibility of correcting poten- was similar in patients treated with apixaban or dalteparin, tially modifiable risk factors for thrombosis; (4) consider- suggesting that this drug may be a good alternative. In the ing accessibility and costs; and (5) considering the personal recently published CARAVAGGIO study, patients with physical and psychological impact of each therapeutic active cancer and DVT or PE were randomized to receive option, compared with therapeutic abstention [74]. apixaban or dalteparin for 6 months. The rate of rethrombo- Despite the importance of SDM, its application to clinical sis was 5.6% in the group treated with apixaban compared practice is limited for several reasons. For example, tools for 13 Clinical and Translational Oncology (2021) 23:697–708 705 Fig. 1  Considerations after 6 months SDM have not yet been developed or validated in the field of considerations into account and ensuring the patient’s full cancer and VTD patients. Other limiting factors are the high and informed participation in the decision-making process. workload, lack of time for consultations and difficulties in increasing education and awareness about the importance of Acknowledgements The authors are grateful for the editorial assis- tance of Beatriz Gil-Alberdi of HealthCo (Madrid, Spain) in the pro- thrombosis and anticoagulation in oncology. Undoubtedly, duction of this manuscript. the research and development ongoing in these areas will help to integrate the available tools in a practical way in Funding SEMI and SEOM not received external financial support for highly complex scenarios for oncological patients for whom this project. oncospecific and supportive treatments coexist. Compliance with ethical standards Conflict of interest The authors declare that they did not have any con- Conclusions flict of interest related with this project. Ethical approval The study has been performed in accordance with the VTD is one of the main cardiovascular complications of ethical standards of the Declaration of Helsinki and its later amend- cancer patients, and therefore, it is of great importance to ments. This article does not contain any studies with human partici- ensure its correct diagnosis and management to minimize pants or animals performed by any of the authors. its impact on morbidity and mortality. The improvements Informed consent For this type of study, formal consent is not required. in oncological treatments in the last decade has increased the survival of cancer patients, and therefore, it is of vital importance to periodically review the evidence for the situ- Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- ation and treatment of thrombotic events, not only in the tion, distribution and reproduction in any medium or format, as long acute phase but also in the long term, especially where the as you give appropriate credit to the original author(s) and the source, scientific evidence has been weak. In reaching the deci- provide a link to the Creative Commons licence, and indicate if changes sion on whether or not to continue anticoagulation beyond were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated 6 months, it is important to review all risk factors involved otherwise in a credit line to the material. If material is not included in (i.e., those related to the disease, the patient and any prior the article’s Creative Commons licence and your intended use is not thrombotic event) and to establish a risk/benefit balance for permitted by statutory regulation or exceeds the permitted use, you will that decision-making. If the decision is reached to continue need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. with anticoagulant treatment, then the choice of the par- ticular therapy to use must be made, which includes VKA, LMWH, or most recently DOAC, taking all factors and 13 706 Clinical and Translational Oncology (2021) 23:697–708 References 17. Ng TL, Smith DE, Mushtaq R, Patil T, Dimou A, Yang S, et al. ROS1 gene rearrangements are associated with an elevated risk of peridiagnosis thromboembolic events. J Thorac Oncol. 1. Arnold M, Rutherford MJ, Bardot A, Ferlay J, Andersson TM, 2019;14:596–605. Myklebust TA, et al. Progress in cancer survival, mortality, and 18. Olson SR, DeLoughery TG, Shatzel JJ. Cyclin-dependent incidence in seven high-income countries 1995–2014 (ICBP kinase inhibitor-associated thromboembolism. JAMA Oncol. SURVMARK-2): a population-based study. Lancet Oncol. 2019;5:141–2. 2019;20:1493–505. 19. Khorana AA, Dalal M, Lin J, Connolly GC. Incidence and pre- 2. Strongman H, Gadd S, Matthews A, Mansfield KE, Stanway S, dictors of venous thromboembolism (VTE) among ambulatory Lyon AR, et al. Medium and long-term risks of specific cardiovas- high-risk cancer patients undergoing chemotherapy in the United cular diseases in survivors of 20 adult cancers: a population-based States. Cancer. 2013;119:648–55. cohort study using multiple linked UK electronic health records 20. Moore RA, Adel N, Riedel E, Bhutani M, Feldman DR, Tabbara databases. Lancet. 2019;394:1041–54. NE, et al. High incidence of thromboembolic events in patients 3. Prandoni P, Lensing AW, Piccioli A, Bernardi E, Simioni P, treated with cisplatin-based chemotherapy: a large retrospective Girolami B, et al. Recurrent venous thromboembolism and bleed- analysis. J Clin Oncol. 2011;29:3466–73. ing complications during anticoagulant treatment in patients with 21. Debbie Jiang MD, Alfred Ian Lee MD. Thrombotic risk from cancer and venous thrombosis. Blood. 2002;100:3484–8. chemotherapy and other cancer therapies. Cancer Treat Res. 4. Lee AY, Levine MN, Baker RI, Bowden C, Kakkar AK, Prins M, 2019;179:87–101. et al. Low-molecular-weight heparin versus a coumarin for the 22. Schutz FA, Je Y, Azzi GR, Nguyen PL, Choueiri TK. Bevaci- prevention of recurrent venous thromboembolism in patients with zumab increases the risk of arterial ischemia: a large study in cancer. N Engl J Med. 2003;349:146–53. cancer patients with a focus on different subgroup outcomes. Ann 5. Lee AYY, Kamphuisen PW, Meyer G, Bauersachs R, Janas MS, Oncol. 2011;22:1404–12. Jarner MF, et al. Tinzaparin vs warfarin for treatment of acute 23. Anderson FA Jr, Spencer FA. Risk factors for venous thrombo- venous thromboembolism in patients with active cancer: a rand- embolism. Circulation. 2003;107:I9–16. omized clinical trial. JAMA. 2015;314:677–86. 24. Cushman M. Epidemiology and risk factors for venous thrombo- 6. Key NS, Khorana AA, Kuderer NM, Bohlke K, Lee AYY, Arcelus sis. Semin Hematol. 2007;44:62–9. JI, et al. Venous thromboembolism prophylaxis and treatment in 25. Kearon C, Ageno W, Cannegieter SC, Cosmi B, Geersing GJ, patients with cancer: ASCO clinical practice guideline update. J Kyrle PA, et al. Categorization of patients as having provoked or Clin Oncol. 2019;15:JCO1901461. unprovoked venous thromboembolism: guidance from the SSC of 7. Olmos VP, Ramos Gallo MJ, Rebollo MA, Ortega DB, Docampo ISTH. J Thromb Haemost JTH. 2016;14:1480–3. LI, Romera-Villegas A, et al. Management of venous thromboem- 26. Khorana AA, Connolly GC. Assessing risk of venous thromboem- bolism in oncological patients: Spanish clinical practice guide- bolism in the patient with cancer. J Clin Oncol. 2009;27:4839–47. lines Consensus SEACV-SEOM. Med Clin. 2015;144(Suppl 27. Kearon C, Akl EA, Ornelas J, Blaivas A, Jimenez D, Bounameaux 1):3–15. H, et al. Antithrombotic therapy for VTE disease: CHEST guide- 8. Samuelson Bannow BT, Lee A, Khorana AA, Zwicker JI, Noble line and expert panel report. Chest. 2016;149:315–52. S, Ay C, et al. Management of cancer-associated thrombosis in 28. Palareti G, Antonucci E, Mastroiacovo D, Ageno W, Pengo V, Poli patients with thrombocytopenia: guidance from the SSC of the D, et al. The American College of Chest Physician score to assess ISTH. J Thromb Haemost JTH. 2018;16:1246–9. the risk of bleeding during anticoagulation in patients with venous 9. Lee AYY. When can we stop anticoagulation in patients with thromboembolism. J Thromb Haemost JTH. 2018;16:1994–2002. cancer-associated thrombosis? Blood. 2017;2017:128–35. 29. Trujillo-Santos J, Nieto JA, Tiberio G, Piccioli A, Di Micco P, 10. Zwicker JI, Bauer KA. How long is long enough? Extended Prandoni P, et al. Predicting recurrences or major bleeding in can- anticoagulation for the treatment of cancer-associated deep vein cer patients with venous thromboembolism. Findings from the thrombosis. J Clin Oncol. 2014;32:3596–9. RIETE Registry. Thromb Haemost. 2008;100:435–9. 11. Chew HK, Wun T, Harvey D, Zhou H, White RH. Incidence of 30. Louzada ML, Carrier M, Lazo-Langner A, Dao V, Kovacs MJ, venous thromboembolism and its effect on survival among patients Ramsay TO, et al. Development of a clinical prediction rule with common cancers. Arch Intern Med. 2006;166:458–64. for risk stratification of recurrent venous thromboembolism in 12. Blom JW, Osanto S, Rosendaal FR. The risk of a venous throm- patients with cancer-associated venous thromboembolism. Cir- botic event in lung cancer patients: higher risk for adenocarci- culation. 2012;126:448–54. noma than squamous cell carcinoma. J Thromb Haemost JTH. 31. Delluc A, Miranda S, den Exter P, Louzada M, Alatri A, Ahn S, 2004;2:1760–5. et al. Accuracy of the Ottawa score in risk stratification of recur- 13. Ahlbrecht J, Dickmann B, Ay C, Dunkler D, Thaler J, Schmidinger rent venous thromboembolism in patients with cancer-associated M, et al. Tumor grade is associated with venous thromboembo- venous thromboembolism. A systematic resview and meta-anal- lism in patients with cancer: results from the Vienna Cancer and ysis. Haematologica. 2019;105:1436–42. Thrombosis Study. J Clin Oncol. 2012;30:3870–5. 32. Prins MH, Lensing AWA, Prandoni P, Wells PS, Verhamme P, 14. Abdol Razak NB, Jones G, Bhandari M, Berndt MC, Metharom P. Beyer-Westendorf J, et al. Risk of recurrent venous thrombo- Cancer-associated thrombosis: an overview of mechanisms, risk embolism according to baseline risk factor profiles. Blood Adv. factors, and treatment. Cancers (Basel). 2018;10:380. 2018;2:788–96. 15. Ades S, Kumar S, Alam M, Goodwin A, Weckstein D, Dugan M, 33. Timp JF, Braekkan SK, Versteeg HH, Cannegieter SC. Epi- et al. Tumor oncogene (KRAS) status and risk of venous throm- demiology of cancer-associated venous thrombosis. Blood. bosis in patients with metastatic colorectal cancer. J Thromb Hae- 2013;122:1712–23. most JTH. 2015;13:998–1003. 34. Chen YW, Khorana AA. The association between race and 16. Zer A, Moskovitz M, Hwang DM, Hershko-Klement A, Fridel L, venous thromboembolism risk after initiation of chemotherapy: Korpanty GJ, et al. ALK-rearranged non-small-cell lung cancer an analysis of the SAVE-ONCO trial control arm. Am J Hematol. is associated with a high rate of venous thromboembolism. Clin 2017;92:E101–E103103. Lung Cancer. 2017;18:156–61. 13 Clinical and Translational Oncology (2021) 23:697–708 707 35. Connors JM. Thrombophilia testing and venous thrombosis. N 51. Dentali F, Barco S, Pegoraro S, Di Minno MND, Mastroiacovo D, Engl J Med. 2017;377:1177–87. Pomero F, et al. Residual vein obstruction in patients diagnosed with 36. Moran J, Bauer KA. Managing thromboembolic risk in acute isolated distal deep vein thrombosis associated with active patients with hereditary and acquired thrombophilias. Blood. cancer. J Thromb Thrombolysis. 2018;46:404–8. 2020;135:344–50. 52. Kearon C, Parpia S, Spencer FA, Schulman S, Stevens SM, Shah 37. Carrier M, Le Gal G, Wells PS, Rodger MA. Systematic review: V, et al. Long-term risk of recurrence in patients with a first unpro- case-fatality rates of recurrent venous thromboembolism and voked venous thromboembolism managed according to D-dimer major bleeding events among patients treated for venous throm- results; a cohort study. J Thromb Haemost JTH. 2019;17:1144–52. boembolism. Ann Intern Med. 2010;152:578–89. 53. Posch F, Riedl J, Reitter EM, Crowther MJ, Grilz E, Quehenberger 38. Chee CE, Ashrani AA, Marks RS, Petterson TM, Bailey KR, Mel- P, et al. Dynamic assessment of venous thromboembolism risk in ton LJ 3rd, et al. Predictors of venous thromboembolism recur- patients with cancer by longitudinal D-dimer analysis: a prospective rence and bleeding among active cancer patients: a population- study. J Thromb Haemost JTH. 2020;18:1348–56. based cohort study. Blood. 2014;123:3972–8. 54. Jara-Palomares L, Solier-Lopez A, Elias-Hernandez T, Asensio- 39. Galanaud JP, Sevestre MA, Pernod G, Genty C, Richelet S, Kahn Cruz MI, Blasco-Esquivias I, Sanchez-Lopez V, et al. D-dimer and SR, et al. Long-term outcomes of cancer-related isolated distal high-sensitivity C-reactive protein levels to predict venous throm- deep vein thrombosis: the OPTIMEV study. J Thromb Haemost boembolism recurrence after discontinuation of anticoagulation for JTH. 2017;15:907–16. cancer-associated thrombosis. Br J Cancer. 2018;119:915–21. 40. Tufano A, Ageno W, Di Micco P, Niglio A, Rosa V, Ballaz A, 55. Kahale LA, Hakoum MB, Tsolakian IG, Matar CF, Terrenato I, et al. Outcomes during anticoagulation in patients with symp- Sperati F, et al. Anticoagulation for the long-term treatment of tomatic vs. incidental splanchnic vein thrombosis. Thromb Res. venous thromboembolism in people with cancer. Cochrane Data- 2018;164:69–74. base Syst Rev. 2018;6:CD006650. 41. Riva N, Ageno W, Schulman S, Beyer-Westendorf J, Duce R, 56. Chai-Adisaksopha C, Iorio A, Crowther MA, de Miguel J, Salgado Malato A, et al. Clinical history and antithrombotic treatment E, Zdraveska M, et al. Vitamin K antagonists after 6 months of low- of incidentally detected splanchnic vein thrombosis: a multi- molecular-weight heparin in cancer patients with venous thrombo- centre, international prospective registry. Lancet Haematol. embolism. Am J Med. 2018;131:430–7. 2016;3:e267–e275275. 57. Francis CW, Kessler CM, Goldhaber SZ, Kovacs MJ, Monreal M, 42. Decousus H, Bourmaud A, Fournel P, Bertoletti L, Labruyere Huisman MV, et al. Treatment of venous thromboembolism in can- C, Presles E, et al. Cancer-associated thrombosis in patients with cer patients with dalteparin for up to 12 months: the DALTECAN implanted ports: a prospective multicenter French cohort study study. J Thromb Haemost JTH. 2015;13:1028–35. (ONCOCIP). Blood. 2018;132:707–16. 58. Jara-Palomares L, Solier-Lopez A, Elias-Hernandez T, Asensio- 43. Debourdeau P, Farge D, Beckers M, Baglin C, Bauersachs RM, Cruz M, Blasco-Esquivias I, Marin-Barrera L, et al. Tinzaparin Brenner B, et al. International clinical practice guidelines for the in cancer associated thrombosis beyond 6 months: TiCAT study. treatment and prophylaxis of thrombosis associated with central Thromb Res. 2017;157:90–6. venous catheters in patients with cancer. J Thromb Haemost JTH. 59. Easaw JC, Shea-Budgell MA, Wu CM, Czaykowski PM, Kassis J, 2013;11:71–80. Kuehl B, et al. Canadian consensus recommendations on the man- 44. Delluc A, Le Gal G, Scarvelis D, Carrier M. Outcome of central agement of venous thromboembolism in patients with cancer. Part venous catheter associated upper extremity deep vein thrombosis in 2: treatment. Curr Oncol. 2015;22:144–55. cancer patients. Thromb Res. 2015;135:298–302. 60. Raskob GE, van Es N, Verhamme P, Carrier M, Di Nisio M, Garcia 45. Saposnik G, Barinagarrementeria F, Brown RD Jr, Bushnell CD, D, et al. Edoxaban for the treatment of cancer-associated venous Cucchiara B, Cushman M, et al. Diagnosis and management of cer- thromboembolism. N Engl J Med. 2018;378:615–24. ebral venous thrombosis: a statement for healthcare professionals 61. Young AM, Marshall A, Thirlwall J, Chapman O, Lokare A, Hill C, from the American Heart Association/American Stroke Association. et al. Comparison of an oral factor Xa inhibitor with low molecular Stroke. 2011;42:1158–92. weight heparin in patients with cancer with venous thromboem- 46. den Exter PL, Hooijer J, Dekkers OM, Huisman MV. Risk of recur- bolism: results of a randomized trial (SELECT-D). J Clin Oncol. rent venous thromboembolism and mortality in patients with cancer 2018;36:2017–23. incidentally diagnosed with pulmonary embolism: a comparison 62. Li A, Garcia DA, Lyman GH, Carrier M. Direct oral anticoagulant with symptomatic patients. J Clin Oncol. 2011;29:2405–9. (DOAC) versus low-molecular-weight heparin (LMWH) for treat- 47. van der Hulle T, den Exter PL, Planquette B, Meyer G, Soler S, ment of cancer associated thrombosis (CAT): a systematic review Monreal M, et al. Risk of recurrent venous thromboembolism and and meta-analysis. Thromb Res. 2019;173:158–63. major hemorrhage in cancer-associated incidental pulmonary embo- 63. McBane RD 2nd, Wysokinski WE, Le-Rademacher JG, Zemla T, lism among treated and untreated patients: a pooled analysis of 926 Ashrani A, Tafur A, et al. Apixaban and dalteparin in active malig- patients. J Thromb Haemost JTH. 2016;14:105–13. nancy-associated venous thromboembolism: the ADAM VTE trial. 48. Di Nisio M, Lee AY, Carrier M, Liebman HA, Khorana AA, Sub- J Thromb Haemost JTH. 2020;18:411–21. committee on H, et al. Diagnosis and treatment of incidental venous 64. Agnelli G, Becattini C, Meyer G, Munoz A, Huisman MV, Connors thromboembolism in cancer patients: guidance from the SSC of the JM, et al. Apixaban for the treatment of venous thromboembolism ISTH. J Thromb Haemost JTH. 2015;13:880–3. associated with cancer. N Engl J Med. 2020;382:1599–607. 49. Napolitano M, Saccullo G, Malato A, Sprini D, Ageno W, Imberti D, 65. McBane RD 2nd, Loprinzi CL, Ashrani A, Lenz CJ, Houghton et al. Optimal duration of low molecular weight heparin for the treat- D, Zemla T, et al. Extending venous thromboembolism secondary ment of cancer-related deep vein thrombosis: the Cancer-DACUS prevention with apixaban in cancer patients: the EVE trial. Eur J Study. J Clin Oncol. 2014;32:3607–12. Haematol. 2020;104:88–96. 50. Donadini MP, Ageno W, Antonucci E, Cosmi B, Kovacs MJ, Le Gal 66. Farge D, Frere C, Connors JM, Ay C, Khorana AA, Munoz A, et al. G, et al. Prognostic significance of residual venous obstruction in 2019 international clinical practice guidelines for the treatment and patients with treated unprovoked deep vein thrombosis: a patient- level meta-analysis. Thromb Haemost. 2014;111:172–9. 13 708 Clinical and Translational Oncology (2021) 23:697–708 prophylaxis of venous thromboembolism in patients with cancer. 72. Jameson JL, Longo DL. Precision medicine—personalized, prob- Lancet Oncol. 2019;20:e566–e581581. lematic, and promising. N Engl J Med. 2015;372:2229–34. 67. Frere C, Benzidia I, Marjanovic Z, Farge D. Recent advances in the 73. Beach MC, Sugarman J. Realizing shared decision-making in prac- management of cancer-associated thrombosis: new hopes but new tice. JAMA. 2019;322:811. challenges. Cancers (Basel). 2019;11:71. 74. Giuse NB, Kusnoor SV, Koonce TY, Naylor HM, Chen SC, Blas- 68. Khorana AA, Noble S, Lee AYY, Soff G, Meyer G, O’Connell C, ingame MN, et al. Guiding oncology patients through the maze of et al. Role of direct oral anticoagulants in the treatment of cancer- precision medicine. J Health Commun. 2016;21(Suppl 1):5–17. associated venous thromboembolism: guidance from the SSC of the 75. Noble S. Venous thromboembolism and palliative care. Clin Med ISTH. J Thromb Haemost JTH. 2018;16:1891–4. (Lond). 2019;19:315–8. 69. Hutchinson A, Rees S, Young A, Maraveyas A, Date K, Johnson MJ. 76. Noble S, Sui J. The treatment of cancer associated thrombosis: does Oral anticoagulation is preferable to injected, but only if it is safe and one size fit all? Who should get LMWH/warfarin/DOACs? Thromb effective: an interview study of patient and carer experience of oral Res. 2016;140(Suppl 1):S154–S159159. and injected anticoagulant therapy for cancer-associated thrombosis in the select-d trial. Palliat Med. 2019;33:510–7. Publisher’s Note Springer Nature remains neutral with regard to 70. Marik PE, Cavallazzi R. Extended anticoagulant and aspirin treat- jurisdictional claims in published maps and institutional affiliations. ment for the secondary prevention of thromboembolic disease: a sys- tematic review and meta-analysis. PLoS ONE. 2015;10:e0143252. 71. Collins FS, Varmus H. A new initiative on precision medicine. N Engl J Med. 2015;372:793–5. 13