If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Thromboprophylaxis should be considered in patients with ALK/ROS1+ stage IV lung cancer requiring systemic treatment.
•
Cautiously reintroduce an effective antiangiogenic agent after thrombosis.
•
Anticoagulation should be maintained for >6 months in patients with metastatic cancer on active treatment.
•
If rethrombosis, rule out active cancer, infratherapeutic dose, or anticoagulation failure.
•
Treat thrombosis with LMWH not requiring dose titration (CrCl>30) or stop (CrCl<15).
Abstract
Background
Cancer patients suffer high risk of venous thromboembolism (VTE). Cancer-associated VTE (CAT) causes hospitalization, morbidity, delayed cancer treatment, and mortality; therefore, exceptional CAT prevention and management are imperative.
Methods
This review offers practical recommendations and treatment algorithms for eight complex, clinically relevant situations posing great uncertainty regarding management and requiring an urgent decision: VTE prophylaxis in ambulatory cancer patients with pancreatic pancreas (1) or lung cancer with molecular alterations (2); optimal management of VTE during antineoplastic treatment with antiangiogenics (3) or chemotherapy (4); protracted VTE treatment, determinants; (5) drugs used (6), and optimal VTE management in situations of high bleeding risk (7) or recurrent VTE (8).
Results
With the evidence available, primary thromboprophylaxis in patients with lung cancer harbouring ALK/ROS1 translocations or pancreatic cancer receiving ambulatory chemotherapy must be appraised. If antiangiogenic therapy can yield a clear benefit and the patient recovers from a grade 3 thrombotic event, it can be cautiously re-introduced in selected cases, provided that the person agrees to assume the risk after being duly informed. Anticoagulation maintenance beyond 6 months is recommended in individuals with metastatic tumours, on active treatment, or at high risk for recurrent VTE without bleeding risk. In such cases, LMWH and DOACs are safe, being mindful that the latter could entail a higher risk of bleeding; consequently, they should be used judiciously in more haemorrhagic tumours, such as gastrointestinal cancers. In cases of recurrent VTE, the presence of active cancer, infra-therapeutic dose, and anticoagulant treatment failure must be ruled out. In individuals with platelet counts of 25,000–50,000 and VTE liable to recur who need anticoagulation, full-dose LMWH and transfusion support can be contemplated to reach values of > 50,000. In CAT unlikely to recur, decreasing the LMWH dose by 25–50% is recommended. Renal impairment associated with thrombosis must be treated with LMWHs; there is no need to adjust the dose in patients with CrCl > 30; with CrCl = 15–30, dose adjustment is advised, and suspended when CrCl is < 15.
Conclusion
We provide useful advice for complex, clinically relevant situations that clinicians treating CAT must face devoid of any unequivocal, strong, evidence-based recommendations.
Venous thromboembolism (VTE) is a leading cause of morbidity and mortality among cancer patients. They develop more VTE than those without cancer and experience higher rates of VTE recurrence and haemorrhagic complications [
Optimal prevention and treatment strategies for cancer-associated thrombosis (CAT) comprise an area of ongoing research. Despite the growing number of randomized clinical trials (RCT) in CAT treatment in recent years, there remains a paucity of evidence for many issues commonly encountered in clinical practice. Indeed, different cancer therapies; increasing diagnoses of incidental events, and the presence of other comorbidities (thrombocytopenia, wasting, or altered renal and liver function) complicate decision making, posing unique challenges for clinicians to provide optimal care for this high-risk population. Furthermore, evidence-based clinical practice guidelines cannot establish robust recommendations for these puzzling scenarios.
This article aims to address issues pertaining to CAT for which there is limited or no evidence to inform best practice in urgent and critical situations in which the thrombotic event and its management can impact cancer progression.
2. Methods
With the aim of agreeing on a reasoned proposal to approach critical situations for which there are no data from clinical trials and only a low level of evidence; that generally require urgent management, and that affect prognosis in patients with CAT, a 12-member expert panel was formed. Its members had to meet the following criteria: (1) medical oncologist; (2) specialized in thrombosis and cancer with proven clinical experience; (3) ≥ 2 publications on thromboembolic disease in the last five years. Subsequently, the panel voted on each of the subjects to be included in the article based on clinical interest, specific relevance in oncological patients, and availability of some degree of scientific evidence. With these considerations, six of the fourteen items were discarded: (1) catheter-associated thrombosis, (2) splanchnic thrombosis, (3) thrombosis in individuals with hepatic insufficiency, (4) thrombosis in patients with renal impairment, (5) thrombosis in subjects with unresected mucosal primary cancer, (6) thrombosis in patients with extreme weight. The eight controversial items surrounding VTE ultimately selected were articulated with eight questions and grouped into four major areas: (1) prophylaxis in ambulatory cancer patients; (2) optimal management during oncological treatment; (3) protracted treatment (> 6 months), and (4) the best possible management in individuals at high risk for haemorrhagic or recurrent VTE.
The available literature was structured following the PICOS format, i.e., population (P), intervention (I), comparison (C), outcome (O), and study design (S) Annex Table 1. reflects the PICOS framework for the objective of this review. The quality of evidence and strength of the recommendation were gauged using the ACP's Guideline Grading System, which is adopted from the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) system (see Annex Table 2) [
]. A contextualized systematic review of the narrative component of the articles was then undertaken. This format seeks to incorporate a reproducible bibliographic search method to reach a systematic level of evidence regarding the practical aspects of managing this disease. The narrative component to make it easier for the reader to obtain updated, intuitive clinical answers, as well as new perspectives, without falling into the rigid aspects of a formal systematic review. Two senior medical oncologists (PJF and EG) conducted a systematic exploration of PubMed and Google Scholar; key terms were used to generate comprehensive lists of key words using the Boolean operators ‘AND’ and ‘OR’ (see Annex Table 1), and controversial issues in CAT were identified. Articles written in English were selected without any restriction regarding time, study design, or type of publication, as were articles addressing preclinical studies, with the belief that they might open new avenues of CAT treatment. The authors of the search took advantage of the references of these articles to manually search for other relevant publications. The 2019 American society of medical oncology (ASCO) [
Annex Table 1. Definition of questions using PICOS framework.
Question #1
In what situations might primary ambulatory thromboprophylaxis be contemplated for a patient at high risk for VTE, such as pancreatic cancer? With what treatment?
Population:
Ambulatory cancer patients
Intervention:
Primary thromboprophylaxis
Comparison:
Placebo
Outcome:
Thrombosis & overall survival
Study design:
All study designs
Search terms:
(thromboprophylaxis OR prophylaxis) AND ambulatory AND cancer AND placebo AND (thrombosis OR survival)
In what situations might primary ambulatory thromboprophylaxis be contemplated in a person with an intermediate-high risk for VTE, such as lung or molecularly selected cancer (ALK/ROS1 translocation carriers)?
Population:
Ambulatory cancer patients at high risk of thrombosis
Intervention:
Primary thromboprophylaxis
Comparison:
Placebo
Outcome:
Thrombosis & overall survival
Study design:
All study designs
Search terms:
(thromboprophylaxis OR prophylaxis) AND ambulatory AND cancer AND (Khorana OR ALK OR ROS1 OR high risk) AND (thrombosis OR survival)
What is the optimal management for venous thromboembolic disease during treatment with antiangiogenics (tyrosine kinase inhibitors (TKI) or monoclonal antibodies (MABs))?
Population:
Cancer patients
Intervention:
Anticoagulant therapy
Comparison:
All
Outcome:
Recurrent thrombosis, bleeding & survival
Study design:
All study designs
Search terms:
Cancer AND thrombosis AND (heparin OR anticoagulant) AND (antiangiogenic OR tyrosine kinase inhibitors OR bevacizumab OR ramucirumab OR aflibercept) AND (Bleeding OR rethrombosis OR survival)
What is the anticoagulant drug of choice for prolonged treatment of thrombosis beyond 6 months and when should the risk be re-evaluated?
Population:
Cancer patients
Intervention:
Long-term anticoagulation (> 6 months); low-molecular weight heparin; vitamin K antagonist; direct factor Xa inhibitor; direct oral anticoagulant; dabigatran; apixaban; rivaroxaban; edoxaban; direct thrombin inhibitor; aspirin
Comparison:
Others
Outcome:
Rethrombosis, bleeding & survival
Study design:
All study designs
Search terms:
cancer AND thrombosis AND (heparin OR "low-molecular weight heparin" OR "vitamin K antagonist" OR warfarin OR "direct factor Xa inhibitor" OR "direct oral anticoagulant" OR dabigatran OR apixaban OR rivaroxaban OR edoxaban OR "direct thrombin inhibitor" OR aspirin) AND (extended OR protracted) AND (Bleeding OR rethrombosis OR survival)
How should thrombosis be managed in a patient with a high bleeding risk?
Population:
Cancer
Intervention:
Anticoagulant therapy
Comparison:
All
Outcome:
Rethrombosis, bleeding & survival
Study design:
All study designs
Search terms:
cancer AND thrombosis AND (heparin OR "low-molecular weight heparin" OR "vitamin K antagonist" OR warfarin OR "direct factor Xa inhibitor" OR "direct oral anticoagulant" OR dabigatran OR apixaban OR rivaroxaban OR edoxaban OR "direct thrombin inhibitor" OR aspirin) AND ("high risk" AND (bleeding OR hemorrhage)) AND (Bleeding OR rethrombosis OR survival)
The two senior medical oncologists assessed the evidence available, including RCTs, meta-analyses, and other studies. After grading the evidence, the other experts voted to issue the recommendation, bearing in mind the severity of the clinical issue, magnitude of risk and benefit, quality of the evidence, viability, cost, and preferences. A summary of the ACP grading system guideline is presented in Annex Table 2. After calculating the results, if there was < 75% consent or > 20% dissent regarding an item in particular, the subject was discussed and recirculated among the panellists until a consensus was reached. Addressing the weak evidence currently available and in the absence of quantitative inferences of the many issues the clinician faces every day, qualitative suggestions (or quantitative ones when available) based on expert consensus were adopted. Finally, the panellists made a critical review and formulated recommendations that were summarized in decision trees.
3. Results
1. In what situations might primary ambulatory thromboprophylaxis be contemplated for a patient at high risk for VTE, such as pancreatic cancer? With what treatment?
If we heed the leading national and international clinical guidelines, we find that there is no consensus in certain situations, yet none of them routinely recommend primary thromboprophylaxis in cancer patients receiving ambulatory treatment, pointing out that each case be assessed individually. The SEOM guideline [
] recommends evaluating thromboprophylaxis with low-molecular-weight heparin (LMWH) or with direct oral anticoagulants (DOACs) (apixaban or rivaroxaban) in high-risk cancer patients (Khorana ≥ 2 or when any other risk scale indicates high risk) who are going to begin onco-specific treatment, gauging bleeding risk and close monitoring (strong recommendation, moderate-quality evidence). The ASCO guidelines [
] point in a similar direction (high-risk patients, Khorana ≥ 2 and low bleeding risk) (weak recommendation, moderate-high quality evidence for apixaban and rivaroxaban, intermediate-quality evidence for LMWH). The ASH guideline [
] establish recommendations for intermediate-high risk patients according to validated scales, high-risk tumours (such as pancreatic cancer), and with low bleeding risk. In this guideline, a specific recommendation is made for the use of prophylaxis with LMWH in individuals with advanced pancreatic cancer receiving systemic cancer treatment who have a low bleeding risk. Likewise, the NCCN guidelines [
There are RCTs that attempt to evaluate the need for thromboprophylaxis in high-risk cancer patients, although at times, the difficulty lies in being able to determine which ones are high-risk. Most clinical trials use the Khorana index for this purpose, its limitations notwithstanding. Since then, part of the effort has focused on constructing multivariable models that might support the development of cost-effective primary thromboprophylaxis strategies, either based on the Khorana score or by integrating biomarkers. The validation of the ONKOTEV model, which adds the presence of vascular compressions and metastatic disease to the dichotomized Khorana score, comprises one such recent development. Another validated alternative is a more concise predictive model that only includes the combination of a biomarker consistently associated with thrombosis (D-dimer) with the location of the primary tumour [
]. Pending testing of these tools as part of enhanced eligibility criteria in RCTs, whether their discriminatory capacity suffices as a selection method has yet to be elucidated.
Meanwhile, the AVERT (apixaban) and CASSINI (rivaroxaban) studies examined the efficacy of DOACs in subjects with cancer with an intermediate-high risk of VTE, based on their Khorana score, receiving oncological treatment for at least 6 months. The incidence of VTE events was seen to decline by 2.8% in the CASSINI study and by up to 6% in the AVERT study, with an increase in bleeding versus placebo of only 1% and 1.7%, respectively. Moreover, the CASSINI study included a considerable number of individuals with pancreatic cancer (n = 274, 32% of the total), detecting the appearance of VTE in 9.6% of the patients compared to 13% in the control group [
Rivaroxaban thromboprophylaxis in ambulatory patients with pancreatic cancer: Results from a pre-specified subgroup analysis of the randomized CASSINI study.
]. One important aspect when interpreting the CASSINI trial is that it was carried out using lower limb US screening of all patients, excluding 4.5% of the cases with pre-existing thrombosis. This may have decreased the thrombotic risk profile of the population that was randomized.
At present, advanced pancreatic cancer is the neoplasm for which there is the greatest evidence regarding primary thromboprophylaxis; furthermore, it is a scenario in which this aspect arises most often. Specific clinical trials have been performed in individuals with pancreatic cancer, including the FRAGEM and CONKO-004 trials [
], with patients with metastatic or locally advanced cancer, assessed prophylaxis with dalteparin associated gemcitabine for at least 12 weeks and revealed a significant reduction in VTE (from 15 to 5% in the study arm). The CONKO-004 RCT [
Primary thromboprophylaxis in ambulatory pancreatic cancer patients receiving chemotherapy: a systematic review and meta-analysis of randomized controlled trials.
] that looked at 1003 cases of pancreatic cancer from 5 RCTs (PROTECHT, SAVE ONCO, CONKO, FRAGEM, and CASSINI). It demonstrated a significant decrease in VTE risk (relative risk: 0.31) with an estimated NNT (Number Needed to Treat) of 11.9 patients to prevent one event. Similar reductions of VTE were seen in studies of LMWH compared to those that examined DOACs. As for bleeding risk, a high relative risk of 1.08 was observed.
With these data, according to the conclusions of the meta-analysis, current evidence endorses the benefit of thromboprophylaxis in individuals with pancreatic cancer receiving ambulatory chemotherapy (Fig. 1A) (strong recommendation, high-quality evidence).
Fig. 1Algorithm for ambulatory thromboprophylaxis in patients with pancreatic cancer (1A) and lung cancer (1B) 1A).
2. In what situations might primary ambulatory thromboprophylaxis be contemplated in a person with an intermediate-high risk for VTE, such as lung or molecularly selected cancer (ALK/ROS1 translocation carriers)?
The leading clinical practice guidelines concur in not routinely recommending primary thromboprophylaxis in ambulatory cancer patients and evaluating each case individually, as commented in the previous section. They suggest appraising thromboprophylaxis with DOACs or LMWH in patients at high risk for thrombosis based on validated scales (Khorana ≥ 2) and who are going to receive antineoplastic therapy.
Different RCTs have examined the potential benefit of primary ambulatory thromboprophylaxis in cancer patients. A systematic Cochrane meta-analysis that collected data from 9 RCT comparing LMWH against placebo in this situation concludes that LMWH lowers the risk of symptomatic thrombosis by 46% compared to placebo (relative risk 0.54, 95% confidence Interval (CI) 0.38–0.75) [
]. Among the most important publications, two RCTs specific for patients with lung cancer are noteworthy, TOPIC-2 and FRAGMATIC, that have proven a risk reduction of thrombosis in patients treated with LMWH vs. placebo (3.5 vs 10.2% and 5.5 vs 9.7%, hazard ratio 0.57), with no significant differences in bleeding risk and without benefit in overall survival [
]. As for DOACs, the CASSINI (rivaroxaban) and AVERT (apixaban) RCTs confirmed the decrease in the risk of thrombosis with no differences in survival with DOACs compared to placebo and, in the case of apixaban, at the expense of greater bleeding risk. Nonetheless, the patient sample with lung cancer was very small (approximately 30 cases) [
While trials with DOACs attempted to lower the NNT of a population with an enriched thrombotic risk chosen based on their Khorana score, this approach also fails to allow for the problem of the variability of thrombotic risk within the same neoplasm by virtue of the different biological profiles of the neoplasms. Be that as it my, this limitation is glaring in the other trials on thromboprophylaxis in lung cancer, such as the TOPIC-2 study [
As molecular screening has been implemented in more tumours, the association between thrombotic risk and specific mutational profiles of certain tumours has been revealed. The most convincing data to date come from patients with lung cancer with ALK/ROS1 alterations (7% of the total). In fact, some guidelines recommend assessing thromboprophylaxis in patients with tumours harbouring ALK/ROS1 translocations [
]. According to various cohorts, these individuals have an accumulated risk of VTE of between 30 and 43% and an almost four-fold thrombotic risk compared to tumours without ALK translocation. Up to 40% of all VTE are pulmonary embolisms (PE) and, in some series, they are associated with lower overall survival with respect to patients without thrombosis [
]. Similarly, individuals with ROS1-translocated lung cancer exhibited an accumulated VTE incidence of up to 50% (41–48%), much higher than patients with lung cancer without molecular alterations. In both subgroups, the highest incidence of thrombosis (up to 60% of the events recorded) take place during diagnosis and progression [
ROS1-rearranged Non–small-cell lung cancer is associated with a high rate of venous thromboembolism: analysis from a phase II, prospective, multicenter, two-arms trial (METROS).
At the same time, in these molecular subgroups, the classical predictive factors for thrombosis are unrelated to the risk of presenting a VTE and, based on different hypotheses, the studies suggest that the increased thrombotic risk has to do with the molecular driver per se. Therefore, the thrombotic risk scales validated thus far do not appear to be pertinent to this subgroup of patients, as they would not portray the true risk of thrombosis, merely associating those factors that are broadly considered [
ROS1-rearranged Non–small-cell lung cancer is associated with a high rate of venous thromboembolism: analysis from a phase II, prospective, multicenter, two-arms trial (METROS).
With the lack of data regarding this subgroup of molecularly selected patients in the RCTs available and the difficulty involved in performing them, and given the scant incidence of these alterations, with the retrospective data that do exist, it can be concluded that the incidence of VTE in ALK/ROS1-translocated tumours is much higher than for individuals with lung cancer without molecular drivers (approximately 40% vs 10%, respectively). The long-estimated survival of lung cancer patients with ALK or ROS1 translocation increases the cumulative risk of a thrombotic event, particularly during diagnosis and progression. Consequently, with the evidence available, it would be wise to appraise the possibility of primary thromboprophylaxis in these patients on a case-by-case basis (Fig. 1B) (strong recommendation, moderate-quality evidence) [
3. What is the optimal management for venous thromboembolic disease during treatment with antiangiogenics (tyrosine kinase inhibitors (TKI) or monoclonal antibodies (MABs?
The ASCO and SEOM clinical guidelines do not contemplate the use of TKI with antiangiogenic activity as a risk factor for VTE in people with cancer and propose stratifying recommendations for thromboprophylaxis based on the presence or absence of other risk factors [
]. As there is no evidence that supports the VTE management recommendations during treatment with MABs and TKI antiangiogenics, some of the specific guidelines indicate that intervention in this regard should follow the rules set forth in the prescribing information pertaining to the drug [
The following recommendations are made for the management of VTE in patients being treated with antiangiogenics (strong recommendation, moderate-quality evidence) based on the Common Terminology Criteria for Adverse Events [
In the case of grade 3 VTE (uncomplicated pulmonary embolism [venous], non-embolic cardiac mural [arterial] thrombus, medical intervention indicated, according to Common Terminology Criteria for Adverse Events, CTCAE) during active treatment with bevacizumab, MAB should be discontinued, and antithrombotic therapy initiated. Following stabilization, if there is no bleeding history associated with bevacizumab and the tumour does not invade large vessels, treatment with bevacizumab could be resumed. Shord et al. [
] add a minimum safety interval prior to initiating or reintroducing bevacizumab of two weeks since the beginning of antithrombotic therapy.
•
If the VTE is grade 4 by CTCAE (PE, cerebrovascular event, arterial insufficiency, or arterial thrombotic event); haemodynamic or neurologic instability; urgent intervention indicated), bevacizumab should be permanently suspended.
•
These same recommendations are applicable for ramucirumab.
•
If the patient is being treated with TKI.
•
If the VTE is grade 3, treatment should be withdrawn and the anticoagulation that is clinically indicated should be started. After two weeks and in the absence of haemorrhagic complications, the TKI could be safely resumed.
•
In VTE grade 4, the drug should be permanently discontinued.
•
VTE recurrence or progression in properly anticoagulated patients should result in the permanent discontinuation of antiangiogenic treatment with TKI.
•
Patients who develop an arterial thrombotic event while receiving an antiangiogenic TKI should suspend treatment.
The most important aspects to be considered are reflected in Fig. 2:
•
Prior to initiating TKI antiangiogenic therapy, predisposing cardiovascular risk factors (CVRF) (hypertension, hyperlipidaemia, and diabetes) must be controlled.
•
A prior history of arterial thrombotic event in the previous 6–12 months is not an absolute contraindication to the use of TKI antiangiogenics. Precaution and close supervision are recommended, given that these patients were excluded from RCTs with these agents, and we have no safety data.
•
Prophylaxis with low-dose aspirin is reasonable in individuals at high risk and in those with a prior arterial thrombotic event.
•
If there is a clear benefit to antiangiogenic therapy, it can be cautiously reintroduced in selected cases and if the patient is willing to accept the risk after being duly informed.
Fig. 2Algorithm for VTE management in patient on antiangiogenic treatment
4. Should primary prophylaxis be specifically addressed for individuals on cisplatin-based chemotherapy?
Chemotherapy is a known predisposing factor for the development of an arterial or venous thrombotic event. Platinum derivatives, especially cisplatin, have been correlated with an incidence of thrombosis of up to 20% [
]. There are numerous causes for the increased incidence of VTE in patients being treated with platins in addition to other known factors in patients with active cancer, including damage to the vascular endothelium, platelet and certain coagulation factor activation, and secondary renal dysfunction. The risk is lower for carboplatin and lower still for oxaliplatin [
Thromboembolism in patients with advanced gastroesophageal cancer treated with anthracycline, platinum, and fluoropyrimidine combination chemotherapy: a report from the UK national cancer research institute upper gastrointestinal clinical studies group.
]. These documents do not recommend primary thromboprophylaxis across the board in ambulatory patients on chemotherapy in general (strong recommendation, high-quality evidence), although it can be contemplated in special, high-risk circumstances, such as in patients with a Khorana score ≥ 3, those without active bleeding, or bleeding risk (strong recommendation, moderate-quality evidence).
As far as the treatment of an established thrombosis is concerned, patients being treated with cisplatin would be framed within the definition of active cancer, for which, in the absence of any major contraindication, anticoagulant therapy should be maintained beyond six months (weak recommendation, moderate-quality evidence).
Thus, we can conclude that, though cisplatin has been associated with a greater probability of thrombotic event, whether venous or arterial, in and of itself, it is not an independent factor in decision-making regarding primary thromboprophylaxis.
5. What factors determine continuing anticoagulant treatment beyond 6 months?
The international guidelines consulted coincide in that anticoagulant treatment with LMWH or DOACs beyond 6 months can be maintained in cases in which there is a high risk of recurrent VTE, those with active cancer, or with systemic treatment, taking into account bleeding risk and regularly evaluating the indication [
]. The ASH guideline suggests continuing anticoagulant therapy indefinitely versus a specific period of anticoagulation in patients with active cancer and VTE [
]. However, risk factors for recurrent VTE are not well-defined for the oncological population. Attempts have been made to validate the Ottawa score, with contradictory results; hence, it is not recommended [
Accuracy of the Ottawa score in risk stratification of recurrent venous thromboembolism in patients with cancer-associated venous thromboembolism. A systematic review and meta-analysis.
] is a prospective study with nadroparin that concludes that after 6 months of treatment, the absence of residual VTE identifies a group of patients at low risk for recurrence, for whom it would be safe to discontinue treatment. Its main limitation is that it only includes patients with symptomatic VTE and that the diagnosis of residual VTE may have interobserver biases. In accordance with the evidence available, maintaining anticoagulation after 6 months in patients with active cancer or systemic treatment is recommended (strong recommendation, low-quality evidence). In subjects with metastatic tumours with temporarily controlled disease (response to chemotherapy or surgery for metastases) in whom an aggressive clinical course is foreseen based on the known prognosis of the neoplasm, prolonging anticoagulant therapy can be justified on a case-by-case basis, as long as there is not an elevated bleeding risk (thrombopenia, renal impairment, treatment with antiplatelet agents...) (weak recommendation, low-quality evidence). The decision must be agreed upon with the patient and contingent on good treatment tolerance.
In short, anticoagulation maintenance beyond 6 months is recommended in patients with metastatic tumours, on active treatment, or at high risk for recurrent VTE (upper gastrointestinal tumours, lung cancer, prior extensive PE, residual VTE...) without bleeding risk and that tolerate anticoagulant treatment well (Fig. 3).
] contemplates dose reduction after one month of anticoagulation. The incidence of bleeding decreased over time, 3.6%, 1.1%, and 0.7% at months 1, 2–6, and 7–12, respectively. Similarly, the rate of VTE recurrence also declined, being 5.7%, 3.4%, and 4.1% at months 1, 2–6, and 7–12, respectively. The probability of recurrent VTE was 7.2% at 6 months and 9.3% at 12 months.
], the rate of bleeding risk was higher in the first 6 months compared to the following months (0.9 and 0.6%), as well as the incidence of VTE recurrence (4.5 and 1.1%). The limitations of both studies are that they are not RCTs and, due to cancer mortality, few patients completed 12 months of treatment, 32% in DALTECAN and 55% in TiCAT.
With DOACs, the HOKUSAI RCT confirms the safety of edoxaban and dalteparin for up to 12 months in patients with CAT [
]. Approximately half of the patients continued treatment for more than 6 months (56% of the edoxaban arm and 52% of the dalteparin arm) and beyond 12 months, 38% with edoxaban and 29% with dalteparin [
]. If we compare the first and second semester, the risk of VTE recurrence decreased with both edoxaban (4.4 and 0.7%) and dalteparin (6.7 and 1.1%), as well as the risk of major bleeding (edoxaban: 4.4 and 1.7%; dalteparin: 2.5 and 1.1%) and the risk of clinically significant bleeding (edoxaban: 14.4 and 3.7%; dalteparin: 9.4 and 3.7%). The SELECT-D study compared rivaroxaban with dalteparin for VTE in cancer over 6 months [
Comparison of an oral factor Xa Inhibitor with low molecular weight heparin in patients with cancer with venous thromboembolism: results of a randomized trial (SELECT-D).
Treatment of cancer-associated venous thromboembolism: 12-month outcomes of the placebo versus rivaroxaban randomization of the SELECT-D Trial (SELECT-D: 12m).
]. In the second 6 months of treatment, VTE recurrence was seen to be higher with placebo vs rivaroxaban (3.6 and 0.8%, respectively), albeit the rate of major and clinically significant bleeding was higher with rivaroxaban compared to placebo (4 and 0%; 5 and 0%, respectively).
Several retrospective studies confirm that anticoagulation maintained for more than 6 months exhibits a low risk of recurrence and bleeding, such as the study conducted at the Vancouver General Hospital [
Long-term treatment of cancer-associated thrombosis (CAT) beyond 6 months in the medical practice: USCAT, a 432-patient retrospective non-interventional study.
], 82.5% of the participants maintained VTE treatment for more than 6 months.
When choosing to continue to treat VTE beyond 6 months, we must consider a series of variables associated with a higher risk of recurrence, including: (1) the type of thrombotic event (extension, site, functional repercussions); (2) tumour characteristics (type of cancer, existence of an active tumour, systemic treatment); (3) patient characteristics (bleeding risk, comorbidity, clinical status, and preferences) [
] (weak recommendation, high-quality evidence). The ASH guideline suggests LMWH, as well as DOACs in individuals with active cancer and VTE who require long-term anticoagulation (> 6 months) [
Both DOACs and LMWH are considered safe long-term treatments. Nevertheless, a meta-analysis in patients with cancer has confirmed that DOACs are associated with a discreet increase in bleeding compared to LMWH (clinically relevant, non-major bleeding, 10.4 vs 6.4%), particularly in subjects with gastrointestinal cancer [
In the case of LMWH, doses could be decreased only in situations of low thrombotic risk and high bleeding risk, such as thrombocytopenia and renal impairment [
] (weak recommendation, high-quality evidence). As for DOACs, the same dose should be maintained (weak recommendation, high-quality evidence), albeit a dose reduction might be contemplated in certain situations, given the increased bleeding risk of these drugs (Fig. 3).
7. What is the optimal management of recurrent VTE in the anticoagulated cancer patient?
Recurrent VTE in this situation implies an increase or the appearance of a VTE in the presence of the recommended therapeutic dosage of anticoagulant treatment. The leading causes of recurrent VTE are the presence of active cancer, treatment at infratherapeutic doses, and anticoagulant treatment failure [
The recommendations can be summarized as follows (Fig. 4):
•
Verify the dosage of the anticoagulant treatment and treatment compliance, that the underlying cancer is not active, or that we are dealing with heparin-induced thrombocytopenia (strong recommendation, low-quality evidence).
•
In thromboprophylaxis or treatment with vitamin K agonists, change to LMWH at therapeutic dose or DOACs (strong recommendation, low-quality evidence).
•
If LMWH at therapeutic dose, contemplate increasing the dose by 20-25% (strong recommendation, low-quality evidence).
•
Switching from LMWH to DOAC or vice versa might also be considered in patients with a recurrence of venous thrombosis or progression of previous VTE in whom escalated the dose of LMWH is undesirable due to bleeding risk (weak recommendation, low-quality evidence). Still, faced with this situation, at present, the risk-benefit of the intervention is highly uncertain.
•
If we cannot resort to the above, evaluate inserting a retrievable, non-permanent vena cava filter, while maintaining the previous treatment and removing it once the cause of recurrent VTE is controlled (weak recommendation, low-quality evidence) [
The anticoagulant treatment should be maintained as long as the cancer is active, or the patient is on onco-specific therapy (strong recommendation, low-quality evidence).
Fig. 4Treatment algorithm for recurrence of thrombosis based on treatment phase Legends: Abbreviations: AVK, antivitamin K; LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant.
3.1 How should thrombosis be managed in a patient with a high bleeding risk?
We will group the recommendations into four situations that are common in oncological patients and associated with high bleeding risk: intracranial primary or metastatic tumours, thrombocytopenia, renal impairment, and cancers with a high bleeding risk.
In patients with brain cancer and thrombosis, anticoagulation with full-dose LMWH is recommended.
], individuals with intracranial tumours have a higher risk of thrombotic complications and intracranial haemorrhage (ICH), although an intracranial tumour is not an absolute contraindication for anticoagulation. The limited data suggest that therapeutic anticoagulation does not increase the risk of ICH in individuals with brain metastases, albeit it can increase risk among patients with primary cerebral tumours. Nevertheless, the lack of anticoagulation long term has been associated with a higher risk of recurrent VTE in people with glioblastoma and it is unclear whether antineoplastic therapy modifies the risk-benefit ratio. Most of the participants in the PRODIGE trial underwent surgery or radiotherapy and an excess of bleeding was detected in the dalteparin treatment arm [
PRODIGE: a randomized placebo-controlled trial of dalteparin low-molecular-weight heparin thromboprophylaxis in patients with newly diagnosed malignant glioma.
]. A recent meta-analysis concluded that therapeutic anticoagulation should be pondered in patients with brain tumours with thrombosis (weak recommendation, moderate-quality evidence) [
]. All of this points to the need to be painstaking when deciding the intensity of anticoagulation in patients with a primary cerebral tumour predicated on each individual case. Moreover, special attention should be paid to situations of high bleeding risk, such as hypervascular metastases of thyroid or renal cancers, melanoma, choriocarcinoma, hepatocarcinoma, or brainstem gliomas, in which the action is more controversial, with the possibility of lowering the anticoagulant dose by 25-50% (weak recommendation, moderate-quality evidence) [
Patients with > 50,000 who require anticoagulant therapy do not generally require dose titration, whereas anticoagulation is contraindicated when figures are < 25,000 (strong recommendation, moderate-quality evidence) [
]. In subject with intermediate platelet counts of 25,000–50,000 platelets, weighing the risk-benefit ratio requires that thrombotic risk be considered individually. Thus, in subjects with high-risk, acute-phase thrombosis (< 14 days), such as symptomatic thromboses (both proximal deep vein thrombosis (DVT), as well as PE of segmental or more proximal branches) or in those with a history of recurrent or progressive thrombosis, transfusion support is necessary to maintain a platelet count of > 50,000 that makes full-dose anticoagulant therapy possible (strong recommendation, weak evidence). When the risk of thrombosis is deemed minor (e.g., distal DVT, incidental subsegmental PE, or other lower risk features), the recommendation would be to lower the dose of LMWH by 25–50%. In the case of subacute/ chronic VTE, the dose could be adjusted to 50% [
] (Fig. 5A). Finally, given the elevated risk of complications, the use of a vena cava filter is reserved for acute VTE and active, life-threatening bleeding or severe, prolonged thrombopenia in which adequate anticoagulation is not achieved with transfusion support.
Fig. 5Management algorithm for patient with thrombocytopenia (5A) and kidney failure (5B)
Renal impairment associated with creatinine clearance (CrCl) < 50 ml/min is another pathology associated with high rates of thrombotic recurrence, bleeding complications, and morbi-mortality in cancer patients in general.
Given this increased bleeding risk, it is preferable to use LMWH that displays renal-dependent clearance but allows for dose titration that can be measured based on anti-Xa activity. LMWHs do not require dose titration for patients with CrCl > 30, whereas with CrCl = 15–30 dose-adjustments are recommended for some heparins, as well as withdrawing them altogether with CrCl < 15. With tinzaparin, no anti-Xa activity accumulation is observed up to CrCl < 20 ml/min. International guidelines do not recommend DOACs in renal impairment. Care should be exercised when prescribing anticoagulation in any patient with severe renal impairment and should be done on a case-by-case basis (weak recommendation, moderate-quality evidence) [
In situations with cancers associated with high bleeding risk, such as gastrointestinal and genitourinary tumours, cancers in unresected mucosae or in patients with nephrostomy, the current recommendation is LMWH, based on the SELECT-D [
Comparison of an oral factor Xa Inhibitor with low molecular weight heparin in patients with cancer with venous thromboembolism: results of a randomized trial (SELECT-D).
] RCTs with DOAC that showed high bleeding risk in some of these neoplasms with edoxaban and rivaroxaban, respectively (weak recommendation, moderate-quality evidence).
In patients with brain cancer and thrombosis, anticoagulation with full-dose LMWH is recommended, barring contraindication.
•
In hypervascular cerebral metastases of cancers of the thyroid, kidney, melanoma, choriocarcinoma, hepatocarcinoma, or in brainstem gliomas, action is more controversial given the high risk of complications; a 25–50% reduction if the anticoagulant dose can be used.
•
In individuals requiring therapeutic or prophylactic anticoagulation and who have platelet counts > 50,000/mm3, no dose adjustment is required, whereas in patients with < 25,000 platelets, anticoagulation would be contraindicated.
•
In cases of platelet counts of between 25,000 and 50,000 and VTE at high risk of recurrence, full-dose LMWH and transfusion support can be contemplated to reach values of > 50,000, as long as there are no contraindications. In VTE associated with a low risk of recurrence, the recommendation would be to decrease the LMWH dose by 25–50%.
•
Renal impairment associated with thrombosis must be treated with LMWHs that do not require dose titration in patients with CrCl > 30, whereas with CrCl = 15–30, dose adjustment is recommended in some, as well as discontinuation when CrCl is < 15.
•
In cancers having a high risk of bleeding, for instance, gastrointestinal and genitourinary tumours, cancers in unresected mucosae, or in patients with nephrostomies, the current recommendation is anticoagulant treatment with LMWH.
4. Conclusion
We provide useful advice, practical recommendations, and treatment algorithms for eight complex, clinically relevant situations that clinicians treating CAT must face posing great uncertainty regarding management and requiring an urgent decision
PJF reported honoraria for consulting/advisory role from Bristol, Pfizer, MSD, Mylan, Baxter; speaking from HRA Pharma, LeoPharma, Rovi, Sanofi; travel grants from IPSEN, all outside of the scope of this work. EG reported honoraria for consulting/advisory role from Sanofi, Janssen, Astellas, Bayer, Ipsen, Pfizer, Roche, Novartis, Eisai, EUSA Pharma, BMS, AstraZeneca, Merck, Rovi, Daiichi Sankyo, Techdow; speaking from Astellas, Janssen, Sanofi, Bayer, Ipsen, Pfizer, Roche, BMS, Novartis, Rovi, Daiichi Sankyo, Leo Pharma, Menarini, Eisai, MSD, Boehringer Ingelheim, Merck, EUSA Pharma; grant support from Astellas, Janssen, Sanofi, Bayer, Ipsen, Ferrer, Pfizer, Roche, GSK, BMS and travel/accommodation expenses from Astellas, Janssen, Sanofi, Bayer, Ipsen, Roche, Novartis, Pierre Fabre, Pfizer and Eisai; all outside of the scope of this work. AC reported honoraria for consulting/advisory role from Bristol, Pfizer, MSD oncology, LeoPharma, Kyowa Kirin, Hoffmann- L. Roche and Boehringer Inhelgeim; all outside of the scope of this work. MCR reported honoraria for speaking from Eisai, Merck, LeoPharma; travel grants from Lilly, Roche, PharmaMar, all outside of the scope of this work. JMM reported honoraria for speaking from Roche, LeoPharma, Rovi, BMS, all outside of the scope of this work. AR reported honoraria for consulting/advisory role from Novartis; speaking from LeoPharma, Sanofi, BMS, MSD, Roche, Pierre Fabre; all outside of the scope of this work. RSM reported honoraria for speaking from LeoPharma outside of the scope of this work. ACB reported travel grants from Ipsen, Roche and Novartis and honoraria for consulting/advisory role from Roche, Rovi, LEO Pharma, Pfizer, Esteve, all outside of the scope of this work. AM MCR reported honoraria for consulting, lectures and advisory board: Sanofi, Celgene, Astrazeneca, Roche, Leo Pharma, Servier, Pfizer, Bristol-Myers Squibb, Daiichi Sankyo, Bayer, Halozyme, Amgen, Rovi, Merck Sharp & Dohme and Lilly; research funding: Sanofi, LEO Pharma, Celgene; Travel, Accommodations: Roche, Merck Serono, Amgen, Celgene; all outside of the scope of this work. FAA, JMB, DCL, MVG, reported no disclosures.
Ethics approval and consent to participate
The current study has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. For this type of study, no formal consent is required.
Consent for publication
Not applicable.
Availability of data and materials
Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.
Funding
This work is not funded.
Acknowledgment
We thank Priscilla Chase Duran for editing the manuscript.
Authors' information
P.J.F., A.C.B., and A.M. are coordinators of the international registry of thrombosis and cancer TESEO (NCT03855592) and A.M. is the coordinator of the SEOM Thrombosis Group.
References
Key N.S.
Khorana A.A.
Kuderer N.M.
Bohlke K.
Lee A.Y.Y.
Arcelus J.I.
et al.
Venous thromboembolism prophylaxis and treatment in patients with cancer: ASCO clinical practice guideline update.
Rivaroxaban thromboprophylaxis in ambulatory patients with pancreatic cancer: Results from a pre-specified subgroup analysis of the randomized CASSINI study.
Primary thromboprophylaxis in ambulatory pancreatic cancer patients receiving chemotherapy: a systematic review and meta-analysis of randomized controlled trials.
ROS1-rearranged Non–small-cell lung cancer is associated with a high rate of venous thromboembolism: analysis from a phase II, prospective, multicenter, two-arms trial (METROS).
Thromboembolism in patients with advanced gastroesophageal cancer treated with anthracycline, platinum, and fluoropyrimidine combination chemotherapy: a report from the UK national cancer research institute upper gastrointestinal clinical studies group.
Accuracy of the Ottawa score in risk stratification of recurrent venous thromboembolism in patients with cancer-associated venous thromboembolism. A systematic review and meta-analysis.
Comparison of an oral factor Xa Inhibitor with low molecular weight heparin in patients with cancer with venous thromboembolism: results of a randomized trial (SELECT-D).
Treatment of cancer-associated venous thromboembolism: 12-month outcomes of the placebo versus rivaroxaban randomization of the SELECT-D Trial (SELECT-D: 12m).
Long-term treatment of cancer-associated thrombosis (CAT) beyond 6 months in the medical practice: USCAT, a 432-patient retrospective non-interventional study.
PRODIGE: a randomized placebo-controlled trial of dalteparin low-molecular-weight heparin thromboprophylaxis in patients with newly diagnosed malignant glioma.
Venous thromboembolism (VTE) is a common complication among patients with cancer. The 6-month risk of VTE complication for cancer patients is 12-fold higher and as much as 23-fold higher in those receiving chemotherapy or targeted therapy when compared to patients without cancer [1]. Furthermore, the 12-month cumulative incidence for VTE has increased three-fold over the past two decades in this patient population [1]. Hence, cancer-associated thrombosis (CAT) is a frequent complication that most clinicians caring for patients with cancer must routinely manage.
00085-1&id=gr1.jpg){kind=link}
00085-1&id=gr2.jpg){kind=link}
00085-1&id=gr3.jpg){kind=link}
00085-1&id=gr4.jpg){kind=link}
00085-1&id=gr5.jpg){kind=link}