Current status of treatment of cancer-associated venous thromboembolism

Patients with cancer are prone to develop venous thromboembolism (VTE) that is the second leading cause of mortality among them. Cancer patients with VTE may encounter higher rates of VTE recurrence and bleeding complications than patients without cancer. Treatment of established VTE is often complex in patients with cancer. Treatment of cancer-associated VTE basically comprises initial treatment, long-term treatment, treatment within 6 months, treatment beyond 6 months, treatment of recurrent VTE, and treatment in special situations. Decision of antithrombotic therapy, selection of anticoagulants, duration of anticoagulation, decision of adjuvant therapy, and adjustment of regimen in special situations are the major problems in the treatment of cancer-associated VTE. Therapeutic anticoagulation is the key of the key in the treatment of cancer-associated VTE. In addition to the efficacy and safety of low-molecular-weight heparin (LMWH) that has been fully demonstrated, direct oral anticoagulants (DOACs) are increasingly showing its advantages along with the accompanying concern in the treatment of cancer-associated VTE. The latest ASCO, ITAC and NCCN guidelines agree with each other on most aspects with respect to the treatment of cancer-associated VTE, whereas differ on a few issues. Encompassing recent randomized controlled trials, clinical trials, and meta-analyses, as well as the comparison of the latest authoritative guidelines including the NCCN, ASCO, and ITAC guidelines in this field, the objective of this review is to present current overview and recommendations for the treatment of cancer-associated VTE.


Introduction
Cancer-associated venous thromboembolism (VTE) is a common and life-threatening condition in adult patients (≥18 years) with cancer [1]. Patients with cancer are four to seven times more likely to develop VTE than patients without cancer [2]. The factors that are responsible for the increase of cancer-associated VTE incidence basically include cancer type and anticancer-associated treatment [2,3]. The statistical data showed that VTE was observed in 8.4% of nearly six million hospitalizations of 3,146,388 individual patients with cancer in the United States over a nearly two-decade period [4]. The incidence of cancer-associated VTE is still increasing worldwide [2].
Cancer-associated VTE is an important cause of morbidity for patients with cancer [5]. Patients with cancer may require critical care for the complication of their malignancy. Pulmonary embolism is one of the complications that can lead to the intensive care unit (ICU) admission for patients with cancer [6]. VTE is the second leading cause of mortality among patients with cancer after disease progression [2]. A large scale survey revealed that in-hospital mortality was observed in 5.5% of cancer patients without a VTE diagnosis, whereas in 15.0% of those with VTE, including 19.4% with a pulmonary embolism, in the United States [4].
Nevertheless, the treatment of established VTE in cancer patients is complicated [2]. Patients with cancer may encounter higher rates of VTE recurrence and bleeding complications than patients without cancer [5]. In addition to the efficacy and safety of LMWH that has been acknowledged, DOACs have demonstrated a favorable risk-benefit ratio and advantages over parenteral anticoagulants, meanwhile, they also pose challenges with respect to oral administration, drug-drug interactions, and bleeding risk [2]. Comprehensive treatment of cancer-associated VTE should be focused on the identification of patients who will mostly benefit from the treatment to reduce its recurrence and mortality [5]. Decision of antithrombotic therapy, selection of anticoagulants, duration of anticoagulation, decision of adjuvant therapy, and adjustment of regimen in special situations constitute the major problems in the treatment of cancer-associated VTE.

Overview
To date, the pharmacological anticoagulant agents most frequently used for cancer-associated VTE are LMWH and DOACs. In a real-world analysis, from September 2018 to January 2020, LMWH was the most common initial anticoagulation treatment for VTE in patients with cancer. Nevertheless, DOACs had a lower risk of recurrent venous thromboembolism compared with LMWH or warfarin [7]. In a retrospective cohort study, LMWH and UFH were the most common initial treatments (35.2 and 27.4%, respectively) of cancer-associated VTE, followed by DOACs (9.6%). Most of patients (71.4%) who received DOACs sticked to the postdischarge DOACs medication, whereas only 24.1, 43.5, and 0.1% of patients receiving LMWH, warfarin, and UFH remained on the same anticoagulant after discharge, respectively. DOACs were the most common initial post-discharge outpatient option. Persistence and adherence in outpatients appeared higher in patients using DOACs or warfarin versus LMWH or UFH [8]. In another retrospective cohort study, from 2000 through 2003, warfarin ± injectable was used in approximately 90% of cases. After 2003, there was a steady decline in warfarin use (90% in 2003 to 25% in 2017) corresponding with an increase in LMWH use (11% in 2003 to 55% in 2017). The regimen of DOACs ± injectable has rapidly increased from < 1% in 2014 to 20% in 2017 [9]. In two recent meta-analyses, recurrent VTE incidence was significantly decreased with DOACs compared to LMWH [10,11], with comparable [10] or increased [11] bleeding risks. VTE recurrence was also decreased with DOACs compared to VKA, with comparable [10] or decreased [11] bleeding risks. LMWH was associated with significantly reduced VTE recurrence compared with VKA [11].

Treatment of established Cancer-associated VTE within 6 months
Treatment of established cancer-associated VTE within 6 months consists of initial treatment (first 5-10 days) and early maintenance (up to 6 months) in the International Initiative on Thrombosis and Cancer (ITAC) guidelines [2], whereas of initial treatment (first 5-10 days) and long-term treatment (up to 6 months) in the American Society of Clinical Oncology (ASCO) guidelines [5]. Initial treatment of cancerassociated VTE is often defined as acute treatment especially the first 5 to 10 days of anticoagulant treatment [1,2,5,12]. In the 2018 Cochrane database of systematic reviews with respect to the initial treatment of cancer-associated VTE, LMWH was possibly superior to adjusted-dose UFH [12].
With respect to the frequency of usage of LMWH, in a retrospective study, clinically relevant bleeding (CRB) was higher in the twice-daily enoxaparin group compared with the once-daily group at 30 days (5.3% vs 2.4%, P = 0.587). The composite outcome of CRB, VTE, and mortality rates was higher in the once-daily enoxaparin group than twice-daily group at all time points [13].
Besides pharmacological anticoagulant agents, mechanotherapy may play a role in the case of contraindications to pharmacological anticoagulation. With respect to IVCF, a cohort study demonstrated that among 33, 740 cancer patients with acute lower extremity DVT and bleeding risk factors who underwent IVCF placement, 4492 patients (5.1%) developed a new PE after the initial DVT diagnosis. A significant improvement was observed in PE-free survival for patients who received IVCF compared with those who did not [17]. In another cohort study, a significant lower risk of PE-related mortality (0.8% vs 4.0%; p = 0.04) were observed in patients receiving IVCF due to contraindication against anticoagulation compared with those who did not, without significant between-group difference regarding the major bleeding rate (6.1% vs 5.7%; p = 0.85). However, recurrent VTE rates were higher in patients who received IVCF compared with those who did not (7.3% vs 3.2%; p = 0.05) [18].

Treatment of established Cancer-associated VTE beyond 6 months
Treatment of established cancer-associated VTE beyond 6 months denotes long-term treatment in the ITAC guidelines [2], whereas treatment after 6 months in the ASCO guidelines [5]. The anticoagulation beyond 6 months often depends on a case-by-case basis. The 2018 Cochrane database of systematic reviews with respect to the long-term treatment of cancer-associated VTE showed that LMWH probably resulted in an important reduction in VTE compared with VKA. DOACs may likely reduce VTE compared with LMWH, but may increase major bleeding risk [20].
A retrospective study reviewed 322 cancer patients from 6 months through 24 months after the diagnosis of cancer-associated VTE. Anticoagulation was continued in 222 patients (68.9%) after 6 months' anticoagulant treatment. Patients who continued anticoagulation beyond 6 months mostly had advanced cancer or anticancer therapy compared with those who discontinued anticoagulation [24]. In a meta-analysis, long-term treatment of DOACs were associated with a significant decrease of recurrent VTE and similar bleeding risk compared to long-term LMWH and VKA [10].

Treatment of catheter-related VTE in patients with Cancer
The evidence regarding the treatment of catheter-related VTE in cancer patients are scarce. A prospective study assessed the efficacy and safety of rivaroxaban in 70 cancer patients with upper extremity deep vein thrombosis due to CVC (CVC-UEDVT). After 12 weeks of followup, the preservation of CVC function was 100%. The recurrent VTE event was observed in one episode of fatal PE (1.43%) at 12 weeks. Nine patients (12.9%) experienced a total of eleven bleeding episodes [25].

Treatment of recurrent Cancer-associated VTE
Poor adherence, temporary cessation, inadequate dosage, cancer progression, and heparin-induced thrombocytopenia (HIT) can lead to recurrent VTE in cancer patients [26]. A study investigated the management of recurrent VTE in 212 cancer patients with breakthrough VTE which was defined as the first objectively verified VTE event after the initiation of anticoagulant therapy. The anticoagulation regimen changes after the breakthrough VTE were: dose unchanged in 33%, dose increased in 31%, switched to another agent in 24%, and other management in 11%. After 3 months of follow-up, additional VTE recurrence was less common with LMWH than with VKA (HR, 0.28; 95% CI, 0.11-0.70) [27].

Treatment of Cancer-associated VTE in special situations
Several special clinical situations deserve extra attention in the treatment of cancer-associated VTE. Although advanced age is a risk factor for bleeding, anticoagulation should be offered to elderly patients without contraindications [5]. However, for cancer patients with intracranial malignancy, thrombocytopenia, renal impairment, obesity and pregnancy, anticoagulation treatment should be adjusted as per the specific situation [2,5].

Intracranial malignancy
Patients with intracranial malignancy are at increased risk of thrombotic complications and intracranial hemorrhage (ICH) simultaneously. A recent metaanalysis reviewed seven studies to determine the difference of ICH risk between glioma patients who underwent anticoagulation for established VTE and those who did not. The OR for ICH in glioma patients with VTE who received therapeutic anticoagulation compared with those who did not was 3.66 (95% CI, 1.84-7.29; I 2 = 31%) [28]. However, the presence of primary or metastatic intracranial malignancy being active or not is not an absolute contraindication to anticoagulation [5]. A retrospective comparative cohort study compared the cumulative incidence of ICH between the use of DOACs and LMWH for 12 months in 172 patients with brain tumors and VTE. In the primary brain tumor cohort (n = 67), the cumulative incidence of any ICH was 0% in the DOACs group and 36.8% (95% CI, 22.3-51.3%) in the LMWH group, respectively. In the brain metastases cohort (n = 105), the incidence of major ICH was11.1% (95% CI, 0.5-40.6%) vs 17.8% (95% CI,10.2-27.2%) in DOACs and LMWH groups, respectively [29].

Thrombocytopenia
Thrombocytopenia does not reduce the risk of VTE. It is not uncommon for thrombocytopenic patients with cancer to have an indication for anticoagulant therapy. Since bleeding risk due to anticoagulation increase when platelets are less than 50 × 10 9 /L, management options may include no change, temporarily withholding anticoagulant treatment, reducing dosage, switching to other regimens, and platelet transfusion in this situation [30]. In a retrospective study, the conventional dose of dalteparin 200 units/kg once daily was adopted for patients with platelets more than 50 × 10 9 /L. For patients with thrombocytopenia (platelets of 25-50 × 10 9 /L) (n = 75), the dosage of dalteparin was reduced to 100 units/kg once daily. For patients with platelets less than 25 × 10 9 /L (n = 77), dalteparin was withheld unless platelets recovered to exceed the 25 × 10 9 /L threshold. During a two-year observation, new or recurrent VTE was documented in 2.6, 0, and 2.2% of patients with platelet counts of< 25, 25-50, and > 50 × 10 9 /L, respectively. (p > 0.9 for all comparisons). Acute blood loss or major bleeding events were documented in 10.5, 12.5, and 15.6% of patients with platelet counts of< 25, 25-50, and > 50 × 10 9 /L, respectively (p > 0.9 for all comparisons) [31]. In the RIETE study, 166 (1.1%) had platelets less than 50 × 10 9 /L (severe thrombocytopenia), 711 (4.6%) had platelets of 50-99 × 10 9 /L (mild thrombocytopenia) and 14,460 (94.3%) had platelets more than 100 × 10 9 /L (normal count). Most patients received LMWH at conventional dose for initial therapy, whereas 62% of those with severe thrombocytopenia received LMWH less than150 IU/kg/day, and 42% of them received LMWH less than 100 IU/kg/day. The mortality rate progressively decreased along with the increase of platelet counts (12, 9.4 and 3.3% at 10 days, 27, 18 and 9.4% at 30 days, respectively), whereas the major bleeding rates did not differ among three groups (1.2, 2.5 and 1.3% at 10 days, 2.4, 4.4 and 2.2% at 30 days, respectively) [32].

Obesity
Anticoagulation dose has been understudied in large or obese patients. In a retrospective study with 7602 cancer patients receiving thromboprophylaxis, body mass index (BMI) was found to be a significant predictor of VTE (OR = 1.094, 95% CI 1.021-1.172, p = 0.011). Cancer patients with high BMI may pose a risk of breaking through standard thromboprophylaxis dosage [35].

Pregnancy
Pregnancy is another special situation in the treatment of cancer-associated VTE. The evidence regarding this situation is also scarce. A prospective study compared the maternal morbidity and mortality between 29 pregnant cancer patients with high-risk of VTE receiving parenteral thromboprophylaxis for 3 months and 23 ones with low-risk of VTE not receiving thromboprophylaxis. Among 29 patients receiving thromboprophylaxis, no one exhibited VTE, adverse effects of thromboprophylaxis or death after 3 months of thromboprophylaxis [36].

Comparison among different therapeutic anticoagulant agents
There are a large number of studies with respect to the comparison of efficacy and safety between different frequently-used anticoagulant agents for established cancer-associated VTE. DOACs vs LMWH is the most frequent comparison, followed by DOACs vs VKA and LMWH vs VKA.

DOACs vs LMWH
In recent years, with respect to DOACs vs LMWH, besides SELECT-D trial [37], six meta-analyses indicated that DOACs were more efficacious in the treatment of cancer-associated VTE with an increase of bleeding risk, compared with LMWH [11,20,[38][39][40][41], whereas a retrospective study [7] and another five meta-analyses [10,15,16,42,43] indicated that DOACs were more efficacious with a comparable bleeding risk, compared with LMWH. In the ADAM VTE trial, major bleeding and VTE recurrence rates were both lower in cancer patients with VTE receiving oral apixaban than those receiving dalteparin [44]. The Caravaggio trial (apixaban) [14], Hokusai VTE Cancer trial (edoxaban) within 12 months [21] and over 12 months [22] reported that DOACs were noninferior to LMWH (dalteparin) with similar bleeding risk in the treatment of cancer-associated VTE, whereas two meta-analyses [45,46] showed that DOACs were noninferior to LMWH with an increase of bleeding risk. In addition, the AMPLIFY trial indicated that DOAC (apixaban) was more efficacious than LMWH (enoxaparin) followed by VKA (warfarin) with less bleeding risk [47].

DOACs vs VKA
With regard to DOACs vs VKA in recent years, one retrospective study [7] and two meta-analyses [10,42] reported that DOACs were more efficacious in the treatment of cancer-associated VTE with similar bleeding risk compared to VKA, whereas another two metaanalyses [11,38] indicated that DOACs were more efficacious with a decrease of bleeding risk, compared to VKA. Besides, the Hokusai-VTE trial reported that DOAC (edoxaban) was as effective as VKA (warfarin) for the treatment of patients with cancer-associated VTE with less bleeding risk [48].

LMWH vs VKA
As for LMWH vs VKA, two meta-analyses [11,20] and one retrospective study [34] reported that LMWH was more efficacious in the treatment of cancer-associated VTE with similar bleeding risk compared with VKA, whereas the CATCH trial [49] and one meta-analysis indicated that the therapeutic efficacy for cancerassociated VTE and bleeding risk were both similar between LMWH and VKA [38].
Anyway, no significant difference was found in allcause mortality among cancer patients with established VTE treated with DOACs, LMWH and VKA to date [1,2,5]. Although the preference for oral anticoagulants over injection is moderately important, most patients do not think it is difficult, painful, or inconvenient to use LMWH regimen [1]. So far, LMWH is still the first choice for the treatment especially the initial treatment of cancer-associated VTE [1,2,5]. Dalteparin, edoxaban, and rivaroxaban have the highest priority in the selection of anticoagulants [1,5]. The comparison among different randomized clinical trials regarding DOACs vs LMWH in the treatment of cancer-associated VTE is in Table 1.

Guidelines
According to the latest guidelines, anticoagulation options recommended for treatment of cancerassociated VTE include one-agent regimens (e.g. LMWH or DOACs) (monotherapy) and regimens of more than one agent (e.g. LMWH+DOACs) (combination therapy) [1]. LMWH, UFH, fondaparinux, rivaroxaban or apixaban can be used for the initial 5 to 10 days of anticoagulation in cancer patients with acute diagnosed VTE in the absence of severe renal impairment. LMWH is preferred over others [1,2,5]. With respect to the usage frequency of LMWH, once daily is recommended unless a twicedaily regimen is necessary in the ITAC guidelines [2], whereas twice-daily dosing is considered to be more efficacious than once-daily dosing in the ASCO guidelines [5]. DOACs can be used in cancer Table 1 Comparison among different randomized clinical trials assessing the efficacy and safety of DOACs in the treatment of cancer-associated VTE SELECT-D [37] Hokusai VTE-Cancer [22] ADAM-VTE [44] Caravaggio [14] Trial design patients without severe renal impairment, gastrointestinal or genitourinary malignancy or bleeding risk, previous small bowel surgery, or strong drugdrug interactions [1,2,5]. Thrombolysis may only be recommended to cancer patients with VTE without contraindications on a case-by-case basis, by clinicians experienced with parenteral or CDT [2], whereas the insertion of IVCF is usually offered to cancer patients with life-threatening VTE and absolute or relative contraindications to anticoagulant therapy, or recurrent VTE despite optimal anticoagulation [2,5]. Incidental VTE should be treated in the same manner as symptomatic VTE, whereas the treatment of subsegmental PE or SPVT should be discussed on a case-by-case basis [5]. Unless there is contraindication, LMWH or DOACs should be used for a minimum of 6 months for established acute cancer-associated VTE [2,5]. Anticoagulation beyond 6 months can be offered to patients with active cancer [5]. Termination or continuation of anticoagulation after 6 months' treatment of cancer-associated VTE should be based on an intermittent individual assessment of the benefit-risk ratio, tolerability, patient preference, and cancer activity [2,5].
For the treatment of symptomatic acute catheterrelated VTE in cancer patients, anticoagulant treatment is recommended for a minimum of 3 months. LMWH is recommended over other anticoagulants in this setting. As long as the CVC is in place, anticoagulation is recommended in cancer patients with established catheterrelated VTE [2].
For patients with recurrence of cancer-associated VTE after standard doses of anticoagulant therapy, treatment adherence, HIT, or mechanical compression due to malignancy progression should be reassessed [5]. Switching to an alternative anticoagulant regimen or increasing the dose of LMWH usually are the optional management [2,5]. Three options may be considered: (1) for LMWH, increase the dose by 20-25% or switch to DOACs; (2) for DOACs, switch to LMWH; and (3) for VKA, switch to LMWH or DOACs [2]. An IVCF in addition to LMWH can be reserved for recurrent cancer-associated VTE as a last resort [2,5]. LMWH or DOACs should be offered for patients with established VTE and primary or metastatic intracranial malignancy [2,5]. For cancer patients with platelet count more than 50 × 10 9 /L, anticoagulation can be used for the treatment of established VTE without the evidence of bleeding. For patients with platelet count below 50 × 10 9 /L, treatment decisions as well as dosage should be deliberated with the extreme caution on a case-by-case basis, whereas platelet count below 20 × 10 9 /L is an absolute contraindication to anticoagulation [2,5]. For cancer patients with severe renal failure, UFH followed by VKA or LMWH adjusted to anti-Xa level are recommended in this situation [2,5]. In addition, an external compression device can be applied [2]. For cancer patients with a BMI greater than 40 kg/m 2 or a weight greater than 120 kg, LMWH is likely preferred over DOACs. If DOACs are used, the monitoring of drug-specific peak and trough levels are advised [5]. For cancer patients with pregnancy, LMWH is recommended for the treatment of established VTE. VKA and DOACs should be avoided in this setting [2].
The comparison of recommendations regarding treatment of established cancer-associated VTE between the latest ASCO [5] and ITAC guidelines [2] are in Table 2. The comparison of contraindications to therapeutic anticoagulant therapy of cancerassociated VTE between the latest National Comprehensive Cancer Network (NCCN) [1] and ASCO [5] guidelines are in Table 3. The comparison of recommendations regarding anticoagulation of cancerassociated VTE in special situations between the latest ASCO [5] and ITAC [2] guidelines are in Table 4. Based on the latest guidelines [1,2,5], the currently recommended treatment procedure of cancerassociated VTE is demonstrated in Fig. 1.

Treatment of Cancer-associated VTE with COVID-19
Unfortunately, we are now living in the era of COVID-19. Thrombosis is one of the multiple potential mechanisms for cardiac injury due to COVID-19 [50], putting cancer patients who are already at a high risk of VTE at higher VTE risk. Nevertheless, in a retrospective cohort analyses of 398 consecutive patients hospitalized with COVID-19, the cumulative incidence of thrombotic events was 18.2% (95% CI, 10.2 to 27.9%) in non-cancer cohort (n = 353) and 14.2% (95% CI, 4.7 to 28.7%) in the active cancer cohort (n = 45) at day 28. The cumulative incidence of major and fatal bleeding was 20.8% (95% CI, 12.1 to 31.0%) in the noncancer group and 19.5% (95% CI, 5.5 to 39.8%) in the cancer cohort at day 28 [51]. Therefore, applying the current guidelines to cancer patients with COVID-19 is challenging due to their high risk for both thrombosis and bleeding. To date, at a guidance level, parenteral anticoagulation (e.g. LMWH) is preferred over oral anticoagulation in the treatment of established VTE in cancer patients with COVID-19 [52]. Table 2 Comparison of recommendations regarding treatment of established cancer-associated VTE between the latest ASCO [5] and ITAC [2] guidelines ASCO ITAC Duration Initial treatment (typically first 5-10 days) + long-term treatment (6 months) + treatment after 6 months Initial treatment (typically first 5-10 days) + early maintenance (6 months) + long-term treatment (

Conclusions
Cancer-associated venous thromboembolism is a concerning issue that increases both morbidity and mortality for patients with cancer. For established cancer-associated VTE including the incidental or asymptomatic one, the decision of treatment should be made based on the risk-benefit ratio, whereas the selection among anticoagulants should be made based on anticoagulant efficacy, bleeding risk assessment, renal or hepatic function, drug-drug interactions, clinical setting, convenience of use, monitoring, FDA approval, cost, drug availability and patient preference. In addition to LMWH, DOACs have shown a predominant role in the treatment of cancer-associated VTE, despite its adverse effects. To date, LMWH and DOACs are most highly recommended anticoagulants for the treatment of cancer-associated VTE. A minimum of 6 months of treatment should be offered to patients with cancerassociated VTE, whereas the continuation or discontinuation of treatment should be made on a case-by-case basis with intermittent assessment of risk-benefit ratio of anticoagulation. For recurrent cancer-associated VTE,  Thrombocytopenia Anticoagulation is absolutely contraindicated when platelet count is persistently below 20 × 10 9 /L, and relatively contraindicated when platelet count is persistently below 50 × 10 9 /L For established VTE, full doses of anticoagulant can be used when platelet count is>50 × 10 9 /L and should be deliberated case-by-case when platelet count is≤50 × 10 9 /L; prophylactic anticoagulation can be used when platelet count is>80 × 10 9 /L Renal impairment For moderate to severe renal impairment, LMWH adjusted to anti-Xa level or UFH followed by VKA are recommended the original anticoagulant regimen should be replaced or a retrievable IVCF can be applied. However, IVCF may only be regarded as a last resort in the case of absolute contraindications to anticoagulation or recurrence of cancer-associated VTE. The future direction of treatment of cancer-associated VTE should be focused on how to reduce its recurrence as well as all-cause mortality without bringing much risk, adverse events or discomfort to patients.