Risk Factors for Major Bleeding During Anticoagulation Therapy in Cancer-Associated Venous Thromboembolism　― From the COMMAND VTE Registry ―

Abstract

Background: Patients with cancer-associated venous thromboembolism (VTE) are at high risk for recurrent VTE and are recommended to receive prolonged anticoagulation therapy if they are at a low risk for bleeding. However, there are no established risk factors for bleeding during anticoagulation therapy.

Methods and Results: The COMMAND VTE Registry is a multicenter retrospective registry enrolling 3,027 consecutive patients with acute symptomatic VTE among 29 Japanese centers. The present study population consisted of 592 cancer-associated VTE patients with anticoagulation therapy. We constructed a multivariable Cox proportional hazard model to estimate the hazard ratio (HR) and 95% confidence interval (CI) of the potential risk factors for major bleeding. During a median follow-up period of 199 days, major bleeding occurred in 72 patients. The cumulative incidence of major bleeding was 5.8% at 3 months, 13.8% at 1 year, 17.5% at 2 years, and 28.1% at 5 years. The most frequent major bleeding site was gastrointestinal tract (47%). Terminal cancer (adjusted HR, 4.17; 95% CI, 2.22–7.85, P<0.001), chronic kidney disease (adjusted HR, 1.89; 95% CI 1.06–3.37, P=0.031), and gastrointestinal cancer (adjusted HR, 1.78; 95% CI, 1.04–3.04, P=0.037) were independently associated with an increased risk of major bleeding.

Conclusions: Major bleeding events were common during anticoagulation therapy in real-world cancer-associated VTE patients. Terminal cancer, chronic kidney disease, and gastrointestinal cancer were the independent risk factors for major bleeding.

Cancer is a strong risk factor for the development of venous thromboembolism (VTE).¹ Patients with cancer-associated VTE have been reported to have a markedly higher risk of recurrence as well as of bleeding, compared with those without active cancer, leading to difficulty in achieving a good risk-benefit balance with anticoagulation therapy.² Optimal management strategies for patients with cancer-associated VTE have become an issue of major concern that challenges clinicians in daily clinical practice. The current VTE guidelines recommend prolonged anticoagulation therapy for patients with active cancer, if they are not at a high risk for bleeding.^3,4 Thus, identification of high-risk patients for bleeding during anticoagulation therapy is clinically relevant in determining the optimal duration of anticoagulation therapy.

Previous studies have reported several useful risk scores, such as the RIETE score⁵ and VTE-BLEED score,⁶ to predict the risk of major bleeding in VTE patients during anticoagulation therapy. However, these risk scores include active cancer as a score component. Thus, all patients with active cancer are classified as a high-bleeding-risk group, which means that these scores might not be applicable for risk stratification of patients with active cancer. There is a paucity of data on risk factors for bleeding specific to patients with cancer-associated VTE. Furthermore, risk factors for bleeding beyond the acute phase could be different from those in the acute phase, because treatment strategies other than anticoagulation therapy, including thrombolysis, could have a great influence on the risk for bleeding in the acute phase. Thus, we sought to identify the risk factors for major bleeding during anticoagulation therapy in patients with cancer-associated VTE using a large observational real-world database in Japan.

Methods

Study Population

The COMMAND VTE (COntemporary ManageMent AND outcomes in patients with Venous ThromboEmbolism) Registry is a physician-initiated, multicenter retrospective cohort study that enrolled consecutive patients with acute symptomatic VTE that was objectively confirmed by imaging examinations or autopsy among 29 centers in Japan between January 2010 and August 2014. The design of the registry has been reported in detail.^7–9 The relevant review board or ethics committee in all 29 participating centers approved the research protocol (Supplementary Appendix 1). This study followed the Declaration of Helsinki and the ethical standards of the responsible committees on human experimentation. It was registered with UMIN (UMIN identifier: UMIN000021132).

We enrolled 3,027 consecutive patients with acute symptomatic VTE after screening 19,634 consecutive patients with suspected VTE for eligibility at each institution. There were 695 patients with active cancer after excluding those without active cancer (n=2,332) (Figure 1). We also excluded patients with major bleeding (n=15), death (n=17), and lost to follow-up (n=10) within 10 days after the diagnosis, and those without anticoagulation therapy beyond 10 days after the diagnosis (n=61). Finally, the present study population consisted of 592 patients with anticoagulation therapy beyond the acute phase.

Figure 1.

Study flow chart. VTE included both PE and/or DVT. DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.

Data Collection and Definition of the Patients’ Characteristics

Data for the patients’ characteristics were collected from hospital charts or hospital databases according to the prespecified definitions, using an electronic case report form in a web-based database system. The physicians at each institution were responsible for data entry, and data were automatically checked for any missing or contradictory input or values out of the expected range. Additional editing checks were performed at the general office of the registry.

Patients with active cancer were defined as those on treatment for cancer such as chemotherapy or radiotherapy, those scheduled to undergo cancer-surgery, those with metastasis to other organs, and/or those with terminal cancer (expected life expectancy ≤6 months) at the time of the diagnosis of VTE. Patients with advanced cancer were defined as those with metastasis to other organs and/or those with terminal cancer at the time of the diagnosis. Upper gastrointestinal (GI) cancer was defined as esophageal, gastric, or duodenal cancer, and lower GI cancer was defined as colon cancer. A history of major bleeding was diagnosed if the patient had a history of International Society of Thrombosis and Hemostasis (ISTH) major bleeding, which consisted of a reduction in the hemoglobin (Hb) level by at least 2 g/dL, transfusion of at least 2 units of blood, or symptomatic bleeding in a critical area or organ.¹⁰ Chronic kidney disease (CKD) was diagnosed if there was persistent proteinuria or if the estimated glomerular filtration rate (eGFR) was <60 mL/min/1.73 m² for >3 months. The eGFR was calculated by the equation reported by the Japan Association of Chronic Kidney Disease Initiative [man: 194×Scr^−1.094×age^−0.287, woman: 194×Scr^−1.094×age^−0.287×0.739]. Anemia was diagnosed if Hb <13 g/dL for men and <12 g/dL for women. Thrombocytopenia was diagnosed if platelet level was <100×10⁹/L. Detailed definitions of other patients’ characteristics are given in Supplementary Appendix 2.

Clinical Follow-up and Endpoints

Collection of the follow-up information was mainly conducted through a review of the hospital charts according to the prespecified definitions, and additional follow-up information was collected through contact with the patients, relatives, and/or referring physicians by phone and/or mailed questions. The median follow-up period was 199 (interquartile range [IQR], 77–627) days.

The outcome measurement in the present study was ISTH major bleeding during anticoagulation therapy beyond 10 days after the diagnosis, which occurred before the first discontinuation of anticoagulation therapy. Discontinuation of anticoagulation was defined as withdrawal of anticoagulation therapy lasting >14 days for any reason, such as bleeding events, drug side effects, physician’s judgment in the absence of adverse events, and non-adherence by the patient. The independent clinical events committee (Supplementary Appendix 3), which was unaware of the patients’ characteristics, reviewed the study outcomes, and classified the causes of deaths as due to a pulmonary embolism (PE), cardiac event, cancer, bleeding event, other non-cardiac event, or unknown cause.¹¹ Death was judged to be bleeding related (fatal bleeding) if it followed an intracranial hemorrhage or bleeding episode leading to hemodynamic deterioration. Final classification of the causes of death was made on the basis of the full consensus of the independent clinical events committee. The time in a therapeutic range (TTR) was calculated by the Rosendaal method,¹² according to a therapeutic international normalized ratio (INR) range of 1.5–2.5, which is recommended in the Japanese guidelines.¹³

Statistical Analysis

We present categorical variables as numbers and percentages and continuous variables as the mean with standard deviation or the median with the IQR based on their distribution. Categorical variables were compared using the chi-squared test when appropriate; otherwise, Fisher’s exact test was used. Continuous variables were compared using Student’s t-test or Wilcoxon’s rank sum test based on their distribution. We used the Kaplan-Meier method to estimate the cumulative incidence. We constructed a multivariable Cox proportional hazard model to estimate the hazard ratio (HR) and the 95% confidence interval (CI) of the potential risk factors for major bleeding during anticoagulation therapy. Patients were censored at the time of first discontinuation of anticoagulation therapy. Consistent with previous reports,^4,14–21 and with consideration of the clinical relevance, we selected 7 clinically relevant variables (age, history of major bleeding, CKD, anemia, thrombocytopenia, terminal cancer, and GI cancer) as indicated in Table 1 and Table 2. We also constructed a multivariable Cox proportional hazard model including separate variables of upper and lower GI cancer instead of GI cancer. Furthermore, to explore differences in risk factors between the short term and long term, we described the differences in patients’ characteristics between those with major bleeding within 3 months and those with bleeding beyond 3 months. In addition, to explore the drug–drug interaction between warfarin and 5-fluorouracil (5-FU)-related drugs, which consisted of fluorouracil, capecitabine, tegafur, tegafur-uracil, tegafur-gimeracil-oteracil potassium, and doxifluridine, we estimated the cumulative incidence of major bleeding for patients with and without 5-FU-related drugs among warfarin users, and assessed the difference with a log-rank test. Given the relatively short term follow-up period in the present study population, we constructed an additional multivariable Cox proportional hazard model as a sensitivity analysis to estimate the HR and 95% CI of the potential risk factors for major bleeding at 2 years. All statistical analyses were performed with EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, version 3.5.0),²² or JMP version 14.0.0 (SAS Institute Inc., Cary, NC, USA). All statistical analyses were two-tailed and P<0.05 was considered statistically significant.

Table 1. Patients’ Characteristics

	Total (n=592)	Major bleeding (n=72)	No major bleeding (n=520)	P value
Baseline characteristics
Age (years)*	66.3±12.1	65.2±16.1	66.5±11.5	0.42
≥75 years	147 (25%)	25 (35%)	122 (24%)	0.054
Women	349 (59%)	38 (53%)	311 (60%)	0.31
Body weight (kg)	55.9±12.2	54.9±11.6	56.1±12.3	0.45
Body mass index (kg/m2)	22.3±4.0	22.1±4.3	22.3±4.0	0.66
≤18.5 kg/m2	100 (17%)	14 (19%)	86 (17%)	0.65
Comorbidities
Hypertension	177 (30%)	23 (32%)	154 (30%)	0.79
Diabetes mellitus	78 (13%)	11 (15%)	67 (13%)	0.71
Dyslipidemia	84 (14%)	8 (11%)	76 (15%)	0.54
CKD*	89 (15%)	16 (22%)	73 (14%)	0.10
Dialysis	3 (0.5%)	1 (1.4%)	2 (0.4%)	0.32
Liver cirrhosis	6 (1.0%)	0 (0%)	6 (1.2%)	1.00
Chronic lung disease	48 (8.1%)	7 (9.7%)	41 (7.9%)	0.76
Heart failure	8 (1.4%)	0 (0%)	8 (1.5%)	0.60
History of myocardial infarction	5 (0.8%)	1 (1.4%)	4 (0.8%)	0.48
History of stroke	29 (4.9%)	4 (5.6%)	25 (4.8%)	0.77
Atrial fibrillation	22 (3.7%)	2 (2.8%)	20 (3.8%)	1.00
Connective tissue disease	18 (3.0%)	1 (1.4%)	17 (3.3%)	0.71
Varicose vein	13 (2.2%)	1 (1.4%)	12 (2.3%)	1.00
History of VTE	33 (5.6%)	6 (8.3%)	27 (5.2%)	0.27
History of major bleeding*	46 (7.8%)	8 (11%)	38 (7.3%)	0.37
Transient risk factors for VTE	161 (27%)	15 (21%)	146 (28%)	0.25
Cancer status
Scheduled to be operated	65 (11%)	10 (14%)	55 (11%)	0.52
Under chemotherapy	338 (57%)	37 (51%)	301 (58%)	0.36
Distant metastasis	180 (30%)	22 (31%)	158 (30%)	1.00
Terminal stage*	86 (15%)	14 (19%)	72 (14%)	0.28
Presentation
PE with or without DVT	337 (57%)	43 (60%)	294 (57%)	0.84
DVT only	255 (43%)	29 (40%)	226 (44%)
Proximal DVT	186/255 (73%)	26/29 (90%)	160/226 (71%)	0.043
Laboratory tests when diagnosed
Anemia*	439 (74%)	58 (81%)	381 (73%)	0.24
Thrombocytopenia*	46 (7.8%)	6 (8.3%)	40 (7.7%)	1.00
D-dimer (μg/mL) (n=557)	13.4 (6.2–25.2)	14.7 (6.4–29.6)	13.3 (6.2–25.0)	0.65
Thrombophilia	18 (3.0%)	2 (2.8%)	16 (3.1%)	1.00
eGFR (mL/min/1.73 m2) (n=590)	70.9 (55.5–86.3)	72.6 (51.8–83.1)	70.5 (56.0–86.6)	0.58
Treatment in the acute phase
Initial parenteral anticoagulation therapy	513 (87%)	63 (88%)	450 (87%)	0.99
Thrombolysis	47 (7.9%)	3 (4.2%)	44 (8.5%)	0.25
Inferior vena cava filter use	164 (28%)	24 (33%)	140 (27%)	0.32
Ventilator support	5 (0.8%)	0 (0%)	5 (1.0%)	1.00
Percutaneous cardiopulmonary support	1 (0.2%)	0 (0%)	1 (0.2%)	1.00
Treatment beyond the acute phase
Direct oral anticoagulant	19 (3.2%)	1 (1.4%)	18 (3.5%)	0.021
Heparin	18 (3.0%)	6 (8.3%)	12 (2.3%)
Warfarin	555 (94%)	65 (90%)	490 (94%)
TTR for INR 1.5–2.5 (%) (n=517)	61.5 (37.4–81.8)	50.0 (34.3–79.6)	62.7 (38.7–82.1)	0.11
Maximum INR value during the follow-up period (n=517)	2.91 (2.34–4.11)	3.54 (2.67–4.99)	2.88 (2.29–3.93)	<0.001
Maximum INR value >2.5 during the follow-up period (n=517)	351 (68%)	49 (78%)	302 (67%)	0.10
INR value at the time of major bleeding event (n=43)	–	2.80 (1.93–4.36)	–	–
Concomitant medications at discharge
Corticosteroids	95 (16%)	11 (15%)	84 (16%)	0.99
Nonsteroidal anti-inflammatory drugs	92 (16%)	15 (21%)	77 (15%)	0.25
Proton pump inhibitors/H2-blockers	274 (46%)	35 (49%)	239 (46%)	0.77
Statins	52 (8.8%)	3 (4.2%)	49 (9.4%)	0.18
Antiplatelet drugs	35 (5.9%)	4 (5.6%)	31 (6.0%)	1.00

Categorical variables are presented as numbers and percentages. Continuous variables are presented as the mean±standard deviation, or the median (IQR) based on their distributions. Categorical variables were compared with the chi-square test when appropriate; otherwise, Fisher’s exact test was used. Continuous variables were compared by Student’s t-test or Wilcoxon’s rank sum test based on the distributions. Patients with active cancer were defined as those on treatment for cancer such as chemotherapy or radiotherapy, those scheduled to undergo cancer-surgery, those with metastasis to other organs, and/or those with terminal cancer (expected life expectancy ≤6 months) at the time of the diagnosis of VTE. A history of major bleeding was diagnosed if the patient had a history of International Society of Thrombosis and Hemostasis (ISTH) major bleeding, which consisted of a reduction in the hemoglobin (Hb) level by at least 2 g/dL, transfusion of at least 2 units of blood, or symptomatic bleeding in a critical area or organ. Anemia was diagnosed if Hb <13 g/dL for men and <12 g/dL for women. CKD was diagnosed if there was persistent proteinuria or if the eGFR was <60 mL/min/1.73 m² for >3 months. The eGFR was calculated based on the equation reported by the Japan Association of Chronic Kidney Disease Initiative [man: 194×Scr^−1.094×age^−0.287, woman: 194×Scr^−1.094×age^−0.287×0.739]. The TTR was calculated by the Rosendaal method, according to a therapeutic INR range of 1.5–2.5, which is recommended in the Japanese guidelines. Thrombocytopenia was diagnosed if platelet level was <100×10⁹/L. Thrombophilia included protein C deficiency, protein S deficiency, antithrombin deficiency, and antiphospholipid syndrome. *Potential risk factors for major bleeding in the multivariable Cox regression model. CKD, chronic kidney disease; DVT, deep vein thrombosis; eGFR, estimated glomerular filtration rate; INR, international normalized ratio; IQR, interquartile range; PE, pulmonary embolism; TTR, time in therapeutic range; VTE, venous thromboembolism.

Table 2. Sites of Cancer

	Total (n=592)	Major bleeding (n=72)	No major bleeding (n=520)
Gastrointestinal tract*	123 (21%)	19 (26%)	104 (20%)
Upper	49/123 (40%)	5/19 (26%)	44/104 (42%)
Lower	74/123 (60%)	14/19 (74%)	60/104 (58%)
Lung	99 (17%)	9 (13%)	90 (17%)
Blood	57 (9.6%)	7 (9.7%)	50 (9.6%)
Ovary	53 (9.0%)	6 (8.3%)	47 (9.0%)
Uterus	46 (7.8%)	6 (8.3%)	40 (7.7%)
Prostate	36 (6.1%)	7 (9.7%)	29 (5.6%)
Pancreas	28 (4.7%)	2 (2.8%)	26 (5.0%)
Breast	25 (4.2%)	2 (2.8%)	23 (4.4%)
Liver/gall bladder	21 (3.5%)	2 (2.8%)	19 (3.7%)
Kidney/ureter/bladder	19 (3.2%)	2 (2.8%)	17 (3.3%)
Brain	16 (2.7%)	1 (1.4%)	15 (2.9%)
Other	32 (5.4%)	6 (8.3%)	26 (5.0%)
Multiple	37 (6.3%)	3 (4.2%)	34 (6.5%)

Variables are presented as numbers and proportions (%). Upper gastrointestinal cancer was defined as esophageal, gastric, or duodenal cancer, and lower gastrointestinal cancer was defined as colon cancer. *Potential risk factors for major bleeding in the multivariable Cox regression model.

Results

Patients’ Characteristics, Major Bleeding and All-Cause Death

In the present study population, the mean age was 66 years, women comprised 59% of the subjects, and the mean body weight and body mass index were 55.9 kg and 22.3 kg/m², respectively. During the median follow-up period of 199 (IQR, 77–627) days, major bleeding events occurred in 72 patients. Of them, 13 patients (18%) died due to bleeding (fatal bleeding). The cumulative incidence of major bleeding was 5.8% at 3 months, 13.8% at 1 year, 17.5% at 2 years, and 28.1% at 5 years (Figure 2). Major bleeding events occurred within 3 months after the diagnosis of VTE in 31 patients (43%), and beyond 3 months in 41 patients (57%). The incidence of major bleeding in patients with advanced cancer was concentrated within 1 year (Supplementary Figure 1). The cumulative incidence of all-cause death is described in Supplementary Figure 2.

Figure 2.

Kaplan-Meier curves for major bleeding beyond 10 days after the diagnosis of venous thromboembolism.

There were no significant differences in the baseline characteristics, comorbidities, or laboratory tests at the time of the diagnosis between patients with and without a major bleeding event (Table 1). Among warfarin users, there was no significant difference in the median TTR for the therapeutic INR range of 1.5–2.5 between the 2 groups, with a numerically lower TTR in patients with major bleeding, and the patients with major bleeding had a significantly higher maximum INR value during the follow-up period than those without major bleeding. Patients with major bleeding within 3 months more frequently had terminal cancer than those beyond 3 months, although there were no significant differences in the other 6 potential variables between those with major bleeding within 3 months and those beyond 3 months (Supplementary Table 1). In addition, among warfarin users with available date on 5-FU-related drugs (n=525), there was no significant difference in the incidence of major bleeding during anticoagulation therapy between patients with and without 5-FU-related drugs (log-rank P=0.99) (Supplementary Figure 3).

Sites of Cancer and Sites of Bleeding

In the present study population, the most frequent site of cancer was GI tract (n=123, 23%), followed by lung (n=99, 17%), blood (n=57, 9.6%), and ovary (n=53, 9.0%) (Table 2). Among warfarin users with available data on 5-FU-related drugs, the influence of the site of cancer on major bleeding seemed to be almost consistent with the main result, irrespective of 5-FU-related drugs (Supplementary Table 2). The most frequent site of major bleeding was GI tract (47%), followed by intracranial (17%) and genitourinary tract (11%) (Table 3). The site of bleeding with the highest mortality rate was intracranial (42% of events). Major bleeding tended to occur from the sites of the cancer, however, the sites of the cancer and the sites of major bleeding were not necessarily concordant (Supplementary Table 3).

Table 3. Sites of Bleeding

	Major bleeding (n=72)	Fatal bleeding (n=13)
Gastrointestinal tract	34 (47%)	5/34 (15%)
Intracranial	12 (17%)	5/12 (42%)
Genitourinary tract	8 (11%)	1/8 (13%)
Intrathoracic/intra-abdominal	3 (4.2%)	1/3 (33%)
Respiratory	3 (4.2%)	1/3 (33%)
Surgery-related/iatrogenic	2 (2.8%)	0/2 (0%)
Subcutaneous/intramuscular/intra-articular	1 (1.4%)	0/1 (0%)
Other	10 (14%)	0/10 (0%)

Variables are presented as numbers and proportions (%). Iatrogenic bleeding includes bleeding events related to thoracentesis, arthrocentesis, or central venous catheter insertion.

Risk Factors for Major Bleeding

The univariate Cox proportional hazard model demonstrated that terminal cancer (HR, 3.86; 95% CI, 2.08–7.16, P<0.001) and CKD (HR, 1.87; 95% CI 1.07–3.26, P=0.028) were associated with an increased risk for major bleeding (Table 4). The multivariable Cox proportional hazard model demonstrated that terminal cancer (adjusted HR, 4.17; 95% CI, 2.22–7.85, P<0.001), CKD (adjusted HR, 1.89; 95% CI 1.06–3.37, P=0.031), and GI cancer (adjusted HR, 1.78; 95% CI, 1.04–3.04, P=0.037) were independently associated with an increased risk for major bleeding (Table 4). The multivariable Cox proportional hazard model including separate variables of upper and lower GI cancer demonstrated that terminal cancer (adjusted HR, 4.13; 95% CI, 2.20–7.76, P<0.001), lower GI cancer (adjusted HR, 1.97; 95% CI, 1.08–3.59, P=0.027), and CKD (adjusted HR, 1.87; 95% CI, 1.05–3.34, P=0.031) were independently associated with an increased risk for major bleeding (Supplementary Table 4). The multivariable Cox proportional hazard model at 2 years also demonstrated consistent results with the main analysis (Supplementary Table 5).

Table 4. Univariate and Multivariable Analyses for the Risk Factors for Major Bleeding

	Univariate		Multivariable
	HR (95% CI)	P value	HR (95% CI)	P value
Age (per year)	0.99 (0.97–1.01)	0.30	0.98 (0.96–1.00)	0.082
History of major bleeding	1.75 (0.84–3.66)	0.14	1.75 (0.83–3.71)	0.14
CKD	1.87 (1.07–3.26)	0.028	1.89 (1.06–3.37)	0.031
Anemia	1.65 (0.92–2.96)	0.094	1.51 (0.84–2.72)	0.17
Thrombocytopenia	1.25 (0.54–2.88)	0.60	1.50 (0.65–3.50)	0.34
Terminal cancer	3.86 (2.08–7.16)	<0.001	4.17 (2.22–7.85)	<0.001
Gastrointestinal cancer	1.47 (0.87–2.48)	0.15	1.78 (1.04–3.04)	0.037

A history of major bleeding was diagnosed if the patient had a history of ISTH major bleeding, which consisted of a reduction in the Hb level by at least 2 g/dL, transfusion of at least 2 units of blood, or symptomatic bleeding in a critical area or organ. Anemia was diagnosed if Hb <13 g/dL for men and <12 g/dL for women. CKD was calculated diagnosed if there was persistent proteinuria or if the eGFR was <60 mL/min/1.73 m² for >3 months. The eGFR was calculated based on the equation reported by the Japan Association of Chronic Kidney Disease Initiative [man: 194×Scr^−1.094×age^−0.287, woman: 194×Scr^−1.094×age^−0.287×0.739]. Thrombocytopenia was diagnosed if platelet level was <100×10⁹/L. Patients with terminal cancer were defined as those with an expected life expectancy of ≤6 months at the time of the diagnosis of VTE. CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 1.

Discussion

The main findings of the present study were as follows. (1) Patients with cancer-associated VTE have a considerably high risk for major bleeding during anticoagulation therapy, and more than half of the major bleeding events occurred beyond 3 months. (2) During the course of anticoagulation therapy, nearly half of the major bleeding sites were GI tract. (3) Terminal cancer, CKD, and GI cancer were independent risk factors for major bleeding during anticoagulation therapy.

Patients with cancer-associated VTE have been reported to have not only a 3-fold increased risk for recurrent VTE,^23,24 but also a 2- to 6-fold increased risk for major bleeding during anticoagulation therapy.^2,25 A previous study from the RIETE registry showed that major bleeding occurred in 4.2% and 0.8% of cancer-associated VTE patients within 3 months and beyond 3 months of anticoagulation therapy, respectively.²⁶ Compared with that report, the present study showed a high incidence of major bleeding not only within 3 months (5.2%) but also beyond 3 months (6.9%). The difference in the incidence of major bleeding beyond 3 months between the previous report²⁶ and the present study might be partly due to the longer follow-up period in the present study, and a follow-up rate of 95% at 1 year compared with 74% beyond 3 months in the previous study. Although the bleeding risk has been reported to be especially high in the short term just after the initiation of anticoagulation therapy,¹⁴ the present study showed that major bleeding events were also common over the long term during anticoagulation therapy in cancer-associated VTE patients, and we should pay more attention to the long term risk for major bleeding even in patients whose condition is stable during anticoagulation therapy.

The previous study from the RIETE registry also showed that the most frequent site of major bleeding during anticoagulation therapy was GI tract (48%), followed by genitourinary tract (18%) and intracranial (11%).²⁶ Consistent with that report, the present study also showed that GI major bleeding was the most frequent, accounting for nearly half of the sites of bleeding. On the other hand, a previous study showed that GI cancer was not independently associated with an increased risk of incidence of major bleeding,²⁷ which was inconsistent with the results of the present study. This difference could be partly due to differences in the patients’ characteristics, including sites of cancer with a high prevalence of GI cancer in the present study. In the current era of direct oral anticoagulants (DOACs), the Hokusai VTE Cancer study also showed that VTE patients with GI cancer had an increased risk for major bleeding with edoxaban, mainly due to upper GI bleeding.²⁸ Considering the higher risk for bleeding in VTE patients with GI cancer, clinicians should be cautious of GI bleeding during anticoagulation therapy especially in these patients.

The risk for bleeding can also vary widely according to the cancer status. Previous studies reported that metastatic cancer was an independent risk factor for major bleeding during anticoagulation therapy.^28–30 In line with those reports, the present study also showed that terminal cancer was a strong independent risk factor for major bleeding, suggesting that more advanced cancers have an especially higher risk for bleeding. Although anticoagulation therapy lasting more than 3 months has been reported to be associated with a lower risk of VTE recurrence,³¹ clinicians should also obtain a good risk-benefit balance with prolonged anticoagulation therapy in patients with advanced cancer. A further study is warranted to clarify the treatment strategies of anticoagulation therapy for patients with advanced cancer, including patients with a terminal stage during palliative care.

Study Limitations

First, the present study was based on observational cohort data and the decision regarding the duration of anticoagulation therapy beyond the acute phase was at the attending physician’s discretion. Therefore, we could not deny the potential for selection bias and unmeasured confounding factors. Second, the risk of bleeding may be influenced by other factors that change during the course of therapy such as renal or liver function, cancer status, and use of concomitant medications. Third, low-molecular-weight heparin for VTE was not covered by the Japanese national insurance. Therefore, warfarin was mostly used as the anticoagulation therapy beyond the acute phase in patients with active cancer, however, low-molecular-weight heparin is recommended over warfarin in the Western guidelines. Fourth, the present study was conducted before the introduction of DOACs for VTE in Japan.³² Thus, it should be interpreted with caution whether the present study results can be extrapolated to patients treated with DOACs. Fifth, demographics, practice patterns and clinical outcomes in patients with VTE in Japan may be different from those outside Japan. Finally, although a previous report suggested that an elevated white blood cell count could increase the risk of major bleeding as well as recurrent VTE and all-cause death,³³ we could not evaluate the association between white blood cell count and bleeding events.

Conclusions

Major bleeding events were common during anticoagulation therapy in real-world cancer-associated VTE patients. Terminal cancer, CKD, and GI cancer were the independent risk factors for major bleeding.

Acknowledgments

We appreciate the support and collaboration of all co-investigators participating in the COMMAND VTE Registry. We are indebted to the independent clinical research organization (Research Institute for Production Development, Kyoto, Japan) for technical support. We also express our gratitude to Mr. John Martin for his assistance with the manuscript.

Funding

The COMMAND VTE Registry is supported by an independent clinical research organization (Research Institute for Production Development, Kyoto, Japan) and research funding from Mitsubishi Tanabe Pharma Corporation. The research funder had no role in the design and conduct of the study, the collection, management, analysis, and interpretation of the data, or the preparation, review, or approval of the manuscript.

Data Availability

The data, analytic methods, and study materials will not be made available to other researchers for purposes of reproducing the results or replicating the procedure. However, if the relevant review board or ethics committee approve the data sharing and all investigators of the COMMAND VTE Registry give their consent, the deidentified participant data will be shared on a request basis through the principal investigator. Study protocol and statistical analysis plan will also be available. The data will be shared as Excel files via E-mail during the proposed investigation period.

Disclosures

Dr. Yamashita received lecture fees from Daiichi-Sankyo, Bristol-Myers Squibb, Pfizer, and Bayer Healthcare; Dr. Morimoto received lecture fees from Mitsubishi Tanabe Pharma and Pfizer Japan and consultant fees from Asahi Kasei, Bristol-Myers Squibb, and Boston Scientific; Dr. Ono is a member of Circulation Journal ’ Editorial Team; Dr. Kimura serves as an advisory board member for Abbott Vascular and Terumo Company and is a member of Circulation Journal ’ Editorial Team. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.

IRB Information

The relevant review boards or ethics committees at all 29 participating centers approved the research protocol (Kyoto University Hospital Ethics Committee, approval no. R0493; Kurashiki Central Hospital Ethics Committee, approval no. 2179; Kinki University Ethics Committee, approval no. 28-056; Kokura Memorial Hospital Ethics Committee, approval no. 16051804; Kobe City Medical Center General Hospital Ethics Committee, approval no. zn160801; Japanese Red Cross Otsu Hospital Ethics Committee, approval no. 373; Kyoto Medical Center Ethics Committee, approval no. 16-021; Osaka Red Cross Hospital Ethics Committee, approval no. 699; Shiga Medical Center Ethics Committee, approval no. 20160603:154: Contemporary Management And Outcomes In Patients With Venous Thromboembolism Registry; Japanese Red Cross Wakayama Medical Center Ethics Committee, approval no. 414; Kitano Hospital Ethics Committee, approval no. P16-05-16; Shizuoka General Hospital Ethics Committee, approval no. SGHIRB#2016011; Osaka Saiseikai Noe Hospital Ethics Committee, approval no. 28-3; Shizuoka City Shizuoka Hospital Ethics Committee, approval no. 1606; Shimada Municipal Hospital Ethics Committee, approval no. 28-1; Shiga University Ethics Committee, approval no. 28-078; Hyogo Prefectural Amagasaki General Medical Center Ethics Committee, approval no. 28-9; Kishiwada City Hospital Ethics Committee, approval no. H28-4-25 Kishibyourin No.1: Contemporary Management And Outcomes In Patients With Venous Thromboembolism Registry; Tenri Hospital Ethics Committee, approval no. 753; Kyoto Okamoto Memorial Hospital Ethics Committee, approval no. 2016-07; Nishikobe Medical Center Ethics Committee, approval no. 2016-04; Nara Hospital, Kinki University Faculty of Medicine Ethics Committee, approval no. 370; Hikone Municipal Hospital Ethics Committee, approval no. 28-6; Kansai Electric Power Hospital Ethics Committee, approval no. 28-13; Mitsubishi Kyoto Hospital Ethics Committee, approval no. 16-3; Koto Memorial Hospital Ethics Committee, approval no. 2016: Contemporary Management And Outcomes In Patients With Venous Thromboembolism Registry; Obama Municipal Hospital Ethics Committee, approval no. 28-2; Hirakata Kohsai Hospital Ethics Committee, approval no. H28-5-19: Contemporary Management And Outcomes In Patients With Venous Thromboembolism Registry; Kyoto University Hospital Ethics Committee; approval no. R0493-1).

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-20-0223

References

• 1.   Blom JW, Doggen CJM, Osanto S, Rosendaal FR. Malignancies, prothrombotic mutations, and the risk of venous thrombosis. J Am Med Assoc 2005; 293: 715–722.
• 2.   Prandoni P, Lensing AWA, Piccioli A, Bernardi E, Simioni P, Girolami B, et al. Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis. Blood 2002; 100: 3484–3488.
• 3.   Konstantinides SV, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galiè N, et al. 2014 ESC Guidelines on the diagnosis and management of acute pulmonary embolism. Eur Heart J 2014; 35: 3033–3080.
• 4.   Kearon C, Akl EA, Ornelas J, Blaivas A, Jimenez D, Bounameaux H, et al. Antithrombotic therapy for VTE disease. Chest 2016; 149: 315–352.
• 5.   Ruíz-Giménez N, Suárez C, González R, Nieto JA, Todolí JA, Samperiz ÁL, et al. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism: Findings from the RIETE Registry. Thromb Haemost 2008; 100: 26–31.
• 6.   Klok FA, Hösel V, Clemens A, Yollo WD, Tilke C, Schulman S, et al. Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment. Eur Respir J 2016; 48: 1369–1376.
• 7.   Kim K, Yamashita Y, Morimoto T, Kitai T, Yamane T, Ehara N, et al. Risk factors for major bleeding during prolonged anticoagulation therapy in patients with venous thromboembolism: From the COMMAND VTE Registry. Thromb Haemost 2019; 119: 1498–1507.
• 8.   Yamashita Y, Morimoto T, Amano H, Takase T, Hiramori S, Kim K, et al. Anticoagulation therapy for venous thromboembolism in the real world: From the COMMAND VTE Registry. Circ J 2018; 82: 1262–1270.
• 9.   Nishimoto Y, Yamashita Y, Morimoto T, Saga S, Amano H, Takase T, et al. Clinical characteristics and outcomes of venous thromboembolisms according to an out-of-hospital versus in-hospital onset: From the COMMAND VTE Registry. Circ J 2019; 83: 1377–1384.
• 10.   Schulman S, Anger SU, Bergqvist D, Eriksson B, Lassen MR, Fisher W. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in surgical patients. J Thromb Haemost 2010; 8: 202–204.
• 11.   Faller N, Limacher A, Méan M, Righini M, Aschwanden M, Beer JH, et al. Predictors and causes of long-term mortality in elderly patients with acute venous thromboembolism: A prospective cohort study. Am J Med 2017; 130: 198–206.
• 12.   Rosendaal FR, Cannegieter SC, Van der Meer FJM, Briet E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost 1993; 69: 236–239.
• 13.   JCS Joint Working Group. Guidelines for the diagnosis, treatment and prevention of pulmonary thromboembolism and deep vein thrombosis (JCS 2009): Digest version. Circ J 2011; 75: 1258–1281.
• 14.   Nieto JA, Solano R, Ruiz-Ribó MD, Ruiz-Gimenez N, Prandoni P, Kearon C, et al. Fatal bleeding in patients receiving anticoagulant therapy for venous thromboembolism: Findings from the RIETE registry. J Thromb Haemost 2010; 8: 1216–1222.
• 15.   Ruíz-Giménez N, Suárez C, González R, Nieto JA, Todolí JA, Samperiz ÁL, et al. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism: Findings from the RIETE registry. Thromb Haemost 2008; 100: 26–31.
• 16.   O’Brien EC, Simon DN, Thomas LE, Hylek EM, Gersh BJ, Ansell JE, et al. The ORBIT bleeding score: A simple bedside score to assess bleeding risk in atrial fibrillation. Eur Heart J 2015; 36: 3258–3264.
• 17.   White RH, Beyth RJ, Zhou H, Romano PS. Major bleeding after hospitalization for deep-venous thrombosis. Am J Med 1999; 107: 414–424.
• 18.   Pisters R, Lane DA, Nieuwlaat R, De Vos CB, Crijns HJGM, Lip GYH, et al. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: The Euro Heart survey. Chest 2010; 138: 1093–1100.
• 19.   Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998; 105: 91–99.
• 20.   Gage BF, Yan Y, Milligan PE, Waterman AD, Culverhouse R, Rich MW, et al. Clinical classification schemes for predicting hemorrhage: Results from the National Registry of Atrial Fibrillation (NRAF). Am Heart J 2006; 151: 713–719.
• 21.   Kuijer PMM, Hutten BA, Prins MH, Büller HR. Prediction of the risk of bleeding during anticoagulant treatment for venous thromboembolism. Arch Intern Med 1999; 159: 457–460.
• 22.   Kanda Y. Investigation of the freely available easy-to-use software “EZR” for medical statistics. Bone Marrow Transplant 2013; 48: 452–458.
• 23.   Hansson PO, Sörbo J, Eriksson H. Recurrent venous thromboembolism after deep vein thrombosis: Incidence and risk factors. Arch Intern Med 2002; 160: 769–774.
• 24.   Louzada ML, Carrier M, Lazo-Langner A, Dao V, Kovacs MJ, Ramsay TO, et al. Development of a clinical prediction rule for risk stratification of recurrent venous thromboembolism in patients with cancer-associated venous thromboembolism. Circulation 2012; 126: 448–454.
• 25.   Hutten BA, Prins MH, Gent M, Ginsberg J, Tijssen JGP, Büller HR. Incidence of recurrent thromboembolic and bleeding complications among patients with venous thromboembolism in relation to both malignancy and achieved International Normalized Ratio: A retrospective analysis. J Clin Oncol 2000; 18: 3078–3083.
• 26.   Trujillo-Santos J, Nieto JA, Ruíz-Gamietea Á, López-Jiménez L, García-Bragado F, Quintavalla R, et al. Bleeding complications associated with anticoagulant therapy in patients with cancer. Thromb Res 2010; 125: S58–S61.
• 27.   Trujillo-Santos J, Nieto JA, Tiberio G, Piccioli A, Di Micco P, Prandoni P, et al. Predicting recurrences or major bleeding in cancer patients with venous thromboembolism: Findings from the RIETE registry. Thromb Haemost 2008; 100: 435–439.
• 28.   Raskob GE, van Es N, Verhamme P, Carrier M, Di Nisio M, Garcia D, et al. Edoxaban for the treatment of cancer-associated venous thromboembolism. N Engl J Med 2017; 378: 615–624.
• 29.   Prandoni P, Trujillo-Santos J, Surico T, Valle FD, Piccioli A, Monreal M. Recurrent thromboembolism and major bleeding during oral anticoagulant therapy in patients with solid cancer: Findings from the RIETE registry. Haematologica 2008; 93: 1432–1434.
• 30.   Angelini DE, Radivoyevitch T, McCrae KR, Khorana AA. Bleeding incidence and risk factors among cancer patients treated with anticoagulation. Am J Hematol 2019; 94: 780–785.
• 31.   Khorana AA, McCrae KR, Milentijevic D, Fortier J, Nelson WW, Laliberté F, et al. Duration of anticoagulant therapy and VTE recurrence in patients with cancer. Support Care Cancer 2019; 27: 3833–3840.
• 32.   Yamashita Y, Morimoto T, Toyota T, Shiomi H, Makiyama T, Ono K, et al. Asian patients versus non-Asian patients in the efficacy and safety of direct oral anticoagulants relative to vitamin K antagonist for venous thromboembolism: A systemic review and meta-analysis. Thromb Res 2018; 166: 37–42.
• 33.   Trujillo-Santos J, Di Micco P, Iannuzzo M, Lecumberri R, Guijarro R, Madridano O, et al. Elevated white blood cell count and outcome in cancer patients with venous thromboembolism: Findings from the RIETE registry. Thromb Haemost 2008; 100: 905–911.