Data from randomized control trials examining edoxaban and rivaroxaban, published in 2018, changed the landscape of anticoagulation for cancer-associated venous thromboembolism (VTE).1,2 These drugs were welcome alternatives to low-molecular-weight heparins, the previously established optimal therapy over warfarin.3 Enthusiasm, however, has been limited by the observed increase in major bleeding with edoxaban and clinically relevant nonmajor bleeding (CRNMB) with rivaroxaban in comparison to dalteparin. After publication of the Hokusai VTE Cancer and SELECT-D studies, the International Society for Thrombosis and Haemostasis published guidelines with recommendations for these drugs but with a notable concern regarding the potential for higher risk of bleeding, especially in patients with luminal gastrointestinal malignancies.4
The Caravaggio study was an open-label, randomized, multinational clinical trial studying apixaban compared to dalteparin for cancer-associated VTE. Participants had active cancer and symptomatic (or incidental, up to 20%) proximal deep vein thrombosis (DVT) or pulmonary emboli. Patients could have received initial nonstudy parental anticoagulation for a maximum of 72 hours before enrollment. The primary outcome of recurrent VTE was defined as proximal DVT (symptomatic or incidental), symptomatic upper extremity DVT, or pulmonary embolism (symptomatic, incidental, or fatal) within six months. The principal safety outcome was major bleeding occurring during the study period and within 72 hours of last study drug administered.
At 119 sites, 1,170 patients were randomized, making it the largest study of cancer-associated VTE. A total of 1,155 patients, 576 treated with apixaban and 579 treated with dalteparin, were included in the modified intention to treat analysis. The primary efficacy outcome of recurrent VTE occurred in 5.6 percent with apixaban and 7.9 percent with dalteparin after six months (HR, 0.63; 95% CI, 0.37-1.07; p<0.001 for noninferiority, p=0.09 for superiority). The primary safety outcome of major bleeding occurred in 3.8 percent with apixaban and 4.0 percent with dalteparin (HR, 0.82; 95% CI, 0.40-1.69; p=0.60). Major gastrointestinal bleeding (not a prespecified outcome) was similar between apixaban and dalteparin, occurring in 1.9 percent and 1.7 percent of patients, respectively. CRNMB was not statistically different and occurred in 9.0 percent of apixaban- and 6.0 percent of dalteparin-treated participants (HR, 1.42; 95% CI, 0.88-2.30).
In Brief
Apixaban now has robust data to support its use in patients with cancer-associated VTE and must be put into perspective with previous studies (Table 1), but there are limitations to the conclusions we can draw from these comparisons. While each drug was compared against the same regimen of dalteparin, differences in inclusion or exclusion criteria, and recruitment of patients at different times in different settings led to small, but potentially meaningful differences in patient populations. For example, the Caravaggio study excluded patients with primary brain tumors or known cerebral metastases and acute leukemia, and the SELECT-D study excluded patients with upper gastrointestinal malignancies after initial safety analyses demonstrated a concern for higher bleeding rates with rivaroxaban. Patients with atypical VTE (upper extremity, cerebral venous sinus, and splanchnic vein thrombosis), which is seen more commonly in patients with cancer, were excluded from all studies except the ADAM-VTE study.5 Heterogeneity in the studies can also be observed by examining the percent treated who had colorectal cancer, which varied up to 15 percent, and those with lung cancer, which varied up to 10 percent between studies. The rates of major bleeding with dalteparin were very similar between most of the studies, but interestingly there was more variability with recurrent VTE across the studies. Now that a robust arsenal of anticoagulants is available for the treatment of cancer-associated VTE, studies should focus on refining the use and selection of these drugs for specific cancers and specific circumstances as a means to optimize benefits and minimize harm.
Direct Oral Anticoagulant Randomized Controlled Trials in Cancer-Associated VTE
| Study . | Published Date . | Study Drug . | Study Size (No. of Patients) . | Follow-up (Months) . | Comparison of Selected Malignancies Among DOAC-treated Groups . | Recurrent VTE . | Major Bleeding . | ||
|---|---|---|---|---|---|---|---|---|---|
| Study Drug | Dalteparin | Study Drug | Dalteparin | ||||||
| Caravaggio | 4/2020 | Apixaban | 1,155 | 6 | Colorectal – 21%Upper GI – 4.0%Lung – 18.2%Genitourinary – 11.5%Hematologic – 5.7% | 5.6% | 7.9% | 3.8% | 4.0% |
| ADAM-VTE5 | 10/2019 | Apixaban | 300 | 6 | Colorectal – 12.2%Upper GI – 4.8%Lung – 21.8%Genitourinary – 8.7%Hematologic – 8.6% | 0.7% | 6.3% | 0% | 1.4% |
| SELECT-D2 | 7/2018 | Rivaroxaban | 406 | 6 | Colorectal – 27%Upper GI – 5%Lung – 11%Genitourinary –10.3 %Hematologic – 6.9% | 4.0% | 11% | 6% | 4.0% |
| Hokusai VTE Cancer1 | 2/2018 | Edoxaban | 1,046 | 12 | Colorectal – 15.9%Upper GI – 6.3%Lung – 14.8%Genitourinary – 12.5%Hematologic – 10.7% | 7.9% | 11.3% | 6.9% | 4.0% |
| Study . | Published Date . | Study Drug . | Study Size (No. of Patients) . | Follow-up (Months) . | Comparison of Selected Malignancies Among DOAC-treated Groups . | Recurrent VTE . | Major Bleeding . | ||
|---|---|---|---|---|---|---|---|---|---|
| Study Drug | Dalteparin | Study Drug | Dalteparin | ||||||
| Caravaggio | 4/2020 | Apixaban | 1,155 | 6 | Colorectal – 21%Upper GI – 4.0%Lung – 18.2%Genitourinary – 11.5%Hematologic – 5.7% | 5.6% | 7.9% | 3.8% | 4.0% |
| ADAM-VTE5 | 10/2019 | Apixaban | 300 | 6 | Colorectal – 12.2%Upper GI – 4.8%Lung – 21.8%Genitourinary – 8.7%Hematologic – 8.6% | 0.7% | 6.3% | 0% | 1.4% |
| SELECT-D2 | 7/2018 | Rivaroxaban | 406 | 6 | Colorectal – 27%Upper GI – 5%Lung – 11%Genitourinary –10.3 %Hematologic – 6.9% | 4.0% | 11% | 6% | 4.0% |
| Hokusai VTE Cancer1 | 2/2018 | Edoxaban | 1,046 | 12 | Colorectal – 15.9%Upper GI – 6.3%Lung – 14.8%Genitourinary – 12.5%Hematologic – 10.7% | 7.9% | 11.3% | 6.9% | 4.0% |
Abbreviations: DOAC, direct oral anticoagulant; GI, gastrointestinal; VTE, venous thromboembolism.
References
Competing Interests
Dr. McBane and Dr. Houghton were non-coauthor participants as site coinvestigators for the Caravaggio study.
