Cancer patients treated with the oral anticoagulant apixaban (Eliquis) had a lower rate of venous thromboembolism but a higher rate of major bleeding, according to data from the AVERT study.
In the placebo-controlled, double-blind trial, 574 ambulatory cancer patients who were at moderate to high risk of thromboembolism (Khorana risk score of 2 or more) and were starting chemotherapy were randomized to either apixaban 2.5 mg twice daily or to placebo for 180 days. Over the 210-day study period, 12 patients (4.2%) in the apixaban group experienced a venous thromboembolism as did 28 patients (10.2%) in the placebo group, an adjusted 61% reduction in risk associated with anticoagulant therapy. The number needed to treat to prevent one venous thromboembolism was 17, Marc Carrier, MD, of the University of Ottawa, and his coauthors reported in the Dec. 4 edition of the New England Journal of Medicine.
“The treatment of venous thromboembolism with therapeutic anticoagulation is challenging in patients with cancer, because it often involves daily injections of low-molecular-weight heparin and is associated with a high risk of thromboembolism recurrence and serious bleeding complications,” they wrote. As an oral agent, apixaban offers a more convenient alternative.
The authors added that their study found more favorable benefits from anticoagulant therapy than had been seen in previous studies and suggested that this may be the result of using a different agent and a twice-daily dosing regimen.
In the AVERT study, the lower incidence of thromboembolism in the treatment arm was largely because of a reduction in pulmonary embolisms; there were 5 cases in the apixaban group, compared with 16 in the placebo group. The apixaban group experienced 7 cases of deep-vein thrombosis, and the placebo group experienced 12 cases.
During the treatment period, the placebo group had 20 venous thrombembolisms and the apixaban group had 3.
However the incidence of major bleeding was twice as high in the apixaban group: 10 patients (3.5%), compared with 5 (1.8%) in the placebo group (P = .046). The difference between the two groups was mostly based on an increased incidence of gastrointestinal bleeding, hematuria, and gynecologic bleeding among patients treated with apixaban.
None of the major bleeds affected critical organs in any patients. Most were category 2 bleeds, and three cases were judged to be clinical emergencies.
There were 62 deaths overall in the study – 35 in the apixaban group and 27 in the placebo group – and 87% of these deaths were related to the cancer.
Many patients in the study had advanced cancer, which was also the most common cause of death, the authors said. However, there was one death from pulmonary embolism in the placebo group. The dominant cancer types in the study participants were lymphoma, gynecologic, pancreatic, and lung cancers. Two-thirds of the patients in each group had a Khorana risk score of 2, and one patient in each group had a score of 5.
A different trial design and larger study would be needed to examine the impact of treatment on mortality and outcomes related to specific tumor types and chemotherapy regimens, the authors said.
They stressed that only 5.9% of patients in the study had renal dysfunction, so the study results cannot necessarily be applied to these patients more generally, especially as they are known to be at higher risk of bleeding.
The study was supported by the Canadian Institutes of Health Research and Bristol-Myers Squibb–Pfizer Alliance. Thirteen authors declared honoraria, grants or personal fees from the pharmaceutical industry unrelated to the study. Two declared grants from the study funders for the study; ten authors had no conflicts of interest to declare.
SOURCE: Carrier M et al. N Engl J Med. 2018 Dec 4. doi: 10.1056/NEJMoa1814468