ASH 2018: Direct Oral Anticoagulants Safely Reduce Clotting Risk In Outpatient Cancer Setting
But patient adherence was a problem in the trial
December 4, 2018—San Diego, California—Ambulatory cancer patients given rivaroxaban oral anticoagulation therapy are at a significantly lower risk of developing harmful clots while on treatment, according to the results of the CASSINI trial, presented here at the 2018 American Society of Hematology Annual Meeting & Exposition, which took place from December 1 to 4.
According to study presenter Alok Khorana, MD, of the Cleveland Clinic, the CASSINI study was designed to address an unmet need in clot prevention: Both chemotherapy and clots are more likely to occur in the outpatient setting, yet anticoagulant therapy is typically administered only in the hospital setting and not prescribed for home use.
The CASSINI trial is the first to investigate the use of oral anticoagulant therapy in ambulant cancer patients at high-risk for clots. It was funded by Bayer, initial developers of rivaroxaban, and Janssen Pharmaceutica, the current marketers.
Previous studies investigated the use of heparin in the home setting to reduce risk of clotting in cancer patients; however, as Dr. Khorana suggested during a press briefing, the injection-based delivery and high cost of heparin make this therapy difficult to use from a patient adherence perspective.
Dr. Khorana and his collaborators performed a randomized, double-blind, placebo-controlled, parallel-group, multicenter study, enrolling a total of 1080 patients. Enrolled patients had various cancers and had venous thromboembolism (VTE) Khorana risk scores of 2 or higher. A screen for deep vein thrombosis (DVT) eliminated 49 patients, and 190 were removed for other reasons, prior to randomization of the remainder (n = 841), in which study participants were assigned 1:1 to a 180-day course of either daily oral rivaroxaban 10 mg or placebo. Of the 841 patients randomized, 32.6% had pancreatic cancer, 83% were white, and 50.9% were male.
The CASSINI trial failed to achieve statistical significance in its primary prespecified efficacy endpoint, a composite of objectively confirmed symptomatic or asymptomatic lower extremity or proximal DVT, symptomatic or incidental pulmonary embolism, and VTE-related death.
The primary efficacy endpoint occurred within the 180-day observation period in 25 of 420 patients who received rivaroxaban (5.95%) and in 37 of 421 patients receiving placebo (8.79%), resulting in a number needed to treat of 35 (hazard ratio 0.66; 95% confidence interval 0.40–1.09, P = .101).
According to Dr. Khorana, the primary endpoint lacked statistical power in the intention-to-treat group because many patients discontinued therapy prior to the prespecified 180 days. Indeed, as many as half of those given placebo and nearly 44% of those assigned to receive rivaroxaban discontinued therapy before completing the 180-day regimen.
A prespecified analysis that sought to detect the primary composite endpoint during the on-treatment period detected significant differences between the rivaroxaban and control groups; whereas 11 of 420 patients receiving rivaroxaban (2.62%) experienced the primary endpoint within the on-treatment period, 27 of 421 patients receiving placebo (6.41%) did (hazard ratio 0.40; 95% confidence interval 0.20–0.80, P = .007). This translated to a number needed to treat of 26.
Taken together, Dr. Khorana said these results suggest rivaroxaban is likely to remain effective so long as patients continue to remain adherent to therapy. As many as 38.7% of patients with VTE experienced a thrombotic event after discontinuing rivaroxaban therapy.
“Pending confirmation by a second, similarly designed trial, these findings could lead to a change in guidelines regarding primary prevention of VTE in cancer patients initiating systemic therapy,” Dr. Khorana concluded.
Risk of major bleeding, the primary safety endpoint, was not significantly different between treatment groups; major bleeding occurred in 8 of 405 patients (1.98%) in the rivaroxaban group and in 4 of 404 patients (0.99%) in the placebo group (hazard ratio 1.96; 95% confidence interval 0.59–6.49, P = .265). The number needed to harm was 101. Incidence of clinically relevant non-major bleeding also did not significantly differ between treatment groups (hazard ratio 1.34; 95% confidence interval 0.54–3.32, P = .53). In this context, the number needed to harm was 135.
“With any blood thinner, safety is an issue,” Dr. Khorana told Elsevier’s PracticeUpdate. “We only saw a low incidence of major and non-major bleeding events, and no substantial concern for additional adverse events.”
Click on any of these tags to subscribe to Topic Alerts. Once subscribed, you can get a single, daily email any time PracticeUpdate publishes content on the topics that interest you.
Visit your Preferences and Settings section to Manage All Topic Alerts