Similar bleeding with DOAC as with LMWH in those with cancer-associated thrombosis

Introduction and methods

Direct oral anticoagulants (DOACs) have a similar efficacy profile compared to subcutaneous injected low-molecular-weight-heparin (LMWH) in patients treated for cancer-associated thrombosis (CAT) [1]. However, safety concerns were raised by clinical trials that reported increased rates of major bleeding events, mostly occurring at the gastrointestinal tract, in patients with CAT using the DOAC edoxaban or rivaroxaban [2,3]. Recent guidelines have favored the use of DOACs in patients with CAT, but have advised caution or exclusion in patients with gastrointestinal cancer [1, 4-7].

This subanalysis of the Caravaggio study compared the sites of major and clinically relevant non-major bleeding (CRNMB) events, associated cancer sites, clinical presentation, and course of major bleeding in patients with CAT receiving either the DOAC apixaban or the LMWH dalteparin.

The Caravaggio study was a multinational, randomized, open-label, non-inferiority study with blind adjudication of the study outcomes. Patients (n=1,155) with cancer or diagnosed with cancer within 2 years before study inclusion and presenting with symptomatic or incidental proximal lower limb deep vein thrombosis (DVT) or pulmonary embolism (PE) were included. Patients were randomized (1:1) to apixaban, 10 mg orally twice a day for 7 days followed by 5 mg bid or 200 IU/kg dalteparin subcutaneously, daily, for 1 month followed by 150 IU/kg od (maximum daily dose was 18,000 IU). Treatment duration was 6 months. Recurrent venous thromboembolism (VTE) occurred in 5,6% of patients treated with apixaban compared to 7.9% of patients receiving dalteparin (HR 0.63, 95%CI: 0.37-1.07). Major bleeding was present in 3.8% of patients with apixaban and in 4.0% in those treated with dalteparin (HR 0.82, 95%CI: 0.40-1.69) [8].

Major bleeding events were classified using the European Medicines Agency (EMA) definition. These included clinically overt bleeding associated with a decline of ≥2 g/dL in hemoglobin, bleeding that required 2 or more units of blood transfusion, bleeding that occurred at a critical site, fatal bleeding, and bleeding events resulting in surgical intervention. CRNMB events were defined as acute clinically overt bleeding that did not meet the EMA definitions, but required medical intervention by a health caretaker.

Main results

• Major bleeding events were present in 22 patients treated with apixaban compared to 23 patients on dalteparin. Number of bleeding risk factors were similar in both treatment groups.

• There was 1 bleeding event in a critical site in the group treated with apixaban compared to 5 in the dalteparin group. Bleeding events that needed surgical intervention occurred in 4 patients using apixaban and no patient treated with dalteparin. Fatal bleedings occurred in 2 patients that received dalteparin.

• There were 52 CRNMB events in patients treated with apixaban compared to 35 events in those receiving dalteparin. The median time to a CRNMB was 48 days in the apixaban group and 28 days in the dalteparin group.

• Major bleeding events that required hospitalization occurred in 8 patients in the apixaban group compared to 13 in the dalteparin group.

• The majority of major bleedings were gastrointestinal bleedings, 11 in the apixaban group compared to 10 in the dalteparin. The number of major bleedings in the upper and lower gastrointestinal tract were comparable in the two treatment groups.

• 7 Patients with cancer of the gastrointestinal tract treated with apixaban experienced a major bleeding compared to 9 patients receiving dalteparin.

• The clinical presentation of bleeding was severe or fatal in 6 patients on apixaban vs. 5 on dalteparin, while the clinical course was life-threatening or fatal in 5 patients treated with apixaban and 7 patients receiving dalteparin.

Conclusion

References

Find this article online at Thromb Haemost