Risk of recurrence and bleeding in patients with cancer‐associated venous thromboembolism treated with rivaroxaban: A nationwide cohort study

This nationwide cohort study of 476 patients with cancer‐associated VTE initiating rivaroxaban showed a 6 months recurrence rate of 6.1%. Major bleeding occurred in 1.9% at 6 months follow‐up.

Abstract

Background

Rivaroxaban could be an attractive alternative to low molecular weight heparin for the treatment of cancer‐associated venous thromboembolism (VTE) but the safety and effectiveness remain unclear. We examined risk of recurrent VTE and major bleeding associated with rivaroxaban treatment of cancer‐associated VTE.

Methods

Through linkage of nationwide Danish registries, we identified all adults with cancer‐associated VTE initiating treatment with rivaroxaban, 2012‐2017. We estimated rates and absolute risk of the primary outcome of recurrent VTE and major bleeding; all‐cause mortality was studied as a secondary outcome.

Results

We identified 8901 patients with cancer‐associated VTE of whom 476 (5.3%) redeemed a prescription for rivaroxaban within 30 days of VTE diagnosis (mean age 71.5 years, 41% females, 57% with pulmonary embolism). Median time from cancer diagnosis to rivaroxaban prescription was 31 days (interquartile range 12‐73 days). Most frequent cancers were gastrointestinal (26.1%), genitourinary (23.3%), and hematological cancer (12.6%). Few had distant metastases (7.1%). At 6 months, recurrent VTE occurred in 6.1% (15.1 events per 100 person‐years) with the highest absolute risks for genitourinary cancer (8.1%), gastrointestinal cancer (7.3%), and breast cancer (6.5%). Major bleeding occurred in 1.9% (5.3 events per 100 person‐years), in particular, in genitourinary cancer (4.5%) and lung cancer (4.2%). Eighty deaths (17.8%) occurred during follow up.

Conclusion

In this clinical practice setting, rivaroxaban was rarely used for cancer‐associated VTE. However, among those who received rivaroxaban, the treatment appeared safe and effective with rates comparable to previous studies of selected populations.

http://bit.ly/2X4NqCp