Background

Venous thromboembolism (VTE) is the second leading cause of mortality in cancer patients. National guidelines recommend treatment with a low-molecular-weight heparin (LMWH) for at least 6 months following a cancer-associated thrombosis (CAT). The direct oral anticoagulants (DOAC) have become attractive options for the management of CAT as they are orally administered and require no routine monitoring. However, current guidelines do not support the routine use of DOAC for CAT, given the lack of direct comparison between LMWH and DOAC. The objective of this study was to compare the efficacy and safety of LMWH to DOAC for the treatment of CAT.

Methods

This was a retrospective, single-center study conducted at the Ohio State University. Adult cancer patients with confirmed VTE treated with therapeutic doses of either a LMWH or DOAC between December 2010 and January 2016 were included. Patients were excluded if they had a history of thrombophilia, heparin-induced thrombocytopenia, creatinine clearance (CrCl) < 30mL/min, required mechanical or pharmacologic thrombolysis, or anticoagulation started > 3 months after the initial VTE date. Medical records were reviewed for 6 months following the start of anticoagulation or until the occurrence of the primary outcome, death, or cessation of therapy. Predefined data were systematically collected, including patient demographics, treatments, risk factors, recurrent VTE, bleeding events, and death. The primary efficacy outcome was VTE recurrence at 6 months and the primary safety outcome was major bleeding at 6 months. Secondary outcomes included clinically-relevant-non-major bleeding (CRNMB), overall mortality, and risk factors for recurrent VTE or bleeding. Major bleeding and CRNMB were defined in accordance with the International Society on Thrombosis and Hemostasis criteria. Baseline characteristics were compared by t-test, Wilcoxon rank sum test, or chi-square test, as appropriate. The VTE and major bleeding outcomes were assessed by univariable logistic regression for LMWH versus DOAC and for other potential risk factors. One multivariable logistic regression model was fit to each outcome.

Results

Two hundred ninety patients treated with LMWH and 190 patients treated with DOAC were included. The majority of patients were treated with enoxaparin (n=287) and rivaroxaban (n=167). Mean age was 58 years, 16% had a history of VTE, 7% had a history of bleeding, and 53% had metastatic disease. Statistical differences in baseline characteristics included: high thrombotic risk cancers (57% LMWH vs 43% DOAC), hematologic cancers (19% LMWH vs 32% DOAC), vena cava filters (20% LMWH vs 12% DOAC), baseline hemoglobin and plateletcounts(10.9g/dL vs 11.4g/dL and 237 K/µL vs 252 K/µL for LMWH vs DOAC respectively). The median follow up time was 160 days and 153 days for the LMWH and DOAC groups respectively. There was no difference in the rate of recurrent VTE at 6 months between groups. Patients treated with LMWH experienced more major bleeding and CRNMB episodes and had more deaths at 6 months (Table 1).A majority of patients died from cancer progression (67% LMWH, 73% DOAC), however, 3 patients died of a major bleed (2 LMWH, 1 DOAC), and 1 LMWH patient died from recurrent PE.

We used the multivariable analysis to identify several risk factors associated with increased risk of recurrent VTE: platelet count < or = 150 K/µL, younger age, vena cava filter insertion after VTE, and history of warfarin failure. Risk factors associated with bleeding included the presence of vena cava filter and baselineCrCl< 50 mL/min. In adjusted analysis, the use of LWMH showed a trend toward a higher risk of major bleeding (Table 2).

Discussion

In patients with CAT, we found no difference in VTE recurrence in patients treated with DOAC compared to LMWH, while LMWH was associated with increased bleeding. Baseline demographics were unbalanced and increased cancer related mortality in the LMWH group suggests more advanced disease in this group. To account for these confounding factors, we used the multivariable logistic regression model and found that LWMH is associated with a trend towards increased bleeding, although not statistically significant.

Conclusion

In our cohort of patients with CAT, treatment with DOAC appears to be as effective and potentially safer when compared to LMWH. Randomized controlled trials for direct comparison are ongoing and results are highly anticipated.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution