Direct oral anticoagulant use in patients with myeloproliferative neoplasms

Myeloproliferative neoplasms (MPN), which include polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF), are associated with a hypercoagulable state. As a result, venous thrombosis and cardiovascular events cause significant morbidity and mortality in such patients, and management of these challenges is important when treating patients with MPN. However, expert consensus recommendations for anticoagulant selection and duration in these patients is lacking. Vitamin K antagonists (VKAs) are recommended as standard of care for the management of thrombosis in MPN; however, there are some limitations such as the need for frequent monitoring, dietary restrictions, drug interactions, and a narrow therapeutic window.

Use of direct oral anticoagulants (DOACs) such as apixaban, rivaroxaban, and edoxaban, which overcome many of the issues associated with warfarin, has increased significantly in both the general population and patients with cancer. However, research into the safety and efficacy of DOACs in MPN specifically is minimal.

Joan How and colleagues¹ published their findings in the Blood Cancer Journal, on a retrospective cohort study examining real-world practice patterns of DOAC use in patients with MPN. Their findings are summarized below.

The results were also presented at the 63rd American Society of Hematology (ASH) Annual Meeting and Exposition.² Check out the top abstracts in MPN based on the recommendations from the MPN Hub Steering Committee members here.

Study design

• In this retrospective single center study, 133 patients with MPN (PV, ET, MF, or MPN not otherwise specified [NOS]) who had been prescribed a DOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) were identified for inclusion.
• Patients prescribed a DOAC for perioperative prophylaxis or prophylaxis during long travel, as well as patients in which the DOAC initiation date was not available, were excluded.
• Thrombotic and bleeding outcomes on DOAC treatment (and up to 3 days after DOAC discontinuation) were evaluated.

Baseline characteristics

Baseline characteristics of patients are shown in Table 1. The indications for DOAC included venous thromboembolism (VTE, n = 75), atrial fibrillation (AF, n = 46), and other events (including stroke and arterial thromboembolism, n = 12). Of note, patients on a DOAC for treatment of VTE were significantly younger (p < 0.001) than those with AF, and more likely to be female (p = 0.04), have prior venous rather than arterial thrombosis (p = 0.001), and have rivaroxaban selected (p = 0.02). VTE subtypes included deep vein thrombosis (43%), pulmonary embolism (31%), splanchnic vein thrombosis (21%), and other events (4.4%).

Table 1. Baseline characteristics of patients with MPN by DOAC indication*

Results

Practice patterns of DOAC use

The practice patterns of DOAC use are shown in Table 2. For patients who completed a finite course of DOAC (12%), the median course duration was 6 months. Dose reduction in 10% of the cohort occurred after a median of 21 months of treatment.

Table 2. Practice patterns of DOAC use*

Incidence of thrombosis and bleeding

The median follow-up for patients was 37 months, and during this time there were 12 thrombotic events that occurred on DOACs.

• The 1-year cumulative incidence of thrombosis on a DOAC was 5.5% (1.5–9.5%).
• There was no significant difference in thrombosis incidence between the VTE and AF groups (5 vs 6 thrombotic events, respectively; p = 0.47).
• Of the 12 patients who had thrombotic events, eight patients were on concurrent aspirin at the time of thrombosis, and 11 were on cytoreductive therapy. One patient was on a reduced DOAC dose at the time of thrombotic event.

There were 28 bleeding events that occurred in patients with MPN on a DOAC; six events were major bleeds, which resulted in death in four patients, and 22 events were clinically relevant non-major bleeding (CRNMB) events.

• Bleeding events included gastrointestinal bleeds (n = 11), hematoma (n = 7), epistaxis, intracranial bleeds, and other bleeding evens (n = 3, each).
• The 1-year cumulative incidence of bleeding on DOAC was 12.3% (6.4–18.2%).
• There was no significant difference in bleeding incidence between the VTE and AF groups (15 and 13 events, respectively; p = 0.86).

Risk factors for thrombosis and bleedings

A univariate analysis was done to evaluate 16 risk factors for thrombosis and bleeding. These risk factors included age, indication for anticoagulation use, history of thrombosis, aspirin use, MPN subtype, driver mutation, type and dosing of DOAC, gender, hematologic parameters at DOAC initiation, body mass index, concomitant thrombophilia, use of cytoreduction, and prior anticoagulation use.

Prior history of thrombosis (hazard ratio [HR], 5.0; 1.4–25.0) and use of dabigatran or edoxaban (HR, 5.0; 1.3–19.5) were significantly associated with increased risk of thrombosis. This remained significant on multivariate analysis.

Age <65 years was also a significant risk factor for recurrent thrombosis on DOAC in multivariate analysis (p = 0.04).

Patients with white blood cell (WBC) counts in the top two quartiles of the cohort had a significantly increased risk of bleeding (HR, 3.03; p = 0.01). No other significant risk factors were identified for increased bleeding risk; however, use of dabigatran or edoxaban trended towards increased bleeding risk (p = 0.07).

Conclusion

DOAC use in patients with MPN at this tertiary center showed variations in practice patterns. VTE was found to be the primary indication for DOAC use, followed by AF. The duration of anticoagulation was indefinite in most patients. Whilst the incidence of thrombosis on DOACs in this study is similar to that demonstrated in other studies, the bleeding risk of 12.3% is considerably higher. Patients with MPN represent a complex patient group and rigorous further evaluation of DOAC use in this population is required to better assess their safety and make therapeutic recommendations.