Comparing Two Types of Blood Thinners to Prevent Blood Clots in Patients with Cancer – The CANVAS Trial [Internet]

Excerpt

Background: Venous thromboembolic events (VTEs) are common in patients with cancer, and long-term anticoagulation with a low-molecular-weight heparin (LMWH) is recommended. The effectiveness of a direct oral anticoagulant (DOAC) compared with an LMWH for preventing recurrent VTE in patients with cancer was uncertain at the time the trial began.

Objective: This pragmatic trial compared DOAC-based and LMWH-based anticoagulation strategies for prevention of recurrent VTE in patients with cancer.

Methods: We conducted an unblinded, hybrid, comparative effectiveness, noninferiority trial with both a randomized and a preference cohort. Between December 2016 and April 2020, 638 participants were randomized and followed for 6 months or until death at 67 US oncology practices. Patients had a diagnosis of cancer, including any invasive solid tumor, lymphoma, multiple myeloma, or chronic lymphocytic leukemia, and a diagnosis of VTE within 30 days of enrollment. Patients who accepted randomization were assigned in a 1:1 ratio to receive either a DOAC or an LMWH. Patients who declined randomization selected their preferred anticoagulation approach (preference cohort). Physicians and patients could choose among any DOAC or any LMWH, and starting doses were suggested but not mandated. The primary study outcome was the recurrent VTE event rate at 6 months. The study was designed to establish noninferiority of anticoagulation with a DOAC vs an LMWH, defined by the upper limit of the 2-sided 90% CI for the difference of a DOAC relative to an LMWH of less than 3%. The minimal clinically important difference (MCID) was developed with input from stakeholders, including physicians, nurses, and patients. The width of the MCID that would be acceptable to physicians (oncologists) was ascertained through a series of meetings held at the Alliance for Clinical Trials in Oncology meeting in Chicago during the protocol-development process. The meeting asked open-ended questions and presented stakeholders with a series of options. The study team reviewed these thresholds and recommendations during protocol design. The primary analysis was conducted for randomized participants who received assigned treatment. Secondary outcomes included rates of death, major bleeding, and health-related quality of life during the 6-month study period.

Results: Between December 2016 and December 2017, 137 participants declined randomization and were treated according to their preference; they were followed for 6 months or until death. The preference cohort was closed to accrual when prespecified thresholds for accrual to the preference as opposed to the randomized cohort were met. The final date of follow-up was November 2020. Among the 330 participants randomized to DOAC and the 308 randomized to LMWH who received study protocol-assigned therapy, 6.1% of participants in the DOAC group and 8.8% of participants in the LMWH group (a difference of −2.7% [90% CI, −6.1% to 0.7%]) had recurrent VTE within 6 months. The noninferiority criterion for the primary end point at 6 months was met. The statistical analysis plan specified that superiority testing was to be performed sequentially if noninferiority were met. The study was underpowered to assess superiority given attenuated accrual because of the onset of the COVID-19 pandemic and publication of another trial demonstrating noninferiority of DOAC therapy. At least 1 episode of major bleeding occurred in 5.2% of participants in the DOAC group and 5.6% of participants in the LMWH group (90% CI, −3.3 to 2.5). Patients with cancer randomized to DOAC therapy had similar rates of death to those assigned to LMWH therapy. Neither quality of life nor patients' perceptions of treatment benefits differed between groups. Patients on DOACs reported lower treatment burden, however, than those assigned to LMWHs. Participants in the preference cohort who declined randomization were more likely to receive DOAC than LMWH anticoagulant therapy.

Conclusions: Among adult patients with cancer and VTE, the use of a DOAC was noninferior to an LMWH for preventing recurrent VTE, with similar rates of bleeding over 6 months of follow-up.

Limitations: This study has several limitations. First, neither participants nor their physicians were blinded to treatment assignment. Second, superiority was not assessed because of lower-than-planned enrollment. Third, the pragmatic trial design allowed treatment with any anticoagulant for up to 30 days from the time of VTE diagnosis to start of therapy to allow procurement of assigned treatment; although this design can be viewed as a limitation, it simulates daily clinical practice and the impediments to care that clinicians face, including the need for prior authorizations for drug coverage, clinic visit scheduling, and transitions of care from emergency departments and inpatient settings.