The impact of low-dose anticoagulation therapy on peripheral artery disease: insights from the VOYAGER trial

Even with modern medical therapy, patients with peripheral artery disease (PAD) remain at very high risk for major adverse cardiovascular events (MACE) and major adverse limb events (MALE). Several previous studies aimed at improving PAD-related outcomes have evaluated more intensive single antiplatelet therapies, dual antiplatelet regimens, and/or the addition of anticoagulation therapy. As elegantly reviewed by Kaplovitch and Anand,1 these attempts were largely underwhelming, and most trials failed to achieve the fine balance of reducing MACE/MALE without increasing bleeding risk. Recently, the COMPASS trial showed that in chronic stable PAD patients, low-dose rivaroxaban (2.5 mg twice daily) added to aspirin potently reduced both MACE [5% vs. 7%, hazard ratio (HR) 0.72] and MALE (1% vs. 2%, HR 0.54) compare to aspirin alone.2 This was accompanied by an increase in major bleeding, but not fatal bleeding or intracranial haemorrhage. Despite the fact that this trial was stopped early for its favourable net clinical benefit, the uptake of low-dose rivaroxaban amongst chronic PAD patients has remained low, presumably due to cost concerns and/or the perception that stable outpatients may not ‘need’ additional risk reduction, especially if it comes at the price of increased bleeding. The VOYAGER trial set out to determine if low-dose rivaroxaban would demonstrate an acceptable risk–benefit profile in PAD patients with more advanced disease who require revascularization (a group that might also be at higher bleeding risk).3 This international, multicentre, double-blinded trial examined low-dose rivaroxaban (2.5 mg twice daily) plus aspirin vs. placebo plus aspirin in symptomatic PAD patients following either endovascular or surgical revascularization. The primary efficacy endpoint was a composite of clinically relevant MACE and MALE components, and the principal safety outcome was major bleeding. Patients were enrolled within 10 days of a successful intervention and a second antiplatelet agent (clopidogrel) was allowed to be administered for up to 6 months at the discretion of the investigator. The trial included 6564 subjects from 34 countries who were followed for a median of 28 months with only ∼1% of patients lost to follow-up. Results showed significantly lower events in the rivaroxaban group for the primary outcome (17.3% vs. 19.9%, HR 0.85), with a particularly notable improvement in acute limb ischaemia compared to the placebo group (5.2% vs. 7.8%, HR 0.67). The Kaplan–Meier curves separated within the first 3 months and continued to improve over time with a number needed to treat of 39 at the terminal endpoint. Secondary outcomes including the need for recurrent revascularization (‘take backs’) and re-hospitalization (‘bounce backs’) were also significantly lower, though there was no improvement in all-cause mortality (HR 1.08, P = 0.34). Thrombolysis in myocardial infarction (TIMI) major bleeding (2.65% vs. 1.87%, P = 0.07) and fatal bleeding (six in each group) did not differ significantly, although the secondary safety outcome of International Society on Thrombosis and Haemostasis (ISTH) major bleeding was clearly higher in the rivaroxaban group at 3 years (5.94% vs. 4.06%, HR 1.42, P = 0.007). Importantly, there was no interaction by type of intervention performed (open vs. endovascular), nor by clopidogrel usage. The authors estimated that for every 10 000 patients treated for a year with low-dose rivaroxaban plus aspirin, 181 MACE/MALE events would be prevented at the cost of 29 bleeding events. There is both clinical and genetic evidence to support the rationale for the use of low-dose anticoagulation in this population. Narula et al.4 described the histopathological characteristics of lower extremity arterial occlusive disease and cited differences between the femoral and popliteal (FEM-POP) and infrapopliteal (INFRA-POP) vessels in PAD. There were significantly more atherosclerotic lesions noted in the FEM-POP arteries, while the INFRA-POP arteries were frequently occluded with acute or chronic thrombi. Interestingly, thrombotic occlusion was more likely to occur in the absence of significant atherosclerosis in the INFRA-POP arteries, where layered thrombi suggested repeated thrombo-embolic events. This is unlike the coronary arteries where atherosclerotic plaque rupture (and associated exposure of tissue factor) may play a greater role in promoting clinical events. It is also known that PAD shares some, but not all, pathophysiological determinants with other vascular disorders, such as coronary artery disease (CAD) and cerebrovascular disease (CVD).5 For example, Klarin et al.6 recently performed an elegant genomics study using data from the Million Veteran Program (MVP) and UK Biobank, which highlighted the similarities and differences among CAD, CVD, and PAD patients. Using genome-wide and phenome-wide association studies, the authors identified four loci which were uniquely associated with PAD, including a novel association with the key coagulation cascade member, factor V (F5 p. R506Q). This variant is associated with venous thromboembolism as it renders factor V resistant to degradation, thus causing carriers to become hypercoagulable. These human genetics findings may explain the role of increased thrombosis in the pathology of PAD and support the rationale for using low-dose anticoagulants as a key adjunct therapy. Regarding economic feasibility, Cowie et al.7 recently presented a Markov model demonstrating the cost effectiveness of lifetime treatment with rivaroxaban in combination with aspirin for stable chronic PAD patients based on COMPASS results. However, the VOYAGER trial showed that PAD patients requiring acute intervention also have a lower rate of subsequent surgical re-intervention and re-hospitalization when prescribed rivaroxaban therapy (with minimal perioperative bleeding). These data suggest that future economic modelling might show even greater cost effectiveness amongst this at-risk group, particularly in light of their high rate of recurrent limb events and medical resource consumption.8 As researchers consider future studies related to low-dose anticoagulants in PAD, a clear focus is emerging on identifying those who are most likely to benefit from therapy, and those least likely to have a bleeding complication. Unfortunately, no genotyping was performed in the VOYAGER trial, but forward-looking pharmacogenomics efforts may identify individuals predisposed to thrombotic complications that could be prioritized for treatment. Additionally, powerful new machine learning approaches (especially those performed in an automated fashion within the electronic health record), should help us build more accurate disease classification and prediction models, thus flagging patients at risk for haemorrhage who may elect to avoid therapy.9 Taken together, the VOYAGER study is a well-designed and executed trial which established the concept that low-dose rivaroxaban added to aspirin can reduce MACE and MALE without increasing fatal bleeding in symptomatic PAD patients requiring revascularization. Mechanistic and genetic evidences provide a plausible rationale for targeting the coagulation cascade, and the overall favourable net clinical benefit observed with this approach suggests it may become standard of care for eligible patients undergoing lower extremity revascularization procedures. Conflict of interest: N.J.L. has received consulting fees from Janssen, Sanofi, and Bristol Meyers Squibb. This study was supported by the National Institutes of Health (R35 HL144475) and the Anticoagulation Forum (2020 Ansell Fellowship Program) to N.J.L. Biography: Dr Eri Fukaya, MD, PhD, is a Clinical Assistant Professor and Vascular Medicine specialist in the Division of Vascular Surgery at Stanford University. Her interests are in clinical and translational research on the diagnosis, treatment, and prevention of peripheral artery disease and chronic venous disease. Biography: Dr Nicholas J. Leeper, MD, is a Professor of Medicine and Surgery and Chief of Vascular Medicine at Stanford University. He is a vascular biologist and geneticist interested in developing new translational therapies for patients with atherosclerotic cardiovascular disease.