Lesson learned from the COMPASS and VOYAGER PAD trials: dual pathway inhibition as a new antithrombotic paradigm?

This editorial refers to ‘Low dose rivaroxaban and aspirin among patients with peripheral artery disease: a meta-analysis of the COMPASS and VOYAGER trials’, by S.S. Anand et al. Peripheral artery disease (PAD) is one of the most common cardiovascular disease worldwide, with rising incidence and prevalence and a high burden of morbidity and mortality.1 Antiplatelet therapy represents a cornerstone of the medical treatment of PAD, with long-term single antiplatelet therapy (with either aspirin or clopidogrel) recommended by the 2017 ESC Guidelines for symptomatic, stable PAD patients.1 In recent years, dual pathway inhibition (DPI)—i.e., a strategy which combines an antiplatelet and an anticoagulant drug—has emerged as a promising approach to improve the treatment of several cardiovascular diseases2 and represented a long-awaited major advancement for the medical treatment of PAD, as demonstrated by the COMPASS and VOYAGER PAD trials.3,4 The COMPASS enrolled patients with stable cardiovascular disease [i.e. coronary artery disease (CAD), PAD, or both], randomized to Rivaroxaban 2.5 mg twice daily plus aspirin, Rivaroxaban 5 mg twice daily, or aspirin alone.3 Later on, the VOYAGER PAD trial randomized patients with lower extremity arterial disease (LEAD) with recent peripheral revascularization to either Rivaroxaban 2.5 mg twice daily plus aspirin or aspirin alone (the use of Clopidogrel was allowed, for up to 6 months, upon investigator choice).4 In both trials, the DPI strategy significantly reduced the primary efficacy outcomes compared to aspirin, although leading to an increase of major bleeding in the COMPASS trial, and ISTH-defined (but not TIMI-defined) major bleeding in the VOYAGER PAD trial. In their meta-analysis, Anand et al.5 expand these findings, by combining the subgroup of patients with PAD randomized to DPI or aspirin in the COMPASS trial, with the data from the VOYAGER PAD. The results showed that treatment with low dose rivaroxaban plus aspirin reduced the composite outcome of CV death, myocardial infarction, ischaemic stroke, acute limb ischaemia, or major vascular amputation by 21% compared to aspirin alone, with moderate heterogeneity between trials. A significantly increased risk of ISTH major bleeding, but not severe bleeding (fatal or symptomatic bleeding into a critical organ), was also observed among patients treated with DPI. Noteworthy, similar results were observed when analysing only the subgroup of patients with LEAD: major adverse limb events (including acute limb ischaemia or major vascular amputation) were significantly reduced by 27% in this subgroup, with no heterogeneity between the two studies. According to their estimates, the authors calculated that for every 1000 LEAD patients treated with low dose rivaroxaban and aspirin, 13 major cardiovascular or limb events are prevented, at the cost of eight more major bleeds, one of which being a fatal bleed. This meta-analysis comes with some interesting findings and few shortcomings. The results certainly contribute to our knowledge on the efficacy and safety of a DPI strategy in patients with PAD (and particularly LEAD), showing that the effect is consistent across different and heterogeneous clinical scenarios (from stable PAD, to LEAD patients that underwent recent revascularization). On the other side, the study-level design of this meta-analysis limited the ability to explore individual-level determinants of the ischaemic and bleeding risk on a much larger cohort of LEAD patients than those enrolled in the original trials, including those factors—perhaps responsible for some of the between-study heterogeneity—associated with the different clinical scenario explored by the two trials. Taken together, these findings confirm that DPI represents a rational choice in patients with PAD; however, the increase in the risk of bleeding still represents a concern and may limit its broad applicability in real-world patients—especially those with high haemorrhagic risk. Indeed, findings from the large REACH registry showed that roughly one-third of PAD patients (and slightly less than 40% patients with PAD and CAD) would have been ineligible to enrolment in the COMPASS, with a high bleeding risk being the most frequent reason for exclusion.6 The ongoing XATOA registry study will provide further insights on the adoption of a DPI-based regimen in real-world patients with PAD, CAD, or both, and will describe the incidence of ischaemic and bleeding events.7 These data, along with other large observational studies, are urgently needed to shed light on the efficacy and safety of DPI in the real-world setting, and to inform treatment choices. For now, an optimal approach to antithrombotic therapy in PAD patients would ideally require an individualized evaluation of the risk–benefit profile, and an accurate stratification of the ischaemic and bleeding risks. Unsurprisingly, the 2021 ESC consensus on antithrombotic treatment in aortic and peripheral arterial diseases reports that DPI may be considered for patients with asymptomatic LEAD and another atherosclerotic disease, and should also be considered for LEAD patients that are symptomatic or undergo revascularization. For all the indications, the consensus reports that the DPI strategy should be considered in the absence of high bleeding risk, thus reinforcing the need for a careful evaluation of the risk-benefit balance in this clinical scenario.8 Unfortunately, comprehensive risk assessment models are still lacking in this setting. However, the application of pre-specified criteria for the risk stratification showed promising results in the REACH study, in which patients with multiple risk factors showed a disproportionate increase in the incidence of ischaemic events compared to serious bleedings, thus expecting a more favourable risk–benefit profile from the adoption of DPI.9 This confirms that risk stratification can identify subgroups of patients with a potentially better risk–benefit profile for DPI, although further studies are required to confirm these findings. While the results of this meta-analysis are in line with the 2021 ESC Consensus recommendations, the future of antithrombotic treatment of PAD still depends on those unmet needs that await responses (Figure 1). COMPASS and VOYAGER PAD marked a turning point in our understanding and treatment of atherosclerotic disease, and the lessons learned from these trials provided solid ground to establish DPI as a potential new paradigm, beyond platelet inhibition, for the secondary and tertiary prevention of cardiovascular disease. However, we still need to find reliable and reproducible tools for accurate stratification of ischaemic and haemorrhagic risks, in order to identify the population most likely to benefit from the DPI-based approach. Moreover, further research is needed to identify the best DPI-based strategy that maximizes the benefit-risk ratio, thus improving both safety profile and ischaemic risk protection. Novel anticoagulant agents, including those targeting Factor XI, may represent a safer option in terms of bleeding, thus representing interesting candidates to expand our options for DPI-based approaches in the future.10 Finally, the role of thromboinflammation on the progression of atherosclerotic diseases, as well as the anti-inflammatory properties of antithrombotics, and their influence on the natural history of cardiovascular disease, still represent an exciting—although largely unknown—research area.11 Disentangling the interplay between thrombosis, inflammation, and antithrombotics is crucial to refine our treatment strategies and represents the keystone of cardiovascular disease prevention. While all these issues are still open, the results available so far likely represent only the beginning of a long, promising, and hopefully successful road for DPI. State of the art and future perspectives on the use of dual pathway inhibition in patients with PAD. State of the art and future perspectives on the use of dual pathway inhibition in patients with PAD. Conflict of interest: S.B. received research grant from MSD, outside the topic of this manuscript. The other authors have nothing to disclose. The opinions expressed in this article are not necessarily those of the Editors of the European Journal of Preventive Cardiology or of the European Society of Cardiology.