Anti-thrombotic options for secondary prevention in patients with chronic atherosclerotic vascular disease: what does COMPASS add?

The publication of the recent COMPASS study1–3 raises key clinical questions: What is the potential for clinically worthwhile improvements in outcomes, in the context of currently proven secondary prevention measures? What are the therapeutic implications of atherothrombotic disease across different vascular territories? Can we extrapolate from studies in acute vascular disease to long-term chronic vascular disease? What does the COMPASS regimen add to current secondary prevention therapies for chronic arterial vascular disease? How do the benefits and risks of the combination of very low dose of rivaroxaban and aspirin compare with those seen with other anti-thrombotic combinations in patients with stable vascular disease? How does the new dual pathway anti-thrombotic therapy compare with the impact of other accepted pharmacological secondary prevention measures in stable vascular disease [hypertension control, reducing low density lipoprotein (LDL) cholesterol, angiotensin-converting enzyme (ACE) inhibitors]? For the prevention of ischaemic clinical events in those with chronic atherosclerosis (secondary prevention), current guidelines recommend lifestyle change (tobacco avoidance, a healthy diet, and regular exercise), drugs (LDL cholesterol lowering primarily with statins, beta-blockers, glucose control, angiotensin-converting enzyme inhibitors/blockers, and anti-platelet drugs), and coronary revascularization in high-risk individuals. These secondary prevention strategies are based on robust evidence and can substantially reduce cardiovascular (CV) complications. However, the residual risks of events due to cardiovascular disease (CVD) remain significant for patients with existing (CVD)4–6 even among populations with very high adherence to prescribed therapies. For example, among patients with multiple CV risk factors or established disease and well treated with secondary prevention therapies, contemporary studies indicate a 15–20% rate of CV death, myocardial infarction (MI), or stroke at 3 and 4 years, respectively.7 , 8 ‘The implications of these observations are that there is scope for further improvement and potential opportunities for novel strategies’. Atherothrombotic disease underlies the diverse clinical manifestations that are determined by the vascular territory affected, the extent of vascular obstruction, and the characteristics of the perfused territory. Increased risks for vascular complications are observed in all vascular territories affected (coronary, cerebrovascular, or peripheral arteries), and the clinical complications are principally due to thrombotic obstructions of specific vessels.5 , 9–11 The pathophysiological mechanisms include the direct consequences of vascular obstruction to a coronary or peripheral artery, or the embolic consequences of disruption of an atheromatous or an intimal lesion with thrombosis and peripheral embolization.5 , 9–14 In addition, patients may sustain the late consequences of myocardial, cerebral, peripheral limb, and other vascular bed injury. Although the clinical presentation may be the result of vascular occlusion in one arterial vascular bed, subsequent events can be the result of occlusion of other vascular beds. For example, peripheral artery occlusion can present as intermittent claudication, but subsequent complications can occur in a coronary artery initiating an MI or in a cerebrovascular artery leading to a stroke.5 , 9–11 Many key risk factors are similar for these different clinical manifestations across vascular beds,6 and similar secondary prevention measures are recommended in guidelines.15 , 16 However, the impact of some therapies may differ depending on whether the primary clinical event is due to coronary, cerebrovascular, or peripheral artery disease (PAD). For example, the potent anti-platelet therapy, ticagrelor was clearly beneficial after an acute coronary syndrome (ACS) and in those with a history of MI17 , 18 but benefits have not been observed in patients with PAD19 or cerebrovascular disease.20 ‘The implications are that while atherothrombotic disease manifests across diverse arterial vascular territories, therapies may differ in their clinical impact on different arterial beds’. Several anti-thrombotic combinations have been tested in the context of an acute vascular event and in most instances following presentation with MI or ACS. These include combined anti-platelet therapies,21 , 22 more potent P2Y12 antagonists combined with aspirin23 , 24 and other anti-thrombotic combinations.25–27 These treatments are administered during or soon after the acute vascular event, with continued treatment for 12 months or longer.17 , 18 , 20–28 In the context of an acute vascular event, thrombus may still be present in the artery, and there may be evidence of plaque disruption at more than one site. In addition, mechanical interventions (e.g. stents) require specific anti-thrombotic therapies. Although activation of platelets and of the coagulation pathways is implicated in thrombus formation in both acute and chronic settings, their relative contributions may differ. Potent anti-platelet combinations have been shown to be particularly effective in preventing stent thrombosis and in reducing vascular events during the in-hospital phase and in the early months after an ACS. However, they also increase bleeding both in the acute and long-term phases. For example, a meta-analysis of prolonged dual anti-platelet therapy (DAPT) showed an increase in major bleeding with DAPT (1.85% vs. 1.09%, risk ratio (RR) 1.73, P = 0.004).29 ‘This suggests that anti-thrombotic regimens designed for the acute phase (e.g. ACS) may not necessarily be optimal in terms of the balance of risk vs. benefit for chronic stable vascular patients’. Single anti-platelet therapy (predominantly aspirin) is the currently accepted anti-thrombotic long-term therapy for patients with stable vascular disease15 , 16 , 30 , 31 and is used by the majority of patients.32 The concept of combining anti-platelet therapy and anti-coagulation is not new. For example, the WAVE study of warfarin and anti-platelet therapy vs. anti-platelet therapy alone in patients with chronic stable PAD did not reduce major adverse cardiovascular events (MACE) or major adverse limb events (MALE), but increased life threatening and intracranial bleeds.12 In a meta-analysis of 10 trials (5938 patients) of warfarin plus aspirin vs. aspirin in patients with recent MI, there was no impact on mortality, but MI [RR 0.56, confidence interval (CI) 0.48–0.69], and stroke (RR 0.46, CI 0.27–0.77) were reduced.13 However, there was an increased risk of major (RR 2.48, CI 1.67–3.68) and minor bleeding (RR 2.65, CI 2.14–3.69). Because of the increased bleeding risk, including that of intracranial haemorrhage (ICH), the lack of an effect on mortality, and the need for monitoring the intensity of anti-coagulation, combined use of warfarin and aspirin in those with stable vascular disease has not been widely adopted in clinical practice. Clopidogrel has been compared with aspirin in patients with stable CV disease in the CAPRIE study.33 In 19 185 patients with symptomatic atherosclerotic disease clopidogrel reduced the risk of MACE (CV mortality/MI/stroke) by 8.7% compared to aspirin (5.83% vs 5.32%; P = 0.043), but there was no reduction in mortality or stroke. The benefit of clopidogrel was most evident in the PAD subgroup but less clear among those with prior coronary or cerebrovascular disease. Additionally, there was no significant reduction in the secondary outcome of ischaemic stroke, MI, vascular death, or amputation (P = 0.076).33 Clopidogrel plus aspirin was tested against aspirin in the CHARISMA trial.14 In 15 603 patients with clinically evident CV disease or CV disease risk factors without clinically evident disease, adding clopidogrel to aspirin did not significantly reduce the risk of stroke, MI, or CV death compared with aspirin alone (hazard ratio (HR) 0.93, CI 0.83–1.05).14 However, in a secondary analysis, among those with prior CV disease the combination of clopidogrel and aspirin produced a significant benefit in MACE (6.9% clopidogrel plus aspirin vs. 7.9% with aspirin plus placebo; RR 0.88; CI 0.77–0.998; P = 0.046).14 There were numerically more severe bleeds using the GUSTO criteria (1.7% vs. 1.5%), and fatal bleeds (0.3% vs. 0.2%) and significantly more moderate bleeds (2.1% vs. 1.3%, RR 1.62, CI 1.27–2.08; P < 0.001) among those treated with combination therapy. The TRA2P34 and PEGASUS1 8 trials were conducted more recently and in the context of more aggressive secondary prevention measures including widespread use of statins and ACE inhibitors. In the TRA2P study, 26 449 patients with a history of MI, ischaemic stroke, or PAD were randomized to vorapaxar (2.5 mg daily) or placebo on background therapy of physician chosen single or dual anti-platelet therapies.34 Cardiovascular death, MI, or stroke occurred in 9.3% of the vorapaxar patients and 10.5% of the placebo patients (HR 0.87, CI 0.80–0.94; P < 0.001). GUSTO major or severe bleeding was significantly increased (4.2% vs. 2.5%, HR 1.66, CI 1.43–1.93, P < 0.001) and intracranial bleeds were significantly increased (1.0% vs. 0.5%, HR 1.94, CI 1.39–2.70, P < 0.001) among patients randomized to vorapaxar.34 In the PAD subgroup, vorapaxar reduced limb events including acute limb ischaemia and urgent revascularization.35 In PEGASUS (21 162 patients who had an MI 1–3 years previously), either 60 mg b.i.d. or 90 mg b.i.d. of ticagrelor plus aspirin was tested against aspirin plus placebo. The 60 mg b.i.d. dose of ticagrelor decreased the risk of CV death, MI, or stroke, (HR 0.84, CI 0.74–0.95; P = 0.004) and similarly for the 90 mg b.i.d. dose of ticagrelor (HR 0.85, CI 0.75–0.96; P = 0.008) vs. aspirin and placebo.18 The rates of thrombolysis in myocardial infarction (TIMI) major bleeding were higher with ticagrelor (2.3% with 60 mg, 2.6% with 90 mg) than with placebo (1.06% P < 0.001 for each comparison). Bleeding leading to study drug discontinuation and bleeding leading to transfusions were each increased three- to five-fold with the dual anti-platelet combinations (HR 3.08–5.79), but there was no significant difference in intracranial or in fatal bleeding.18 , 36 Ticagrelor (both doses combined) in the patients with PAD reduced MALE (HR 0.65, CI 0.44–0.95; P = 0.026). In the SOCRATES trial, 13 199 patients were randomized to either ticagrelor or aspirin within 24 h of ischaemic stroke or transient ischaemic attack (TIA). There was no significant difference in safety or efficacy.20 In the EUCLID trial, 13 885 patients with symptomatic PAD were randomized to ticagrelor or clopidogrel, in the absence of aspirin. Ticagrelor did not reduce MACE compared with clopidogrel (HR 1.02, CI 0.92–1.13), and there was no difference in acute limb ischaemia nor in bleeding.19 The recent COMPASS trial randomized 27 395 patients with stable atherosclerotic vascular disease (coronary and/or PAD) in whom neither DAPT nor full anti-coagulation were needed.37 Two relatively ‘low’ doses of a non-vitamin K oral anti-coagulant (rivaroxaban) were tested against aspirin: either rivaroxaban 2.5 mg b.i.d. (a quarter of the dose for stroke prevention in atrial fibrillation) in combination with aspirin, or rivaroxaban alone, 5 mg b.i.d. (half of the dose used for stroke prevention). The combination of rivaroxaban 2.5 mg b.i.d. and aspirin was previously tested.1 , 37 Compared with aspirin alone, the combination significantly reduced CV death, MI, or stroke (4.1% vs. 5.4%, HR 0.76, 95% CI 0.66–0.86; P < 0.0001), but it increased major bleeding 3.1% vs. 1.9%, HR 1.70, 95% CI 1.40–2.05; P < 0.0001).1 , 37 There was no significant increase in intracranial or fatal bleeds. The rate of deaths was lower in the combination arm 313 (3.4%) compared with 378 (4.1%) in the aspirin arm (HR 0.82, 95% CI 0.71–0.96; P = 0.01). There was a trend for reduced MI, and when added to sudden death and resuscitated cardiac arrest or unstable angina (all coronary events), this post hoc composite was significantly reduced.3 In addition, in the PAD subgroup there was approximately a halving of MALE including amputations.2 Major adverse limb events included the development of new acute or chronic limb ischaemia or amputations with a vascular cause. In the rivaroxaban plus aspirin arm, MALE events were reduced from 2.2% to 1.2% (HR 0.54, CI 0.35–0.89) and limb amputations were markedly reduced (HR 0.30, CI 0.11–0.80). The reductions in all amputations for vascular complications were also consistent (HR 0.40, CI 0.20–0.79; P = 0.007).2 In contrast, rivaroxaban alone compared with aspirin did not significantly reduce the primary outcome, and increased major bleeding.1 Considering the impact of the combined COMPASS regimen vs. earlier anti-thrombotic strategies, the magnitude of the relative risk reduction in CV death, MI, and stroke was apparently greater for the COMPASS rivaroxaban 2.5 mg b.i.d. plus aspirin regimen, than for the other therapies (Table 1). In any comparison of effects of different agents in separate trials, caution must be exercised, but in COMPASS the combined regimen also demonstrated a significant reduction in CV death and all-cause mortality (which was not seen with the other strategies) (Table 1). Reductions in MALE and in stroke were also significant for the COMPASS combined regimen (Table 1). Efficacy: anti-thrombotic secondary prevention strategies in stable vascular disease CV, cardiovascular; MI, myocardial infarction; NS, non significant. Ischaemic stroke. Acute limb ischaemia, revascularization for ischaemia, or amputation: ticagrelor 90 mg hazard ratio (HR) = 0.54 (0.34–0.85) P = 0.008, ticagrelor 60 mg HR = 0.75 (0.50–1.14), P = 0.18. Acute limb ischaemia. Efficacy: anti-thrombotic secondary prevention strategies in stable vascular disease CV, cardiovascular; MI, myocardial infarction; NS, non significant. Ischaemic stroke. Acute limb ischaemia, revascularization for ischaemia, or amputation: ticagrelor 90 mg hazard ratio (HR) = 0.54 (0.34–0.85) P = 0.008, ticagrelor 60 mg HR = 0.75 (0.50–1.14), P = 0.18. Acute limb ischaemia. All anti-thrombotic strategies increase bleeding (Table 2). The absolute increase in major bleeding in COMPASS was 1.2% over 23 months (∼0.63% per annum).1 However, the temporal patterns of benefit and bleeding risk may differ with different strategies. In PEGASUS, TIMI major bleeding was increased with ticagrelor 60 mg at each time landmark, but with the greatest hazard in the first year (Year 1: HR 3.22; Year 2: HR 2.07; Year 3: HR 1.65).38 The COMPASS trial ended earlier than expected following a recommendation (for efficacy) from the Data Safety Monitoring Committee. In COMPASS, dual treatment showed excess major bleeding mainly in the first year of treatment (HR 2.32, CI 1.75–3.07) but without a significant excess thereafter (Year 2: HR 1.19, CI 0.84–1.68; Year 3: HR 1.05, CI 0.63–1.75).3 Bleeding: anti-thrombotic secondary prevention strategies in stable vascular disease ICH, intracranial haemorrhage; NS, non significant. Estimated from data provided in the article. GUSTO moderate bleeding in symptomatic patients. TIMI major bleeding. Non-coronary artery bypass grafting-related TIMI major bleeding. GUSTO moderate or severe. Bleeding: anti-thrombotic secondary prevention strategies in stable vascular disease ICH, intracranial haemorrhage; NS, non significant. Estimated from data provided in the article. GUSTO moderate bleeding in symptomatic patients. TIMI major bleeding. Non-coronary artery bypass grafting-related TIMI major bleeding. GUSTO moderate or severe. Randomized trial evidence suggests that adding an anti-platelet agent or an anti-thrombin to aspirin may be of benefit in patients with stable CVD, particularly those with coronary artery disease (CAD) or PAD. On a review of the evidence one option is to use clopidogrel plus aspirin based on the CHARISMA trial.14 However, based on the magnitude of benefit in the trial data, the preferred options for clinicians to consider are the PEGASUS regimen (ticagrelor 60 mg b.i.d. plus low dose aspirin for patients who are 1–3 years post-MI and have high-risk features such as age, PAD, diabetes, chronic kidney disease, recurrent MIs, and multivessel disease),18 , 36 or the COMPASS regimen combining very low dose rivaroxaban and low dose aspirin in a broad range of stable CAD or PAD patients who have high-risk features.1–3 The latter regimen is particularly attractive given its impact on reducing CV mortality and all-cause mortality. It also markedly reduced MALE. The COMPASS regimen1–3 and ticagrelor 60 mg (PEGASUS)18 both reduced ischaemic strokes without significant increases in ICH. The combination of efficacy and mortality improvements with the dual pathway COMPASS regimen makes this option clinically attractive. More intensive anti-thrombotic regimens increase the risks of bleeding (Table 2) so more intensive anti-thrombotic therapy is not for all patients with stable vascular disease. Fatal bleeding and ICH are not significantly increased (except ICH in TRA2P) but other major bleeding is increased (ranging from 43% relative increase with aspirin plus clopidogrel in CHARISMA, to 70% in COMPASS, to 169% for ticagrelor 90 mg plus aspirin in PEGASUS, Table 2). Thus risks, benefits, and net clinical benefits must be balanced for more potent anti-thrombotic regimens. In COMPASS the combination of major, critical organ bleeding, and outcome events demonstrated a net clinical benefit in favour of the rivaroxaban 2.5 mg b.i.d. plus aspirin strategy (4.7% CV death, stroke, MI, fatal bleeding, or symptomatic critical organ bleeding vs. 5.9% with aspirin alone (HR 0.80, CI 0.70–0.91). A retrospective analysis of the REACH registry suggests that approximately half of CAD and two-thirds of PAD patients would qualify for the COMPASS trial.39 The impact of the dual pathway COMPASS regimen is of at least similar magnitude to that seen with all other accepted secondary prevention regimens [aspirin, lipid lowering, blood pressure (BP) lowering, and ACE inhibitors] (Table 3 and Figure 1),40–44 and is additive to these treatments. The impact of the dual mechanism COMPASS regimen in the context of proven secondary prevention regimens (aspirin, lipid lowering, blood pressure lowering, and angiotensin-converting enzyme inhibitors). MACE, major adverse cardiovascular events. Comparison of the effects of guideline indicated secondary prevention pharmacological therapies for patients with vascular disease ACE, angiotensin-converting enzyme; BP, blood pressure; CV, cardiovascular; LDL, low density lipoprotein; MACE, major adverse cardiovascular events; MI, myocardial infarction; NS, non significant. Major coronary event. Non-fatal MI. Comparison of the effects of guideline indicated secondary prevention pharmacological therapies for patients with vascular disease ACE, angiotensin-converting enzyme; BP, blood pressure; CV, cardiovascular; LDL, low density lipoprotein; MACE, major adverse cardiovascular events; MI, myocardial infarction; NS, non significant. Major coronary event. Non-fatal MI. The implications of the COMPASS results for clinical practice are worth considering. Patients with stable vascular disease and additional risk features have substantial rates of CV events over time, including deaths, strokes, MIs, and MALE. This is despite currently applied lifestyle and secondary prevention measures (Figure 2). Addition of anti-thrombotic therapy (generic clopidogrel or ticagrelor or vorapaxar or very low dose rivaroxaban) to aspirin (compared with aspirin alone) reduces major adverse cardiac events, but the magnitude of treatment effects differ by type of event prevented and the types of patients who benefit, with the respective strategies, and the bleeding risks also differ. What is unknown is how the various strategies would compare in trials where they are directly compared. Such trials are desirable given the magnitude of the global public health burden of CVD. Diagram to illustrate the potential role of newer therapies in patients with elevated vascular risk despite current secondary prevention measures. Patients identified with chronic coronary or peripheral arterial disease and comorbidities or markers of elevated vascular risk (e.g. polyvascular disease, recurrent cardiovascular (CV) events, diabetes, heart failure, and renal dysfunction). Patients may also be identified following dual anti-platelet therapy (DAPT) therapy after myocardial infarction (MI), acute coronary syndrome (ACS), or stent implantation. Newer therapy options include proprotein convertase subtilisin kexin 9 (PCSK9) inhibitors in patients with persistent low density lipoprotein (LDL) elevation, modification of inflammation in those with elevated C reactive protein (CRP) (canakinumab, not yet or dual pathway COMPASS regimen yet BP, blood secondary prevention therapies may include reduction with proprotein convertase subtilisin kexin 9 (PCSK9) and/or modification of inflammation in the vascular to risks of to lower cholesterol and to inflammation are not necessarily but combined therapies need to be including their The dual mechanism COMPASS regimen dose with rivaroxaban 2.5 mg b.i.d. plus low dose aspirin as an anti-platelet is of at least similar or greater benefit to that seen in studies of other anti-thrombotic regimens in reducing the composite of CV death, MI, or stroke. it reduced MALE and CV and mortality. Although it increased the risk of bleeding, the net clinical benefit was The dual mechanism COMPASS regimen is of benefit to that seen with accepted secondary prevention regimens (aspirin, lipid lowering, lowering, and ACE and the potential clinical impact of using these proven drugs is The data from the respective trials need to be in the context of the populations the use of secondary prevention therapies and the increased use of such therapies over must be in trial but the time in the of the trials that the effects may be of and from and from from and and from and from and has from and from during the of the from The and other with and of and from the of the of and of and from the of and the of Cardiovascular and from the and from and from and from and from and and from the and from and the and from the and and from and in has and for from and