Use of risk scores to identify lower and higher risk subsets among COMPASS‐eligible patients with chronic coronary syndromes. Insights from the CLARIFY registry

Abstract Background The COMPASS trial showed a reduction of ischemic events with low‐dose rivaroxaban and aspirin in chronic coronary syndromes (CCS) compared with aspirin alone, at the expense of increased bleeding. Hypothesis The CHA2DS2VaSc Score, REACH Recurrent Ischemic (RIS), and REACH Bleeding Risk Score (BRS) could identify patients with a favorable trade‐off between ischemic and bleeding events, among COMPASS‐eligible patients. Methods We identified the COMPASS‐eligible population within the CLARIFY registry (>30.000 patients with CCS). High‐bleeding risk patients (REACH BRS > 10) were excluded, as in the COMPASS trial. Patients were categorized as low (0–1) or high (≥ 2) CHA2DS2VaSc; low (0–12) or intermediate (13–19) REACH RIS, and low (0–6) or intermediate (7–10) REACH BRS. Ischemic outcome was the composite of cardiovascular death, myocardial infarction or stroke. Bleeding was defined as serious bleeding (haemorrhagic stroke, hospitalization for bleeding, transfusion). Results The COMPASS‐eligible population comprised 5.142 patients with ischemic and bleeding outcome of 2.3 (2.1–2.5) and 0.5 (0.4–0.6) per 100 patient‐years, respectively. Patients with intermediate REACH RIS (n = 1934 [37.6%]) had the higher ischemic risk (3.0 [2.6–3.4]) with similar bleeding risk (0.5 [0.4–0.7]) as the overall population. Patients with low CHA2DS2VaSc (n = 229 [4.4%]) had a very low ischemic risk (0.6 [0.3–1.3]) with similar bleeding risk (0.5 [0.2–1.1]). Conclusions Intermediate REACH RIS identified potential optimal candidates for adjunction of low‐dose rivaroxaban while patients with low CHA2DS2VaSc score .appears unlikely to benefit from the COMPASS regimen. None of the three risk scores predicted the occurrence of serious bleeding.


| INTRODUCTION
In patients with chronic coronary syndromes (CCS) or peripheral arterial disease (PAD), antithrombotic treatment represents a cornerstone of medical therapy. This has been for a long time limited to single antiplatelet therapy alone. However, there remains strong evidence of a high residual thrombotic risk in both CCS and PAD patients. [1][2][3][4] In that sense, several antithrombotic regimens have been evaluated in order to reduce the occurrence of recurrent ischemic events, based on potent antiplatelet therapy, or on the combination of antiplatelet therapy with oral anticoagulant therapy. [5][6][7] The COMPASS 6 trial demonstrated the efficacy of a combination of low-dose rivaroxaban and aspirin in reducing ischemic events in a broad population of high-risk patients with CCS and/or PAD, but with a 70% relative increase in the risk of major bleeding. Recently, the ESC guidelines 8  In the present analysis, we used data from the CLARIFY Registry, a large international registry of more than 30 000 patients with CCS, to evaluate the performance of established ischemic and bleeding risk scores: the CHA 2 DS 2 VaSc score, the REACH Recurrent Ischemic Score (RIS) and the REACH Bleeding Risk Score (BRS) in order to attempt to identify patients with the most favorable trade-off between ischemic and bleeding events, among CCS patients eligible to COMPASS. ECG was recorded at each visit. The rationale, design, and long-term outcomes of CLARIFY registry participants have been described in detail. 9,10,12,13 To be eligible for enrolment, patients had to have any of the following: documented myocardial infarction or history of myocardial revascularization-either by percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) more than 3 months earlier-or documented coronary artery disease with coronary angiography showing at least one coronary stenosis of more than 50%, or chest pain with proven myocardial ischemia (using stress ECG, stress echocardiography, or myocardial imaging). Patients were excluded if they had been hospitalized for CV disease within the last 3 months, had a planned revascularization, or any condition hampering the participation or the 5-year follow-up, including advanced heart failure.
The study was conducted in accordance with the Declaration of Helsinki, and local ethical approval was obtained before recruitment. All patients gave written informed consent.  14,15 According to its validated use in AF, patients with CHA 2 DS 2 VaSc score of 0-1 were categorized in low-intermediate risk, and patients with ≥2 as high risk.

| The REACH Recurrent Ischemic Score
The REACH RIS, 16

| Outcomes definition
For the present analysis, the ischemic outcome was a composite of CV death, myocardial infarction (MI) or stroke. The bleeding outcome was "serious bleeding" a composite of bleeding leading to either admission to hospital and/or transfusion, or haemorrhagic stroke.
Both outcomes are expressed per 100 patients year. high REACH RIS, and thus will not be considered in the analysis.

| Bleeding according to risk scores
The rate of bleeding in the overall COMPASS-eligible population was 0.

| DISCUSSION
In the present analysis, we sought to evaluate the performance of three risk scores to identify within COMPASS-Eligible patients from the CLARIFY registry, those with the most favorable trade-off between ischemic and bleeding events. The main results of our analysis are that a low CHA 2 DS 2 VaSc score (0 or1) identifies, admittedly a small proportion of patients, but at very low ischemic risk and at similar bleeding risk. Thus, these patients appear unlikely to benefit from a combination of low-dose rivaroxaban and aspirin. On the other hand, patients with an intermediate REACH RIS had a markedly higher ischemic risk, but with no apparent increased risk of bleeding, and thus appear to be the optimal candidates for the COMPASS regimen.
Finally, all three risk scores used in this study demonstrated poor performance in identifying patients at higher or lower bleeding risk, and probably should not be used for evaluation of the risk of bleeding in COMPASS-eligible patients.
Bleeding risk remains a major concern when using antithrombotic therapies in clinical practice, with the fear of causing harm with an overt bleed while the ischemic events prevented are clinically "silent" and thus unrecognized, leading to a distorted perception of benefits and harms of treatment. There have been several attempts to identify patients with an optimal benefit-risk ratio for the prescription of low dose rivaroxaban and aspirin using simple clinical criteria such as diabetes, heart failure, number of diseased vascular beds or renal failure. 19,20 Additionally, we 21 and others 20 have found that the combination of multiple enrichment criteria resulted in a dramatic increase of ischemic risk without significant increase in bleeding risk suggesting that these patients may derive the greatest benefit from the COMPASS regimen. However, to date, the performance of risk scores has never been evaluated for this purpose.
This score-based approach for clinical management and risk stratification of patient has already proven to be effective in other clinical settings, such as the use of the GRACE score 22 for risk stratification in acute coronary syndromes, or the PRECISE-DAPT, 23 DAPT 24 or the ABC-CHD 25 scores in decision making regarding duration of dual antiplatelet therapy after PCI, and is recommended by guidelines. 8 The rationale for evaluating both the REACH RIS 16 and BRS 17 is that these two scores have been developed and externally validated in a large registry of more than 68 000 outpatients with chronic vascular disease, 18 somewhat similar to the COMPASS Trial population. The CHA 2 DS 2 VaSc score has been validated for stroke prediction in patients with AF but based on the individual components of this score and its ease of use in clinical practice, several studies have also demonstrated its value for predicting ischemic events or mortality in various settings, independent of the presence of AF. [26][27][28][29][30] If these three scores performed well in identifying patients at high or low risk of ischemic events in the present study, they did not allow accurate evaluation of the risk of bleeding. Such a discrepancy was unexpected, in particular for the REACH BRS but may be explained by the fact that serious bleeding events were relatively rare in the study population (108 events, 0.5 per 100 patients years), and that the COMPASS-eligible population accounted for approximately one out of three patients with CCS, and around half with chronic vascular disease, 31 with higher risk patients being excluded.
Interestingly, several items used in the calculation of these scores are redundant, and their performance only depends on the weighting of each item. The REACH RIS involves more items with a larger score range (from 0 to >29) and may be slightly more difficult to compute, but this may explain its better distribution in the study population and its enhanced performance for prediction of ischemic events.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.