Study design and population

The CLEAR-CAD trial is a prospective, open-label, multicentre, randomised clinical trial enrolling 6444 patients referred to a cardiology outpatient clinic with stable chest pain suspected of CAD. The study specifically excludes patients with a history of CAD on cardiac imaging or those suspected of having acute coronary syndrome. The inclusion and exclusion criteria are presented in Table 1. Patients are selected for inclusion in the study at the end of the first visit to the outpatient clinic, and consenting patients are randomised in a 1:1 ratio to either the upfront CTCA-guided strategy or standard care (Fig. 1).

Table 1 Inclusion and exclusion criteriaFig. 1

Study design (CAC coronary-artery calcium score, CAD coronary artery disease, CTCA computed tomography coronary angiography, OMT optimal medical therapy)

Study proceduresUpfront CTCA-guided strategy

Patients randomised to the upfront CTCA-guided strategy (intervention) undergo CTCA (+CAC) within 6 weeks after randomisation (Fig. 2). The CTCA is assessed using the standardised Coronary Artery Disease-Reporting and Data System (CAD-RADS 2.0) method [12]. The subsequent diagnostic and treatment strategies are based on the CAD-RADS score:

In patients with CAD-RADS 0, CAD is excluded as the cause of anginal chest pain. In these patients no specific cardiac medication is mandated.

In patients with CAD-RADS 1–2, CTCA indicates non-obstructive CAD, excluding obstructive CAD as the cause of anginal chest pain. In these patients, preventive OMT is started, consisting of a lipid-lowering drug. Initiation of platelet aggregation inhibition is at the discretion of the treating cardiologist.

In patients with CAD-RADS ≥ 3, CTCA indicates obstructive CAD. In these patients, the treating cardiologist starts preventive OMT consisting of both a lipid-lowering drug and a platelet aggregation inhibitor and additional anti-anginal medication (OMT+) consisting of a minimum of one of the following: beta-blocker, long-acting nitrate or calcium channel blocker. After 4–6 weeks, anginal symptoms will be evaluated by the cardiologist. In the event of persistent symptoms, additional non-invasive functional imaging will be performed within 3 months. Patients qualify for ICA in the presence of either myocardial ischaemia in at least 10% of the myocardium on nuclear perfusion via SPECT/PET or CMR stress perfusion imaging or at least 2 of 16 segments with severe hypokinesis or akinesis on stress echocardiography. In the absence of ischaemia as defined above and persisting symptoms under optimised medical therapy, ICA may be considered.

In CAD-RADS ≥ 3 patients, a high-risk anatomy is defined as having left main (LM) diameter stenosis of at least 50% and/or a proximal left anterior descending (LAD) diameter stenosis of at least 70%. These patients do not require non-invasive ischaemia detection but are sent for direct ICA and subsequent revascularisation.

In patients with CAD-RADS N, CTCA indicates a (partial) non-diagnostic examination. Management of these patients is described in Appendix C of the Electronic Supplementary Material.

Fig. 2

The intervention group (CTCA-guided) (CAD-RADS Coronary Artery Disease-Reporting and Data System, CT computed tomography, ICA invasive coronary angiography, LAD left anterior descending artery, OMT optimal medical treatment)

Initiation of medical treatment is allowed before CTCA.

Standard care strategy

Patients randomised to the standard care strategy (control) are diagnosed and treated according to the current standard care as outlined in the ESC guidelines for suspected CAD [6]. In current practice, X‑ECG, CAC, CTCA, non-invasive functional imaging tests and ICA are used interchangeably [1, 2]. A slow uptake of CTCA scans in the control group is anticipated during the study [13]. The assessment of CTCA is similar in both groups. Initiation of medical treatment is allowed at any time.

Data collection and follow-up

All data, including the initial diagnostic trajectory and treatment, are documented in the case report forms (CRF) of the web-based electronic data capture system Castor. Follow-up is performed by telephone at 1, 3, 6 and 12 month(s) and every other year until follow-up at 5 years has been completed in all patients. In the event of the suspected occurrence of clinical endpoints, full source clinical data are collected from the hospital and general practitioners.

Assessment of symptoms of angina, dyspnoea and quality of life are assessed by the Seattle Angina Questionnaire (SAQ), the Rose Dyspnea Scale, and the EuroQol-5D (EQ-5D-5L) at the corresponding follow-up time points. Patients are also asked to complete the Institute for Medical Technology Assessment (iMTA), Medical Consumption Questionnaire (iMCQ) and the Productivity Cost Questionnaire (iPCQ) at 3, 6 and 12 months and every other year until follow-up at 5 years has been completed in all patients. All questionnaires are sent by email or post.

Study endpoints and statistical analysisPrimary endpoint

The primary endpoint is the composite of all-cause death and MI according to the fourth Universal definition [14]. The main analyses of the primary endpoint are based on the time from randomisation to the first occurrence of any component of the primary endpoint. These analyses are performed in the full analysis population of all randomised patients under application of the intention-to-treat principle, which means events are counted irrespective of their occurrence relative to implementation of the randomised strategy (upfront CTCA or control). Kaplan-Meier estimates of cumulative risk and cumulative hazard functions are provided to evaluate the timing of event occurrence in the different treatment groups and the consistency of the respective treatment effects for all time points. The hazard ratio (HR), of upfront CTCA versus standard care, and corresponding two-sided 95% confidence intervals (CIs) are estimated based on Cox proportional hazards models.

The study was designed to test the following hypotheses in this hierarchical order to preserve type 1 error rate; (i) the upfront CTCA-guided strategy is non-inferior to standard care in terms of the primary endpoint, (ii) the upfront CTCA-guided strategy is superior to standard care in terms of the primary endpoint.

Non-inferiority of the upfront CTCA regimen (vis-à-vis standard care) in terms of the primary endpoint is declared if the 95% CI of the HR excludes 1.62. Superiority of the upfront CTCA regimen (vis-à-vis standard care) in terms of the primary endpoint is declared if the 95% CI of the HR excludes 1, which is equivalent to p < 0.05 for the log-rank test. Use of 95% CI is equivalent to non-inferiority testing with a one-sided type I error (α) of 0.025 and to superiority testing with a two-sided type I error rate of 0.05.

Based on the literature, the cumulative incidences at 3 years were estimated at 3.0% (control) and at 1.75% (intervention) respectively [5]. Because of an anticipated greater use of CTCA up to 30% in the control group, we estimated that a 3-year event rate in the control group would be equal to 2.82%. Assuming an exponential distribution, a target relative risk at 3 years of 0.6206 (a reduction of the 3‑year cumulative incidence from 2.82 to 1.75%) corresponds to a HR of 0.617. With 5900 fully evaluable patients and a median duration of follow-up of 3 years, the study is designed with 80% power to demonstrate the superiority of the upfront CTCA-guided strategy over standard care. The study is event-driven and the primary analysis will be performed when the number of 135 events has been reached. Follow-up is censored at the last date of known outcome status for the occurrence of death or myocardial infarction or at the global end-of-study date, which is defined as the day of occurrence of the 135 events. To account for an expected rate of lost to follow-up of 5–7%, this study will enrol 6444 patients. With a total of 135 primary endpoints, the study has a 80% power to show non-inferiority of the upfront CTCA-guided strategy to standard care, using a non-inferiority margin of 1.62 for the HR. The non-inferiority margin of 1.62 is approximately equal to the inverse of the expected therapeutic benefit (1.0/0.617).

Ranked secondary endpoints

If the incidence of the primary endpoint is significantly lower in the upfront CTCA group than in the control group (p < 0.05), then the ranked secondary endpoints are tested for superiority of CTCA over standard care in a hierarchical manner at a significance level of 0.05 to preserve the alpha level (Table 2). The time-to-event analysis methods are similar to those described for the primary endpoint. In these analyses, testing for superiority is not preceded by testing for non-inferiority. An overview of additional endpoints and endpoint definitions, is shown in Appendix B of the Electronic Supplementary Material.

Subgroup analysis

The statistical analysis of the subgroups will be documented in the statistical analysis plan. Baseline characteristics that define subgroups of interest include age, sex, pre-test probability according to the Diamond and Forrester score [15], pre-test likelihood based on the Risk Factor-weighted Clinical Likelihood (RF-CL) model [16], cardiovascular disease risk prediction according to the cardiovascular risk using the SIGN guidelines (ASSIGN) score [17], and the Duke Clinical Score [18], angina symptom severity, blood cholesterol level and patient comorbidities such as diabetes, hypertension and renal failure. Subgroups defined by imaging characteristics include the CAC score, CAD-RADS score, coronary lesion location and characteristics, reduced left ventricular function, and ischaemia severity. Subgroups defined by follow-up results include OMT therapy adherence and lifestyle.

Cost-effectiveness analysis

The primary cost-effectiveness outcome is defined as the costs per quality-adjusted life-year (QALY). The economic evaluation of the upfront CTCA-guided strategy in patients with stable chest pain and suspected CAD will be accomplished as a cost-utility analysis from a societal perspective with the costs per QALY as primary outcome. In addition, cost-effectiveness analyses with the costs per MACE and costs per year progression-free of MACE will be performed to optimise the guidelines and treatment. The time horizon exceeds 12 months, so health effects and costs beyond the first year will be (differentially) discounted. A more detailed Health Economic Analysis Plan will be written prior to the start of the economic analyses [19].

Computed tomography coronary angiography

CTCA (+CAC) will be performed using at least a 64-multidetector scanner in all patients. The minimal standards and protocol guidance have been described in the standard operating procedures, which is outlined in Appendix C of the Electronic Supplementary Material. This includes the most important patient aspects (patient information, preparation, instruction), scanner aspects (specific scan—and reconstruction protocols as well as contrast injection protocols), and reporting aspects (interpretation and reporting standards). The standards and scan protocols of all included sites will be assessed and feedback will be provided before start of inclusion for overall high quality scan acquisition, interpretation and reporting. Enrolment will only start after meeting the prescribed minimal standards.

Study oversight and funding

The study was designed in accordance with the principles of the Declaration of Helsinki.

The CLEAR-CAD trial is registered on ClinicalTrials.gov with the unique identifier NCT05344612 (www.clearcad.nl). The study is designed and sponsored by the Amsterdam University Medical Center and Radboud University Medical Center. The Steering Committee is responsible for the study design, trial execution, data analysis and reporting of results. All primary and secondary endpoints are adjudicated by an independent Clinical Event Committee (CEC), blinded to the assigned study group. An independent Data and Safety Monitoring Board (DSMB) provides external oversight to ensure the safety of the study patients. The CLEAR-CAD trial is funded by a research grant from the Care Evaluation and Appropriate Use (‘Zorgevaluatie en Gepast Gebruik’) program of Dutch Organisation for Health Research and Development (ZonMw).

The first patient was randomised on 14 September 2022 and complete enrolment is expected by the end of 2025. At present, 25 sites (of the current 27 participant sites) have started enrolment and randomised approximately 3300 patients (Fig. 3). The main results of the trial are expected to be available by the end of 2027.

Fig. 3

CLEAR-CAD participating centres (CCN Cardiology Center Netherlands, MC medical centre, UMC university medical centre)