Influence of intensive lipid‐lowering on CT derived fractional flow reserve in patients with stable chest pain: Rationale and design of the FLOWPROMOTE study

Abstract Introduction Coronary CT angiography (CTA) derived fractional flow reserve (FFRCT) shows high diagnostic performance when compared to invasively measured FFR. Presence and extent of low attenuation plaque density have been shown to be associated with abnormal physiology by measured FFR. Moreover, it is well established that statin therapy reduces the rate of plaque progression and results in morphology alterations underlying atherosclerosis. However, the interplay between lipid lowering treatment, plaque regression, and the coronary physiology has not previously been investigated. Aim To test whether lipid lowering therapy is associated with significant improvement in FFRCT, and whether there is a dose–response relationship between lipid lowering intensity, plaque regression, and coronary flow recovery. Methods Investigator driven, prospective, multicenter, randomized study of patients with stable angina, coronary stenosis ≥50% determined by clinically indicated first‐line CTA, and FFRCT ≤ 0.80 in whom coronary revascularization was deferred. Patients are randomized to standard (atorvastatin 40 mg daily) or intensive (rosuvastatin 40 mg + ezetimibe 10 mg daily) lipid lowering therapy for 18 months. Coronary CTA scans with blinded coronary plaque and FFRCT analyses will be repeated after 9 and 18 months. The primary endpoint is the 18‐month difference in FFRCT using (1) the FFRCT value 2 cm distal to stenosis and (2) the lowest distal value in the vessel of interest. A total of 104 patients will be included in the study. Conclusion The results of this study will provide novel insights into the interplay between lipid lowering, and the pathophysiology in coronary artery disease.


| INTRODUCTION
Fractional flow reserve (FFR) has emerged as the gold standard for assessment of lesion-specific ischemia and decision-making on coronary revascularization. [1][2][3] It is well documented that revascularization can be safely avoided in lesions with FFR > 0.80, while patients having one or more lesions with FFR ≤ 0.80 may benefit from revascularization. 1,4 In the FAME-2 study of patients with coronary artery disease (CAD) and FFR ≤ 0.80, the incidence of the composite endpoint (death, myocardial infarction, and revascularization) was lower in the revascularization than in the medical therapy group (13.9 vs. 27%). 4 Notably, the driving force for the difference in outcomes was repeat revascularization, while the majority of patients in the medically treated group did not experience any serious cardiac events at 5-years follow-up (all-cause death or myocardial infarction did not occur in 85.9% of the patients). 4 Coronary CT angiography (CTA) derived FFR (FFR CT ) is based on standard acquired CT data set postprocessing algorithms with integration of quantitative anatomical, physiological modeling and computational fluid dynamics. 5 In patients with stable CAD, FFR CT demonstrates high diagnostic performance relative to invasively measured FFR, and in real-world practice FFR CT may favorably change patient management and outcomes. [6][7][8][9][10][11][12] Accordingly, in the recent 2021AHA/ACC/ASE/ CHEST/SAEM/SCCT/SCMR Guideline for the evaluation and diagnosis of chest pain, it is stated that FFR CT in intermediate-high risk patients with chest pain and stenosis 40%-90% (mid-proximal segment on CTA) can be useful for the diagnosis of vessel-specific ischemia and to guide clinical decision-making on revascularization (evidence level, 2A). 13 A negative FFR CT result (>0.80) in patients with intermediate range coronary lesions is associated with favorable clinical outcomes. [10][11][12] A positive FFR CT result (≤0.80) indicates lesion-specific ischemia when positive 1-2 cm distal to a stenosis (similar to FFR) or diffuse ischemia when there is no lesion-specific ischemia but a gradual decline in FFR CT along the length of one or more vessels with distal values ≤0.80 ( Figure 1). 10,14 While patients with lesion-specific ischemia by FFR CT may be referred to ICA for possible revascularization, patients without a focal FFR CT loss have no focal substrate for coronary intervention, and thus are managed by optimal medical therapy alone. 14 Lipid lowering therapy with statins is the cornerstone of contemporary preventive care in patients with CAD. Statin therapy is associated with plaque stabilization through favorable changes in high-risk atherosclerotic phenotype characteristics (APC) and a lower rate of overall atherosclerotic plaque volume progression. [15][16][17][18] These changes include a reduction in the total noncalcified plaque burden including low attenuation density plaques (LAP), and stabilization of thin-cap fibroatheroma. By serial coronary CTA, a significant decrease in APC plaque volumes have been demonstrated over 6-12 months both by regular and potent statin therapy, with more pronounced effects of the latter. 17 In post-hoc analyses, the presence and burden of APC plaques, even in nonobstructive CAD, is associated to the presence and severity of ischemia as assessed by FFR. [19][20][21] One prospective single-center study of 20 patients with stable CAD treated with fixed-dose rosuvastatin 5 mg per day for 18 months after coronary stenting suggested the existence of a negative correlation between low density lipoprotein (LDL) cholesterol lowering and changes in FFR. 22 We designed the prospective,

| Study objectives
The main purpose of study is two-fold: (1) to assess whether lipid lowering therapy in stable CAD patients with FFR CT ≤ 0.80 is associated with improvement in coronary physiology, and (2) to investigate whether there is a dose-response relationship between lipid lowering intensity, plaque regression, and coronary flow recovery.

| Study subjects
Patients referred for nonemergent clinically indicated coronary CTA demonstrating CAD, diameter stenosis ≥50%, FFR CT ≤ 0.80, and no obvious indication for coronary revascularization are eligible for study inclusion (Table 1). In the participating institutions, FFR CT testing is recommended for physiological coronary assessment in patients with stable chest pain and intermediate-risk anatomy as previously described. 10 Anatomical and physiological study eligibility criteria are shown in Table 1 and Figure 2. All patients undergoing clinically indicated FFR CT assessment (including reasons for study exclusion) during the study period are registered in a screening list.

| Patient workflow
Patients randomized to one of two lipid lowering treatment regimens are followed for 18 months (Figure 3). Coronary CTA scans with T A B L E 1 Study eligibility criteria

Inclusion criteria
Exclusion criteria 1. Age >35 years 2. Symptoms suggestive of stable CAD 3. CAD with at least one stenosis with ≥50% lumen reduction determined by the index coronary CTA investigation a 4. Presence of at least two low attenuation plaques (with attenuation <30 Hounsfield units) present in at least two orthogonal planes by CTA 5. Sinus rhythm 6. LDL cholesterol >2.0 mM 7. FFR CT ≤ 0.80 ( Figure 2) 8. Life expectancy >3 years 9. Fertile women must use safe contraception throughout the study period 1. Previous lipid lowering therapy b 2. Known CAD 3. Unstable angina 4. Indication for coronary revascularization 5. BMI > 40 6. Allergy to iodinated contrast media 7. Poor coronary CTA image quality inadequate for FFR CT calculation (determined by core laboratory) 8. Pregnancy (women < 45 years will be screened for pregnancy) 9. Moderate to severe liver failure 10. Estimated glomerular filtration rate <60 ml/min 11. Participation in another clinical trial Anatomical or FFR CT based exclusion criteria: 12. Left main-stenosis ≥50%, 3-VD or high-grade proximal LAD stenosis resulting in direct referral to ICA 13. FFRct ≤ 0.80 2 cm distal to stenosis on CTA in segments 5 and 6 ( Figure 2). 14.

| Coronary CT angiography
CTA is performed using contemporary high-end technology scanners according to the best practice CTA acquisition guidelines. 23 24 Patients with at least one lesion with ≥50% stenosis severity at the index scan will have FFR CT performed.

| Biochemistry
Measurement of total cholesterol, high density lipoprotein cholesterol, LDL cholesterol is measured at baseline and at follow-up visits.
Lipoprotein(a) is measured at the baseline visit. Hepatic enzymes, creatin kinase, creatinine, and hemoglobin is assessed at baseline, and after 3 months. Further special analyses related to inflammation for substudy analyses will be performed (Supporting Information).

| Coronary plaque analysis
Together with the pericoronary adipose tissue (PCAT) attenuation as a marker of coronary artery inflammation, plaque analyses will be performed at a core-lab in vessel segments with diameter ≥2 mm using a semiautomated software (Autoplaque, Cedars-Sinai Medical Center) as previously described. [18][19][20][25][26][27][28] In brief, automated attenuation thresholds will be used for scan-specific plaque differentiation, 26,27 and the vessel lumen, wall and plaque are defined automatically with manual input as required. Total plaque, calcified, and noncalcified (including LAP < 30 HU) will be measured (mm 3 ), and aggregate plaque volume (APV%) will be computed as total plaque volume/vessel volume × 100%.
Plaque burden will be assessed on a per-lesion (diameter stenosis ≥50%), vessel and patient level. Observers will be blinded to all clinical information, timing of the CT scans and FFR CT results.

| CTA derived fractional flow reserve
The FFR CT computation data transferal process has previously been described in detail. 5

| Endpoints and substudy analyses
In our analyses, we will assess whether the changes in FFR CT are associated to the absolute changes in LDL-cholesterol from baseline until 18 months of follow-up (ΔLDL-cholesterol).
The primary endpoint will be the 18-month difference in FFR CT using (1) the FFR CT value 2 cm distal to stenosis in the event of lesion-specific ischemia at baseline, and (2)

| Sample size
We assume that 104 patients will be sufficient to demonstrate a dose-relationship effect between LDL lowering and FFR CT recovery.
Assuming an average temporal average increase in the lowest FFR CT distal value of +0.07 in the intensive lipid treatment group, and +0.04 in usual care lipid lowering group (personal experience following treatment with atorvastatin 40-80 mg over 5-9 months, n = 9), with a noncompliance rate of 15% in both groups, leaving an intensity to treat effect of 0.85 × (0.07 ÷ 0.04) = 0.025, and a standard deviation of the difference = 0.03 in both groups, one can with 99% power detect a statistical difference between the randomization groups (and with 72% power to detect an effect >0.01).

| Data analysis plan
In the primary intention-to-treat analysis plan changes in FFR CT will be compared in a temporal hierarchically fashion. Thus, if the difference in FFR CT estimates at 18 months is of statistically significance, then the difference in FFR CT at 9 months (scan 1) will also be tested. Changes in categorical variables will be analyzed using the Fishers exact test, and means between groups by the Student

| Ethical considerations and safety
This study is conducted in accordance with the Declaration of Helsinki, thus written informed consent will be obtained from each participant at study inclusion. The study is approved by the Ethics Committee for each participating center. Regulations for good clinical practice (GCP) will be followed and monitored by the GCP-units at Aarhus University and University of Southern Denmark Hospitals. All patients in the FLOWPROMOTE study will have a strong guidelinerecommendation for initiating lipid-lowering therapy with statins and/or ezetimibe. 3,13 Statins and ezetimibe, even in combination, have proven safe. 33 The radiation exposure for each CT scan will be approximately 2.5-3.0 mSv, thus in total 10-12.5 mSv for patients participating in the study corresponding to or less than the radiation exposure inflicted by a single rest/stress single-photon emission computed tomography (SPECT). 34 Deferring patients with lesionspecific ischemia from ICA and revascularization will involve only patients with controlled symptoms and without proximal hemodynamically significant lesions in whom revascularization does not improve clinical hard outcomes when compared to medical treatment alone. 35

CONFLICTS OF INTEREST
Adam Updegrove is an employee and shareholder in HeartFlow. HeartFlow. TimFonte is an employee and shareholder in HeartFlow.

DATA AVAILABILITY STATEMENT
Data sharing not applicable-no new data generated.