Reduction in the risk of major adverse cardiovascular events with the BET protein inhibitor apabetalone in patients with recent acute coronary syndrome, type 2 diabetes, and moderate to high likelihood of non-alcoholic fatty liver disease.

Background: Nonalcoholic fatty liver disease (NAFLD) is common among patients with type 2 diabetes mellitus (T2DM) and is associated with increased risk for coronary atherosclerosis and acute cardiovascular (CV) events. We employed the validated, non-invasive Angulo NAFLD fibrosis score (FS) in an intervention study in patients with T2DM and recent acute coronary syndrome (ACS) to determine the association of FS with CV risk and treatment response to apabetalone. Apabetalone is a novel selective inhibitor of the second bromodomain of bromodomain and extra-terminal (BET) proteins, epigenetic regulators of gene expression.
Methods: The Phase 3 BETonMACE trial compared apabetalone with placebo in 2,425 patients with T2DM and recent ACS. In this post hoc analysis, we evaluated the impact of apabetalone therapy on CV risk, defined as a composite of major adverse cardiovascular events (MACE: CV death, non-fatal myocardial infarction [MI], or stroke) and hospitalization for heart failure (HHF) in two patient categories of FS that reflect the likelihood of underlying NAFLD. Patients were initially classified into three mutually exclusive categories according to a baseline Angulo FS <-1.455 (F0-F2), -1.455 to 0.675 (indeterminant), and >0.675 (F3-F4), where F0 through F4 connote fibrosis severity none, mild, moderate, severe, and cirrhosis, respectively. The composite of ischemic MACE and HHF in the placebo group was higher in indeterminant and F3-F4 categories compared to the F0-F2 category (17.2% vs 15.0% vs 9.7%). Therefore, for the present analysis, the former two categories were combined into an elevated NAFLD CVD risk group (FS+) that was compared with the F0-F2 group (lower NAFLD risk, FS0-2).
Results: In 73.7% of patients, FS was elevated and consistent with a moderate-to-high likelihood of advanced liver fibrosis (FS+); 26.3% of patients had a lower FS (FS0-2). In the placebo group, FS+ patients had a higher incidence of ischemic MACE and HHF (15.4%) than FS0-2 patients (9.7%). In FS+ patients, addition of apabetalone to standard of care treatment lowered the rate of ischemic MACE compared with placebo (HR = 0.79; 95% CI 0.60-1.05; p=0.10), HHF (HR = 0.53; 95% CI 0.33-0.86; p=0.01), and the composite of ischemic MACE and HHF (HR = 0.76; 95% CI 0.59-0.98; p=0.03). In contrast, there was no apparent benefit of apabetalone in FS0-2 patients (HR 1.24; 95% CI 0.75-2.07; p=0.40; HR 1.12; 95% CI 0.30-4.14; p=0.87; and HR 1.13; 95% CI 0.69-1.86; p=0.62, respectively). Over a median duration of 26.5 months, FS increased from baseline in both treatment groups, but the increase was smaller in patients assigned to apabetalone than to placebo (p=0.04). Conclusions: Amongst patients with T2DM, recent ACS, and a moderate-to-high likelihood of advanced liver fibrosis, apabetalone was associated with a significantly lower rate of ischemic MACE and HHF and attenuated the increase in hepatic FS over time.

Introduction

Nonalcoholic fatty liver disease (NAFLD) is associated with a 2.26-fold higher rate of incident coronary artery disease and a 1.42-fold higher risk of cardiovascular (CV) mortality, but the reasons for the increased cardiovascular disease (CVD) morbidity and mortality are poorly understood as are optimal pharmacological targets for reducing NAFLD associated CVD [1]. Patients with NAFLD have increased risk for coronary artery calcification as well as development of peripheral arterial and cerebrovascular disease [2,3]. NAFLD is highly correlated with multiple CV risk factors, including increased visceral adiposity, insulin resistance, endothelial dysfunction, atherogenic dyslipidemia (increased serum levels of triglyceride-enriched remnant lipoproteins, low high-density lipoprotein (HDL), and elevations in small, dense low-density lipoproteins (LDL)), as well as heightened inflammatory tone and a pro-oxidative and prothrombotic state [4], [5], [6]. NAFLD is highly prevalent in patients who are obese or have metabolic syndrome and/or type 2 diabetes mellitus (T2DM) [7,8]. NAFLD is estimated to affect 25-30% of the populations of Western nations and its impact on the risk for CV events tends to be underestimated [9].

Insulin resistance predisposes patients to the development of NAFLD. Progressive hepatic steatosis leads sequentially to increased intrahepatic inflammation, steatosis, fibrosis, and ultimately can result in cirrhosis and end-stage liver disease. It has been shown that inflammation and steatosis in the liver correlate with progression of coronary atherosclerosis and increase risk for acute CV events [10]. Hepatic steatosis is frequently accompanied by other forms of visceral steatosis, such as epicardial fat pad expansion which may act directly and adversely on the coronary arteries and myocardium [11].

The gold standard for diagnosis of NAFLD is liver biopsy. Non-invasive diagnostic techniques include imaging and mechanical testing for liver stiffness (Fibroscan), reflecting fibrosis. However, these techniques are not ordinarily incorporated in the design of large CV outcomes trials. The Angulo fibrosis score (FS) uses 6 readily available variables (age, body mass index [BMI], hyperglycemia/diabetes, aspartate aminotransferase/alanine aminotransferase [AST/ALT] ratio, platelet count, and albumin) to calculate a score using an empiric formula [12]. The score was validated in patients with biopsy-verified NAFLD and shown to predict histologically-determined liver fibrosis severity with high negative and positive predictive value. These findings suggest that the Angulo FS is a useful non-invasive tool to distinguish patients with or without advanced fibrosis [12].

Apabetalone is a novel selective bromodomain (BD) 2-inhibitor of bromodomain and extra-terminal (BET) proteins, epigenetic regulators of gene expression. There is evidence in humans that apabetalone beneficially impacts the lipid profile, endothelial dysfunction, and factors that regulate vascular calcification, inflammation, oxidative tone, and thrombotic status [13], [14], [15]. In vitro, apabetalone increases hepatocyte Apolipoprotein A-I (ApoA1), while decreasing the expression of genes that populate pro-inflammatory, pro-atherosclerotic, and pro-thrombotic pathways, including alkaline phosphatase (ALP) and acute phase response (APR) proteins [13,[16], [17], [18]]. Strikingly, chimeric mice with humanized livers respond to apabetalone treatment with less expression of human hepatocyte APR genes [18], supporting the possibility that apabetalone would have favorable effects in patients with NAFLD.

The Phase 3 BETonMACE trial compared apabetalone with placebo in 2,425 patients with T2DM and recent acute coronary syndrome (ACS). Treatment with apabetalone resulted in hazard ratios (HR) of 0.82 (p=0.11) for the primary endpoint of ischemic MACE (CV death, non-fatal myocardial infarction [MI], or stroke) and 0.59 (p=0.03) for the secondary endpoint of hospitalization for heart failure (HHF) [19]. In this post hoc analysis of BETonMACE, we evaluated the association of the Angulo FS with CVD risk and the effect of apabetalone on the composite outcome of ischemic MACE and HHF.

Methods

Study design

The design and primary CVD results of the phase 3 BETonMACE trial (https://www.clinicaltrials.gov NCT02586155) have been reported [19], [20], [21], [22], [23], [24]. The protocol was approved by the responsible institutional review board or ethics committee at each participating site. In brief, eligible participants had an ACS within 7–90 days prior to randomization, a low high-density lipoprotein cholesterol (HDL-C) level, and a diagnosis of type 2 diabetes. Exclusion criteria included any condition which, in the opinion of the investigator, was likely to prevent the subject from complying with the requirements of or completing the study. Qualifying patients who provided written, informed consent were randomized in a 1:1 ratio to receive apabetalone 100 mg orally twice daily or matching placebo, in addition to high-intensity statin therapy with atorvastatin or rosuvastatin and other clinically defined standard of care. The primary outcome was time to the first occurrence of ischemic MACE, defined as a composite of CV death, non-fatal MI, and stroke. The incidence and time to first occurrence of HHF were secondary outcomes.

The analysis population for the previously reported primary results was applied to the current analyses, consisting of all randomized patients who received any amount of study therapy.

FS measurements

The Angulo FS is calculated as FS = -1.675 + (0.037*age [years]) + (0.094*BMI [kg/m2]) + (1.13*hyperglycemia/diabetes [yes=1, no=0]) + (0.99*AST/ALT ratio) – (0.013*platelet count [×109/L]) – (0.66*albumin [g/dl]) [12]. Height was measured at baseline. Weight was measured at baseline, week 100, and last visit on treatment. For the purpose of calculating BMI, the last observations for height and weight were carried forward until the next scheduled measurements were obtained. Liver function tests including AST, ALT, and ALP were obtained at baseline and every 2 weeks until week 12, then every 4 weeks until week 28, then every 12 weeks until last visit on treatment. Other biochemical measurements including platelet count and albumin were obtained at baseline, week 24, week 52, and every 24 weeks until last visit on treatment. All parameter measurements were performed by a central laboratory (ICON, Farmingdale, New York). FS scores were consequently assigned to windows corresponding to baseline (last score prior to randomization) and 0.5 (13 to <39 weeks), 1 (39 to <65 weeks), 1.5 (65 to <91 weeks), 2 (91 to <117 weeks) and 2.5 years (≥117 weeks) after randomization; if a patient had multiple values within a given post-baseline window, the first score was included in the analyses.

Patients were initially classified into three mutually exclusive categories according to a baseline Angulo FS <-1.455 (F0-F2), -1.455 to 0.675 (indeterminant), and >0.675 (F3-F4), where F0 through F4 connote fibrosis severity none, mild, moderate, severe, and cirrhosis, respectively. The composite of ischemic MACE and HHF in the placebo group was higher in indeterminant and F3-F4 categories compared to the F0-F2 category (17.2% vs 15.0% vs 9.7%). Therefore, for the present analysis, the former two categories were combined into an elevated NAFLD CVD risk group (FS+) that was compared with the F0-F2 group (lower NAFLD risk, FS0-2).

Statistical analysis

Baseline characteristics were summarized for all patients who had FS data at baseline by FS category (FS+ vs. FS0-2) and further by assigned treatment group. Data were expressed as counts and percentages for categorical variables, mean (standard deviation [SD]) for approximately normal data, or median (interquartile range [IQR]) for non-normal continuous variables. Group comparisons were performed by chi-square tests for categorical variables, z-tests for normal continuous variables, and non-parametric Mann Whitney Wilcoxon tests for non-normal continuous variables.

Change in FS over time was analyzed by a repeated-measures mixed-effects model with absolute change from baseline as the outcome, random effects for intercept and baseline FS score, and fixed effects for treatment group and time, yielding least square (LS) means and corresponding 95% confidence intervals (CIs) for each treatment group and p-values for treatment group differences. Change in FS score was also modeled jointly with time to all-cause death as a sensitivity analysis to account for competing risk. LS means and p-values at each timepoint (0.5, 1, 1.5, 2, and 2.5 years after randomization) were determined by interaction terms between treatment and time. Results were also generated for subgroups defined by baseline FS category.

Possible heterogeneity in relative and absolute effects of apabetalone treatment on ischemic MACE and HHF end points were assessed according to baseline FS category. We constructed a Cox proportional hazards model with risk group, treatment, and their interaction as predictors. Absolute risk reductions (ARRs) were calculated in terms of events per 100 patient-years of follow-up, with a test for quantitative interaction between risk group and treatment. Changes in biochemical parameters from baseline to each individual measured time point were analyzed using analysis of covariance models (ANCOVA) with baseline measurements serving as covariates. Results from the ANCOVA analyses are reported as least squares (LS) means with standard error. Analyses were performed with R software, version 3.5.1 or higher (R Foundation for Statistical Computing). P-values less than 0.05 were considered statistically significant without adjustment for multiplicity in this post-hoc analysis.

Results

Demographic features

The BETonMACE trial randomized 2,425 patients to either apabetalone or placebo at 190 sites in 13 countries between November 2015 and July 2018. As shown in Fig. 1, after excluding 7 participants who were randomized in error and an additional 71 participants for whom FS could not be calculated, 2,347 participants were included in this analysis. These participants received at minimum one dose of study drug and were followed for a median of 26.5 months. The analysis cohort included 1,729 (73.7%) patients categorized as FS+ with a mean (SD) FS -0.17 (0.83), and 618 (26.3%) patients categorized as FS0-2 with FS -2.40 (0.97). Patient flow through the trial by FS category and treatment assignment is depicted in Fig. 1.

Fig. 1

Patient Flow in the FS subgroups of the BETonMACE trial comparing apabetalone versus placebo.

Patients with FS+ scores compared to those with FS0-2 scores were older by approximately 6 years, more likely to be White (89.1 vs 83.0%), had higher BMI (30.8 vs 28.6), were more likely to be hypertensive (90.6 vs 82.5%), had higher systolic blood pressure (5 mm Hg between group difference), were less likely to smoke (10.2 vs 14.7%), and had longer duration of T2DM (7.8 vs 6.9 years from diagnosis) (Table 1). There were no clinically meaningful differences between FS categories for cardiovascular or diabetes medication usage. Although there were some statistically significant between-group differences in specific blood chemistries (glycated hemoglobin (HbA1c), HDL-C, ALP, ALT, and total bilirubin), these differences were small and unlikely to be clinically significant. Among FS0-2 participants considered by treatment allocation, there were small but significant imbalances in percentage who were White and among those using beta-blockers (Table 2). Among FS+ participants, there were no discernible baseline differences by treatment allocation.

Table 1

Demographics, clinical, pharmacologic, and laboratory characteristics of the BETonMACE trial participants at baseline, according to assigned treatment group, and FS category.

	Full Study Cohort According to Assigned Treatment Group(n = 2,347)			Full Study Cohort According to FS Category(n = 2,347)
	Placebo	Apabetalone	p-value	FS0-2 Patients	FS+ Patients	p-value
Number of Participants	1,167	1,180	–	618	1,729	–
Fibrosis Score	-0.72 (1.29)	-0.81 (1.33)	0.10	-2.40 (0.97)	-0.17 (0.83)	<0.0001
F0 – F2	290 (24.9%)	328 (27.8%)	0.12	618 (100%)	–	–
Indeterminant range	732 (62.7%)	708 (60.0%)	0.19	–	1,440 (83.3%)	–
F3 – F4	145 (12.4%)	144 (12.2%)	0.92	–	289 (16.7%)	–
Demographics
Age, years	62 (55.5 – 68)	62 (55 – 68)	0.07	57 (49 – 62)	63 (57 – 69)	<0.0001
Female, n (%)	301 (25.8%)	295 (25.0%)	0.69	148 (23.9%)	448 (25.9%)	0.36
White, n (%)	1,018 (87.2%)	1,036 (87.8%)	0.73	513 (83.0%)	1,541 (89.1%)	0.0001
Asian, n (%)	19 (1.6%)	20 (1.7%)	0.97	14 (2.3%)	25 (1.4%)	0.24
Other race, n (%)	130 (11.1%)	124 (10.5%)	0.67	91 (14.7%)	163 (9.4%)	0.0004
Body mass index, kg/m2	30.3 (5.0)	30.2 (4.8)	0.53	28.6 (4.2)	30.8 (5.0)	<0.0001
Hypertension history, n (%)	1,059 (90.7%)	1,083 (91.8%)	0.42	510 (82.5%)	1,566 (90.6%)	<0.0001
Smoking status, n (%)	125 (10.7%)	147 (12.5%)	0.21	89 (14.4%)	176 (10.2%)	0.006
Diabetes duration, years	8.7 (7.6)	8.4 (7.6)	0.45	7.3 (6.9)	9.0 (7.8)	<0.0001
Blood Pressure, mm Hg
Systolic	130 (120 – 140)	130 (120 – 140)	0.57	125 (116 – 135)	130 (120 – 140)	<0.0001
Diastolic	77 (70 – 82)	78 (70 – 82)	0.62	76 (70 – 81)	78 (70 – 82)	0.09
Index ACS
Myocardial infarction, n (%)	865 (74.8%)	862 (73.2%)	0.41	506 (82.4%)	1,221 (71.0%)	<0.0001
NSTEMI, n (%)	403 (46.8%)	401 (46.8%)	0.97	189 (37.4%)	615 (50.7%)	<0.0001
STEMI, n (%)	458 (53.2%)	456 (53.2%)	0.97	317 (62.6%)	597 (49.3%)	<0.0001
Unstable angina, n (%)	291 (25.2%)	315 (26.8%)	0.41	108 (17.6%)	498 (29.0%)	<0.0001
Time from index ACS, days	38 (25 – 62)	38 (25 – 63)	0.48	31 (23 – 56)	40 (26 – 64)	<0.0001
Cardiovascular Medications
Atorvastatin, n (%)	599 (51.3%)	600 (50.8%)	0.85	299 (48.4%)	900 (52.1%)	0.13
Rosuvastatin, n (%)	568 (48.7%)	580 (49.2%)	0.85	319 (51.6%)	829 (47.9%)	0.13
Intensive statin therapy, n (%)	1,067 (91.4%)	1,067 (90.4%)	0.44	561 (90.8%)	1,573 (91.0%)	0.95
Ezetimibe, n (%)	30 (2.6%)	32 (2.7%)	0.93	18 (2.9%)	44 (2.5%)	0.73
ACE inhibitors or ARB, n (%)	1,073 (91.9%)	1,089 (92.3%)	0.82	557 (90.1%)	1,605 (92.8%)	0.04
Beta-blockers, n (%)	1,052 (90.1%)	1,076 (91.2%)	0.43	549 (88.8%)	1,579 (91.3%)	0.08
Antiplatelet agents, n (%)	1,156 (99.1%)	1,164 (98.6%)	0.46	612 (99.0%)	1,708 (98.8%)	0.79
Diabetes Medications
Metformin, n (%)	958 (82.1%)	981 (83.1%)	0.54	530 (85.8%)	1,409 (81.5%)	0.02
Insulin, n (%)	446 (38.2%)	437 (37.0%)	0.58	237 (38.3%)	646 (37.4%)	0.70
Sulfonylureas, n (%)	332 (28.4%)	359 (30.4%)	0.32	163 (26.4%)	528 (30.5%)	0.06
DPP4 inhibitors, n (%)	170 (14.6%)	177 (15.0%)	0.81	98 (15.9%)	249 (14.4%)	0.42
SGLT2 inhibitors, n (%)	142 (12.2%)	146 (12.4%)	0.93	82 (13.3%)	206 (11.9%)	0.42
GLP1 receptor agonists, n (%)	43 (3.7%)	40 (3.4%)	0.78	25 (4.0%)	58 (3.4%)	0.50
Biochemical Parameters
eGFR, mL/min/1.73 m2	97.3 (75.2 – 125.1)	99.8 (77.3 – 127.1)	0.08	106.4 (79.5 – 133.9)	95.8 (74.5 – 123.3)	<0.0001
HbA1c, %	7.3 (6.4 – 8.6)	7.4 (6.4 – 8.7)	0.36	7.6 (6.5 – 9.0)	7.3 (6.4 – 8.5)	0.0001
Serum glucose, mg/dL	132.9 (109.9 – 174.2)	136.4 (110.6 – 175.3)	0.33	133.1 (109.3 – 172.9)	135.7 (110.8 – 175.5)	0.32
Total cholesterol, mg/dL	129.9 (111.9 – 155.6)	128.4 (109.4 – 155.2)	0.33	129.2 (112.5 – 153.8)	129.5 (109.8 – 157.0)	0.83
LDL cholesterol, mg/dL	65.0 (48.6 – 85.5)	65.0 (49.1 – 85.1)	0.89	65.0 (50.3 – 82.0)	65.2 (48.3 – 86.2)	0.97
HDL cholesterol, mg/dL	33.6 (30.2 – 37.1)	33.3 (29.8 – 37.1)	0.78	32.9 (29.8 – 36.7)	33.6 (30.2 – 37.1)	0.008
Triglycerides, mg/dL	150.6 (115.1 – 201.9)	147.0 (111.6 – 198.4)	0.14	150.1 (113.6 – 205.5)	147.5 (112.5 – 198.4)	0.25
Alkaline phosphatase, U/L	77.0 (64.0 – 93.0)	78.0 (64.0 – 95.0)	0.38	82.0 (68.0 – 101.8)	76.0 (63.0 – 92.0)	<0.0001
Alanine aminotransferase, U/L	22.0 (17.0 – 30.0)	22.0 (17.0 – 31.0)	0.58	25.0 (18.0 – 34.0)	21.0 (16.0 – 29.0)	<0.0001
Aspartate aminotransferase, U/L	19.0 (15.0 – 23.0)	19.0 (15.0 – 23.0)	0.61	18.0 (15.0 – 22.8)	19.0 (15.0 – 23.0)	0.07
AST / ALT, ratio	0.82 (0.69 – 1.00)	0.82 (0.69 – 1.00)	0.73	0.73 (0.62 – 0.88)	0.86 (0.71 – 1.06)	<0.0001
Albumin, g/L	43.0 (41.0 – 45.0)	43.0 (41.0 – 45.0)	0.58	43.0 (41.0 – 45.0)	43.0 (41.0 – 44.0)	<0.0001
Platelets, 109/L	246.0 (206.0 – 296.5)	252.0 (209.0 – 306.0)	0.10	337.5 (295.0 – 385.0)	228.0 (194.0 – 262.0)	<0.0001
Total bilirubin, umol/L	9.2 (6.9 – 12.1)	9.1 (6.8 – 11.9)	0.53	8.3 (6.4 – 11.0)	9.4 (7.1 – 12.4)	<0.0001
hsCRP, mg/L	2.7 (1.1 – 6.1)[n = 238]	3.0 (1.3 – 6.2)[n = 237]	0.52	3.7 (1.3 – 7.0)[n = 124]	2.7 (1.2 – 5.7)[n = 351]	0.16
NLR, ratio	2.6 (2.0 – 3.3)	2.5 (2.0 – 3.3)	0.09	2.6 (2.0 – 3.3)	2.5 (2.0 – 3.4)	0.21

Abbreviations: FS, fibrosis score; FS0-2, F0 – F2 fibrosis; FS+, indeterminant range or F3 – F4 fibrosis; NSTEMI, non-ST segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarction; ACS, acute coronary syndrome; ACE, angiotensin-converting enzyme; ARB, Angiotensin II receptor blocker; DPP4, dipeptidyl peptidase 4; SGLT2, sodium-glucose cotransporter 2; GLP1, glucagon-like peptide 1; HbA1c, hemoglobin A1C; LDL, low-density lipoprotein; HDL, high-density lipoprotein; AST, aspartate aminotransferase; ALT, alanine aminotransferase; hsCRP, high-sensitivity C-reactive protein; NLR, neutrophil-lymphocyte ratio.

Categorical variables are presented as n (%). Continuous variables are presented as mean (SD) for normal data or median (quartile 1–quartile 3) for non-normal data.

P-values comparing groups at baseline were calculated using chi-square test for categorical variables, z-test for normal continuous variables, and Mann-Whitney Wilcoxon test for non-normal continuous variables. P-values of <0.05 are considered statistically significant and are highlighted in bold.

Table 2

Demographics, clinical, pharmacologic, and laboratory characteristics of the BETonMACE trial participants at baseline, according to FS category and assigned treatment group.

	FS0-2 Patients According to Assigned Treatment Group (n = 618)			FS+ Patients According to Assigned Treatment Group (n = 1,729)
	Placebo	Apabetalone	p-value	Placebo	Apabetalone	p-value
Number of Participants	290	328	–	877	852	–
Fibrosis Score	-2.39 (0.98)	-2.42 (0.96)	0.72	-0.16 (0.82)	-0.19 (0.84)	0.57
F0 – F2	290 (100%)	328 (100%)	–	–	–	–
Indeterminant range	–	–	–	732 (83.5%)	708 (83.1%)	–
F3 – F4	–	–	–	145 (16.5%)	144 (16.9%)	–
Demographics
Age, years	56 (50 – 62)	57 (49 – 62)	0.97	64 (58 – 70)	63 (57 – 69)	0.12
Female, n (%)	69 (23.8%)	79 (24.1%)	0.99	232 (26.5%)	216 (25.4%)	0.64
White, n (%)	231 (79.7%)	282 (86.0%)	0.048	787 (89.7%)	754 (88.5%)	0.45
Asian, n (%)	9 (3.1%)	5 (1.5%)	0.30	10 (1.1%)	15 (1.8%)	0.38
Other race, n (%)	50 (17.2%)	41 (12.5%)	0.12	80 (9.1%)	83 (9.7%)	0.72
Body mass index, kg/m2	28.5 (4.3)	28.7 (4.1)	0.65	30.9 (5.1)	30.7 (4.9)	0.57
Hypertension history, n (%)	238 (82.1%)	272 (82.9%)	0.86	782 (89.2%)	784 (92.0%)	0.052
Smoking status, n (%)	36 (12.4%)	53 (16.2%)	0.23	84 (9.6%)	92 (10.8%)	0.45
Diabetes duration, years	7.3 (7.1)	7.4 (6.7)	0.90	9.1 (7.8)	8.8 (7.9)	0.45
Blood Pressure, mm Hg
Systolic	125 (116 – 136)	124 (115 – 134)	0.28	130 (120 – 140)	130 (120 – 140)	0.70
Diastolic	75 (70 – 80)	77 (70 – 81)	0.43	78 (70 – 82)	78 (70 – 82)	0.89
Index ACS
Myocardial infarction, n (%)	242 (84.6%)	264 (80.5%)	0.22	623 (71.6%)	598 (70.4%)	0.63
NSTEMI, n (%)	92 (38.0%)	97 (36.7%)	0.84	311 (50.2%)	304 (51.3%)	0.77
STEMI, n (%)	150 (62.0%)	167 (63.3%)	0.84	308 (49.8%)	289 (48.7%)	0.77
Unstable angina, n (%)	44 (15.4%)	64 (19.5%)	0.22	247 (28.4%)	251 (29.6%)	0.63
Time from index ACS, days	32 (24 – 56)	31 (23 – 56)	0.64	41 (27 – 63)	40 (26 – 65)	0.67
Cardiovascular Medications
Atorvastatin, n (%)	146 (50.3%)	153 (46.6%)	0.40	453 (51.7%)	447 (52.5%)	0.77
Rosuvastatin, n (%)	144 (49.7%)	175 (53.4%)	0.40	424 (48.3%)	405 (47.5%)	0.77
Intensive statin therapy, n (%)	265 (91.4%)	296 (90.2%)	0.73	802 (91.4%)	771 (90.5%)	0.54
Ezetimibe, n (%)	6 (2.1%)	12 (3.7%)	0.35	24 (2.7%)	20 (2.3%)	0.72
ACE inhibitors or ARB, n (%)	261 (90.0%)	296 (90.2%)	0.97	812 (92.6%)	793 (93.1%)	0.76
Beta-blockers, n (%)	248 (85.5%)	301 (91.8%)	0.02	804 (91.7%)	775 (91.0%)	0.66
Antiplatelet agents, n (%)	288 (99.3%)	324 (98.8%)	0.80	868 (99.0%)	840 (98.6%)	0.61
Diabetes Medications
Metformin, n (%)	243 (83.8%)	287 (87.5%)	0.23	715 (81.5%)	694 (81.5%)	0.98
Insulin, n (%)	107 (36.9%)	130 (39.6%)	0.54	339 (38.7%)	307 (36.0%)	0.28
Sulfonylureas, n (%)	77 (26.6%)	86 (26.2%)	1.00	255 (29.1%)	273 (32.0%)	0.20
DPP4 inhibitors, n (%)	50 (17.2%)	48 (14.6%)	0.44	120 (13.7%)	129 (15.1%)	0.43
SGLT2 inhibitors, n (%)	40 (13.8%)	42 (12.8%)	0.81	102 (11.6%)	104 (12.2%)	0.77
GLP1 receptor agonists, n (%)	10 (3.4%)	15 (4.6%)	0.61	33 (3.8%)	25 (2.9%)	0.41
Biochemical Parameters
eGFR, mL/min/1.73 m2	105.2 (78.7 – 135.7)	107.6 (80.3 – 132.5)	0.80	94.6 (72.5 – 122.1)	97.1 (75.8 – 125.1)	0.09
HbA1c, %	7.6 (6.5 – 9.0)	7.7 (6.5 – 9.1)	0.77	7.2 (6.4 – 8.5)	7.3 (6.4 – 8.5)	0.49
Serum glucose, mg/dL	131.4 (110.6 – 177.5)	134.1 (108.3 – 168.9)	0.92	133.5 (109.9 – 173.5)	136.9 (111.5 – 176.9)	0.20
Total cholesterol, mg/dL	127.8 (114.2 – 152.8)	130.9 (110.1 – 154.3)	0.75	130.7 (110.6 – 157.8)	127.4 (109.4 – 155.5)	0.20
LDL cholesterol, mg/dL	63.8 (50.0 – 78.9)	66.1 (50.7 – 84.1)	0.31	65.4 (48.0 – 86.5)	65.0 (48.3 – 86.2)	0.47
HDL cholesterol, mg/dL	32.5 (29.4 – 37.0)	32.9 (29.8 – 36.3)	0.76	33.6 (30.2 – 37.1)	33.6 (29.8 – 37.1)	0.71
Triglycerides, mg/dL	149.7 (119.6 – 206.6)	151.0 (111.4 – 205.5)	0.68	150.6 (114.3 – 200.4)	145.3 (111.6 – 195.7)	0.13
Alkaline phosphatase, U/L	82.5 (69.0 – 102.0)	81.0 (67.8 – 101.0)	0.55	76.0 (62.0 – 91.0)	77.0 (63.0 – 93.0)	0.22
Alanine aminotransferase, U/L	25.0 (18.0 – 34.0)	25.0 (18.0 – 34.0)	0.95	22.0 (16.0 – 29.0)	21.0 (16.0 – 29.0)	0.32
Aspartate aminotransferase, U/L	18.0 (15.0 – 22.0)	19.0 (15.0 – 23.0)	0.41	19.0 (15.0 – 23.0)	19.0 (15.0 – 23.0)	0.88
AST / ALT, ratio	0.73 (0.62 – 0.88)	0.73 (0.62 – 0.89)	0.69	0.86 (0.71 – 1.06)	0.87 (0.72 – 1.07)	0.53
Albumin, g/L	43.0 (41.0 – 46.0)	43.0 (41.0 – 45.0)	0.95	43.0 (41.0 – 44.0)	43.0 (41.0 – 44.0)	0.42
Platelets, 109/L	332.0 (293.0 – 384.8)	340.5 (299.8 – 385.0)	0.34	228.0 (189.0 – 262.0)	228.0 (196.0 – 262.0)	0.73
Total bilirubin, umol/L	8.0 (6.4 – 11.0)	8.6 (6.4 – 10.9)	0.38	9.5 (7.1 – 12.4)	9.3 (7.0 – 12.3)	0.34
hsCRP, mg/L	4.0 (1.3 – 7.4)[n = 61]	3.7 (1.3 – 6.8)[n = 63]	0.77	2.3 (1.1 – 5.6)[n = 177]	3.0 (1.3 – 5.7)[n = 174]	0.38
NLR, ratio	2.7 (2.1 – 3.5)	2.4 (1.9 – 3.2)	0.001	2.5 (1.9 – 3.3)	2.6 (2.0 – 3.4)	0.79

Abbreviations: FS, fibrosis score; FS0-2, F0 – F2 fibrosis; FS+, indeterminant range or F3 – F4 fibrosis; NSTEMI, non-ST segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarction; ACS, acute coronary syndrome; ACE, angiotensin-converting enzyme; ARB, Angiotensin II receptor blocker; DPP4, dipeptidyl peptidase 4; SGLT2, sodium-glucose cotransporter 2; GLP1, glucagon-like peptide 1; HbA1c, hemoglobin A1C; LDL, low-density lipoprotein; HDL, high-density lipoprotein; AST, aspartate aminotransferase; ALT, alanine aminotransferase; hsCRP, high-sensitivity C-reactive protein; NLR, neutrophil-lymphocyte ratio.

Categorical variables are presented as n (%). Continuous variables are presented as mean (SD) for normal data or median (quartile 1–quartile 3) for non-normal data.

P-values comparing groups at baseline were calculated using chi-square test for categorical variables, z-test for normal continuous variables, and Mann-Whitney Wilcoxon test for non-normal continuous variables. P-values of <0.05 are considered statistically significant and are highlighted in bold.

Impact of apabetalone on cardiovascular outcomes in FS+ and FS0-2 participants

In the placebo group, the incidence of ischemic MACE and HHF was greater among patients categorized as FS+ (15.4%) compared to those categorized as FS0-2 (9.7%) (Table 3). As shown in Table 3 and Figs. 2–4, the addition of apabetalone to standard of care treatment lowered the rate of ischemic MACE in FS+ patients (HR = 0.79; 95% CI 0.60-1.05; p=0.10) (Fig. 2), HHF (HR = 0.53; 95% CI 0.33-0.86; p=0.01) (Fig. 3), and the composite of ischemic MACE and HHF (HR = 0.76; 95% CI 0.59-0.98; p=0.03) (Fig. 4). In contrast, there was no apparent benefit of apabetalone in FS0-2 patients [HR 1.24, 95% CI 0.75-2.07, p=0.40; HR 1.12, 95% CI 0.30-4.14, p=0.87; and HR 1.13, 95% CI 0.69-1.86, p=0.62 for ischemic MACE (Fig. 2), HHF (Fig. 3), and the composite of ischemic MACE and HHF (Fig. 4), respectively]. The interaction p-values of treatment and FS category on risk of ischemic MACE, HHF, and the composite of ischemic MACE and HHF were 0.14, 0.28, and 0.16, respectively (Table 3).

Table 3

Kaplan-Meier estimates of time to first occurrence of ischemic MACE (CV death, non-fatal MI, or stroke), HHF, and the composite of ischemic MACE and HHF, according to FS category and assigned treatment group. A log-rank test was used for the formal hypothesis testing and a Cox proportional hazards model to estimate the HR with 95% CI. P-values of <0.05 are considered statistically significant.

		FS0-2 Patients According to Assigned Treatment Group (n = 618)			FS+ Patients According to Assigned Treatment Group (n = 1,729)			Cox Model Interaction p-valueof HR
		Placebo(n=290)	Apabetalone(n=328)	p-value	Placebo(n=877)	Apabetalone(n=852)	p-value
Ischemic MACE(CV Death,Non-Fatal MI, or Stroke)	No. of Events (%)	25 (8.6%)	34 (10.4%)	–	114 (13.0%)	90 (10.6%)	–	–
	HR (95% CI)	1.24 (0.75 – 2.07)		0.40	0.79 (0.60 – 1.05)		0.10	0.14
HHF	No. of Events (%)	4 (1.4%)	5 (1.5%)	–	43 (4.9%)	22 (2.6%)	–	–
	HR (95% CI)	1.12 (0.30 – 4.14)		0.87	0.53 (0.33 – 0.86)		0.01	0.28
Composite of Ischemic MACEand HHF	No. of Events (%)	28 (9.7%)	35 (10.7%)	–	135 (15.4%)	102 (12.0%)	–	–
	HR (95% CI)	1.13 (0.69 – 1.86)		0.62	0.76 (0.59 – 0.98)		0.03	0.16

Abbreviations: MACE, major adverse cardiovascular events; CV, cardiovascular; HHF, hospitalization for heart failure; FS, fibrosis score; FS0-2, F0 – F2 fibrosis;

FS+, indeterminant range or F3 – F4 fibrosis; HR, hazard ratio; CI, confidence interval.

Cox model interaction p-values of HR indicate differences by FS category in the effect of apabetalone on HR.

Categorical variables are presented as n (%).

P-values for categorical variables were calculated using chi-square test.

Fig. 2

Kaplan-Meier estimate of time to first occurrence of ischemic MACE (CV death, non-fatal MI, or stroke) in (A): FS+ patients; and, (B): FS0-2 patients. A log-rank test was used for the formal hypothesis testing. A Cox proportional hazards model was used to estimate the HR with 95% CI. P-values of <0.05 are considered statistically significant.

Fig. 4

Kaplan-Meier estimate of time to first occurrence of the composite of ischemic MACE (CV death, non-fatal MI, or stroke) and HHF in (A): FS+ patients; and, (B): FS0-2 patients. A log-rank test was used for the formal hypothesis testing. A Cox proportional hazards model was used to estimate the HR with 95% CI. P-values of <0.05 are considered statistically significant.

Fig. 3

Kaplan-Meier estimate of time to first occurrence of HHF in (A): FS+ patients; and, (B): FS0-2 patients. A log-rank test was used for the formal hypothesis testing. A Cox proportional hazards model was used to estimate the HR with 95% CI. P-values of <0.05 are considered statistically significant.

Changes in FS over time

At baseline, there were no significant differences in FS score between patients assigned to apapbetalone or placebo in either FS category (Table 2). We followed change in FS over time to explore if and how apabetalone influences the score and whether FS associates with the clinical efficacy of apabetalone. Among the patients with a baseline score, 2,142 had at least one post-baseline score and were therefore included in the analyses of change. The LS mean change in FS from baseline to 2.5 years are summarized in Fig. 5A for the full analysis cohort, and in Fig. 5B and 5C for the FS+ and FS0-2 categories respectively. At 0.5-years, participants in both treatment groups of both FS categories showed increases in FS that were primarily attributable to increased AST (Supplementary Tables S1-S4). The increase from baseline was smaller in the FS+ category than in the FS0-2 category. After the 0.5-year timepoint, FS remained relatively stable in both treatment groups of both FS categories. However, over the course of the trial, considering the full study cohort, the increase in FS from baseline was significantly smaller in patients assigned to apabetalone treatment than in those assigned to placebo (p=0.04, Fig. 5A). This reflected a significantly smaller increase in FS from baseline with apabetalone than with placebo in the FS+ category (p=0.02, Fig. 5B), with no difference between treatment groups in the change in FS from baseline in the FS0-2 category (p=0.62, Fig. 5C). The factor that primarily accounted for smaller increases in FS from baseline with apabetalone than placebo in the full analysis cohort and in the FS+ category was a smaller increase in AST/ALT ratio (Supplementary Tables S1-S4).

Fig. 5

Change in FS over time in (A): the full study cohort, according to assigned treatment group; (B): FS+ patients, according to assigned treatment group; and, (C): FS0-2 patients, according to assigned treatment group. Change in FS over was analyzed using a repeated-measures mixed-effects model with absolute change from baseline as the outcome, random effects for intercept and baseline FS score, and fixed effects for treatment group and time, modeled jointly with time to all-cause death as a sensitivity to account for competing risk. Data are presented as least squares (LS) means with 95% confidence intervals (CI). LS means and p-values at each timepoint were determined by interaction terms between treatment and time. P-values of <0.05 are considered statistically significant.

Safety

Overall, 68% of the analysis cohort reported at least one adverse event, and 28% reported at least one serious adverse event. There were no discernible imbalances between FS categories or treatment groups (Table 4). Slightly more patients in the FS+ and FS0-2 categories discontinued apabetalone than placebo (Table 5). There was a small excess in the number of patients with alanine aminotransferase elevations in the apabetalone group apparent in both FS categories (Table 5). There was no apparent imbalance in aspartate aminotransferase or gamma-glutamyltransferase elevations. Apabetalone therapy did not correlate with increased risk for diabetes, hypertension, diarrhea, or any type of chest pain.

Table 4

Adverse events (AEs) according to assigned treatment group, and FS category. Data are presented as n (%).

	Full Study Cohort According to Assigned Treatment Group		Full Study Cohort According to FS Category
	Placebo (n=1,167)	Apabetalone (n=1,180)	FS0-2 Patients (n=618)	FS+ Patients (n=1,729)
Patients with at least one adverse event*	792 (68%)	813 (69%)	428 (69%)	1,177 (68%)
Patients with at least one serious adverse event*	324 (28%)	346 (29%)	158 (26%)	512 (30%)
Patients with at least one adverse event leading to study drug discontinuation*	65 (5.6%)	110 (9.3%)	42 (6.8%)	133 (7.7%)
Frequent adverse events*,†
Alanine aminotransferase increased	17 (1.5%)	63 (5.3%)	23 (3.7%)	57 (3.3%)
Angina pectoris	72 (6.2%)	72 (6.1%)	35 (5.7%)	109 (6.3%)
Aspartate aminotransferase increased	8 (0.7%)	20 (1.7%)	5 (0.8%)	23 (1.3%)
Diabetes mellitus	58 (5.0%)	75 (6.4%)	35 (5.7%)	98 (5.7%)
Diarrhea	41 (3.5%)	42 (3.6%)	31 (5.0%)	52 (3.0%)
Gamma-glutamyltransferase increased	12 (1.0%)	10 (0.8%)	4 (0.6%)	18 (1.0%)
Hypertension	70 (6.0%)	68 (5.8%)	32 (5.2%)	106 (6.1%)

Includes treatment-emergent adverse events only, defined as those occurring after the first dose and within 14 days of the last dose of the study drug.

Defined as occurring with a frequency of 5% or more in any of the FS categories or treatment groups.

Table 5

Adverse events (AEs) according to FS category and assigned treatment group. Data are presented as n (%).

	FS0-2 Patients According to Assigned Treatment Group		FS+ Patients According to Assigned Treatment Group
	Placebo (n=290)	Apabetalone (n=328)	Placebo (n=877)	Apabetalone (n=852)
Patients with at least one adverse event*	205 (71%)	223 (68%)	587 (67%)	590 (69%)
Patients with at least one serious adverse event*	72 (25%)	86 (26%)	252 (29%)	260 (31%)
Patients with at least one adverse event leading to study drug discontinuation*	13 (4.5%)	29 (8.8%)	52 (5.9%)	81 (9.5%)
Frequent adverse events*,†
Alanine aminotransferase increased	4 (1.4%)	19 (5.8%)	13 (1.5%)	44 (5.2%)
Angina pectoris	21 (7.2%)	14 (4.3%)	51 (5.8%)	58 (6.8%)
Angina unstable	6 (2.1%)	18 (5.5%)	33 (3.8%)	38 (4.5%)
Aspartate aminotransferase increased	2 (0.7%)	3 (0.9%)	6 (0.7%)	17 (2.0%)
Diabetes mellitus	16 (5.5%)	19 (5.8%)	42 (4.8%)	56 (6.6%)
Diarrhea	13 (4.5%)	18 (5.5%)	28 (3.2%)	24 (2.8%)
Gamma-glutamyltransferase increased	2 (0.7%)	2 (0.6%)	10 (1.1%)	8 (0.9%)

Includes treatment-emergent adverse events only, defined as those occurring after the first dose and within 14 days of the last dose of the study drug.

Defined as occurring with a frequency of 5% or more in any of the FS categories or treatment groups.

Discussion

In this post hoc analysis of the BETonMACE trial we calculated the Angulo FS, originally developed as a non-invasive surrogate for histologic liver fibrosis changes in NAFLD, based on the rationale that post-ACS T2DM patients are likely to have NAFLD. Strikingly, 73.7% of patients with T2DM and recent ACS had scores consistent with a moderate to high likelihood of advanced liver fibrosis, while only 26.3% had scores consistent with a low likelihood of advanced fibrosis [12]. Moreover, the baseline Angulo FS identified patients who were at high risk for further CV events despite proven, standard-of-care therapies. Importantly, patients with an elevated baseline FS were at a higher risk of ischemic MACE and HHF and showed a CV benefit from treatment with apabetalone, while no such benefit was apparent among those with lower baseline FS.

With recognition that the interaction of treatment and FS category on the composite of ischemic MACE and HHF was not statistically significant, a greater treatment benefit of apabetalone in FS+ patients is plausible because FS+ patients were older, had a higher prevalence of hypertension, and had a longer duration of diabetes. Each of these characteristics is associated both with elevated CV risk and with increased BET dysregulation and heightened response to BET-inhibition treatment [25], [26], [27], [28], [29], [30].

Several studies have found that high FS associates with incidence of coronary heart disease [31,32] as well as the angiographic complexity [33], [34], [35], risk of MACE [34,36], and death [37] in patients with established coronary heart disease. In an analysis of the IMPROVE-IT trial that compared ezetimibe with placebo in patients with ACS treated with simvastatin [36], high FS (>0.67, n=2,106) vs low FS (<-1.455, n=5,440) was associated with 30% increased risk of MACE. Analogous to the findings for apabetalone in the current analysis, the data from IMPROVE-IT suggested that high FS was an effect modifier for the benefit of ezetimibe. In patients with high FS, ezetimibe conferred a 3.7% absolute reduction in risk of MACE compared to placebo (HR 0.85 [0.74-0.98]), while no benefit of ezetimibe was evident in the low FS group (HR 1.01 [0.91-1.12]; p-interaction = 0.053) [36].

The observed benefit of apabetalone in patients with high FS, serving as a surrogate for NAFLD, may be important as the already-high prevalence of NAFLD is predicted to increase across populations in the future, augmenting the need for effective, targeted treatments [38], [39], [40], [41], [42], [43]. Although no drugs are currently approved for treating NAFLD, PPAR activators, which may also favorably influence a dysregulated transcriptional profile, have shown promise [44,45]. In addition, exercise, weight loss, and bariatric surgery have been shown to regress steatosis or even resolve distorted hepatic cellular architecture [46], [47], [48], [49], [50]. Considerable investigation is ongoing worldwide to find safe and efficacious therapies to treat NAFLD so as to prevent hepatic injury and attenuate risk for CVD. Many classes of drugs are being tested for efficacy and safety [51], [52], [53].

The observation that FS was lower under treatment with the BET protein inhibitor apabetalone compared with placebo in FS+ patients is novel. However, the difference in FS between treatment groups was small. Further studies are needed to determine if a favorable effect of apabetalone on FS is accompanied by favorable changes in hepatic imaging, mechanical properties, or histology, whether parallel reductions in steatosis are observed in extra-hepatic tissues including the heart [54,55], in turn whether reducing steatosis-driven inflammation attenuates the risk of CVD events [56,57].

Epigenetic BET-inhibition by apabetalone may constitute a novel approach to slow or even stop the progression of hepatic steatosis and fibrosis, and may explain the CVD reduction by apabetalone observed in the FS+ category. Inflammatory stimuli drive human hepatocyte overexpression of acute phase reactants in vitro, creating a positive feedback loop that can promote local liver and systemic inflammation. By preventing BET protein – chromatin associations at specific translational start sites, apabetalone can reduce the transcription of multiple induced APR genes [18]. Additionally, it has been shown that the expression of key APR genes is lower in the liver of endotoxemic mice and in the plasma of coronary artery disease (CAD) patients following apabetalone treatment [18]. Furthermore, the abundance of pro-inflammatory mediators in the plasma of patients from multiple phase 2 clinical studies, is significantly lower in CAD patients treated with apabetalone [14,15,18,58]. Plasma ALP, sourced primarily from the liver, is a robust and independent predictor of all-cause mortality, as it promotes vascular calcification and arterial stiffness, vascular inflammation, destabilization of atherosclerotic plaques, and oxidative stress. Apabetalone has also been shown to reduce hepatocyte ALP in vitro [13], and ALP protein levels are reduced in CAD and chronic kidney disease (CKD) patient plasma following apabetalone treatment [58,59]. Finally, the BETi I-BET151 has been shown to improve NASH and liver fibrosis in the STAM mouse NASH model [60]. Thus, precedence exists for BETi efficacy in NASH, and apabetalone itself is predicted to benefit liver function through its epigenetic regulation of dysfunctional transcription that arises in NAFLD and CAD.

There are several limitations to this study. First, we used a validated scoring system based on six demographic and laboratory parameters that has been shown to associate with the prevalence of NAFLD determined by biopsy, imaging, or non-invasive biomechanical properties [12]. Based on a validation group of 253 subjects with NAFLD and advanced fibrosis, the Angulo model has a positive predictive value of 82% to diagnose fibrosis and a negative predictive value of 88% to exclude fibrosis [12]. However, we did not obtain imaging or biomechanical data to indicate whether the FS categories defined in this cohort corresponded to differences in measures of liver fibrosis or steatosis. Second, substantial data associate FS with the prevalence of NAFLD, but it is less clear whether changes in FS reflect changes in the clinical course of NAFLD. Accordingly, it is uncertain whether the acute improvement in FS with apabetalone compared with placebo reflects modulation of the underlying pathologic processes leading to NAFLD and its progression, or simply modulation of the biochemical contributors to FS. To make this distinction would require correlation of FS with concurrent hepatic histology, imaging, or biomechanics. Third, the number of patients with FS score consistent with low likelihood of NAFLD was relatively small and the assessment of treatment effect by apabetalone vs. placebo in this group may therefore have been underpowered. Finally, this analysis is post-hoc and therefore should be considered hypothesis-generating. The primary clinical outcome of the BETonMACE trial did not attain statistical significance, so any inference of efficacy in a subgroup must be viewed as exploratory. Notwithstanding these limitations, the CVD reduction observed in the FS+ category is consistent with apabetalone's BET-inhibition mode of action. A prospective, placebo-controlled evaluation of hepatic and CV effects of apabetalone in patients with NAFLD is warranted.

CRediT authorship contribution statement

Peter P. Toth: Writing – original draft. Gregory G. Schwartz: Writing – original draft. Stephen J. Nicholls: Writing – review & editing. Aziz Khan: Formal analysis. Michael Szarek: Formal analysis. Henry N. Ginsberg: Writing – review & editing. Jan O. Johansson: Writing – original draft. Kamyar Kalantar-Zadeh: Writing – review & editing. Ewelina Kulikowski: Writing – review & editing. Ken Lebioda: Writing – review & editing. Norman C.W. Wong: Writing – review & editing. Michael Sweeney: Writing – review & editing. Kausik K. Ray: Writing – review & editing.

Declaration of Competing Interest

GGS has received research support to the University of Colorado from AstraZeneca, Resverlogix, Roche, Sanofi, and The Medicines Company. He is coinventor of pending US patent 62/806,313 (“Methods for Reducing Cardiovascular Risk”) assigned in full to the University of Colorado. SJN reports research grants from AstraZeneca, Amgen, Anthera, Eli Lilly, Esperion, Novartis, Cerenis, The Medicines Company, Resverlogix, InfraReDx, Roche, Sanofi-Regeneron and LipoScience and honoraria from AstraZeneca, Akcea, Eli Lilly, Anthera, Omthera, Merck, Takeda, Resverlogix, Sanofi-Regeneron, CSL Behring, Esperion, Boehringer Ingelheim. PPT reports speaker bureau compensation from Amgen, Esperion, Novo-Nordisk; consultant compensation from Amarin, Amgen, Kowa, and Novartis. AK, JJ, EK, MS, and NCWW are employees of Resverlogix Corporation. KKZ has received honoraria from Abbott, Abbvie, ACI Clinical, Akebia, Alexion, Amgen, Ardelyx, Astra-Zeneca, Aveo, Braun, Cara Therapeutics, Chugai, Cytokinetics, Daiichi, DaVita, Fresenius, Genentech, Haymarket Media, Hospira, Kabi, Keryx, Kissei, Novartis, Pfizer, Regulus, Relypsa, Resverlogix, Dr. Schaer, Sandoz, Sanofi, Shire, Vifor, UpToDate, and ZS-Pharma. HNG has received research support from AstraZeneca, Amgen, Pfizer and honoraria from Astrazeneca, Amgen, Pfizer, Merck, Kowa, Resverlogix, Regeneron, Sanofi. KKR reports honoraria from Resverlogix, Aegerion, Amgen, Pfizer, AstraZeneca, Cerenis, Akcea, The Medicines Company, Kowa, Novartis, Cipla, Eli Lilly, Algorithm, Takeda, Boehringer Ingelheim, Abbvie, Silence Therapeutics, Dr. Reddys, Bayer, Daiichi Sankyo, Esperion, Zuelling Pharma, Sanofi-Regeneron and Merck and grants from Sanofi-Regeneron and Merck.