
==== Front
Am J Prev Cardiol
Am J Prev Cardiol
American Journal of Preventive Cardiology
2666-6677
Elsevier

S2666-6677(24)00083-7
10.1016/j.ajpc.2024.100715
100715
Commentary
Bempedoic Acid in Secondary Prevention
Mousavi Idine idine.mousavi@bcm.edu
a⁎
Nambi Vijay abc
Abushamat Layla A. ab
Al-Kindi Sadeer G. d
Shapiro Michael D. e
Sperling Laurence f
Virani Salim S. agh
Minhas Abdul Mannan Khan ab
a Department of Medicine, Baylor College of Medicine, Houston, TX, USA
b Division of Atherosclerosis and Vascular Medicine, Department of Medicine, Baylor College of Medicine, Houston, TX, USA
c Section of Cardiology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, Texas
d Department of Cardiology, Houston Methodist Hospital, TX, USA
e Section on Cardiovascular Medicine, Wake Forest University School of Medicine, Winston-Salem, NC, USA
f Division of Cardiology, Emory Clinical Cardiovascular Research Institute, Emory University School of Medicine, Atlanta, GA, USA
g Aga Khan University, Karachi, Pakistan
h Baylor College of Medicine and Texas Heart Institute, Houston, TX, USA
⁎ Corresponding author at: 1 Baylor Plaza, Houston, TX 77030, USA. idine.mousavi@bcm.edu
06 8 2024
9 2024
06 8 2024
19 10071528 4 2024
18 7 2024
2 8 2024
© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Highlights

• Bempedoic acid has been shown to reduce major adverse cardiovascular outcomes in patients unable to take statins due to statin-associated side effects.

• Analysis of CLEAR Outcomes trial data reveals possible differences in baseline characteristics between the primary and secondary prevention subgroups.

• Further research is needed to optimize use of bempedoic acid and clarify its impact on cardiovascular outcomes in diverse patient populations.

Graphical abstract

Image, graphical abstract
==== Body
pmcBempedoic acid (BA) is an oral adenosine triphosphate citrate lyase inhibitor that was shown to reduce the risk of major adverse cardiovascular events (MACE) in patients unable to take statins due to statin-associated side effects (SASEs) who are at high cardiovascular risk or have established atherosclerotic cardiovascular disease (ASCVD) compared to placebo in the Cholesterol Lowering via Bempedoic Acid, an ACL-Inhibiting Regimen (CLEAR) Outcomes, a randomized clinical trial of approximately 14,000 patients [1]. Most of the trial population (∼70 %) in CLEAR Outcomes comprised of patients with established ASCVD; however, a prespecified subgroup analysis showed a lower hazard ratio for the primary MACE endpoint in primary prevention compared to the secondary prevention population. Recently, a detailed analysis of the primary prevention subgroup was published which showed significant net benefit with regards to MACE outcomes and overall mortality in this population [2]. In the following commentary, we explore available data from CLEAR Outcomes to investigate the possible differential response to treatment with BA among primary vs. secondary prevention participants.

Although recent evidence challenges the current paradigm of classifying ASCVD risk into binary primary and secondary preventive groups given the heterogeneity of risk within these categories, it is expected that, on average, patients who have already experienced ASCVD events are at higher risk than those who have yet to experience an initial event and will consequently derive greater benefit from additional low-density lipoprotein cholesterol (LDL-C) lowering and risk factor control [3]. Furthermore, a prespecified analysis of CLEAR Outcomes data revealed a greater clinical benefit with BA in patients experiencing an increasing number of total events [4]. Given that all patients that experienced multiple events are in the secondary prevention group, this raises the question that if BA does indeed provide greater benefit in patients with increasing numbers of cardiovascular events, what may explain that BA favored the primary prevention cohort compared to the secondary prevention population in the CLEAR Outcomes trial, the latter at baseline higher risk for future MACE?

As a nonstatin lipid-lowering agent, BA may be more commonly used in the secondary preventive setting to achieve more stringent guideline-directed risk factor control. Thus, it is important to explore factors that may explain any potential treatment effect heterogeneity between the primary and secondary prevention strata in CLEAR Outcomes. A recent post-hoc analysis that reconstructed secondary prevention data found smaller reductions in MACE and a potential increased risk of mortality in this group using Bayesian logistic regression analyses [5]. This analysis highlights the need to investigate if true effect heterogeneity exists between these groups or whether these observations represent findings of chance. Therefore, we deduced prevalence and outcome data in the secondary prevention group by subtracting select data points reported in a separate analysis of the primary prevention group from the corresponding values in the overall trial [2]. We reported data in the secondary prevention group only if the data were available in both the overall trial and the primary prevention analysis which are summarized in Table 1. In addition to outcome data, we also reported the prevalence of select comorbidities in both groups.Table 1 Baseline Characteristics and Outcome, Lipid and Inflammatory Biomarker Data Stratified by Primary and Secondary Preventive Cohorts.

Table 1Characteristic/Outcome	Prim Prev - BA (N = 2100)	Prim Prev - Placebo (N= 2106)	Secondary Prev - BA (N = 4892)	Secondary Prev - Placebo (N = 4872)	Total - BA	Total - Placebo	
Patients with CKD	146 (7 %)	155 (7.4 %)	1291 (26 %)	1289 (26.4 %)	1437 (20.6 %)	1444 (20.7 %)	
Baseline Statin Use	394 (19 %)	417 (19.8 %)	1207 (25 %)	1156 (24 %)	1601 (22.9 %)	1573 (22.5 %)	
Additional Lipid-Lowering Therapies	141 (6.7 %)	261 (12.4 %)	516 (11 %)	827 (17 %)d	657 (9.4 %)	1088 (15.6 %)	
Diabetes	1369 (65.2 %)	1412 (67 %)	1775 (36 %)	1817 (37 %)	3144 (45 %)	3229 (46.3 %)	
Inadequately Controlled Diabetes	569 (27 %)	593 (28.2 %)	787 (16 %)	776 (16 %)	1356 (19.4 %)	1369 (19.6 %)	
LDL-C at 6 mo (mg/dL)	108.2	138.6	-	-	107	136	
% Change in LDL-C from baseline	-23.9 %	-2.87 %	-	-	-21.1 %	-0.8 %	
% Change in hsCRP from baseline	-26.8 %	3.28 %	-	-	-22.2 %	2.4 %	
Renal Impairment	216 (10.3 %)	170 (8.1 %)	586 (12 %)	429 (8.7 %)	802 (11.5 %)	599 (8.6 %)	
Primary Endpoint (4-comp MACE)*	111 (5.3 %)	161 (7.6 %)	708 (14.5 %)	766 (15.7 %)	819 (11.7 %)	927 (13.3 %)	
Secondary Endpoint (3-comp MACE)⁎⁎	83 (4.0 %)	134 (6.4 %)	492 (10 %)	529 (10.8 %)	575 (8.2 %)	663 (9.5 %)	
CV Deaths	37 (1.8 %)	65 (3.1 %)	232 (4.7 %)	192 (4 %)	269 (3.8 %)	257 (3.7 %)	
All Cause Mortality	75 (3.6 %)	109 (5.2 %)	359 (7.3 %)	311 (6.4 %)	434 (6.2 %)	420 (6 %)	
Abbreviations: CKD: chronic kidney disease. hsCRP: high-sensitivity C-reactive protein. LDL-C – low-density lipoprotein-cholesterol. CV deaths – cardiovascular deaths. MACE – major adverse cardiovascular events.

% Change in LDL-C from baseline was calculated at 6 months.

% Change in hsCRP from baseline was calculated at 6 months in the overall trial and 12 months in the primary prevention subgroup based on available data.

⁎ Time to first occurrence of death from cardiovascular causes, nonfatal MI, nonfatal stroke, or coronary revascularization.

⁎⁎ Time to first occurrence of composite endpoint of death from cardiovascular causes, nonfatal MI, or nonfatal stroke.

1 Baseline Comorbidities in Primary vs. Secondary Prevention in CLEAR Outcomes

We observed that the distribution of comorbidities differed between the primary and secondary prevention subgroups. Diabetes was more prevalent in the primary prevention subgroup (65 % vs. 36 %) whereas chronic kidney disease (CKD) was more prevalent in the secondary prevention subgroup (26 % v. 7 %). These observations raise important questions regarding the impact of comorbidities on the efficacy of non-statin therapies such as BA. The observed differences in baseline comorbidities may have differentially affected both the residual risk of adverse cardiovascular outcomes and the pharmacodynamics of BA between the primary and secondary prevention cohorts, potentially altering its risk-benefit profile. A recent prespecified subgroup analysis of the efficacy of BA on a composite 4-point MACE endpoint by glycemic status revealed that there was no statistically significant effect modification across glycemic strata however the study was not formally powered to assess interaction analyses. There were greater cardiovascular event rates and greater absolute reductions noted in patients with diabetes (AR: 2.4 %, 95 % CI: 0.7-4.0 %, p = 0.0063) [6]. The greater absolute reduction in patients with diabetes in the prespecified analysis was largely attributed to a higher baseline absolute risk of cardiovascular events. Given that patients with diabetes are at higher absolute risk for ASCVD events compared to those without diabetes, studies of other lipid-lowering therapies (LLTs) in large-scale outcome trials have consistently shown greater absolute reductions of cardiovascular events from LDL-C lowering in individuals with diabetes compared to those without diabetes [7,8]. Although some studies have shown similar relative reductions, The Improved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE-IT) trial demonstrated both greater absolute and relative benefit among patients with diabetes compared to those without diabetes. Could a higher prevalence of diabetes with a corresponding greater absolute risk reduction of BA in diabetic patients have contributed to favoring the primary prevention subgroup?

CKD is a known risk-enhancing factor for the development of ASCVD and substantially increases risk for recurrent ASCVD events. While LLTs have shown benefit across varying stages of CKD, evidence suggests the relative risk reduction per LDL-C lowering may be reduced as renal function declines, consistent with a prior meta-analysis of 28 randomized trials of statin-based therapies [9,10]. Randomized evidence of LLTs in the CKD population remains limited and includes the Study of Heart and Renal Protection (SHARP) trial, which showed that simvastatin/ezetimibe combination therapy reduced the risk of MACE compared with placebo in CKD patients with eGFR categories G3a–G5 [11]. However, SHARP was underpowered to separately detect the efficacy of statins across different CKD stages. The effect of LLTs on MACE risk in patients with varying stages of kidney function remains an important area of further study. We consider the point that as the secondary prevention subgroup had a higher prevalence of CKD, could it be that patients with baseline kidney impairment derive less net benefit from BA? In CLEAR Outcomes, when stratified by baseline eGFR category, results for the primary endpoint were as follows: eGFR <60 mL/min/1.73m2 (15.1 % vs 16.9 %, HR: 0.89, 95 % CI: 0.74, 1.07); ≥60 to ≤90 (11.4 % vs 13.1 %, HR: 0.85, 95 % CI: 0.76, 0.96) and ≥ 90 (8.7 % vs 9.5 %, HR: 0.91, 95 % CI: 0.70, 1.18). Approximately 62 % of the patients were in ≥60 to ≤90 category. Data on further stratification by eGFR were not available and interpretation of these results are limited. Given the possible heterogenous effects of LLTs on MACE risk in patients with varying levels of kidney function, investigating the efficacy of BA on ASCVD events in patients with varying stages of CKD represents an opportunity for further study to guide the use of BA in this population. The presence of high-risk comorbidities may warrant additional non-statin anti-atherosclerotic therapies for prevention of cardiovascular events, depending on the specific comorbidities in question and the patient's individual clinical scenario. These observations emphasize the importance of investigating potential effect modification by comorbidity status on the efficacy of BA in diverse patient populations.

2 Outcomes and Inflammatory and Lipid Biomarkers - Limitations and Future Directions

Regarding reported outcome data, we observed that absolute rates of cardiovascular deaths were numerically higher in the BA arm in the secondary prevention subgroup compared to placebo (4.7 % v. 4 %). Similar findings were observed for all-cause mortality (7.3 % v. 6.4 %). It is important to note that these are differences in absolute rates derived from aggregate data and may not be statistically significant. In a separate analysis of primary prevention patients, BA was associated with significant reduction (HR: 0.70, 95 % CI 0.55-0.89; p = .002) in the primary 4-component MACE endpoint along with reductions in myocardial infarction (MI), cardiovascular death, and all-cause mortality. The limitations of drawing conclusions from these data have been noted previously [12]. However, given the magnitude and consistency of reduction in adverse cardiovascular outcomes in the primary prevention group, a corresponding analysis of the secondary prevention group is important in further clarifying which patient groups stand the most to benefit from BA. Regarding lipid and inflammatory biomarkers, the least-squares mean percent reduction in LDL-C at 6 months in the BA arm of the primary prevention group was 21.3 % compared with placebo, similar to the mean percent change observed in the overall trial between the BA arm and placebo at 6 months (20.3 %). However, it is difficult to ascertain whether differences in statistical methods may have influenced reported reductions in LDL-C levels. Notably, in a subgroup analysis of the phase 3 CLEAR Harmony trial, LDL-C lowering with BA among those with ASCVD at baseline was (least squares mean difference (LSMD) -18.6 % (-20.6 % to -16.7 %) compared to those without ASCVD (LSMD: -24.8 % (-47.1 % to -2.6 %) p-interaction = 0.43) [13]. However, the limited sample size of primary prevention patients (n = 51) vs. secondary prevention (n = 2098) and non-significant interaction p-value limits the application of this data as an explanation for a potential difference in treatment effect in CLEAR Outcomes. Based on available data, LDL-C lowering by BA compared to placebo in the primary prevention group and overall cohort were similar in CLEAR Outcomes, however these data warrant further investigation. Percent reduction of high-sensitivity C-reactive protein (hsCRP) levels was 21.5 % after 12 months compared to placebo in the primary prevention subgroup. In the overall trial, there was a 21.6 % reduction in median hsCRP levels reported at 6 months compared to placebo [1]. Percent reduction of LDL-C and hsCRP levels could not be reported for the secondary prevention subgroup from available data. Additionally, it has been previously suggested that the addition of lipid lowering therapies (LLTs) may have been used differently in the primary vs. secondary prevention subgroups given that patients in the placebo group were more likely to receive add-on therapy than those in the bempedoic acid group (15.6 % vs. 9.4 %, data not shown) [1]. With available data, we believe that both the primary and secondary prevention groups did not differ greatly in the absolute difference in percentage of additional LLTs between the BA and placebo arms of each subgroup (5.7 % in primary prevention vs. 6 % in secondary prevention). This suggests that the addition of LLTs is less likely to have contributed to the attenuation of benefit seen in the secondary prevention group; however, we could not ascertain the strength nor categories of these agents. Secondary analyses of clinical trials of nonstatin LLTs such as ezetimibe and the agents evolocumab and bococizumab have evaluated the impact of these agents on residual inflammatory risk as indicated by hsCRP levels [[14], [15], [16]]. In similar fashion, a close analysis of the secondary prevention group that provides both levels of LDL-C and hsCRP reduction and data on how and what specific anti-atherosclerotic agents were used would be helpful and provide further clarification.

We note that these observations may likely represent a finding of chance rather than a true difference between these two populations. Moreover, our observations rely on aggregate data without statistical adjustment for baseline differences, limiting their strength and interpretability. The lack of direct subgroup data may result in overlooked confounding factors or biases and differences in endpoint rates between arms cannot be definitively stated without formal statistical testing as numerical imbalances do not necessarily reflect statistically significant differences. Of note, the interaction p-value between the primary and secondary prevention groups for the primary outcome in CLEAR Outcomes was significant (p = 0.03) yet the inability to analyze individual-level data limits our current understanding and our observations highlight specific areas for further study. Given that BA is an important tool in reducing ASCVD risk in patients unable to take statins due to SASEs, these questions would benefit from further investigation. A close analysis of the secondary prevention subgroup, including additional analyses by comorbidity status, would provide clarity of the role that these factors may play in modifying treatment effects with this agent. In doing so, clinicians would gain valuable insight into patient-specific factors that aid in optimizing patient selection for those that stand the most to benefit from BA and subsequently improve cardiovascular outcomes.

Funding

This manuscript was not funded.

Disclosures

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this manuscript. The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter discussed in the manuscript.

CRediT authorship contribution statement

Idine Mousavi: Writing – review & editing, Writing – original draft, Conceptualization. Vijay Nambi: Writing – review & editing. Layla A. Abushamat: Writing – review & editing. Sadeer G. Al-Kindi: Writing – review & editing. Michael D. Shapiro: Writing – review & editing. Laurence Sperling: Writing – review & editing. Salim S. Virani: Writing – review & editing. Abdul Mannan Khan Minhas: Writing – review & editing, Writing – original draft, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix Supplementary materials

Image, application 1

Image, application 2

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.ajpc.2024.100715.
==== Refs
References

1 Nissen SE Lincoff AM Brennan D Bempedoic Acid and cardiovascular outcomes in statin-intolerant patients N Engl J Med 388 15 2023 1353 1364 36876740
2 Nissen SE Menon V Nicholls SJ Bempedoic Acid for primary prevention of cardiovascular events in statin-intolerant patients JAMA 330 2 2023 131 140 37354546
3 Saba PS Al KS Nasir K Redefining Cardiovascular Risk Assessment as a Spectrum J Am Coll Cardiol 83 2024 574 576 38296401
4 Nicholls SJ Nelson AJ Lincoff AM Impact of Bempedoic Acid on total cardiovascular events: a prespecified analysis of the clear outcomes randomized clinical trial JAMA Cardiol 2024 Published online January 17
5 Sayed A Brophy JM. Effect of bempedoic acid on mortality and cardiovascular events in primary and secondary prevention: A post-hoc analysis of the CLEAR-outcomes trial Int J Cardiol 2024
6 Ray KK Nicholls SJ Li N Efficacy and safety of bempedoic acid among patients with and without diabetes: prespecified analysis of the CLEAR Outcomes randomised trial Lancet Diabetes Endocrinol 12 1 2024 19 28 38061370
7 Giugliano RP Cannon CP Blazing MA Benefit of Adding ezetimibe to statin therapy on cardiovascular outcomes and safety in patients with versus without diabetes mellitus: results from improve-it (improved reduction of outcomes: vytorin efficacy international Trial) Circulation 137 2018 1571 1582 29263150
8 Ray KK Colhoun HM Szarek M Effects of alirocumab on cardiovascular and metabolic outcomes after acute coronary syndrome in patients with or without diabetes: a prespecified analysis of the ODYSSEY OUTCOMES randomised controlled trial Lancet Diabetes Endocrinol 7 2019 618 628 31272931
9 Grundy SM Stone NJ Bailey AL AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines J Am Coll Cardiol 73 24 2019 e285 e350 30423393
10 Herrington W Emberson J Impact of renal function on the effects of LDL cholesterol lowering with statin-based regimens: a meta-analysis of individual participant data from 28 randomised trials Lancet Diabetes Endocrinol 4 10 2016 829 839 27477773
11 Baigent C Landray MJ Reith C The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial Lancet 377 2011 2181 2192 21663949
12 Kazi DS. Bempedoic Acid for High-Risk Primary Prevention of Cardiovascular Disease: Not a Statin Substitute but a Good Plan B JAMA 330 2 2023 123 125 37354548
13 Ray Kausik K. Bays Harold E Catapano Alberico L. Safety and Efficacy of Bempedoic Acid to Reduce LDL Cholesterol N Engl J Med 380 11 2019 1022 1032 30865796
14 Bohula EA Giugliano RP Leiter LA Inflammatory and Cholesterol Risk in the FOURIER Trial Circulation 138 2 2018 131 140 29530884
15 Pradhan AD Aday AW Rose LM Ridker PM. Residual Inflammatory Risk On Treatment with PCSK9 Inhibition and Statin Therapy Circulation 138 2 2018 141 149 29716940
16 Bohula EA Giugliano RP Cannon CP Achievement of dual low-density lipoprotein cholesterol and high-sensitivity C-reactive protein targets more frequent with the addition of ezetimibe to simvastatin and associated with better outcomes in IMPROVE-IT Circulation 132 13 2015 1224 1233 26330412
