Treatment With Icosapent Ethyl to Reduce Ischemic Events in Patients With Prior Percutaneous Coronary Intervention: Insights From REDUCE‐IT PCI

Background Patients who undergo percutaneous coronary intervention (PCI) are at increased risk for recurrent cardiovascular events despite aggressive medical therapy. Methods and Results This post hoc analysis focused on the subset of patients with prior PCI enrolled in REDUCE‐IT (Reduction of Cardiovascular Events With Icosapent Ethyl–Intervention Trial), a multicenter, randomized, double‐blind, placebo‐controlled trial of icosapent ethyl versus placebo. Icosapent ethyl was added to statins in patients with low‐density lipoprotein cholesterol <100 mg/dL and fasting triglycerides 135–499 mg/dL. The primary end point was a composite of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, coronary revascularization, or unstable angina requiring hospitalization. There were 8179 patients randomized in REDUCE‐IT followed for a median of 4.9 years, and 3408 (41.7%) of them had a prior PCI with a median follow‐up of 4.8 years. These patients were randomized a median of 2.9 years (11 days to 30.7 years) after PCI. Among patients treated with icosapent ethyl versus placebo, there was a 34% reduction in the primary composite end point (hazard ratio [HR], 0.66; 95% CI, 0.58–0.76; P<0.001; number needed to treat4.8 years=12) and a 34% reduction in the key secondary composite end point of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke (HR, 0.66; 95% CI, 0.56–0.79; P<0.001; NNT4.8 years=19) versus placebo. Similarly, large reductions occurred in total coronary revascularizations and revascularization subtypes. There was also a 39% reduction in total events (rate ratio, 0.61; 95% CI, 0.52–0.72; P<0.001). Conclusions Among patients treated with statins with elevated triglycerides and a history of prior PCI, icosapent ethyl substantially reduced the risk of recurrent events during an average of ~5 years of follow‐up with a number needed to treat of only 12. Registration URL: https://www.clinicaltrials.gov; Unique identifier: NCT01492361.

Peterson et al REDUCE-IT PCI P atients who undergo percutaneous coronary intervention (PCI) are at increased risk for subsequent cardiovascular events when compared with patients with other cardiovascular risk factors. 1 In recent years, efforts to improve stent design, lower low-density lipoprotein cholesterol, and modify inflammation and platelet activity have resulted in some reductions in repeat events among patients who undergo coronary stenting. [2][3][4] Yet, many patients still experience recurrent events, especially those with diabetes and elevated triglycerides. [5][6][7][8] The REDUCE-IT (Reduction of Cardiovascular Events With Icosapent Ethyl-Intervention Trial) trial was designed to test the effectiveness of icosapent ethyl 4 g/day (a highly purified form of eicosapentaenoic acid [EPA]) versus placebo among patients with established cardiovascular disease or diabetes and additional risk factors. 9,10 The significant reduction in first and total major adverse cardiovascular events (MACE) among patients who were treated with icosapent ethyl was out of proportion to the degree of reduction in triglycerides. [11][12][13][14] These large reductions occurred in patients with diabetes, patients with only modestly elevated triglycerides, patients in the United States, and patients across numerous other prespecified subgroups. [15][16][17][18] Treatment with icosapent ethyl also substantially reduced instances of first and subsequent revascularization events. [19][20][21] The aim of the present post hoc analysis of the REDUCE-IT trial was to study the effects of icosapent ethyl versus placebo among patients who have been treated previously with PCI.

METHODS
The data that support the findings of this study may be made available from the corresponding author on reasonable request.

Patient Population and Treatment
The design of the REDUCE-IT trial has been published previously. 9 REDUCE-IT was a double-blind, multicenter, placebo-controlled, randomized trial comparing the effects of icosapent ethyl in high-risk patients treated with statins with persistently elevated triglycerides. After a screening period of up to 60 days, patients were randomized to receive icosapent ethyl 4 g daily (2 g twice daily) versus a matching placebo.
Patients were enrolled in REDUCE-IT if they were at least 45 years of age and had established cardiovascular disease or at least 50 years of age and had diabetes and additional risk factors. In this present post hoc analysis, patients were analyzed only if they had a prior PCI, such as balloon angioplasty or stenting (drug-eluting or bare-metal stents). Patients were included regardless of the amount of time elapsed between PCI and enrollment, though planned coronary intervention (such as PCI or coronary bypass surgery) was an exclusion criterion. Patients could be (re)evaluated for participation in the trial (starting with Visit 1.1) after their recovery from the intervention/surgery. Of note, randomization to icosapent ethyl versus placebo was stratified according to cardiovascular risk (established cardiovascular disease versus diabetes plus risk), geographic region, and ezetimibe use. In addition to prior PCI, all patients had been treated with a stable dose of statin for at least 4 weeks and had low-density lipoprotein cholesterol under 100 mg/dL as well as

CLINICAL PERSPECTIVE
What Is New?
• Icosapent ethyl greatly reduced first occurrences of cardiovascular events among patients who had elevated triglycerides despite statin therapy and a history of prior percutaneous coronary intervention; number needed to treat 4.8 years =12. • There were also significant reductions in total ischemic events (first and subsequent), cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, hospitalization for unstable angina, and coronary revascularization.
What Are the Clinical Implications?
• Patients with prior percutaneous coronary intervention and elevated triglycerides despite statin therapy are at extremely high risk for recurrent cardiovascular events. • Icosapent ethyl could benefit a large proportion of patients with a history of prior percutaneous coronary intervention, and such patients should be screened for eligibility. serum triglycerides from 135-499 mg/dL. Other key inclusion and exclusion criteria for REDUCE-IT have been published previously. All sites received ethics approval from relevant institutional review boards, and informed consent was obtained.

Statistical Analysis
In this post hoc analysis, we analyzed patients enrolled in REDUCE-IT who had a prior PCI. The primary and key secondary end points for this analysis were the same as the main REDUCE-IT trial. The primary composite end point was the first occurrence of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, coronary revascularization, or unstable angina requiring hospitalization. The key secondary composite end point (or hard MACE end point) was the first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. The intention-to-treat principle guided all analyses. Baseline characteristics were compared among groups using the Wilcoxon rank sum test for continuous variables and Chi-square test for categorical variables. Hazard ratios (HRs) and 95% CIs were generated using Cox proportional-hazard models that included risk stratum (established cardiovascular disease versus diabetes plus cardiovascular risk factors), geographic region, and ezetimibe use as covariates. It has been shown in other analyses that patients benefited from icosapent ethyl versus placebo regardless of baseline triglyceride levels, so this was not included as a covariable in this analysis. With Kaplan-Meier analysis, we compared the time to events among patients randomized to icosapent ethyl versus placebo, with log-rank P values also stratified by risk stratum, geographic region, and ezetimibe use.
As with other REDUCE-IT analyses, we employed various statistical methods in comparing the risk for total (first and subsequent) events among patients treated with icosapent ethyl versus placebo. 17 We used the negative binomial regression model to calculate rates and rate ratios (RRs) for total cardiovascular events. In supportive analyses, the modified Wei-Lin-Weissfeld method (Li and Lagakos modification taking into account death as a terminating event) was applied to calculate HRs for the time to the first and second, and a negative binomial model for rate ratios of third and greater events. 22 As a sensitivity analysis, the Gray's test was applied to the primary composite end point considering noncardiovascular death as a competing event. In addition to the primary and key secondary end points, results for additionally prespecified secondary end points in the original testing hierarchy are presented. Further post hoc explorations included time to total coronary revascularization and various revascularization subtypes (eg, elective, emergent, and urgent) as well as a coronary-specific composite end point of myocardial infarction, coronary revascularization, or unstable angina. All statistical analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC).

Baseline Characteristics
Of the 8179 patients enrolled in REDUCE-IT, 3408 (41.7%) had a prior PCI. In the 2559 patients with reported dates of PCI, the median time from PCI was 2.9 years, ranging from 11 days to 30.7 years. There were 675 (26.4%) patients with a PCI ≤1 year before randomization and 1884 (73.6%) with a PCI more than 1 year before randomization. Among patients in this study, the median age was 63 years, 20.7% were female, 96.3% were on moderate-or high-intensity statin therapy, and the median triglyceride level was 218 mg/ dL (Q1, Q3; 178.5 mg/dL, 274.5 mg/dL). There were no significant differences in baseline characteristics among patients randomized to icosapent ethyl versus placebo (Table 1).
Patients treated with icosapent ethyl experienced a significant 40% reduction in the risk of repeat coronary revascularization versus those treated with placebo (17.1% versus 27.6%; HR, 0.60; 95% CI, 0.51-0.70; P<0.001), with similar reductions in elective and urgent revascularization. There was also a significant reduction in the combined coronary end point of myocardial infarction, coronary revascularization, or unstable angina requiring hospitalization (HR, 0.65; 95% CI, 0.56-0.75; P<0.001) ( Figure 2). There was no significant difference in the safety or efficacy of icosapent ethyl versus placebo among patients taking single-or dual-antiplatelet therapy or a combined antithrombotic regimen ( Figure S1). In addition, there were similar reductions in cardiovascular end points among women and men randomized to icosapent ethyl versus placebo ( Figure S2).
Testing in patients with prior PCI across the original prespecified hierarchical end points showed significant reductions in the primary and key secondary end points as well as in the following end points: cardiovascular death or nonfatal myocardial infarction; fatal or nonfatal myocardial infarction; urgent or emergent coronary revascularization; cardiovascular death; hospitalization for unstable angina; fatal or nonfatal stroke; and all-cause mortality, myocardial infarction, or stroke ( Figure 3). There were similar reductions in the primary and key secondary end points when accounting for noncardiovascular death as a competing risk factor ( Figure S3). It should be noted that although the patients enrolled in REDUCE-IT with cardiovascular risk factors and no history of PCI were a somewhat heterogeneous group, they had fewer cardiovascular events and derived a smaller in magnitude but still significant benefit from treatment with icosapent ethyl versus placebo ( Figure S4).

DISCUSSION
Among the 3408 patients in REDUCE-IT with a prior PCI, icosapent ethyl taken 4 g daily (2 g twice daily) versus placebo resulted in a significant 34% reduction in the primary end point and a significant 34%  After an Acute Coronary Syndrome During Treatment With Alirocumab). [25][26][27] The patient population in this subgroup analysis of REDUCE-IT reflects a large proportion of patients who undergo PCI in US, Canadian, and European registries. [28][29][30][31] Representative qualities include moderately elevated baseline triglycerides, with 50% of patients being <218 mg/dL and 96.3% on a moderate-or highintensity statin. The distribution of age, sex, and typical comorbidities also highly reflects contemporary populations undergoing PCI. 1 Furthermore, this trial enrolled patients from November 2011 to August 2016 as the latest generation of drug-eluting stents were employed, ameliorating the usual difficulty of interpreting clinical events in trials that had events during periods that used prior PCI technologies. 2 Contemporary guidelines and consensus statements consistently recommend the use of icosapent ethyl in this patient population. 32,33   reduction in MACE at 4 g per day in REDUCE-IT. 11,42 In each of these trials, and even recently among patients with COVID-19, high-dose EPA in the form of icosapent ethyl has been well tolerated. 43,44 It may be that any benefit from omega-3 fatty acid treatment is most directly tied to EPA alone, whereas other omega-3 fatty acids, such as docosahexaenoic acid, may attenuate the benefit of EPA. 16,45 In REDUCE-IT EPA, the on-treatment levels of EPA correlated directly with reduction in the primary, key secondary, and individual cardiovascular end points. 46 This suggests a dose-dependent class effect specific to EPA rather than a broader impact associated with omega-3 fatty acid supplementation alone. In REDUCE-IT, the average patient taking icosapent ethyl experienced a 386% increase in serum EPA levels. Although some of the patients in STRENGTH receiving an EPA and docosahexaenoic acid mixture experienced a similar degree of elevation of EPA levels, they did not reap the same benefit. 34 Further investigation may inform why this same benefit was not seen in the presence of simultaneous high-dose docosahexaenoic acid supplementation. However, it is known that EPA and docosahexaenoic acid have very different tissue distribution and disparate effects on membrane stabilization and fluidity, formation of cholesterol rafts and crystals, rates of lipid oxidation in lipoproteins and cells, inflammatory modulation, transcriptional regulation, and endothelial function. [47][48][49][50][51] Although the precise molecular mechanism of benefit from icosapent ethyl/EPA still requires some elucidation, the EVAPORATE (Effect of Vascepa on Improving Coronary Atherosclerosis in the People With High Triglycerides Taking Statin Therapy) trial has recently shed some important light on the gross vascular mechanism of benefit. 52,53 A total of 80 patients who had atherosclerotic coronary plaques with at least 20% stenosis on multidetector coronary computed tomography and a median baseline fasting triglyceride level of 259 mg/dL were randomized to icosapent ethyl 4 g daily versus placebo. Final follow-up imaging at 18 months showed a significant 17% reduction in low-attenuation plaque volume in patients treated with icosapent ethyl (whereas patients taking placebo nearly doubled their low-attenuation plaque volume). Significant reductions also occurred in fibrofatty, fatty, total noncalcified, and total plaque volumes. Thus, high-dose EPA therapy seems to result in significantly increased plaque stability and even plaque reduction, which could at least partially contribute to this marked reduction in cardiovascular events among high-risk patients.

Limitations
The exploratory nature as well as the lack of adjustment for multiple comparisons limited this post hoc analysis; REDUCE-IT was not powered for this or other subgroup analyses. With the post hoc nature of these analyses, all P values should be considered hypothesis-generating. As noted previously, patients with prior PCI had higher event rates and derived greater benefit from icosapent ethyl versus placebo. Although the time period from PCI to randomization was known in most patients, there was a subset of patients in which this was not known. Among the 24.9% of this subset and the remainder of patients, the distribution of randomization to icosapent ethyl versus placebo was equivalent. Double blinding also eliminated bias arising from this issue. The prerandomization extent of coronary artery disease and revascularization strategy (complete versus incomplete) among patients with a prior PCI was not known. Randomization was not stratified by history of PCI, and because there is potential for confounding, this subgroup finding needs corroboration in future studies. Future investigation will be required to gain a better understanding of whether icosapent ethyl reduced the rates of in-stent restenosis versus de novo plaque events as vessel-and lesionspecific data are not available in REDUCE-IT. Also, had patients been enrolled soon after PCI when risk is highest, the degree of benefit seen here may have been even greater, especially if future studies validate the use of a loading dose. 43 CONCLUSIONS Icosapent ethyl versus placebo resulted in significant and clinically meaningful reductions in cardiovascular events in this post hoc analysis. In patients with a prior PCI, the reductions in first and total primary end point events were 34% and 39%, respectively. There were large reductions in the primary and key secondary (hard MACE) end points, with NNTs 4.8 years of 12 and 19, respectively, and consistent benefit across the hierarchical end points. These data highlight the substantial positive impact of icosapent ethyl on patients in the REDUCE-IT population, including patients with a history of prior PCI.
reports receiving research grants and/or speaking honoraria from Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, EOCI Pharmacomm Ltd, HLS Therapeutics, Janssen, Merck, Novartis, Novo Nordisk, Sanofi, Sun Pharmaceuticals, and the Toronto Knowledge Translation Working Group. He is the President of the Canadian Medical and Surgical Knowledge Translation Research Group, a federally incorporated not-for-profit physician organization. Dr Ballantyne reports receiving consulting fees from Arrowhead, AstraZeneca, Eli Lilly, Matinas BioPharma, Merck, Boehringer Ingelheim, Novo Nordisk, Denka Seiken, and Gilead and grant support (paid to his institution) and consulting fees from Amarin, Amgen, Esperion, Novartis, Regeneron, Sanofi-Synthelabo, and Akcea. Dr Peterson has no disclosures to report.

Figure S3. Cumulative Incidence Plots of A. Primary Composite End Point and B. Key Secondary Composite End Point with Non-CV Death as Competing Risk Among Patients with Prior PCI.
A.