EPA Rises From a Dead Sea of Omega-3: The Landmark REDUCE-IT Trial

Editor's Note: Commentary based on Bhatt DL, Steg G, Miller M, et al. Cardiovascular risk reduction with icosapent ethyl for hypertriglyceridemia. N Engl J Med 2018. Epub ahead of print.

As a treatment target, triglycerides (TGs) have lived in the shadow of low-density lipoprotein cholesterol (LDL-C). The latter took the spotlight as the dominant lipid risk factor because of efficacious therapies that cut rates of heart attacks and strokes. In particular, statin drugs have been central to our strategy for managing atherosclerotic vascular disease (ASCVD) since 1988. The development of PCSK9 inhibitors has been another great success in cardiology driving the focus of treatment on LDL-C.

Lipid particles beyond LDL-C have proved more challenging therapeutic targets. Prior trials using niacin to raise HDL-C or fibrates to lower TGs were disappointing, failing to show a benefit on top of statin therapy in those with ASCVD. However, it has remained an open question whether TG-lowering therapy might be beneficial as an adjunctive ASCVD risk reducing treatment in the patients with elevated TG levels despite statin treatment.

The rationale for targeting TG levels is that elevated TGs reflect elevated levels of atherogenic "remnant-like particles." TGs absorbed from the gut are transported through the circulation in chylomicrons, and TGs synthesized in the liver are packaged into VLDL particles (very low-density lipoproteins). Both particle types undergo lipolysis in the circulation, depleting the TGs and leaving cholesterol-rich particles behind. These "remnant particles" are supposed to be recycled back to the liver. However, when circulating in excess, they can enter the arterial wall and be taken up by intimal macrophages in the vasculature, like LDL, leading to formation of foam cells and progression of plaques. Unlike LDL, remnant particles can be directly taken up by intimal macrophages without oxidation. These particles are also not fully accounted for as part of the estimated LDL-C, and they likely contribute to residual ASCVD risk despite appropriate treatment of LDL-C with a statin.

Current evidence suggests that remnant-like particles (remnants) promote atherosclerotic disease, and currently we do not routinely measure remnants and do not specifically treat them. Statin treatment increases re-uptake of LDL-C (which are the product of VLDL remnant metabolism), but it does not address the production of remnant-like particles. Lowering TG levels may be able to reduce the secretion of VLDLs in the first place, thereby reducing the level of atherogenic remnant particles. Since remnant-like particles are the "left-overs" of distributing TGs to the peripheral tissues, patients with excess TGs put this system into overdrive, creating more remnant-like particles and greater atherosclerotic burden. This population is the most likely to benefit from TG-lowering therapy.

Interestingly, although hypertriglyceridemic patients would seem the most likely to benefit from TG-lowering treatment, they were not the targeted population in most previous trials of omega-3 fatty acids and fibrates. Additionally, most previous omega-3 studies have used relatively low doses, and a mix of EPA and DHA, which have distinct effects on lipid metabolism.

DHA specifically has been noted to raise LDL-C, while EPA does not. Only one trial has examined EPA specifically for cardiovascular events, JELIS (Japan Eicosapentaenoic acid [EPA] Lipid Intervention Study), a Japan-only study which compared EPA plus statin treatment to statin alone. JELIS also used a markedly higher dose of EPA (1.8g/day) compared to other omega-3 studies (240-840mg/day). In contrast to other studies, JELIS did show a reduction in cardiovascular events, suggesting that omega-3 studies may have been hampered by inadequate dosing or by the use of omega-3 mixtures rather than pure EPA. The questionable generalizability of this single-country trial required validation in a large-scale, international randomized control trial.

So elevated TGs may represent an untapped therapeutic target. EPA has been shown to reduce TG levels in multiple studies and reduced cardiac events in the JELIS study. However, the reduction was driven by a soft endpoint: unstable angina. Moreover, additional omega-3 studies failed to replicate JELIS. Therefore, it has remained uncertain if treatment with EPA may be beneficial for patients with ASCVD and residual elevated TGs.

This brings us to REDUCE-IT (Reduction of Cardiovascular Events With Icosapent Ethyl–Intervention Trial), a randomized, double-blind, placebo-controlled trial designed to investigate the effect of high dose EPA in addition to statin treatment in a high cardiovascular risk population specifically with elevated TG levels (between 150 and 500 mg/dL). The trial enrolled over 8,000 patients with either known atherosclerotic disease (CAD, PAD, or cerebrovascular disease) or diabetes mellitus and at least one additional vascular risk factor, with a composite primary endpoint of cardiovascular death, nonfatal MI, nonfatal CVA, coronary revascularization or unstable angina requiring hospitalization.

The requirement for TG levels between 150 and 500 mg/dL selected for participants with abnormal TGs (but not high enough to meet current treatment recommendations) despite appropriate statin treatment (median LDL level was 75 mg/dL), addressing the potential issue that previous omega-3 trials did not specifically enroll patients with elevated TGs. REDUCE-IT also uses a significantly higher dose of EPA than prior omega-3 studies (4g per day), and pure EPA rather than an EPA/DHA mix.

These differences in study design may explain why REDUCE-IT reached its primary endpoint, with a dramatic 25% reduction in events (p < 0.001), and prior omega-3 trials largely did not. Important secondary endpoints were similarly reduced, including the hard endpoint of cardiovascular death. Subgroup analysis was consistent with the whole cohort, and the safety profile of EPA was favorable. The EPA arm showed a significant decrease in TG levels at 1 year (-18.3%), while the placebo arm had a small rise in TG levels (+2.2%). There was also an increase in LDL-C levels in both groups, +3.1% in the EPA group versus +10.2% in the placebo group. The mineral oil placebo pills (used to maintain blinding, since mineral oil has a similar appearance to pure EPA) may have contributed to raising LDL-C levels in the placebo group, however the rise in LDL-C of 7% (5 mg/dL) in the placebo group is unlikely to explain the observed effect.

Overall, REDUCE-IT demonstrated a significant cardiovascular benefit from adding EPA to statin treatment for high risk patients with elevated TG levels. The magnitude of benefit appears to exceed what can be explained by lipid effects, making anti-inflammatory, anti-thrombotic or other effects of EPA real considerations. While there is more to understand about mechanism of benefit, with the landmark REDUCE-IT trial EPA has risen from a dead sea of omega-3, and put TGs into the spotlight alongside LDL-C.

Clinical Topics: Diabetes and Cardiometabolic Disease, Dyslipidemia, Vascular Medicine, Hypertriglyceridemia, Lipid Metabolism, Nonstatins, Novel Agents, Statins

Keywords: Eicosapentaenoic Acid, Cholesterol, LDL, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Fibric Acids, Triglycerides, Niacin, Foam Cells, Fatty Acids, Omega-3, Mineral Oil, Risk Factors, Double-Blind Method, Chylomicrons, Lipolysis, Lipid Metabolism, Cardiovascular Diseases, Cholesterol, Atherosclerosis, Angina, Unstable, Myocardial Infarction, Stroke, Lipoproteins, VLDL, Cerebrovascular Disorders, Diabetes Mellitus, Liver, Metabolic Syndrome X, Dyslipidemias

< Back to Listings