Hello, I'm Gabriel Steg from Hôpital Bichat in Paris, France. Today I'd like to discuss perspectives on the results of recent clinical outcome trials with EPA.
These are my disclosures. Specifically, I want to mention that I've been an executive steering committee member for the REDUCE-IT trial.
Where do we start? What's the background for all of this? It's the JELIS trial, which I think was really instrumental. 10 years ago, there was a review of all the low-dose omega-3 trials that showed no significant cardiovascular benefit. One trial, however, stood out. It was a Japanese trial called JELIS, which showed cardiovascular risk reduction with 1.8 grams per day of EPA in Japanese hypercholesterolemic patients compared to control. Now, there was no placebo in that trial, but it was a large trial of 18,000 patients, and the benefit was seen in the overall population, but also in the primary and the secondary prevention subsets.
That provided the impetus for conducting REDUCE-IT, which, as you know, I'm sure, compared four grams of purified EPA to a mineral oil placebo in a double-blind placebo-controlled manner several years in patients that either established cardiovascular disease or diabetes with risk factors. The results are very clear. There was approximately a 25% reduction in the outcomes both in the overall trial population, but also in the primary and secondary subsets. The benefits appeared consistent for the primary composite outcome of CV death, MI, stroke, coronary revascularization, or unstable angina, or a more conventional triple composite outcome of CV death, MI, and stroke. The absolute benefit is important. Close to 5% for the primary composite. Close to 4% for the key secondary composite. The results were highly statistically significant. A pre-specified hierarchical analysis of secondary outcomes showed consistent benefits on almost all of the outcomes, including a statistically significant reduction in cardiovascular mortality by 20%. All-cause mortality was lowered by 13%, and that is the only secondary outcome that did not reach statistical significance.
Now, what about the safety in balance to the benefits? While safety is overall excellent, the treatment-emergent adverse events were not different between the two treatment arms. There were two highlights for safety, however. First, bleeding was increased with a greater frequency of all bleedings and of bleedings as an SAE. Second, the risk of atrial fibrillation or flutter was also increased modestly, but statistically significantly by icosapent ethyl compared to placebo, close to 7.9% versus 6.1%. Now, that is a signal that's been seen in prior trials of marine omega-3 fatty acids, in which there's been an increase in atrial fibrillation or flutter despite no apparent benefit on cardiovascular outcomes. Interestingly, when we look at the doses of omega-3 fatty acid supplements that were given in these trials, the increase seems greater in those trials that used a higher dose of omega-3 fatty acids above one gram per day.
Now, a subset of particular interest in REDUCE-IT, the results of which have been presented recently is the REDUCE-IT ACS subgroup of patients. In that subgroup of patients, the benefits of icosapent ethyl appeared particularly marked with a 37% relative risk reduction in the primary composite outcome for patients with a recent ACS of less than 12 months and an absolute risk reduction of 9% corresponding to a number needed to treat of only 11. Of note, there was also an increase in the risk of atrial fibrillation or flutter in that population, highly statistically significant. But of note, there was no increase in bleeding in that population, including when one focuses on the subgroup of patients who are on dual antiplatelet therapy at baseline, for whom bleeding rates were 7.7% in the icosapent ethyl arm compared to 9.4% in the placebo arm, so no hint of a signal for increased bleeding in that population.
Now, finally, last year we had the reporting of the RESPECT-EPA trial. RESPECT-EPA is another Japanese trial that compares eicosapentaenoic acid to control with no placebo. It's single-blind, and it selected patients on the basis of having an elevated EPA to arachidonic acid ratio, actually a low ratio of less than 0.4, and compared highly purified EPA icosapent ethyl 1.8 grams per day to standard statin therapy as the control. What the trial reported is a substantial increase in EPA with treatment, a decrease in arachidonic acid, a modest decrease in CRP, essentially no change in LDL cholesterol or HDL cholesterol, and a very modest decrease in triglycerides. In terms of clinical outcomes, what is interesting is that there was a borderline reduction in the primary outcome by approximately 22%, bordering on statistical significance, and a reduction in the secondary outcome of sudden cardiac deaths, MI, and stable angina or coronary vascularization. A signal going in the exact same direction as JELIS and REDUCE-IT despite no placebo being used in JELIS or in RESPECT-EPA. All-cause mortality was numerically lower and cardiovascular mortality was numerically lower, but neither of these reached statistical significance. Safety appeared good. Of note, gastrointestinal disorders appeared increased in that trial. Bleeding was not increased, and new-onset atrial fibrillation was increased, 3.1% versus 1.6%.
To put everything in perspective, I'd like to contrast trial results of the EPA versus the EPA+DHA trials. On this slide, you have a forest plot that represents the major randomized cardiovascular outcome trials of omega-3 fatty acid looking at cardiovascular outcomes. All the trials in blue are the trials that compared combinations of EPA and DHA to control or to placebo. As you can see, the confidence interval for these five trials actually all cross the line of identity and none of these trials showed benefit. Contrasting with this at the bottom are the three trials in red which are the three trials that looked at high-dose EPA. JELIS and RESPECT-EPA used 1.8-gram EPA, REDUCE-IT used 4 grams of EPA per day, and all three trials show clear signal of benefit. What's interesting is that in REDUCE-IT, we looked at whether there was a correlation between the risk reduction for the primary and the key secondary outcome and the on-treatment serum EPA. As you can see, as serum EPA increased, the risk of the primary endpoint decreases. This is true on the left for patients with established cardiovascular disease and on the right for patients with diabetes and risk factors. The secondary prevention cohort and the primary prevention cohort yield identical and consistent results. The higher the EPA, the greater the cardiovascular outcomes benefit.
Of note, it's interesting to look at similar analyses that have been done in another trial called the OMEMI trial which was a post-MI trial done in Scandinavia that didn't find a benefit of overall EPA supplementation on MACE but did see a correlation between the change in EPA and MACE. The greater the change in EPA, the greater the decrease, the lower the MACE rate. Of note, they also saw that there was an increase and a correlation between the change in EPA and the risk of atrial fibrillation. Benefit increases as EPA increases and the risk of atrial fibrillation also increases as EPA increases. In that respect, I'd like to contrast the baseline of EPA levels of four major trials looking at omega-3 supplementation for cardiovascular risk reduction. JELIS started off with a high baseline EPA level in red, approximately 100 as is frequent in Japan. You're considering the diet in Japan that has a lot of oily fish. On treatment, the level of plasma EPA increased to more than 170. In REDUCE-IT, the baseline level was much lower, 26 serum EPA, but increased to close to 150 on treatment with supplementation. In RESPECT-EPA, the baseline level was intermediate between the two previous trials because the Japanese diet has changed over time. There's less fish consumption but still on treatment reached 1.45. In contrast with these three trials that showed signals of benefit, the strength trial did not see any hint of cardiovascular benefit. Of note, the baseline level was similar to what was seen in REDUCE-IT at 21, but the plasma EPA level that was reached on treatment was only 90, so substantially lower than in those three trials that were positive. That may be an explanation for the discrepant results in terms of benefit on cardiovascular outcomes.
To summarize what we've seen, clinical trials using low doses of omega-3 fatty acids for cardiovascular prevention have yielded inconsistent results. The modern clinical trials that used EPA and DHA in combination have never shown cardiovascular benefits. Three trials using high doses of EPA have shown robust cardiovascular benefit, JELIS and RESPECT-EPA in comparison to usual care with no placebo control, and REDUCE-IT in comparison to mineral oil. The safety profile appears good, but clearly, the risk of atrial fibrillation and flutter is increased and the bleeding risk may be increased. The benefit appears strongly correlated to achieved EPA levels but not to triglycerides, LDL cholesterol levels, or high sensitivity CRP.
Thank you very much.
