New STRENGTH analysis puts spotlight on CV impact of omega-3 fatty acids
Click Here to Manage Email Alerts
Higher blood levels of eicosapentaenoic acid 1 year after daily omega-3 carboxylic acid supplementation were not associated with lower CV risk, according to a secondary analysis of the STRENGTH trial.
Presenting data from a post hoc analysis at the American College of Cardiology Scientific Session, Cardiology Today Editorial Board Member Steven E. Nissen, MD, MACC, chief academic officer of the Sydell and Arnold Miller Family Heart, Vascular & Thoracic Institute and Lewis and Patricia Dickey Chair in Cardiovascular Medicine at Cleveland Clinic, and study chair for the STRENGTH trial, said the evidence strongly demonstrated no benefit with daily omega-3 carboxylic acid (Epanova, AstraZeneca), similar to the neutral findings observed in large, lower-dose omega-3 fatty acid trials like VITAL, ASCEND and ORIGIN.
“The top tertile [of participants] had a 443% increase in EPA, and there is no benefit,” Nissen told Healio. “Similarly, the top tertile of [docosahexaenoic acid] showed no evidence of harm; it was neutral, too.”
Deeper analyses
As Healio previously reported, data from STRENGTH, presented at the virtual American Heart Association Scientific Sessions in November, showed that administration of high-dose omega-3 carboxylic acid compared with corn oil placebo did not reduce incidence of major adverse CV events among statin-treated adults with elevated triglycerides at high CV risk.
Those findings were in stark contrast to the REDUCE-IT trial, presented in 2018, which demonstrated that icosapent ethyl (Vascepa, Amarin Pharmaceuticals), a pharmaceutical-grade omega-3 fatty acid containing EPA but not DHA, was superior to placebo for reducing risk for ischemic events in a similar patient group. The STRENGTH study was stopped early by AstraZeneca in January 2020 at the recommendation of the data safety monitoring board due to lack of observed benefit.
“We have two trials [STRENGTH and REDUCE-IT], that give completely disparate results,” Nissen said in an interview. “REDUCE-IT shows this almost unbelievable benefit and STRENGTH is completely neutral. What is going on here? Because REDUCE-IT used purified EPA, one of the thoughts was maybe we did not get enough EPA into people in the STRENGTH trial. The other possibility was, since we have a mixture of EPA and DHA, maybe the DHA was somehow counterbalancing the effects of EPA.”
For the secondary analysis of STRENGTH, Nissen and colleagues analyzed omega-3 fatty acid levels from a subset of 10,382 adults with high CV risk, elevated triglyceride levels and low levels of HDL from 675 centers in 22 countries. Participants were randomly assigned 4 g daily omega-3 carboxylic acid (n = 6,539) or corn oil (n = 6,539) on top of the usual background therapies, including statins. The primary prespecified endpoint was a composite of CV death, nonfatal MI, nonfatal stroke, coronary revascularization or unstable angina requiring hospitalization. Blood levels of omega-3 fatty acid at baseline and 12 months were available for 5,175 participants who received omega-3 carboxylic acid and 5,207 participants who received corn oil placebo.
Among participants who received omega-3 carboxylic acid, median plasma levels at 12 months were 89 g/mL for EPA and 91 g/mL for DHA, with top tertile levels of 151 g/mL and 118 g/mL, respectively.
Compared with corn oil, the adjusted HRs for the primary endpoint were 0.98 (95% CI, 0.83-1.16; P = .81) for the highest tertile of achieved plasma levels of EPA and 1.02 (95% CI, 0.86-1.2; P = .85) for the highest tertile of achieved plasma levels of DHA.
Sensitivity analyses based on changes in plasma and red blood cell levels of EPA and DHA and primary and secondary prevention subgroups showed similar results, Nissen said.
The results were simultaneously published in JAMA Cardiology.
Limitations, study debate
Nissen acknowledged several limitations of the secondary analyses. Analysis by tertiles reduces statistical power, since each group represents only one-third of the population; however, he noted confidence intervals were reasonably narrow. Additionally, there was a moderate correlation between EPA and DHA levels, which precludes completely independent assessment of fatty acids and outcome, he said.
Despite limitations, Nissen said the findings do not support the use of omega-3 carboxylic acid for patients at high CV risk. Omega-3 fatty acid supplementation was also associated with elevated risk for atrial fibrillation, he noted.
“With this unusual benefit seen in REDUCE-IT, you have to look at it carefully,” Nissen told Healio. “It is our belief that it is the comparator that is different. REDUCE-IT used a mineral oil comparator ... we think this is more evidence that REDUCE-IT is a false positive study driven by a comparator that had toxicity.”
In a discussion after the data presentation, Cardiology Today Intervention Section Editor Deepak L. Bhatt, MD, MPH, executive director of interventional cardiovascular programs at Brigham and Women’s Hospital, professor of medicine at Harvard Medical School, and the lead investigator of REDUCE-IT, said one could conclude that the absence of an association in a negative trial “does not reveal much,” other than that the drug studied did not work.
Bhatt also noted that not all omega-3 fatty acids are the same; the pure EPA used in REDUCE-IT is more effective, whereas DHA may counter some of the benefits of EPA, he said, noting that how EPA is delivered and how tissue levels are achieved also matters.
“A really important observation in STRENGTH is that, despite the reduction in triglycerides, there was no impact on major adverse CV events, a very interesting disconnect,” Bhatt said. “Why is there no correlation between higher levels of EPA or DHA and triglyceride reduction?”
Nissen said REDUCE-IT and STRENGTH both reduced triglycerides by almost identical amounts — almost 19%.
“We do not see much of a relationship between achieved EPA levels [and triglyceride levels],” Nissen said. “It may be the threshold for triglyceride reduction is different from the EPA effect. It could be that lower levels of achieved EPA are sufficient to lower triglycerides. We don’t have an answer to that question.”
Bhatt also noted that high-intensity statin use was lower among STRENGTH participants in the higher tertiles of achieved EPA level, and questioned if that could affect the findings.
“Statin use does not explain the results we saw here,” Nissen said. “Our purpose was to find out if this drug, EPA, showed benefits, and whether high levels of DHA showed evidence of harm. We saw neither. We are left with the concern that came up from the very beginning, which is, is the difference between these two trials a difference [due to] the placebo used.”
Bhatt said another possible explanation is “it could just be a different drug, different results.”
Reference:
Perspective
Back to Top
Pradeep Natarajan, MD
The STRENGTH trial previously showed that among statin-treated patients at high CV risk and with elevated triglycerides, those randomly assigned to a carboxylic acid formulation of a combination of EPA and DHA did not have improved CV prognosis vs. corn oil placebo. Since REDUCE-IT, using a similar patient population, showed that icosapent ethyl vs. mineral oil placebo led to markedly improved CV prognosis, this secondary analysis in STRENGTH now evaluates whether on-treatment EPA levels in plasma predicted benefit. However, those who happened to have the highest EPA levels in STRENGTH did not have greater CV risk reduction.
So far, icosapent ethyl remains the sole omega-3 fatty acid formulation demonstrated to achieve improved CV prognosis as demonstrated with REDUCE-IT and JELIS. Icosapent ethyl is a purely EPA formulation. To test the hypothesis that EPA supplementation may improve prognosis, the STRENGTH investigators tested whether those who had higher on-treatment EPA levels had improved prognosis, but there was no difference. To test the hypothesis that DHA supplementation may adversely affect prognosis, the STRENGTH investigators tested whether those who had higher on-treatment DHA levels had worsened prognosis, but again, there was no difference.
However, analyses in the REDUCE-IT trial do show that on-treatment EPA concentrations did predict improved clinical benefit from icosapent ethyl.
On-treatment concentrations of EPA were much higher in REDUCE-IT (median 135 ug/mL) compared with STRENGTH (median 89 ug/mL). Since the STRENGTH intervention was a combination of both EPA and DHA, it is challenging to separately tease out EPA and DHA effects statistically. Much has been said about the effect from the differences in placebo used, but I do not suspect placebo differences could explain the observed CV effect estimates in REDUCE-IT and JELIS.
In the end, we simply have two different omega-3 fatty acid formulations. STRENGTH alone is unable to refute the findings in REDUCE-IT since the interventions were different. REDUCE-IT additionally has the benefit of another randomized controlled trial, JELIS, with similar benefits. Indeed, more work is needed to better understand why icosapent ethyl is uniquely associated with improved CV prognosis compared with any other omega-3 fatty acid formulation. However, regardless of mechanism, there are now two randomized trials supportive of icosapent ethyl.
Collapse
Nissen SE, et al. Joint American College of Cardiology/Journal of the American Medical Association Late-Breaking Clinical Trials. Presented at: American College of Cardiology Scientific Session; May 15-17, 2021 (virtual meeting).
Read more about
Click Here to Manage Email Alerts