Fish Oil Tanks in STRENGTH, Making Waves for REDUCE-IT

Steven Nissen argues REDUCE-IT gave a “false-positive” result. But Deepak Bhatt, that study's PI, calls the assertion “ridiculous.”

Fish Oil Tanks in STRENGTH, Making Waves for REDUCE-IT


(UPDATED) Full results of the STRENGTH trial confirm what has been known since the study was stopped for futility back in January: a carboxylic acid formulation of omega-3 fatty acids (Epanova; AstraZeneca) does not reduce cardiovascular events in patients with elevated triglycerides and low HDL cholesterol.

But Steven Nissen, MD (Cleveland Clinic, OH), chair of the STRENGTH’s executive committee, argues there are far-reaching implications of the negative trial results, which were reported during the virtual American Heart Association 2020 Scientific Sessions and simultaneously published in JAMA.

“We believe that STRENGTH demonstrates that REDUCE-IT is a false-positive result. There is no effect of fish oil on cardiovascular outcomes, just as all of the other trials have shown. REDUCE-IT we think is positive because it used a negative control rather than a neutral control,” Nissen told TCTMD.

REDUCE-IT showed that another omega-3 fatty acid formulation—icosapent ethyl (Vascepa; Amarin)—cut MACE by a relative 25% compared with a placebo pill containing mineral oil.

We believe that STRENGTH demonstrates that REDUCE-IT is a false-positive result. Steven Nissen

And that’s a problem, Nissen said, because mineral oil had pro-inflammatory effects in REDUCE-IT, which exaggerated the treatment benefit observed in the trial. He pointed to a 10.2% increase in LDL cholesterol, a 7.8% increase in apolipoprotein B, and a “really huge” 32% increase in high-sensitivity C-reactive protein (hsCRP) in the placebo arm. That didn’t happen with the placebo in the STRENGTH trial, which consisted of corn oil.

“I don’t think we can believe that fish oil is beneficial unless REDUCE-IT is repeated with a neutral control like corn oil,” Nissen said.

For STRENGTH principal investigator A. Michael Lincoff, MD (Cleveland Clinic), who presented the results today, the findings, “in the context of the increased risk of atrial fibrillation in this and other trials of omega-3 fatty acids, [raise] a question regarding whether or not there is a net benefit or harm with any omega-3 fatty acid preparation.”

But the principal investigator of REDUCE-IT, Deepak Bhatt, MD (Brigham and Women’s Hospital, Boston), pushed back on the critiques of his trial. To TCTMD, he highlighted the differences between the fish oil formulations used in the two trials. Icosapent ethyl, used in REDUCE-IT, is a purified form of eicosapentaenoic acid (EPA), whereas Epanova, used in STRENGTH, is a mixture of EPA and docosahexaenoic acid (DHA). Research has shown that EPA stabilizes cell membranes and DHA destabilizes them, meaning that the DHA in Epanova could have counteracted the beneficial effects of the EPA, explaining the discrepant trial results, Bhatt said.

“All omega-3 fatty acids aren’t created equal. They have different physiochemical properties, different biochemical properties, and therefore could have different effects on human health in different situations.” Even if skeptics aren’t convinced by that argument, he added, they should remember that there was an earlier trial of icosapent ethyl, albeit at a lower dose, that also showed a significant reduction in cardiovascular events. In the JELIS trial, treatment reduced major coronary events by a relative 19%.

“For cardiovascular prevention and treatment, I think the REDUCE-IT executive committee and leadership team remain confident that the results are quite robust and strong,” Bhatt said.

The Futility of STRENGTH

STRENGTH, conducted at 686 sites in 22 countries, enrolled 13,078 statin-treated patients with or at high risk for CVD who had elevated triglycerides (180 to 500 mg/dL) and low HDL cholesterol (< 42 mg/dL for men and < 47 mg/dL for women). The participants were randomized to corn oil placebo or Epanova, a carboxylic acid formulation of omega-3 fatty acids combining EPA and DHA given at a dose of 4 g/day. That formulation is very well absorbed, providing high levels of EPA and modest levels of DHA, Nissen said.

Active treatment was associated with a 268.8% increase in median plasma EPA levels—from 21.0 μg/mL at baseline to 89.6 μg/mL at 1 year. That was accompanied by a 19% decline in triglycerides (from 239 to 191 mg/dL) and a 20% reduction in hsCRP (from 2.13 to 1.69 mg/L).

Through a median follow-up of 42.0 months, however, Epanova did not reduce the risk of CV death, MI, coronary revascularization, or hospitalization for unstable angina, the primary composite endpoint (12.0% vs 12.2%; HR 0.99; 95% CI 0.90-1.09). That’s the same primary endpoint as in REDUCE-IT.

Consistent with REDUCE-IT, fish oil treatment increased risk of new-onset A-fib (2.2% vs 1.3%; HR 1.69; 95% CI 1.29-2.21). There was also a higher rate of GI adverse events (24.7% vs 14.7%).

A Tale of Two Trials

In a discussion about how to reconcile these results with those from REDUCE-IT, Lincoff noted that the increase in EPA levels was about 45% greater in REDUCE-IT than in STRENGTH, adding, however, that “this magnitude of difference seems insufficient to explain the completely different results of the two trials.” Also, treatment in both trials resulted in a reduction in triglycerides of about 19%, suggesting similar biochemical effects, Lincoff said.

Even in participants who attained the highest levels of EPA in STRENGTH, Nissen pointed out, there was no sign of benefit from active treatment. And that means that REDUCE-IT falsely demonstrated a CV benefit where none might exist, he argued. “I think it’s very unfortunate that REDUCE-IT, with all the fanfare that it got, really is not a positive trial. And STRENGTH shows us that.”

But other commentators were not so sure. Karol Watson, MD, PhD (University of California, Los Angeles), said the negative STRENGTH results were surprising in light of REDUCE-IT. But there were differences between the trials, she noted, pointing to the inclusion of DHA in the fish oil formulation used in STRENGTH for one. She acknowledged the debate around the potential harms of the mineral oil placebo used in REDUCE-IT, but didn’t think that would be enough to fully explain the different trial results. “We’ll all have to just guess about what would have happened if perhaps they used corn oil or olive oil or some other comparator. We just don’t know,” Watson said. “But the REDUCE-IT results were so positive that it’s hard to . . . imagine that all of that benefit goes away.”

I think at a certain point it’s just ridiculous to not believe something. Deepak Bhatt

Salim Virani, MD, PhD (Baylor College of Medicine, Houston, TX), also wouldn’t go so far as to say that the neutral result of STRENGTH indicated that REDUCE-IT provided a false-positive result. He listed three main issues to consider when it comes to comparing the trials: the formulations, the placebos, and the trial populations.

Maybe EPA-only formulations are superior to ones containing EPA and DHA, he said. And maybe the effects of mineral oil explained the benefit in REDUCE-IT, although “probably not when you have such a large absolute risk reduction,” Virani said. “Could that have contributed? Possibly so.”

Virani, chair of the American College of Cardiology’s prevention section and leadership council, then considered the trial populations, pointing out that a higher proportion of REDUCE-IT participants had established CAD (71% vs 56%). A higher risk patient population in REDUCE-IT, he said, is reflected in rates of the primary endpoint in the placebo arms of the two trials—22.0% in REDUCE-IT and 12.2% in STRENGTH. “Maybe that very high-risk population is the one where these triglyceride risk-based therapies work,” Virani said.

So before jumping to conclusions about false-positive trial results, the community needs to be thoughtful and take the time to study the nuanced differences between the trials, he said, noting that the results of REDUCE-IT were robust and consistent across multiple endpoints and followed up on the positive findings from JELIS.

Nissen acknowledged that the US Food and Drug Administration specifically examined the impact of the mineral oil placebo on the results of REDUCE-IT but “they were not able to demonstrate without doubt that REDUCE-IT was false-positive.” The agency approved icosapent ethyl in December 2019.

“I think it was a mistake and I do not think that fish oil is beneficial,” Nissen said. “We think we have a public health obligation to get this out there and to tell people, because as you know this EPA drug is being advertised on television, they have 800 reps running around pushing the drug, and we just don’t think that it’s the right answer.”

Lincoff said that he is not routinely prescribing omega-3 fatty acids for his patients, except in those who have very high levels of triglycerides (> 500 mg/dL), saying its disputed "whether or not the risk and benefit are of a favorable ratio.” In response to a question at a press briefing, he said patients should be warned about the risk of A-fib when taking omega-3 fatty acids.

For Bhatt, the answer is clear: “Icosapent ethyl in two different independent trials showed significant reductions in ischemic events. So I think at a certain point it’s just ridiculous to not believe something. And I think the story for EPA at this point is extremely strong and the data are growing.”

Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …

Read Full Bio
  • The trial was funded by AstraZeneca and coordinated by the Cleveland Clinic Coordinating Center for Clinical Research (C5Research).
  • Nissen reports grants from AstraZeneca during the conduct of the study; and grants from Novartis, Abbvie, Silence Therapeutics, Medtronic, MyoKardia, Esperion, Eli Lilly, Amgen, Novo Nordisk, Pfizer, Cerenis, and The Medicines Company outside the submitted work.
  • Bhatt reports receiving research funding or unfunded research support from Abbott, Amarin, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Chiesi, Eisai, Ethicon, Forest Laboratories, Idorsia, Ironwood, Ischemix, Lilly, Medtronic, PhaseBio, Pfizer, Regeneron, Roche, Sanofi, Synaptic, The Medicines Company, FlowCo, Merck, Novo Nordisk, PLx Pharma, and Takeda; being a site co-investigator for Biotronik, Boston Scientific, St. Jude Medical, and Svelte; being a trustee for ACC; serving as an advisory board member, director, or chair for Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, Regado Biosciences; the Boston VA Research Institute, the Society of Cardiovascular Patient Care, TobeSoft; the American Heart Association Quality Oversight Committee; serving on a range of data safety monitoring committees; receiving honoraria for editorial or committee activities for a range of publications and organizations; and receiving royalties from Elsevier.
  • Virani reports research support from the Department of Veterans Affairs and the World Health Federation, as well as an honorarium from the American College of Cardiology.
  • Watson reports no relevant conflicts of interest.
  • Nissen reports grants from AstraZeneca during the conduct of the study; and grants from Novartis, Abbvie, Silence Therapeutics, Medtronic, MyoKardia, Esperion, Eli Lilly, Amgen, Novo Nordisk, Pfizer, Cerenis, and The Medicines Company outside the submitted work.