1. A milestone worth studying
On 13 December 2019, the FDA approved icosapent ethyl as an adjunct to maximally tolerated statin therapy for patients with elevated triglycerides (≥150 mg/dL, fasting not required) who have established cardiovascular disease or diabetes plus at least two additional risk factors. Wang, Verma, Mason and Bhatt, writing in 2020, frame it plainly: as the first drug approved in its class, it gave patients already on maximally tolerated statins a new option for reducing residual risk.
The word to notice is first. Reaching it required a category to fail publicly, repeatedly, for two decades.
2. What came before — and why it failed
Three drug classes were tried against elevated triglycerides: fibrates, niacin and omega-3 fatty acids.
Fibrates looked promising before high-intensity statins became standard. In the contemporary setting, adding a fibrate to a statin produced no comparable benefit. Niacin fared worse: AIM-HIGH and HPS2-THRIVE both found no clinical benefit when added to statin therapy.
Omega-3 trials were mostly negative too. VITAL randomized 25,871 participants to 1 g/day omega-3 (380 mg DHA plus 460 mg EPA) or placebo and found no significant reduction in major cardiovascular events (HR 0.92, 95% CI 0.80–1.06). ASCEND gave 15,480 people with diabetes 1 g/day of a DHA/EPA mixture against an olive-oil placebo over 7.4 years; first serious vascular events were 8.9% versus 9.2% (RR 0.97, 95% CI 0.87–1.08). In acute coronary syndrome patients already on guideline therapy, Alpha Omega and OMEGA also failed to reduce sudden cardiac death or major cardiovascular events.
Read as a block, those results say something specific: 1 g/day of a mixed omega-3 preparation does not work, at least not in populations already receiving good background therapy.
3. Two early positives that could not prove the point
GISSI-P, open-label, randomized 11,324 recent myocardial infarction patients two-by-two to vitamin E or n-3 polyunsaturated fatty acids. At 42 months vitamin E showed no effect, while the n-3 arm reduced death, non-fatal MI and non-fatal stroke — two-way RR 0.90 (95% CI 0.82–0.99). One detail complicates comparison with modern practice: only 4.7% were on lipid-lowering therapy at baseline, rising to 45.5% by 42 months.
JELIS took a different route: 18,645 participants with total cholesterol ≥6.5 mmol/L (251 mg/dL), randomized to 1.8 g/day of highly purified EPA ethyl ester plus statin versus statin alone. Over a mean 4.6 years, major coronary events occurred in 2.8% versus 3.5% (RRR 19%, p=0.011), consistent in the secondary-prevention subgroup (8.7% vs 10.7%) and directionally similar but non-significant in primary prevention (1.4% vs 1.7%, p=0.13). JELIS pointed at the answer — a purified single molecule at a higher dose — but an open-label design, a mixed population and no dedicated outcomes power meant it could not settle it.
4. The four decisions that made REDUCE-IT work
The REDUCE-IT designers drew three lessons from the earlier trials, then added a fourth element of their own.
Dose. Many prior trials used 1–2 g/day, which the authors suggest may be insufficient for Western populations with typically low background omega-3 intake. REDUCE-IT used 4 g/day, taken as 2 g twice daily with food.
Composition and purity. EPA and DHA have different effects on cell membranes. REDUCE-IT used a single prescription-grade molecule: highly purified EPA ethyl ester — not a bulk supplement.
Population enrichment — the decisive one. Participants needed LDL-C already controlled at 41–100 mg/dL and fasting triglycerides of 150–499 mg/dL, with entry at age ≥45 with established cardiovascular disease, or ≥50 with diabetes plus at least one additional risk factor, on stable statin therapy.
A protocol detail shows how much the trial depended on that enrichment: the triglyceride floor was initially set 10% lower (135 mg/dL) to accommodate the analyte’s inherent variability, then raised to 200 mg/dL mid-trial to ensure enough participants sat in range. Success required patients who were, by construction, already well treated and still at risk.
5. The result
Among 8,179 randomized participants, the primary composite endpoint — cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, coronary revascularization or unstable angina — occurred in 17.2% of the icosapent ethyl group versus 22.0% of the placebo group. Relative risk reduction: 25% (HR 0.75, 95% CI 0.68–0.83, p=0.00000001).
Subsequent analyses found the effect extended beyond first events, reducing the total burden of ischemic events by 30% (RR 0.70, 95% CI 0.62–0.78, p<0.0001), and significantly reducing coronary revascularization rates.
Two generalizability checks matter for market sizing. In the CLARIFY registry of stable coronary artery disease patients (excluding the US, where hypertriglyceridaemia is more prevalent), 15.5% met REDUCE-IT entry criteria; in a Canadian cohort of 196,717 ASCVD patients, 25% met the treatment indication on lipid profile alone. And among the 3,146 US-randomized participants the effect was larger: 18.2% versus 24.7% (HR 0.69, 95% CI 0.59–0.80, p=0.000001).
6. The finding that matters most for positioning
Icosapent ethyl’s benefit exceeded what its triglyceride reduction would predict — the authors are explicit about this. It held across the full baseline triglyceride range, including in the roughly 10% of enrolled patients whose baseline triglycerides were normal.
That is the commercially significant finding, because it separates two possible product claims. A treatment that works by lowering triglycerides is a triglyceride drug. A treatment that works through something else — fibrous cap thickness, plaque volume, macrophage accumulation, pro-resolving signalling, endothelial function — is a cardiovascular drug that happens to lower triglycerides.
The supporting evidence includes plaque imaging: EVAPORATE’s 9-month interim showed reduced total plaque volume, suggesting rapid onset, and the 18-month final results showed improvement across multiple measures of plaque volume and composition.
7. A 25% result still had to clear a regulatory gate
Even with p=0.00000001, approval was not automatic. The drug had previously been approved only as a diet adjunct to lower triglycerides in severe hypertriglyceridaemia (TG ≥500 mg/dL), where the objective is reducing pancreatitis risk.
On 14 November 2019, the FDA’s Endocrinologic and Metabolic Drugs Advisory Committee voted 16–0 to recommend approval for a cardiovascular risk-reduction indication. The FDA approved on 13 December 2019; Health Canada followed on 30 December 2019.
Note what the final label says. Triglycerides ≥150 mg/dL, and — importantly — fasting status not specified. The approved indication is broader than the trial’s entry criteria, which is the kind of outcome that only follows a unanimous advisory vote.
8. Reimbursement is the second gate
The review also tracks cost, which is where most approval stories quietly end. The Institute for Clinical and Economic Review put the annual wholesale acquisition cost at 3,699∗∗(netroughly∗∗3,699∗∗(netroughly∗∗1,625), against a value-based benchmark of 6,300–9,200peryear.Aseparatecost−effectivenessanalysisfoundthatat∗∗6,300–9,200peryear.Aseparatecost−effectivenessanalysisfoundthatat∗∗4.16 per day**, icosapent ethyl was the dominant strategy in most scenario analyses — a result the authors call rare for a brand drug, previously associated mainly with aspirin or generic statins in secondary prevention. An approval that clears the evidence bar but fails the cost bar does not reach patients. Both gates were cleared here.
9. What this means for the supply chain
“Omega-3” is not an approvable category. Six trials failed using mixtures, low doses, or both. One succeeded using a single purified molecule at 4 g/day. The differentiation available to a supplier is therefore not the presence of omega-3; it is purity, monomer separation and dose-per-gram.
The comparator problem is a design problem. ASCEND used olive oil, which may itself have favourable lipoprotein effects. A placebo that is not inert can bias a trial toward a null result — a point that recurs across this literature.
The evidence is more mature than the positioning. The ESC, ADA, National Lipid Association, Brazilian Society of Cardiology and AHA now recommend icosapent ethyl for defined residual-risk populations, with the AHA calling it the first non-LDL-targeting lipid-lowering therapy proven to have cardiovascular benefit and a first-line consideration for type 2 diabetes with coronary heart disease. Suppliers who describe omega-3 as a wellness ingredient are describing a different product than the one that earned those recommendations.
One caveat to carry forward. The review’s open access was funded by Amarin Pharma, the manufacturer of the approved product; the lead author chaired REDUCE-IT; and the disclosures list extensive industry relationships. That does not invalidate the trial data — the FDA vote was unanimous — but it is why the primary literature, not any single review, remains the place to check a claim.
The takeaway
The icosapent ethyl pathway is often summarised as “omega-3 finally worked.” The more useful reading is narrower and more demanding: a single purified molecule, at a pharmacological dose, in a patient population defined by controlled LDL-C and a specific residual-risk window, with an outcomes endpoint and enough enrolled patients to prove it.
That is a specification long enough to look like a manufacturing brief — which, for anyone in the omega-3 supply chain, is exactly what it is.
Industry commentary on a published review, intended for a professional non-clinical audience. Not medical advice. The review is Wang X, Verma S, Mason RP, Bhatt DL, “The Road to Approval: A Perspective on the Role of Icosapent Ethyl in Cardiovascular Risk Reduction,” accepted 10 September 2020. Note the paper’s open-access funding and author industry disclosures, detailed in the source. Clinical and regulatory conclusions should be verified against the primary literature, the product’s approved labelling, and qualified professionals.
