Amarin Highlights REDUCE-IT Analysis Linking Lp(a) to Cardiovascular Risk at AHA 2026

Key Takeaways

• Amarin presented two analyses at the American Heart Association (AHA) Scientific Sessions 2026 in Chicago: a clinical analysis of lipoprotein(a) [Lp(a)] in the REDUCE-IT trial and a lab study of how eicosapentaenoic acid (EPA) affects oxidized Lp(a) in endothelial cells.
• Among 6,988 REDUCE-IT participants with statin-controlled LDL cholesterol and elevated triglycerides, Lp(a) was independently associated with major adverse cardiovascular events (MACE), while non-Lp(a) apolipoprotein B (apoB) was not prognostic.
• In the cell experiments, EPA pretreatment reduced IL-6 levels by 31% relative to oxidized Lp(a). The cell findings don’t show a clinical effect, and both studies come from the company that sells icosapent ethyl.

Amarin Corporation, which markets the prescription drug Vascepa/Vazkepa (icosapent ethyl), highlighted new data presented at AHA Scientific Sessions 2026 this past weekend. The company framed the work around Lp(a), a lipoprotein that is increasingly studied as a contributor to cardiovascular risk that persists despite treatment. It also pointed to the recent announcement that the Novartis Lp(a)HORIZON outcomes trial of pelacarsen did not meet its primary endpoint despite substantial Lp(a) lowering.

The first analysis, a post hoc look at REDUCE-IT, evaluated baseline Lp(a) and apoB measurements in 6,988 participants. Lp(a) and LDL particles each carry a single apoB molecule, but the researchers cite emerging evidence that Lp(a) may be more atherogenic per particle. They tested whether, after statin control of LDL cholesterol, Lp(a) was more strongly tied to cardiovascular risk than other apoB-containing lipoproteins. Among participants with statin-controlled LDL cholesterol and elevated triglycerides, Lp(a) was independently associated with MACE, while non-Lp(a) apoB was not prognostic. The results were similar in participants above and below an Lp(a) level of 150 nmol/L, and risk was most evident in the highest Lp(a) quartile. The investigators concluded that Lp(a) is an important contributor to residual risk in this population independent of icosapent ethyl treatment. The abstract appeared in Circulation (2026;154, Suppl 1, Abstract 8761).

The second set of experiments studied oxidized Lp(a) in endothelial cells. Exposure to oxidized Lp(a) changed the expression of more than 600 proteins, including some tied to reactive oxygen species signaling, lipid peroxidation, ferroptosis, inflammation, and matrix degradation. It also raised release of the inflammatory cytokine IL-6 by 27% relative to control, and EPA treatment lowered IL-6 by 31% relative to oxidized Lp(a). EPA pretreatment also increased anti-inflammatory and cytoprotective proteins and reversed several of the oxidized-Lp(a) protein changes. The investigators concluded that EPA may help mitigate processes linked to oxidized Lp(a)-induced endothelial dysfunction.

The company says these findings build on an earlier REDUCE-IT analysis in JACC in which baseline Lp(a) was associated with higher cardiovascular risk despite statin-controlled LDL cholesterol, and icosapent ethyl reduced MACE across Lp(a) levels. CEO Aaron Berg said the drug is available now and has shown event reduction on top of statin therapy at both high and low Lp(a) levels. In the U.S., Vascepa is indicated as an adjunct to maximally tolerated statin therapy in adults with elevated triglycerides and established cardiovascular disease or diabetes plus other risk factors, and for severe hypertriglyceridemia. The label cites increased risks of atrial fibrillation or flutter requiring hospitalization (3% vs 2% in a placebo-controlled trial) and bleeding (12% vs 10%).

The clinical analysis is post hoc and limited to REDUCE-IT participants, who all had elevated triglycerides and were on statins, so it can’t show whether the pattern applies more broadly. It shows an association with risk, not that targeting Lp(a) prevents events. The cell experiments were done in the lab, and the company doesn’t claim they show benefit in patients. The announcement doesn’t give hazard ratios or confidence intervals for the Lp(a) finding.