According to a phase 3 clinical trial, overall survival among patients undergoing surgery for colorectal cancer liver metastases was prolonged by a high-dose omega-3 supplement. However, the treatment showed no significant effect on disease progression, leaving researchers searching for an explanation for the survival benefit.
Phase 3 EMT2 Trial Results in Madrid
Patients who received high-dose supplements of eicosapentaenoic acid (EPA) for at least two years post-surgery experienced a roughly 30% improvement in overall survival compared to those given a placebo. The findings from the EMT2 trial were presented at the European Society for Medical Oncology Annual Meeting 2026 in Madrid.
The multicenter, double-blind trial enrolled 418 patients with a median age of 64 between 2018 and 2023 across 13 UK centers. Participants were randomized in a 1:1 ratio to receive 4 grams daily of purified EPA ethyl ester (icosapent ethyl; Vascepa) or matching mineral oil placebo capsules starting just before surgery.
Estimated five-year overall survival in the intent-to-treat population reached 62% in the EPA arm compared to 45.1% in the placebo group, yielding a hazard ratio of 0.68. At the time of analysis, median overall survival stood at 59 months for the placebo group, whereas the EPA group had not yet reached it.
Discordance Between Survival and Progression
Despite the survival advantage, the trial failed to meet its primary endpoint of progression-free survival. Median progression-free survival stood at 13.5 months in the EPA arm and 12.9 months in the placebo arm, with estimated five-year progression-free survival rates of 23.6% and 24.5% respectively.
Mark A. Hull, MD, of the University of Leeds in the UK, proposed several biological mechanisms to account for this disconnect. These possibilities include anti-cachexia effects, enhanced antitumor immunity driven by EPA’s cyclooxygenase-inhibiting activity, and alterations in the gut microbiome. Hull noted that similar survival benefits without progression-free survival impacts have appeared in certain cancer immunotherapy trials.
Treatment was generally well-tolerated with no excess adverse events, bleeding, or atrial fibrillation reported. Discontinuation rates were similar across both arms, with roughly 40% completing treatment and 22% withdrawing entirely, frequently due to perceived treatment futility after disease progression rather than drug side effects.
Clinical Caution and Unresolved Questions
Specialists not involved in the research urged restraint despite the intriguing data. Jeffrey Meyerhardt, MD, MPH, of Dana-Farber Cancer Institute in Boston, stated that the findings should not alter immediate clinical practice because the discordance between progression and survival remains unexplained.
Meyerhardt pointed to a small, randomized trial of perioperative omega-3 supplements that showed numerically worse long-term survival in the treatment arm. He also emphasized the need to evaluate whether cardiovascular risk reduction contributed to the survival gap, noting that omega-3 supplementation, particularly EPA alone, has been shown to reduce cardiovascular mortality.
Chiara Cremolini, MD, PhD, an associate professor of medical oncology at the University of Pisa in Italy, agreed that confirmatory results are necessary prior to adopting EPA into routine cancer care. The EMT2 trial received funding from Yorkshire Cancer Research, with Amarin supplying the study capsules.
Frequently Asked Questions About the EPA Trial
What specific dosage of EPA was administered to trial participants?
Participants in the active treatment arm received 4 grams daily of purified EPA ethyl ester, marketed as icosapent ethyl or Vascepa, taken as capsules starting at least one day before surgery.
How long did patients continue taking the omega-3 supplements?
Treatment continued for a minimum of two years and a maximum of four years, persisting even after documented cancer progression occurred during follow-up.
Were there differences in baseline health characteristics between the study groups?
Baseline characteristics were generally comparable between arms, though diabetes was twice as common in the EPA group at 16% compared to 8% in the placebo group.