Has the ‘strength’ of fish oil therapy been ‘reduced’? Reconciling the results of REDUCE-IT and STRENGTH

Abstract

The disparate results of STRENGTH1 (stopped for futility) and REDUCE-IT2 (significantly reduced cardiovascular outcomes), despite similar triglyceride reductions in high-risk statin-treated hypertriglyceridaemic patients, have raised questions about omega-3 fatty acids in prevention. We summarize differences that might explain these disparate results (Table 1). REDUCE-IT2 and STRENGTH1: similarities and key differences. REDUCE-IT2 and STRENGTH1: similarities and key differences. REDUCE-IT participants, who received highly purified eicosapentaenoic acid (EPA; 4 g icosapent ethyl), had greater increases in EPA levels than STRENGTH participants, who received omega-3 carboxylic acid [4 g; EPA + docosahexaenoic acid (DHA)]. Structural differences between EPA and DHA and consequent differences in expressed resolvins3 may result in EPA having beneficial antioxidant and endothelial function effects.4 On-treatment EPA levels were inversely associated with cardiovascular outcomes in REDUCE-IT and JELIS5 but not STRENGTH. In REDUCE-IT, mineral oil placebo was associated with increased LDL-C (either directly or perhaps by inhibiting statin absorption) and hs-CRP. No effect on LDL-C was seen with corn oil placebo in STRENGTH, and hs-CRP was reduced by 6.3%. However, the large relative (25%) and absolute (4.8%) risk reduction in cardiovascular events in REDUCE-IT suggests that the placebo alone is unlikely to explain the entire benefit. Furthermore, in post hoc analyses in REDUCE-IT, the beneficial effect of icosapent ethyl was not related to whether patients in the placebo arm had increased, decreased, or unchanged LDL-C. Participants in REDUCE-IT appeared to be ‘higher risk’. Secondary-prevention patients constituted 70.7% of REDUCE-IT participants vs. 55.9% in STRENGTH, and event rates with placebo were 25.5% in REDUCE-IT (4.9-year follow-up) vs. 12% in STRENGTH (3.5-year follow-up). In summary, while some mechanistic and trial-related differences exist, additional research is needed to explain fully the disparate results of these key trials. Both medications are safe, and clinical context should inform clinical practice. Salim S. Virani: Research support: Department of Veterans Affairs, World Heart Federation, Tahir, and Jooma Family; Honorarium: American College of Cardiology (Associate Editor for Innovations, acc.org). Conflict of interest: S.S.V.: research support: Department of Veterans Affairs, World Heart Federation, Tahir, and Jooma Family; Honorarium: American College of Cardiology (Associate Editor for Innovations, acc.org). V.N.: site PI for studies sponsored by Merck and Amgen. C.M.B.: Grant/Research Support: Akcea, Amgen, Esperion, Novartis, Regeneron, NIH, and AHA; Consultant: Althera, Amarin, Amgen, Arrowhead, Astra Zeneca, Corvidia, Denka Seiken, Esperion, Genentech, Gilead, Matinas BioPharma Inc., New Amsterdam, Novartis, Novo Nordisk, Pfizer, Regeneron, and Sanofi-Synthelabo.