Abstract P330: Effect of Randomized Omega-3 Treatment on Downstream Bioactive Lipids and Their Association With Incident Cardiovascular Disease Events: Metabolomic Studies of the Vital and Jupiter Trials

Background: The effect of omega-3 (n-3) treatment on cardiovascular disease (CVD) outcomes in randomized controlled trials remains inconsistent. We hypothesize that the downstream products of n-3 metabolism, including bioactive lipids (BALs), have a heterogeneous relationship with CVD, which may provide insight into response heterogeneity. Methods: Using state of the art LC-MS, we assayed the plasma bioactive lipidome (>10K BALs) across 3,512 individuals at baseline and Y1 or -2, in the VITAL substudies (CVD case control N=1540; CTSC subcohort N=1054; 45% women, median age 70, 20% non-white) and JUPITER CVD case control (N=918) (NCT01169259, NCT00239681). Linear regression revealed BALs that change consistently with n-3 treatment (n3BALs) (discovery in VITAL controls, validation in an independent VITAL CTSC substudy, cumulative FDR .05). n3BALs that are also associated with CVD in VITAL CVD were identified using adjusted conditional logistic regression (p<.05). A BAL score was established using a multivariable adjusted model and validated externally in JUPITER CVD. Y1 or -2 change in BAL score in response to randomized n-3 and nonrandomized fish intake were tested using 2-sample t-test (p<.05). Results: 354 BALs changed in response to randomized n-3 vs placebo. Baseline levels of 8 of these BALs were significantly associated with incident CVD (Figure, panels A, B). When combined into the multivariable BAL score, 5 were inversely and 3 positively associated with the CVD risk. The BAL score was significantly associated with CVD with fully adjusted HR/SD of 1.32, 95%CI [1.12 - 1.55] in external study, Figure panel C. N-3 therapy, statin therapy and baseline fish intake were associated with weak Y1 or -2 reduction in BAL score. Conclusions: Downstream BALs are altered in response to n-3 treatment and associate with both decreased as well as increased risk of CVD. BAL response to n-3 treatment may explain clinical heterogeneity associated with fish oil supplementation and may be utilized for selection of more specific treatment.