Abstract 4140204: Relationship of Oxidized Phospholipids and Lp(a) to Outcomes after Acute Coronary Syndrome: A Post Hoc Analysis of the ODYSSEY OUTCOMES Trial

Background: Oxidized phospholipids (OxPL) are preferentially carried by and contribute to the pro-inflammatory properties of lipoprotein(a) [Lp(a)]. OxPL can be quantitated on all apolipoprotein B-100 containing lipoproteins (OxPL-apoB), a dominant proportion of which are present on Lp(a) particles. Objectives: To assess the effect of PCSK9 inhibition by alirocumab on plasma levels of OxPL-apoB, and the relationship to Lp(a) and major adverse cardiovascular events (MACE) in patients with a recent acute coronary syndrome (ACS) on optimized statin treatment. Methods: OxPL-apoB (Diazyme, Inc) and Lp(a) (TinaQuant, Roche Diagnostics) were measured at baseline (1-12 months after ACS) in a subset of participants in the ODYSSEY OUTCOMES trial with samples available at baseline (prior to randomization to alirocumab or placebo, n=11,630) and log 2 -transformed. Proportional hazards models adjusted for 12 baseline covariates evaluated the association of predictor variables (OxPL-apoB and Lp(a)) with MACE (coronary heart disease death, non-fatal MI, ischemic stroke, and hospitalized unstable angina) and all-cause death, with HRs for doubling of the predictor variable. Results: Baseline OxPL-apoB correlated with Lp(a) (r=0.68, p<0.001). At 4-months, ALI reduced median (IQR) PBO-corrected OxPL-apoB by 15%, P<0.001. In the placebo group, when analyzed in separate models with multivariable adjustment, pre-randomization OxPL-apoB (HR (95% CI) 1.081 (1.026-1.139), p=0.003) and Lp(a) (HR (95% CI) 1.080 (1.041-1.120), p<0.001) each predicted MACE. However, when OxPL-apoB and Lp(a) were placed in the same model only Lp(a) remained significant (HR (95% CI) 1.069 (1.023, 1.118), p=0.003). In the alirocumab group, neither baseline OxPL-apoB nor Lp(a) predicted MACE in individual or joint models. Baseline high sensitivity C-reactive protein predicted all-cause death, but OxPL-apoB and Lp(a) did not. Conclusions: In statin-treated patients with recent ACS assigned to placebo, the increased risk of MACE due to elevated OxPL-apoB appears to be mainly driven by their presence on Lp(a). In patients assigned to alirocumab, these variables were not predictive of MACE, suggesting mitigation of OxPL-apoB and Lp(a)-mediated risk by alirocumab.