Independent and cumulative prognostic information of C-reactive protein and interleukin-6 to predict major adverse cardiovascular events and death in post acute coronary syndrome patients

Abstract Background Interleukin 6 (IL6), produced in response to infections and tissue injury, contributes to host defence through the stimulation of acute phase responses and immune reactions. The inflammatory cascade is characterized by activation of the NLRP3-inflammasome, causing IL1 to IL6 production leading to the clinical biomarker C-reactive protein (CRP) elevation. In patients with acute coronary syndrome (ACS), both IL6 and high-sensitivity (hs) CRP levels are associated with risk of major adverse cardiovascular events (MACE). However, the independent and cumulative prognostic information provided by IL6 and hs-CRP after ACS is unclear. Methods The ODYSSEY OUTCOMES trial compared the PCSK9 inhibitor alirocumab with placebo added to maximum-tolerated statin treatment in patients post-ACS. We assessed in patients from whom biobank samples were available the relation between log IL6 and hsCRP levels (Roche) at baseline and the risk of MACE and all-cause death in proportional hazards models. Results 11,817 patients had both baseline CRP and IL6 data; 1306 had a MACE primary endpoint and 458 died. Median hsCRP, IL6 and low-density lipoprotein (LDL) were 1.54 mg/L, 5.00 pg/mL, and 86 mg/dL, respectively. hsCRP and IL6 had moderate correlation (r=0.52). Adjusted for treatment assignment, age, sex, diabetes, LDL-cholesterol, and time since index ACS, hsCRP was a significant predictor of MACE (p<0.0001) but IL6 was not (p=0.21), while both predicted all-cause death (p<0.0001 for hsCRP and p=0.0003 for IL6). Relationships did not depend on treatment (all interaction p>0.25). Therefore, hsCRP remains significantly related to risk of MACE and death when accounting for IL6. Conclusion In patients with recent ACS, both hsCRP and IL6 predicted all-cause death in univariate and adjusted analyses. IL6 levels predicted MACE in univariate but not in adjusted analyses. In contrast, hsCRP predicted MACE in both univariate and adjusted analyses. Thus, hsCRP appears to convey substantial independent information on inflammatory risk beyond that provided by IL6. This has implications for prognosis and possibly for trials of anti-inflammatory drugs.