From C-Reactive Protein to Interleukin-6 to Interleukin-1: Moving Upstream To Identify Novel Targets for Atheroprotection

Ridker PM
Circ Res. 2016 Jan 8;118(1):145-56. doi: 10.1161/CIRCRESAHA.115.306656.

Evidence is accumulating that vascular inflammation, via a wide array of cell types and signalling molecules such as cytokines, chemokines and adhesion molecules, plays an important role in atherogenesis and thrombogenesis. Most attention to date has been focussed on high sensitivity C-reactive protein (hs-CRP), a biomarker of vascular risk. Due to observations that CRP is unlikely to be an effective target for intervention, studies are now focussing on potential targets more upstream in the inflammatory IL-6 and Il-1 signalling pathway.
This (summary of a) review describes the evidence supporting the shift of focus to study more upstream molecules for anti-inflammatory atheroprotection.
 An important study leading to our thinking about hsCRP in relation to vascular disease was the prospective Physicians Health Study (PHS) in 1997. In this study hsCRP levels were found to be elevated decades before a first acute ischemic event. Also, anti-inflammatory aspirin was more effective in preventing first arterial atherosclerotic events in those with elevated hsCRP. PHS did, however, not conclude that CRP played a causal role in atherosclerosis, because other inflammatory biomarkers, including sICAM-, IL-6 and fibrinogen, also predicted vascular risk.

Many studies have reproduced these observations, and the Emerging Risk Factor Collaboration (ERFC) conducted a meta-analysis of these studies. That revealed that each SD increase in log hsCRP was associated with a multivariate-adjusted relative increase in risk of future coronary disease (1.37, 95%CI: 1.27-1.48) and CV mortality (1.55, 95%CI: 1.37-1.76). These effects are of the same order of magnitude  as that of total cholesterol, HDL-c and blood pressure. Adding hsCRP to risk prediction models outperforms risk scores based on traditional risk factors.

Several studies then showed that statins gave clinical benefit in individuals with low LDL-c levels, but with elevated hsCRP. Statins proved more effective if not only LDL-c but also hsCRP was reduced, with an efficacy proportional to the initial hsCRP elevation. These data established hsCRP as a powerful risk biomarker for primary and recurrent events. This does however not imply a causal role.

CRP is not only produced in the liver as a primary acute phase reactant, and is involved in complement activation and innate immune function, it is also produced as a response to localised inflammation. Functional studies have shown proinflammatory and prothrombotic effects on human endothelial cells and in mice, but some data are contradicting. Recent studies using an antisense oligonucleotide targeted to CRP-production or pharmaceutical-grade CRP infusions did not provide evidence for upstream effects on systemic inflammation as a response to altered CRP production. Nor have population-based Mendelian randomisation studies pointed in the direction of a causal relationship.

* Moving partially upstream to IL-6: positive associations with disease and partial links to causality

The upstream cytokine IL-6 resembles hsCRP in several aspects: IL-6 levels predict future vascular risk in apparently healthy populations, and an ERFC meta-analysis demonstrated a 25% increased risk of future vascular events (RR: 1.25, 95%CI: 1.19-1.32) for each SD increase in log IL-6. Lastly, like CRP, IL-6 levels correlate with endothelial dysfunction, arterial stiffness and extent of subclinical atherosclerosis, and new onset type 2 diabetes. Measurement of IL-6 in clinical setting, however more complicated, among other things due to circadian and postprandial variability and assay stability.

Importantly, IL-6 has different links to causal pathways of atherothrombosis. Il-6 signalling has been linked to plaque initiation and stabilisation, to microvascular flow dysfunction and to adverse outcomes in the setting of acute ischemia. Il-6 is upregulated at the site of coronary occlusion in STEMI. Interestingly, IL-6 can be produced by cardiac myocytes during local hypoxia in the viable border zone of reperfused infarctions. Mendelian randomisation studies further suggest that vascular risk varies largely as a consequence of heritable differences in IL-6 signalling.

Some data suggests however that it may be necessary to move even further upstream for an anti-inflammatory atheroprotective approach. Also, it is uncertain whether directly antagonising IL-6 has the desired and specific effects needed for therapeutic use. Moreover, IL-1 is the main driver of IL-6 signalling. IL-1 inducing factors of the NLRP3 inflammasome however do not affect IL-6 levels, although they are directly related to atherothrombosis.

Still, tocilizumab, a humanised IL-6 receptor antibody, is being evaluated for its effect on vascular events, as compared with a TNF inhibitor in patients with rheumatoid arthritis (RA). IL-6 inhibition may upregulate apolipoprotein B, which leads to increased LDL-c. A dose-dependent effect was observed in initial tocilizumab studies in RA patients. The underlying mechanism remains to be clarified, in order to establish whether a direct iL-6 inhibitory approach would be viable.

* Moving fully upstream to IL-1: can a causal pathway be proven and a therapeutic target validated?

The Il-1 signalling pathway is now considered a major target for immunomodulation and atherothrombotic protection. IL-1 is a proinflammatory mediator in both acute and chronic inflammation and a powerful inducer of innate immunity. It stimulates its own production and synthesis and expression of secondary inflammatory mediators, including Il-6. Two IL-1 proteins exist: IL-1α and IL-1β, which both bind to the type 1 IL-1 receptor. While 1α acts mostly locally, IL-1β is the primary circulating form. IL-1β is produced as a precursor that is cleaved following activation of the NLRP3 inflammasome. Studies on IL-1β-targeted interventions in rare inherited disorders associated with overproduction of IL-1β, suggest that IL-1β rather than IL-1α is the critical culprit.

IL-1β levels cannot be reliably measured in plasma, but experimental studies and pathological data have pointed in the direction of a role for IL-1β in atherogenesis. Several factors associated with atherosclerosis have been demonstrated to activate the crucial IL-1β-producing NLRP3 inflammasome. For instance, deposition of cholesterol crystals can trigger the NLRP3 inflammasome. This has been suggested to have a direct proinflammatory effect on atherogenesis through release of neutrophil extracellular traps, which prime macrophages to produce the precursor pro- IL-1β. These processes have been associated with acute thrombosis and several proatherosclerotic processes.

Several agents targeting IL-1 are available. The Il-1R antagonist anakinra significantly reduced the area under the CRP release curve in patients with NSTE-ACS, implying that IL-1 drives CRP elevation during acute ischaemia. Anakinra inhibits both IL-1α and IL-1β, which may not be ideal.

Canakinumab is a monoclonal antibody that specifically targets IL-1β. In a phase IIb trial in diabetics with high vascular risk, canakinumab lowered IL-6 and CRP in a dose-dependent manner, and to a lesser extent fibrinogen. LDL and HDL were not affected but a small increase in triglycerides was seen. The inflammasome-mediated IL-1β, IL-6 and CRP production was inhibited for several months. In the context of atheroprotection it would be helpful if chronic inhibition of inflammation only needs to be given a few times per year. Since canakinumab does not affect IL-1α function, infection risk should be lower.

The placebo-controlled Canakinumab Anti-Inflammatory Thrombosis Outcomes Study (CANTOS) examines whether IL-1β inhibition with subcutaneous canakinumab every 3 months can reduce recurrent CV events in patients with stable coronary artery disease and persistently elevated hsCRP (>2 mg/L). The CANTOS trial thereby uses hsCRP as a surrogate for IL-1β activity.
The trial enrolled over 10000 post-MI patients worldwide and is due to complete in 2017. The study is powered to detect a 20% relative risk reduction in hard CV events. Since canakinumab does not affect LDL-c levels, CANTOS is the first large study directly addressing the inflammation hypothesis of atherothrombosis.
A logical extension will be to study IL-1β inhibition in the context of acute ischemia, especially in light of very recent data in mice that neutralising IL-1β favoured infarct healing. Anticipated side-effects include an increased risk of infections, and potential benefits must thus outweigh this risk.

Other agents impacting the CRP - IL-6 – Il-1 – axis currently under investigation for a potential CV benefit include low-dose methotrexate, because lower event rates have been observed in patients with RA who received this treatment.
Also colchicine also affects the NRLP inflammasome and can reduce IL-1β expression. In a pilot study in STEMI patients, it reduced area under creatinine kinase-MB curve as well as infarct size, and CV event rates in an open-label randomised study. This needs to be tested in a formal double-blind set-up.

If these studies prove successful, it serves as an example of how close collaboration between clinical, epidemiological and basic scientists can translate fundamental biological principles can be translated into personalised medical practice.

Find this article online at Circ Res