CANTOS: A Seductive Song With Several Verses

Roland Klingenberg; Thomas F. Lüscher


Eur Heart J. 2018;39(38):3508-3510. 

Rubor, Calor, Dolor, and Tumor—this is how the Roman physician and writer Aulus Cornelius Celsus (~25 BC to 50 AD) described inflammatory skin lesions.[1] He could not see atherosclerotic plaques nor did he have any idea of their importance, but his concept can easily be translated into unstable coronary lesions: vulnerable plaques are highly perfused, they are 'red' and 'hot'. They do not hurt as the vasculature is not as densely innervated as the skin, but the chemokines, cytokines, and autocoids responsible for it are there released from infiltrating white blood cells. Finally, white blood cells, cholesterol crystals, debris, and water are responsible for the swelling of the inflamed lesion—in short the tumour. Rudolf Virchow, the eminent German pathologist of the 19th century, considered plaque formation as a result of two distinct processes in the arterial vessel wall 'Fettmetamorphose' (lipid metamorphosis) and 'Endoarteriitis',[2] and coined the concept of 'atherosclerosis as an inflammation induced by cholesterol'. However, it took nearly half a century until Anitschkow demonstrated that cholesterol feeding induced atherosclerosis in rabbits,[3] and even longer to rediscover inflammation in this context.

The demonstration that high levels of C-reactive protein predict adverse outcome after an acute coronary syndrome[4] raised the question of whether inflammation was triggered exclusively by cholesterol and its modified metabolites or may rather represent an independent mechanism in its own right, thereby promoting the atherosclerotic process and its clinical consequences such as myocardial infarction, stroke, and death. Is there a remaining inflammatory risk beyond cholesterol? The JUPITER trial[5] suggested that there is, but did not convince the sceptics. The recent CANTOS trial,[6] however, proved the case. Indeed, the monoclonal antibody targeting interleukin-1β did not change lipid levels, but reduced C-reactive protein and major adverse cardiovascular events (MACE).

C-reactive protein is actually only a read-out of a cascade of inflammatory mediators reflecting activation of the NLRP3 inflammasome by cholesterol crystals, glucose, and pathogens, among others, leading to the formation of pro-inflammatory interleukin-1β and eventually interleukin-6 (Take home figure). Of note, in patients with an acute coronary syndrome and ST-segment elevation, interleukin-6 is markedly increased at the site of the ruptured plaque and coronary occlusion by thrombus formation compared with peripheral levels in the aorta.[7] Interleukin-6 has been recognized as an independent circulating biomarker discriminating risk for adverse cardiovascular events in patients with acute coronary syndromes as well as in stable coronary artery disease, and even apparently healthy individuals.[8–11] Furthermore, Mendelian randomization studies[12] and a meta-analysis of the interaction of gene variants with circulating protein and clinical phenotype involving 82 studies[13] strongly suggest a causal relationship between interleukin-6 receptor activation and coronary artery disease.

Take Home Figure.

The substudy of the CANTOS trial presented in the current issue of the European Heart Journal provides further evidence in support of this concept.[14] Indeed, it demonstrates that circulating interleukin-6 levels, lowered indirectly by administration of the monoclonal antibody canakinumab targeting interleukin-1β, translated into a reduction of adverse clinical events. In this on-treatment analysis at 3 months, based on patients reaching interleukin-6 levels below the median of 1.65 ng/L, the authors observed, after adjustment for clinical characteristics and baseline levels of LDL-cholesterol and interleukin-6, a significant reduction of MACE defined as cardiovascular mortality, recurrent myocardial infarction, or stroke by 32%, as well as MACE plus hospitalization for unstable angina requiring urgent revascularization by 30%, cardiovascular mortality by an impressive 52%, and even all-cause mortality by 48% (Take home figure). Of note, the percentage of patients achieving interleukin-6 levels lower than the median increased dose dependently from 39% in the 50 mg to 50% in the 150 mg and 60% in the 300 mg canakinumab group. This dose dependency of the effect of canakinumab provides further credibility to the findings of this subanalysis.

Due to the nature of the current study conducted as a post-hoc analysis of the CANTOS trial and the indirect interference with the interleukin-6 pathway, the current results are hypothesis-generating at this point; as such, causality remains to be proven prospectively, but the results are indeed suggestive. Several randomized, double-blinded, placebo-controlled clinical trials are underway or have recently been completed to assess the impact of tocilizumab, a monoclonal antibody that selectively blocks the interleukin-6 receptor, in patients with myocardial infarction.[15] Kleveland et al. demonstrated in 117 patients with non ST-segment elevation myocardial infarction (NSTEMI) that tocilizumab compared with placebo attenuated the inflammatory response as assessed by C-reactive protein levels, which was the primary endpoint, and reduced the release of troponin T during the first 3 days after admission.[16] Of note, lipoprotein levels of LDL-cholesterol, HDL-cholesterol, and triglycerides at 3 days, 3 months and 6 months were similar between treatment groups (tocilizumab vs. placebo) with the exception of a moderate temporary increase of total cholesterol at day 3. Conversely, in a trial randomizing 324 patients with rheumatoid arthritis to tocilizumab vs. adalimumab, the latter being a monoclonal antibody against tumour necrosis factor-α, tocilizumab was associated with a significant increase in levels of total cholesterol, LDL-cholesterol, HDL-cholesterol, and triglycerides at 8 weeks of treatment.[17] Furthermore, recent data in patients with rheumatoid arthritis demonstrate that low LDL-cholesterol levels in these patients are due to hypercatabolism of LDL particles, while interleukin-6 receptor blockade by administration of tocilizumab normalizes this abnormality with concomitant increases in lipoproteins.[18] This appears to set rheumatoid arthritis apart from coronary artery disease and may explain the lack of an effect of tocilizumab on lipoproteins in patients with NSTEMI.[15] Clearly, future prospective randomized controlled studies on the efficacy and safety of an interleukin-6 receptor antagonists such as tocilizumab in patients with coronary artery disease would prove the current hypothesis generated within the CANTOS trial. In summary, the study by Ridker et al.[14] adds an important piece to the emerging evidence that interference with the interleukin-6 pathway may translate into clinical benefit for patients with coronary artery disease.