Why the CANTOS Is a Game Changer in Cardiovascular Medicine

On August 27, 2017, in occasion of the European Society of Cardiology congress, the results of the Canakinumab Anti-inflammatory Thrombosis Outcomes Study (CANTOS) were presented and simultaneously published in the New England Journal of Medicine1 and in the Lancet.2 The results of the CANTOS have been covered by medical news at the time of presentation. I would like to discuss in this article the reasons why the CANTOS trial is a game changer in cardiovascular medicine. First, the CANTOS confirms that patients with previous myocardial infarction (MI) and elevated C-reactive protein (CRP) (greater than 2 mg/L) remain at an unacceptably high risk of major adverse cardiac events.1 This high incidence of events, predicted by previous studies and meta-analyses,3 averaged 4%–5% per year in the placebo cohort. Second, the CANTOS introduces a new drug, canakinumab, that has a unique mechanism of action, hitherto unexplored in any large-scale cardiovascular clinical trial. Canakinumab is a human monoclonal antibody blocking interleukin-1β (IL-1β). First in class of IL-1β antibodies,4 it followed anakinra and rilonacept in the IL-1 blockers, also being explored in phase II cardiovascular clinical trials.5 Canakinumab is currently commercially available as Ilaris and indicated for the use of cryopyrin-associated periodic syndromes, a rare genetic disease.5 In the CANTOS trial,1 canakinumab was tested as an add-on, in one of 3 doses (50, 150, or 300 mg every 3 months) in comparison with placebo. Canakinumab 150 mg every 3 months significantly reduced with an HR of 0.85, the primary end point of cardiac death, nonfatal MI, and nonfatal stroke, largely due to a reduction in MI, in patients with previous MI and residual inflammatory risk, defined as plasma levels of CRP greater than 2 mg/L. Given the uniqueness of the mechanism of action of canakinumab, the effects on CRP despite low-density lipoprotein cholesterol baseline levels close to accepted goals and unchanged with canakinumab treatment, and the lack of other treatments targeting the residual inflammatory risk, canakinumab is the only anti-inflammatory drug shown to reduce the risk in patients with cardiovascular disease. Canakinumab offers also the advantage of a subcutaneous injection every 3 months, as compared to the more frequent injections with anakinra or rilonacept. Third, the CANTOS data set is very robust. Although it may appear as 1 single trial, the CANTOS is actually a sort of a combination of 3 separate trials with a common placebo arm: each of the canakinumab dose is indeed compared separately with placebo. It is therefore reassuring to see that a dose response is seen within most if not all end points by which the 50-mg dose appears insufficient, the 150 mg is significantly associated with a benefit, and that for some of the end points, the 300 mg is as well and provides a dose-dependent response. It is also very reassuring that for the primary end point and most of the other efficacy end points, both the 150- and 300-mg dose groups compared favorably with placebo. Fourth, the benefits of canakinumab in the CANTOS are seen across the various components of the primary end point and the other cardiovascular outcomes. Although the benefits are largely driven by a 24% reduction in recurrent MI, there is also a nonsignificant trend for a 10% reduction in cardiovascular death. The effects on stroke seem less pronounced, which is in line with the different pathogenesis of MI and stroke. It is also encouraging to see a 36% reduction in urgent revascularization and a 37% reduction in cardiac arrest. Fifth, the consistency of the benefit signal with canakinumab across the subgroups is also appreciated. Sixth, the on-treatment analysis and CRP responder analysis both confirm the benefit signal. Although the intention-to-treat analysis is the preferred method to test the widespread and real-life applicability of a treatment strategy, the on-treatment analysis is better suited to explore the signals of efficacy and safety, as it takes into account the actual exposure. It is therefore reassuring to see that in the CANTOS,1 the on-treatment analysis is associated with an even more favorable HR favoring canakinumab. In a similar fashion, the CRP responder analysis takes into account whether the subject is displaying the expected biological response to the treatment (CRP reduction) and if accordingly the outcomes are affected: the responder analysis at 3 months show that those who achieve a more than median reduction in CRP to <1.8 mg/L have a 27% reduction in recurrent events (Table 1).TABLE 1.: Key Reasons Why the CANTOS is a Game Changer in Cardiovascular MedicineSeventh, canakinumab is associated with a low rate and predictable adverse events. The annualized rate of serious adverse events was not significantly different in canakinumab 150 mg versus placebo (11.7 vs. 12.0). The serious adverse events occurring more frequently in canakinumab were leukopenia (0.37 vs. 0.24, difference of 0.13), serious infection (3.13 vs. 2.86, difference of 0.27), and fatal infections (0.28 vs. 0.18, difference 0.10). Although none of these differences were statistically significant in the comparison of the canakinumab 150-mg group alone versus placebo, leukopenia and fatal infection were significantly more common in the canakinumab-treated patients when the 3 groups were pooled together. Of note, the number of opportunistic infections, as defined by an ad hoc adjudication committee, was not different in canakinumab versus placebo (0.15 vs. 0.18). Altogether, these data show that, as expected from the literature derived from other IL-1 blockers used in other conditions,5 canakinumab increases the risk of serious and lethal infections not by increasing the likelihood of opportunistic infections, thus reflecting that IL-1 blockers do not interfere with the immune-specific response but rather blunt the innate immune response likely delaying the recognition of the clinical presentation of infection and thus promoting more widespread infection. Although this is certainly a reason for concern, education among providers and patients may help reduce this risk by increasing awareness, prompt recognition, and early treatment. Moreover, the absolute incidence of these infections and of lethal infections is very low and outweighed by benefits in cardiovascular, and cancer-related, mortality. Fear of infection should not prevent the use of this class of drugs in the appropriate patient but rather guide its use by appropriate monitoring and early treatment of infections. Eighth, a rather unexpected clinical finding in the CANTOS, although likely predictable from preclinical studies, is the benefit of canakinumab on lung cancer incidence and cancer-related mortality.2 Canakinumab 150 mg was associated with a reduction in all cancer-related mortality (0.50 vs. 0.64 in placebo). Ninth, as expected, canakinumab reduced the incidence of a wide range of inflammatory arthritis, including osteoarthritis, rheumatoid, and gouty arthritis. Although these are not major outcomes, these are significant reasons for discomfort and disabilities, and barriers to exercise and healthy living in patients with cardiovascular disease. Tenth, although not statistically significant, and not powered for it, canakinumab 150 mg was associated with an 8% difference in all-cause mortality. Although it may appear as a small difference, and it certainly is not significantly different, it should be noted that this derives from nearly 3500 patients treated for a median of 3.5 years. In numbers, 375 of the 3344 patients randomized to placebo had died (11.2%) versus 238 of the 2284 of the canakinumab 150 mg (10.4%), meaning that if the remaining approximately 3000 patients on placebo were now placed on canakinumab, there could be an additional 30 lives saved in 3.5 years. There are obvious barriers to treating all patients with previous MI and CRP >2 mg/L with canakinumab. Cost would be a barrier if Novartis Pharma would commercialize the cardiovascular use of canakinumab as Ilaris, which is only indicated for the use in a rare orphan disease and is extremely costly, but this is unlikely to be the case, as cardiovascular disease is anything but an orphan disease. Cost may remain a barrier, but any calculation taking into account the current Ilaris cost is likely a waste of time until the new indication is reviewed. The appropriate choice of the patient and the decision between additional lipid-lowering versus IL-1β blocking strategies will need to be addressed: these are likely to be based on algorithms that take into account the residual risk. Education would be required for cardiologists on how to screen, monitor, and treat common infections especially in patients with diabetes. The “how to” such as “how long would patients need to be treated,” and “whether patients would be treated to a target CRP” or “if treatment would be continued long term or intermittently,” etc…, will need to be further addressed. Let us not forget that it has taken several decades for us to understand and optimize antihypertensive, lipid-lowering, and glycemic-controlling therapies, we should not assume that the CANTOS can answer all these questions in a single trial. The ideal patient for canakinumab 150 mg, if it were to be approved for this cardiovascular indication, would likely be one that has a high a risk of recurrent MI based not only on elevated CRP but also on the failure of other (first-line) therapies, and who is not at high risk of infection. It would be reasonable to assess response to treatment, in CRP reduction at 3 months, and if there is a reduction to <2 mg/L, continue for 1–3 years and then reassess risk– and cost–benefit ratios. In conclusions, the CANTOS is a game changer in cardiovascular medicine. Canakinumab and IL-1 blockade as a class are here to stay in cardiovascular medicine. Those who have and are continuing to oppose resistance to the inflammatory hypothesis of cardiovascular disease should consider to come to peace with the fact that it is no longer a hypothesis by an actual theory.