Inflammatory Statements About Cardiovascular Risk Reduction: The CANTOS Trial


Written By

Reviewed By


Ridker PM, Everett BM, Thuren T, et al. Antiinflammatory Therapy with Canakinumab for Atherosclerotic Disease. N Engl J Med. 2017;377:1119-31.

We’ve all seen and used the American College of Cardiology 10-year atherosclerotic cardiovascular disease (ASCVD) risk calculator. There are several modifiable risk factors such as blood pressure, cholesterol, and smoking status that, if addressed, can lower ASCVD risk. But are there other modifiable risk factors that we are failing to account for and address? New evidence suggests systemic inflammation may be one.


To date, lowering levels of atherogenic cholesterol (e.g., low-density lipoprotein cholesterol [LDL-c]) using statins has been the focus of ASCVD risk reduction. But we’ve known for some time now that inflammatory biomarkers (e.g., high-sensitivity C-reactive protein [hsCRP]) can also be used to predict future atherosclerotic events.1 Some have argued that statins are effective, at least in part, because they reduce systemic inflammation, measured by hsCRP, in addition to LDL-c.2 However, the hypothesis that reducing systemic inflammation, independent of LDL-c, leads to a reduction of ASCVD risk has not yet been proven. The Canakinumab Anti-inflammatory Thrombosis Outcomes Study (CANTOS) sought to prove this inflammatory hypothesis.  Canakinumab is an interleukin-1β (IL-1β) inhibitor that reduces serum levels of inflammatory biomarkers such as hsCRP, interleukin-6, and fibrinogen without lowering LDL-c.3


CANTOS was a randomized, double-blind, placebo-controlled trial that examined atherosclerotic CV events rates in patients who received canakinumab or placebo. Eligible patients were at least 18 years old with a history of a myocardial infarction (MI) and an elevated hsCRP ≥2 mg/L despite the use of standard secondary prevention strategies, including statins. The study exclusion criteria were related to the risks of receiving canakinumab (i.e., immunocompromised, high risk of developing infection). Patients were randomized to either placebo or one of three doses of canakinumab: 50 mg, 150 mg, or 300 mg. All study treatments were administered subcutaneously every 3 months. The primary composite endpoint consisted of nonfatal MI, nonfatal stroke, and CV death.


The use of canakinumab at the dose of 150 mg, but not 50 mg and 300 mg, led to a statistically significant reduction in the primary composite endpoint (hazard ratio [HR] 0.85, P=0.021). See Table 1.  Overall, serious adverse events were similar in the placebo and canakinumab groups. However, patients receiving canakinumab were more likely to experience a fatal infection or sepsis (P=0.02). Intriguingly, patients receiving canakinumab were less likely to die from cancer (P=0.02), but the incidence rates were very low (0.45% vs. 0.64% for the canakinumab and placebo groups, respectively).


Table 1: Effects of canakinumab on major adverse cardiovascular events


Primary Outcome
(per 100 pt-yrs)

Hazard Ratio
(95% CI)

ARR (%)


Placebo (n=3344)






50 mg (n=2170)





150 mg (n=2284)





300 mg (n=2263)





All doses (n=6717)





ARR: absolute risk reduction; CI: confidence interval; NNT: number needed to treat
a Number of patients who would need to be treated for an adjusted duration of 3 years with canakinumab to prevent one nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death
P=0.021; This result was significant according to the prespecified closed-testing procedure. The threshold P value for the primary end point for the 150-mg dose was 0.02115.
P=0.031; This result was not significant according to the prespecified closed-testing procedure. The threshold P value for the primary end point for the 300-mg dose was 0.01058.


The CANTOS trial found that patients with coronary artery disease (CAD), most of whom were taking statins, experienced a significant reduction in major adverse cardiovascular events (MACE) when given canakinumab. While previous trials have investigated the benefits of reducing inflammation for other CV indications, such as pericarditis with colchicine and prevention of post-MI cardiac remodeling with anakinra, they did not assess the occurrence of CV events.4-5  Thus CANTOS is unique and is the first clinical trial of sufficient size and duration to test this scientific hypothesis.


Strengths of this study included the large sample size, international study population, well-matched baseline characteristics (including LDL-c), and the use of clinically important outcomes. Unfortunately, the results cannot be extrapolated to patients with low or normal hsCRP. It should be noted that patients whose hsCRP was less than 2 mg/L at baseline were excluded, which accounted for 20% of all patients screened.


This trial sought to determine if systemic inflammation is a driver for CV risk. Patients enrolled in the study had elevated hsCRP levels at baseline. Canakinumab not only reduced hsCRP (a surrogate marker) but also the rate of MACE. Enrolling a cohort of patients with low or normal hsCRP at baseline would help clarify the inflammatory hypothesis. If this subgroup experiences a similar (or greater) reduction in MACE, it suggests that canakinumab is beneficial in any CAD patient, and would potentially disprove the systemic inflammation hypothesis. On the other hand, if patients with a low or normal hsCRP show little or no reduction in MACE, one could rightfully conclude that systemic inflammation is an important target for CV risk reduction.


Despite the lack of complete clarity regarding the inflammatory hypothesis, it can be confidently stated that the CV benefit seen with canakinumab in this trial was not related to LDL-c reduction. While nearly all of the patients in the study were taking statins throughout the trial, there were no differences between the two groups’ LDL-c values following treatment with canakinumab or placebo. This is an interesting finding because of what was observed in the Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects with Elevated Risk (FOURIER) trial. In FOURIER, patients with CAD taking statins and whose LDL-c levels were ≥70 mg/dL were prescribed either evolocumab or placebo.6 The study showed that those patients who received evolocumab achieved greater LDL-c lowering and experienced a modest reduction in MACE compared with those who received placebo. It has also been shown that proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors do not affect inflammation, and therefore their benefits are thought to be purely the result of LDL-c reduction.7 Statins are associated with a significant mortality benefit which is thought to be a result of both LDL-c reduction and anti-inflammatory effects.  FOURIER and CANTOS suggest that either an additional LDL-c reduction or a reduction in inflammation (or both) can lead to improvements in CV outcomes.


Is there a role for canakinumab to treat patients with ASCVD? Given the effect size observed in CANTOS and the cost of canakinumab, it would be hard to justify. To put the CANTOS trial’s findings in perspective, the NNT was 194 in patients treated with canakinumab 150mg for 3 years. The NNT in the FOURIER study was 48 when patients were treated with evolocumab for 3 years.6 Both studies showed a significant reduction in MACE, but the NNT indicates that the data are more compelling for evolocumab. A recent cost-effectiveness analysis found that the price of PCSK9 inhibitors would need to be reduced by 71% in order to be cost-effective based on data from FOURIER.8 Yearly costs of canakinumab at the 150 mg dose are estimated to be around $77,000, which is nearly five times the yearly cost of evolocumab.9 Clearly a very substantial cost reduction would be necessary before canakinumab could be considered an economically viable choice.


What does the future hold? The Cardiovascular Inflammation Reduction Trial (CIRT) is currently ongoing. This study will assess the effects of low-dose methotrexate therapy on MACE in patients with CAD and type 2 diabetes or metabolic syndrome.10 If methotrexate proves to be effective and safe, it would likely be a more cost-effective option than canakinumab.


Despite its limitations, the CANTOS trial adds substantial new evidence to support the inflammatory hypothesis. While canakinumab is unlikely to revolutionize therapy due to its high cost, these findings will encourage the investigation of new treatments that target systemic inflammation. We believe this will likely change the management of ASCVD by tailoring treatments based on both hsCRP and LDL-c concentrations. But don’t take our word for it. Tell us what you think!




I believe that the event rates should be in 100 pat-yrs (rather than 1,000 pat-yrs); otherwise the ARR and NNT numbers don't fit.... correct? The fatal infection rate/sepsis is cited as statistically significant but knowing the absolute rates would be important in weighing the risks/benefit. I was an investigator in this trial (which I thought was an excellent study idea), but, I tried hard and failed to enroll a single patient. I think the main objection was a statement in the consent form that mentioned a potential increased risk of malignancy. Even after I assured patients that I could find no data to support such an effect of this previously marketed drug, I still could get no takers. Interesting now to see that malignancy was actually lower!!
Stuart T Haines's picture

Thanks Henry for catching the error in Table 1. Corrected!

Thank you for the correction and the comments. The incidence of fatal infection/sepsis was 0.31 per 100 patient-years across all canakinumab dosing groups, compared to 0.18 per 100 patient-years in the placebo group. As you alluded to, although this difference was significant, the incidence in each group was still quite low. Your insights as a study site attempting to enroll patients are very interesting. The proposed mechanism for the lower cancer mortality rates in the canakinumab group is related to the role of interleukin-1 in progression of certain cancers (primarily lung cancer), which has been demonstrated in prior studies. However, rates of any cancer development were not significantly different between groups, and as noted in the commentary, the overall rates of malignancy in this study were quite low, making it difficult to draw conclusions from these findings alone.

Would love to see more studies looking at cheaper anti-inflammatory alternatives like curcumin and other natural products. There is a whole host of evidence on a small scale for inflammatory conditions with curcumin in particular. Obviously if a drug company picked it up, formulated it and marketed it it would probably cost more than the current $50 a month for a quality brand, but I'm sure way less expensive than $77,000 per year! I also wonder if in our at risk patients if we should be checking CRP, especially in patients unable to tolerate statins so we can look at other options to address inflammation, even from a functional/integrative perspective.