CANTOS: Lowering Inflammation, Not LDL-C, to Reduce CVD

An Interview With Paul Ridker

John M. Mandrola, MD; Paul M. Ridker, MD


September 05, 2017

John M. Mandrola, MD: Hi, everyone. This is John Mandrola from | Medscape Cardiology. I'm here in Barcelona, Spain, at the European Society of Cardiology (ESC) Congress. I'm pleased to have Dr Paul Ridker, who is the primary investigator of the CANTOS trial, a large multicenter, randomized controlled trial testing the inflammation hypothesis for treating atherosclerosis.[1] Welcome, Dr Ridker.

Paul M. Ridker, MD: John, it's a pleasure to be with you.

Dr Mandrola: Tell me about the background of this trial.

Dr Ridker: If we think about preventive cardiology, the first era was mom and pop and apple pie. There was diet, exercise, and smoking cessation. The second era began, I think, in 1994 with the 4S trial.[2] We said, "Wow—statins really work. We need to think as a community of cardiologists about lipid lowering." And the past 25 years has been about the expression of that, ultimately leading to the PCSK9 inhibitors. During that same 25 years, the vascular biology community and a group of translational biologists have been thinking about inflammation, the process by which our bodies repair damage.

As you know well, we showed 20 years ago that if you measured high-sensitivity CRP, you could predict who was at high risk for a heart attack or stroke, even if lipid levels were low.[3,4] It was 19 years ago that we showed that statins were both anti-inflammatory as well as lipid-lowering.[5] Of course, it's now 9 years ago that the JUPITER Trial[6] came out and we said, "Wait a minute—we think statins are very effective, but you have to give them to people with a proinflammatory response as well as a lipid response" and our guidelines changed. Remember: None of that proves the hypothesis that lowering inflammation, per se, would lower cardiovascular risk.

Dr Mandrola: That's because the statins reduced LDL cholesterol (LDL-C) as well as inflammation.

Dr Ridker: Absolutely. We gave a lipid-lowering drug with anti-inflammatory effects. The whole point of CANTOS was to get everybody on a high-dose statin and then say, we're going to give a targeted narrow spectrum anti-inflammatory intervention which will have no effect on LDL-C; no effect on triglycerides or HDL-C; massive reductions in interleukin-6 (IL-6), the key cytokine that plays in this game; as well as the CRP levels, and ask whether there's any benefit.

Dr Mandrola: Tell us about this novel compound that you studied.

Dr Ridker: Canakinumab is a fully human monoclonal antibody that targets interleukin-1 beta. There's a key concept here about the NLRP3 inflammasome. That inflammasome activates IL-1 beta. IL-1 beta has a number of direct vascular effects, but it also leads to an increase of IL-6, which has many vascular effects. And, of course, IL-6 drives the liver to produce fibrinogen, clotting factors, PAI-1, clot-dissolving factors, and our biomarker, the CRP.

Residual Inflammatory Risk

Dr Mandrola: How did you go about answering this question? How was the trial designed?

Dr Ridker: We enrolled over 10,000 patients in 39 countries. We had 1000 collaborators who took post–myocardial infarction (MI) patients. Not everyone. These post-MI patients were already on very aggressive care; nearly 80% had undergone revascularization, and almost everyone was on aggressive lipid lowering. In fact, our LDL-C baseline levels were below those of the FOURIER trial.[7] They were selected for this proinflammatory response, defined as a CRP above 2 mg/L. The median CRP was above 4 mg/L.

That's important because we already know what to do for residual cholesterol risk—that's ezetimibe, that's PCSK9 inhibitors. This is a different patient group. This is residual inflammatory risk selected because the driver of their problem is that high CRP.

Dr Mandrola: There were different doses of the drug.

Dr Ridker: That's right. We have a low-dose 50 mg, intermediate-dose 150 mg, and a high dose of 300 mg given only once every 3 months. This is a very different kind of therapy. It's only being given three or four times a year, all against placebo. The bottom line of the trial was the 150-mg dose and the 300-mg dose delivered no change in LDL-C and no change in HDL-C. Each of those doses gave the same 40% reduction in CRP and IL-6, and they both delivered the same 15% reduction in our hard MACE primary endpoint and a 17% reduction in the broader endpoint that included revascularization for acute ischemia.[1]

What's really interesting as a cardiologist is that we had a 30% reduction in the endpoint of any need for bypass surgery or revascularization at any point in the trial. That's interesting because inflammation is a driver of the progression of this disease, and so that endpoint really was quite instructive.

Cancer Findings

Dr Mandrola: A 15% relative risk reduction. What about some other outcomes? There was an interesting cancer outcome.

Dr Ridker: The cancer issue and the atherosclerosis issue both reflected another biologic concept, which is responder/nonresponder. Even on the vascular side, we probably should start there, as we dosed this every 3 months. We measured their on-treatment CRP and their on-treatment IL-6. We found that the people who got the large reductions in CRP got a 25%-30% reduction in hard MACE, whereas in the people who only got a modest reduction in CRP, the drug was effective but not anywhere near so.

As a clinician I might say, "Well, I could figure out who's most likely to benefit by measuring an on-treatment CRP, just like we measure on-treatment in LDL-C." Lower is better for inflammation as well as for lipids.

You've raised this other really interesting piece of biology. When we started the trial, we knew that atherosclerosis patients were likely to be current or past smokers, and CRP level—most cardiology listeners may not know—is also a pretty good predictor of lung cancer. Why? Because there's inflammation in the lungs. If you smoke a pack of cigarettes, you're inflaming your lung. If you're breathing in diesel fuel, you're inflaming your lung. It's been known for a long time that there are inflammatory cancers.

What we've really learned from CANTOS is that inflammation reduction in the absence of lipid reduction lowers vascular event rates.

What we found here was that the doses of canakinumab led to a 50% reduction in all-cause cancer mortality that was due almost exclusively to a 75% reduction—very dose-dependent in lung cancer incidence.[8] In fact, it was a 77% reduction in death from lung cancer. Be careful here; while it's very interesting biology, we have to replicate that. To me, as an inflammation biologist, it makes great sense, and the cancer world understands the role that inflammation plays in certain cancers. What we've really learned from CANTOS is that inflammation reduction in the absence of lipid reduction lowers vascular event rates. We're not going to argue any longer about the inflammation hypothesis of atherosclerosis. It's time to move forward and figure out whether there are other agents, other ways of doing this.

On the cancer side, we have to step back and say, "With the biology of what worked in these very early cancers, might it work in already diagnosed cancers?" We don't have an answer for that.

Dr Mandrola: I'm just curious: Is this something that just came up or did you have a notion going into the trial that this would be an effect?

Dr Ridker: We actually established an oncology endpoint committee on day one. We had a group of oncologists looking at this from the very beginning. The reason for that is it's been hypothesized for about 20 years that inflammation might drive certain cancers, particularly lung cancer. In fact, a group of vascular biologists led by a guy names Charles Dinarello,[9] who actually cloned IL-1 beta in 1984, has been arguing that maybe we ought to be thinking about this as a cancer drug. That's been out there. The experiment had just never been done.

The Cost Question

Dr Mandrola: Very interesting. Our listeners are going to be wondering about cost. This drug is an orphan drug now; the cost is off the charts. But you've commented that that's probably not relevant, is it?

Dr Ridker: I don't think you can even have an intelligent discussion about cost right now. First, as you correctly point out, the drug is indicated for very rare genetic inherited syndromes—-overexpression IL-1 beta syndromes—-that a very small number of kids in the United States have (it's probably 1000 patients globally). The price for the drug right now—that's an orphan drug cost.

Clearly, if the sponsor chooses to get a label for atherosclerosis, they're going to lose their orphan drug status, and so cost numbers right now don't make sense. The other thing that's a bit difficult right now is that there's no way to compute a number needed to treat (NNT) because—again, as I'm saying—even within this trial of high-risk post-MI patients selected for high CRP, I don't think we're going to treat all of them. As I've shown in the data presented here at the ESC, the magnitude of benefit is pretty large: 25%-30% if you respond to the drug.

If you're a nonresponder, I would wonder why you're taking it; you just get the toxicity. We'll see if the sponsor wants to go in that direction, but if they do, it will dramatically lower the NNT values. It is just good medical care. All of us would rather give a drug to people who respond to it than to everybody.

Dr Mandrola: Now that you mentioned toxicity, tell us about that.

Dr Ridker: Broadly, the drug was actually less toxic than we had concerns about. By that, I mean that there's really nothing for liver, nothing for kidney. There was, however, leukopenia, which is expected. That's one of the things the drugs does, and that did translate into a small but statistically significant increase in fatal infection.

That shouldn't be very surprising when our rheumatology colleagues treat patients with rheumatoid arthritis with a variety of these drugs; that's a risk they take. They've become very good at recognizing early signs of infection, getting people in earlier. If the cardiology community starts to think about using therapies like this, we're going to have to become a little more aggressive. If the patient reports a fever or has some sort of an infection, get them in earlier, get them on antibiotics a little earlier. That's how we handle this in the other environment.

Just the Beginning

Dr Mandrola: One final note. This, you said, will open up other avenues. Just briefly tell us about these other avenues of looking into inflammation treatment.

Dr Ridker: This is systems biology. We've given a drug that alters innate immune function for 5-6 years. We had a reduction in incident rheumatoid arthritis. We had a reduction in osteoarthritis. We had a reduction in gout. We had a reduction in heart disease. We talked about the surprisingly large reductions in cancer, with some hazards on the infectious side.

Now the question's become, how do we apply this clinically? What other kinds of agents might there be? My group's running a trial of low-dose methotrexate, funded by the Natural Heart, Lung, and Blood Institute. It's called the Cardiovascular Inflammation Reduction Trial (CIRT). It was started around the same time as CANTOS. We've randomized about 4500 patients across the United States. We have about 2000 to go. My guess is that it will read out in about 2-3 years. By contrast to canakinumab, this uses methotrexate, which is a very inexpensive generic approach to the same central biologic pathway.

There are some studies being done with a drug called colchicine, which you and I know as something we might give for pericarditis. It's a tough drug to take on a chronic basis, but that's being evaluated. As you might imagine, there's a whole host of drugs that have been sitting in the rheumatology catalogue which are going to be rethought now as, "Wow—if inflammation is something that, on top of aggressive lipid reduction, will give us benefits, maybe we need to rethink the whole approach."

Dr Mandrola: Thank you very much. I'm pleased to have you here. I learned a lot.

Dr Ridker: John, thank you for having me.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.