ֱ

Do You Think Novartis Wanted CANTOS to Succeed?

— In this Revolution and Revelation Milton Packer offers an answer

MedpageToday

Every company wants its clinical trials to succeed. Right? Well, usually.

On June 22 2017, with little fanfare, announced the topline results of CANTOS, a large-scale trial of its interleukin 1-beta antagonist, canakinumab (Ilaris). The drug is used to treat rare inflammatory disorders.

The CANTOS trial tested the efficacy of Ilaris in patients with high C-reactive protein levels following an acute myocardial infarction. It was hoped that the drug might reduce the combined risk of cardiovascular death, nonfatal myocardial infarction and nonfatal stroke.

The trial was big (>10,000 patients) and really expensive. If I made that investment, I would want the trial to succeed.

The investigators wanted success. They hoped to demonstrate that inflammation was important in the genesis of new coronary events. The CANTOS trial was poised to provide that evidence.

But CANTOS started >6 years ago, and kept going and going. The trial was not stopped early because of superior efficacy. After 6 years of no news, most observers thought the trial would be neutral.

Then on June 22, a surprise. Novartis announced that CANTOS had met its primary endpoint. Wow! Trials that continue unstopped for 6 years are rarely successful. This was really unexpected.

I am certain that the investigators were delighted. If the announced topline results are confirmed, they will have demonstrated that inflammation leads to new coronary events. Amazing!

But was Novartis happy?

I forgot to mention one thing about Ilaris. It is expensive. Not a little expensive. Think outrageously expensive.

Remember PCSK9 inhibitors? These drugs markedly lower serum cholesterol. In the FOURIER trial, evolocumab (Repatha) reduced the risk of cardiovascular death, nonfatal myocardial infarction and nonfatal stroke by 15%. Did the sponsor of the trial celebrate? Not after the payers pushed back on the price. Repatha costs about $14,000 per year for one patient. According to the payers, the magnitude of benefit on the primary endpoint was disappointing, and the drug did not reduce cardiovascular death by itself. Their conclusion: it may not be worth $14,000 per year.

Now it seems that both Ilaris and Repatha reduce the risk of major cardiovascular events. Repatha was not warmly welcomed because it cost $14,000 a year.

But Ilaris does not cost $14,000 a year. It costs $16,000 per injection! And for its current indications, the injections are commonly given monthly, so the annual cost is about $200,000. In CANTOS, the drug was given quarterly, and thus, it could cost $64,000 per year.

If Novartis charges $64,000 annually for Ilaris, payers might expect it to reduce cardiovascular events in a major way and to decrease the risk of cardiovascular death. But if the drug did any of these things, the CANTOS trial would probably have been stopped early -- and it wasn't.

My prediction: Ilaris may cost $64,000 for a 15-20% reduction in the risk of a major cardiovascular event, without decreasing cardiovascular death by itself.

Is it worth it? If there was push back from payers for Repatha, I can just imagine what will happen if Ilaris receives a cardiovascular indication.

Remember that Novartis has had trouble selling Entresto for a mere $4500 per year. And that drug actually reduced the risk of cardiovascular death. (I know. I was one of the two principal investigators for the landmark trial). Ilaris is much more expensive and may have fewer benefits.

Is Novartis celebrating right now? I do not know, but I would not make any assumptions. It is one thing to have a mild headache; it is another to have migraine. Remember: be careful what you wish for; you may get it.

Disclosures

Packer has recently consulted for Amgen, Boehringer Ingelhim, Cardiorentis and Sanofi. He was one of the two co-principal investigators for the PARADIGM-HF trial (sacubitril/valsartan) and currently chairs the Executive Committee for the EMPEROR trial program (empagliflozin).