Did Amgen's Repatha cut CV risks enough to make it cost-effective? Analysts say no

WASHINGTON, D.C.—Get ready to negotiate, payers: Amgen has its PCSK9 outcomes data, and it has some new pay-for-performance deals to offer. But, despite some impressive-sounding numbers, the data may not be enough to persuade payers to change their coverage much, analysts said.

Results from the long-awaited Fourier trial, designed to test Amgen’s Repatha as a tool to cut cardiovascular risks, showed that the drug cut heart attack risks by 27% over the course of the study, which followed patients for an average of two years. It cut the risk of stroke by 21%.

Put those two stats together, and Repatha patients saw a 19% reduction in risk for the first year and 33% after that, SVP of global development Elliott Levy said in an interview. And now, Amgen plans to offer a money-back guarantee to payers in hopes of revving up the lagging lauch.

But the drug's performance on the primary endpoint, a composite measure, came in at a 15% reduction, and it didn't deliver a decrease specifically in cardiac death risk. Those important figures triggered skepticism among analysts, pricing experts and investors. After the data's presentation at the American College of Cardiology meeting Friday, Amgen's shares dropped by more than 6%.

Peter Bach, for instance, director of the Center for Health Policy and Outcomes at Memorial Sloan-Kettering, emphasized that the number of CV events prevented would be too small for payers to save much money.

Amgen believes the numbers are strong enough to change clinical practice, though. These results are likely to make their way into all-important treatment guidelines published by medical societies, Levy said: Preventing “1 in 5 or 1 in 4 strokes and heart attacks, with no offsetting safety concerns,” should be plenty to turn Repatha into the standard of care for the type of high-risk patients Fourier enrolled.

Amgen also figures the numbers are impressive enough to make previously reluctant payers take notice, and the company is planning some new proposals, including that refund offer, that will put the Fourier data to work in payer negotiations.

Here's why analysts are skeptical, however. Repatha's 15% risk reduction on the primary endpoint—a composite that included cardiac death, hospitalization and revascularization procedures—means that preventing a CV "event" will be costly, Bernstein analyst Ronny Gal said in a Friday note. 

"At a cost of $7K/year, this translates to $958K per event saved," Gal wrote. "Thus, in our view, payers will continue to restrict access to the drug."

Barclays analysts weren't impressed, either. "We view this news as negative, with the results unlikely to drive significant expansion of the PCSK9 class following heavy restrictions among payers who believed the increased cost relative to current care standards (i.e., statins) not [commensurate] with the benefit," the analysts wrote in a Friday note.

Amgen will be doing its best to prove the analysts wrong by striking new payer deals. If a patient has a heart attack or stroke while taking Repatha as prescribed, Amgen will refund the cost of treatment. The company already has some pay-for-performance deals with payers—including arrangements with Harvard Pilgrim and Cigna—that link the net price of Repatha to expected LDL cholesterol reductions.

RELATED: Most U.S. payers are eyeing outcomes-based drug pricing: Report

So far, PCSK9 drugs, which are monoclonal antibodies that inhibit proprotein convertase subtilisin–kexin type 9, haven’t made the splash in the market that pharma watchers had expected and their makers had hoped. Pharmacy benefit managers and other insurers have put up hurdles to PCSK9 drug use, despite their ability to lower LDL levels dramatically. The payers have been citing a lack of data quantifying their effects on cardiovascular risks. Symphony Health data shows that commercial insurers rejected almost three-fourths of PCSK9 requests last year, and Medicare coverage was denied at a rate of 61%.

Now, payers no longer have that excuse. “They asked for outcomes data and we gave them outcomes data,” an Amgen spokeswoman said.

RELATED: As Repatha beats back atherosclerosis, Amgen chides payers for lack of PCSK9 coverage

The Fourier trial was the first PCSK9 outcomes study to wrap up—a similar trial of Sanofi and Regeneron’s rival drug Praluent is due to finish at the end of this year—and, perhaps more importantly to the broader field of CV research, the first outcomes trial to test the power and safety of lowering LDL to previously unheard-of levels.

The upshot? The data showed that the lower LDL goes, the lower CV risks go, and extremely low LDL levels aren’t unsafe. Some cardiologists and researchers had worried that very low LDL might have deleterious effects, just as excessively low blood pressure can.

“The trial really validates the LDL hypothesis and it proves that lower is better,” Levy said, adding that Fourier “answers the extremely important longstanding question—how low should it go? It shows in high risk patients, there’s no number. We really should be trying to get it as low as possible.”

As for safety, “patients who achieved the lowest levels of LDL, the lowest quartile, saw levels of 19 or less,” Levy said, “and the safety profile in those patients was also pristine.”

RELATED: Is the PCSK9 patent fight giving Amgen's Repatha a boost? Script numbers say so

The Fourier trial tested Repatha in patients at high risk of a cardiovascular event, a group in which Amgen sees a high unmet need, with no real advances in treatment in more than a decade. These patients came into the trial with established atherosclerotic CV disease or peripheral artery disease, or having had a heart attack or stroke. Their LDL levels at baseline were 92 mg/dl, and they were already “optimally treated” on all preventive therapies, Levy said.

For patients in the Repatha arm, LDL levels fell to a median of 30 mg/dl. Their risks, as measured by a composite endpoint known as MACE that includes CV death in addition to myocardial infarction and stroke, dropped by 20% compared with placebo.

The larger reductions in heart attacks and strokes drove the drug’s performance on the MACE measure, and on the primary endpoint, which comprised the MACE numbers as well as hospitalization for unstable angina and coronary revascularization procedures, Levy said. 

The drug had no effect on CV death when measured separately—the heart attack and stroke numbers included fatal episodes—and no effect on angina hospitalizations, and a 22% reduction in risk for revascularization procedures.

Safety-wise, Repatha patients didn’t see some of the side effects that other lipid therapies produce, he said. Muscular issues and liver safety concerns, such as those seen with statin drugs, didn’t crop up, and Repatha patients saw “no meaningful increase” in the rate of new-onset diabetes or neurocognitive effects. Injection site reactions were “rare” in both arms of the study, but slightly higher with Repatha patients, Levy said. 

Cardiologists have long been taught that CV patients have three risk factors that can be handled with treatment: high blood pressure, smoking and high LDL cholesterol, Levy said. Blood pressure has to be treated to a “normal range,” not too high or low, while smoking is toxic and should be stopped altogether.

Till now, with LDL, “no one was sure whether it was more like blood pressure and needed to be in range, or whether we want to eliminate it like tobacco,” Levy said. “This trial proves that it’s a poison, like tobacco.”

Editor's note: This story was updated with comments from experts and analysts.