Amgen on Thursday afternoon that evolocumab (Repatha) had met both its primary composite endpoint (cardiovascular death, non-fatal MI, non-fatal stroke, hospitalization for unstable angina, or coronary revascularization) and the even more rigorous key secondary composite endpoint (cardiovascular death, non-fatal MI, or non-fatal stroke) in the FOURIER trial.
The company also announced that an important FOURIER substudy, the EBBINGHAUS cognitive function trial conducted in FOURIER patients, also achieved its primary endpoint, demonstrating that Repatha was non-inferior to placebo for the effect on cognitive function.
FOURIER is undoubtedly the most highly anticipated cardiovascular trial in several years. The full results of FOURIER are scheduled to be presented in March. It is the first large trial to assess cardiovascular outcomes with a PCSK9 inhibitor, the important and controversial new class of cholesterol-lowering drugs.
FOURIER Background
For many years, statins have been the standard of treatment for people with or at very high risk for cardiovascular disease. (There is wide agreement that they are beneficial for secondary prevention in people with established or known cardiovascular disease; there is no broad consensus on their use for primary prevention in people without known cardiovascular disease.)
Despite the success of the drugs it is widely known that statin-treated patients still have an elevated residual risk for CV disease.
Over the past decade it became clear that a new class of drugs, PCSK9 inhibitors, exerted an even more powerful LDL cholesterol-lowering effect than statins, leading many to believe that they could confer additional cardiovascular benefits when given on top of statins (or in people who have difficulty taking statins).
There are now two PCSK9 inhibitors on the market: evolocumab and alirocumab (Praluent manufactured by Sanofi and Regeneron). Both drugs are monoclonal antibodies and have a retail cost of nearly $15,000 a year (less with discounts negotiated by insurance companies). Both drugs attracted wide attention when they were first approved a few years ago, but they have not been widely used so far. Until now, the drugs have only been shown to lower the surrogate endpoint of LDL cholesterol without good evidence demonstrating the clinical effect of these drugs. Adoption of the drugs has been limited by narrow indications in guidelines and recommendations, physician reluctance and skepticism, and obstacles to reimbursement raised by insurance companies.
Now FOURIER is set to change the terms of the debate.
FOURIER Basics
FOURIER is the first large trial to rigorously assess cardiovascular outcomes of a PCSK9 inhibitor, in this case Amgen's evolocumab.
In it, 27,000 patients with cardiovascular disease and already taking statins were randomized to evolocumab or placebo. The trial ran until at least 1,630 patients had a cardiovascular death, MI, or stroke. The trial has a 90% power to detect at least a 15% relative reduction in the primary endpoint -- the composite of cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization.
The alirocumab outcomes trial, known as , is about a year behind FOURIER. Odyssey Outcomes is expected to enroll 18,600 patients.
FOURIER Timing
The main results of FOURIER are scheduled to be presented at by the TIMI Study Group's Marc Sabatine, of Brigham & Women's Hospital in Boston.
The next day, Saturday, results from an important FOURIER substudy, called EBBINGHAUS, will focus on neurocognitive adverse events in the trial. (See below for more on this topic.)
FOURIER Results & Predictions
Most experts I spoke with in recent weeks believed FOURIER would probably meet its primary endpoint, but they also said that this is a low standard. In order to make a significant dent in clinical practice and in the marketplace, the trial will need to demonstrate a dramatic benefit in hard endpoints. The Amgen announcement suggests that the trial met this standard, at least in part, since it also met the more rigorous secondary endpoint composed of the hard endpoints of CV death, non-fatal MI or non-fatal stroke.
However, it is unclear at this point whether the trial also resulted in a significant reduction in either CV death or all-cause mortality. The statistical analysis plan for the trial called for the CV death endpoint to be analyzed if the primary and the secondary endpoints were positive, and for all-cause mortality to be analyzed if the CV death endpoint was positive.
Without more details it is impossible to reach firm conclusions about safety, but the Amgen announcement does offer some reassurance in this regard.
The announcement about the neurocognitive study will likely put to rest one of the largest threats that had been looming over the field. had raised concerns in some quarters that this could be a threat.
The Amgen announcement made no mention of another issue of concern -- diabetes. Since the JUPITER study, it has been known that statins have been linked to a small increase in the risk of developing diabetes, though it is very unclear what this finding means in clinical practice. It is widely believed that the PCSK9 inhibitors will have a similar effect. FOURIER will provide the first good evidence on this important topic.
There have also been recent concerns, based on a very small number of events, that . Again, this topic was not mentioned in the press release.
Expert Opinions
I asked a number of knowledgeable experts how they were starting to think about FOURIER. Here are there responses, given prior to the Amgen announcement:
Sek Kathiresan (Massachusetts General Hospital, Broad Institute) provided the following calculations/predictions:
- My best guess is that the FOURIER trial will show that evolocumab reduces disease risk in agreement with the cholesterol treatment trialists (CTT) regression line.
- The CTT regression line, based on statin randomized controlled trials, suggests that for every 1 mmol/L (39 mg/dL) LDL lowering, relative risk reduction (RRR) for major vascular events is 22%.
- The baseline LDL in the FOURIER placebo arm is 91.5 mg/dL. Assuming a 50% reduction in LDL with evolocumab therapy, the LDL change will be ~46 mg/dL.
- If you round that to 1 mmol/L, then the RRR should be about 22%.
- From the study design paper, the predicted event for the primary endpoint was 4.5%. A 22% RRR with treatment would mean that the event rate in the treated group would be 3.5%. This would translate into an absolute risk reduction of 1% and a number needed to treat (NNT) of 100.
- The safety endpoints I would be looking out for are neurocognitive, cataract, and type 2 diabetes.
Marilyn Mann (patient advocate): "From my perspective as a patient advocate, I am very interested in seeing the degree of risk reduction in the primary endpoint, as well as the key secondary endpoint of cardiovascular death, MI, and stroke. I am also very interested to see the effect of evolocumab on the secondary endpoints of cardiovascular death and all-cause mortality. In the past, intensification of lipid-lowering therapy, for example by adding ezetimibe [Zetia] or switching from a moderate to high intensity statin, has reduced cardiovascular events but not all-cause mortality. Indeed, in the IMPROVE-IT trial, cardiovascular death was also not reduced by the addition of ezetimibe. I am also eager to learn more about safety, including with respect to neurocognitive events, new-onset diabetes, muscle symptoms, and any other safety issues that became evident during the trial."
Sanjay Kaul (Cedars-Sinai in Los Angeles) pointed out that FOURIER was not stopped prematurely by the Data and Safety Monitoring Board, "which suggests against a large treatment effect. So, I am guessing the trial is likely going to be positive but not persuasively." Kaul also raised the question whether the trial isn't substantially overpowered: "Shouldn't a treatment intervention that reduces LDL by 60%, be powered to detect a 30% reduction instead of 15% reduction in CV outcomes?"
"Why would a sponsor knowingly go for a 27,000-patient trial if it could get it done with 6-7,000 instead?"
"Perhaps," Kaul wondered, "they are hedging their bets against the lipid hypothesis!"
Later, he added, "PCSK9 inhibitors are truly a triumph of genomic medicine which in my opinion has over-promised but under-delivered so far. We should celebrate this victory but we also need to be asking critical questions about whether they represent a conceptual advance or a truly therapeutic advance. And cost and affordability (value) is just as important as science for these drugs to qualify as a true therapeutic advance."
Khurram Nassir (Baptist Health South Florida), who co-authored , also anticipated a relative risk reduction between 20% and 30%. This, he calculated, would result in a NNT to prevent one event in 5 years ranging from 33 to 50. Thus, although the trial will likely demonstrate that PCSK9 inhibition successfully reduces cardiovascular risk, "we have to be pragmatic in acknowledging some real concerns whether it would add value to our key health care system in balancing the benefits with the cost. To put it simply, is the drug cost (at current or even slightly discounted rates) a worthwhile investment from a societal and our mostly private payer's perspective?"
Nassir calculated that even if FOURIER was extremely positive, producing a relative risk reduction of 50%," the annual treatment price would need to be nearly $4,250 to be cost effective," based on the "societal willingness-to-pay limit of $100,000 for each quality adjusted life year gained." For the more likely 20% to 30% reduction in risk, the cost of the drug would "need to be much lower than $4K/year for it to be cost effective from a societal perspective."
Final Observation
A comment from Sek Kathiresan after the announcement sums up the thoughts of these and other experts I have consulted. These appear to represent a broad consensus of opinion:
- We need to wait for full safety/efficacy profile (particularly the signal emerging for cataract).
- The cost undoubtedly needs to come down.
- The biologic implication is likely to be that treating LDL to a very low level is helpful for secondary prevention. Now, we will have three ways to get a patient's LDL to a very low level: statin, ezetimibe, and PCSK9 inhibition.
- From a science perspective, a tremendous breakthrough: gene discovery in 2003 to positive cardiovascular outcomes trial in 2017.