By Dr Malcolm Kendrick
Last month two new cholesterol lowering drugs stepped into the limelight at AHA’s (American Heart Association) annual conference. Neither is likely to make it to market until the end of next year but soon after that they are likely to be coming to a doctor’s surgery near you. Questions are already being asked about their effectiveness, what benefits they will have to show and their side effects.
Of immediate interest is the light they throw on this summer’s statins silly season when senior researcher Sir Rory Collins rode out to attack the BMJ for publishing articles that disagreed with his view that statins were very safe and effective and should be prescribed to millions more. (See previous articles here in the “Statin” category for more details).
At the time it seemed a wild over-reaction but now with two new hot, patented and very expensive cholesterol lowering drug on the way (the patents on statins have all now run out) vigorous action to protect this most lucrative of brands, can be seen as making very good sense.
A major disaster looms
The promotion of evolocumab and alirocumab, extra powerful cholesterol-lowering drugs that are expected to earn billions, dominated the conference. The official AHA app, packed with ads, was provided courtesy of one of them. Both work by blocking an enzyme in the liver known as PCSK9, which has the effect of speeding up the rate the rate “bad” LDL cholesterol is pulled out of the blood stream.
One reason for the massive PR campaign was a possible major threat from the findings of a long-running trial, due to be presented at the conference. Called IMPROVE-IT, there was a chance that could seriously threaten the cholesterol hypothesis. This is the widely held belief that lowering cholesterol automatically cuts the risk of heart disease.
Statins had been licenced on this basis and the key claim of the new drugs is that they lower cholesterol even more effectively. If IMPROVE-IT failed to support the hypothesis, the launch of the new block-busters would be in jeopardy. What happened is revealed in the post by Dr Malcolm Kendrick below.
When a drug class comes off patent, as statins have, the PR for the next generation usually stresses the failure of the older drugs. The new PSCK9 blockers are no exception. The result makes the claims by statin promoters in the summer that statins were virtually side-effect free look widely optimistic.
PR shock: statins side effects affect 50% of patients
A major benefit of the new drugs, it’s claimed, is that they can help the large number of people with high cholesterol who are statin intolerant (suffer serious side effects). Just how large?
Dr Patrick Moriarty director of clinical pharmacology at the University of Kansas Medical Center told the journal Cardiovascular Business last month that as many as half of the patients who present at his clinic are statin intolerant.
‘In other studies, 10 percent to 25 percent of patients report being statin intolerant, making the need for an alternative treatment acute,’ he said. When expert estimates collide in this way, who is poor consumer to believe?
The PCSK9 drugs, however, come with their own side-effects. In a recent trial 52% of those on a lower dose of evolocumab reported side effects which included a cold-like infection, back pain and cystitis. The rise in infections fits with the fact that cholesterol plays a role in the immune system. Experts have already pointed out that driving levels down even further than can be achieved with statins could make patients more vulnerable to infections.
But can these drugs do anything more than just drive down cholesterol? The saga that lead up to the vital IMPROVE-IT trial that tested the benefits of a drug called Ezetimibe, explains why scepticism about the benefits of just lowering cholesterol makes a lot of sense.
Ezetimibe – Tale of the ghost at the statin feast
When Jerome Burne asked me to write a few words about Ezetimibe I foolishly I agreed, thinking it was a simple matter. A few paragraphs and that would be that. However, it soon became clear that you cannot understand anything about Ezetimibe, or the new PCSK9 blockers that are set to succeed it, without understanding the complicated spider’s web of marketing and research that surrounds this drug.
So I shall start at the very beginning, nicking a short passage from Wikipedia to explain what Ezetimibe is and does:
‘Ezetimibe inhibits the absorption of cholesterol from the small intestine and decreases the amount of cholesterol normally available to liver cells, leading them to absorb more from circulation and thus lowering levels of circulating cholesterol. The exact mechanism is not known…’ [Wikipedia]
So because raised cholesterol is regarded as a cause of heart disease and Ezetimibe lowers it, it’s assumed to prevent cardiovascular disease (CVD). But the story quickly gets more complicated. This is what Wikipedia says about another drug called Cholestyramine that acts in a very similar way to Ezetimibe:
‘Cholestyramine removes bile acids from the body by forming insoluble complexes with bile acids in the intestine, which are then excreted in the feces. As a result of this loss of bile acids, more plasma cholesterol is converted to bile acids in the liver to normalize levels. This conversion of cholesterol into bile acids lowers plasma cholesterol levels…..’ [Ref: Wikipedia]
Ezetimibe’s daddy doesn’t work
In other words both Ezetimibe and Cholestyramine block cholesterol absorption from the small intestine. Cholestyramine was being used over fifty years ago to treat high levels of blood cholesterol but it’s no longer used because it has “not proved useful to lower the risk of heart attack or stroke.” [Reference]
As a result of failure of cholesterol lowering Cholestyramine to cut the risk of a heart attack distinguished researchers such as Professors Michael Oliver and John McMichael were still vociferous in their opposition to the lipid (cholesterol) hypothesis,in the mid 1980s.
But then cameth the statins… Finally, in very large scale studies, cholesterol was lowered and the risk of CVD was also lowered, huzzah! Take that you damn doubting Thomases. Professor Michael Oliver admitted that he was wrong, opposition crumbled. The cholesterol hypothesis was now a proven fact. Raised cholesterol (raised LDL actually) causes CVD and lowering it protected against CVD.
So began a Brave New World where cholesterol lowering was the new marketing holy grail and the search was on for new drugs to lower cholesterol and out they tumbled from the laboratories: fluvastatin, pravastatin, simvastatin, cerivastatin (sorry withdrawn for killing people), atorvastatin, rosuvastatin, etc.
Cholesterol hypothesis starts to wobble
But then at the start of the millennium Ezetimibe appeared. As son of Cholestyramine it proved to be a spectre at the statin feast. Why, I wondered at the time, should it show any benefit? After all it worked in almost the same way and lowered cholesterol/LDL by about the same amount. Why would Ezetimibe work when Cholestyramine had not?
However, no one else seemed concerned; the fact it lowered cholesterol was thought to be enough. After it had been licenced a trial called ENHANCE was set up to discover if it actually did reduce the risk of heart attacks and strokes.
Rather than measuring such mundane things as heart attacks and deaths, it measured how ‘thick’ the patient’s artery walls were – otherwise known as theCarotid Intima-Media Thickness (CIMT). Thicker means worse. Rapid progression of CIMT ‘thickness’ is thought to be closely related to risk of future CVD.
ENHANCE found adding Ezetimibe to a statin compared to a statin alone, increased the speed of progression of CIMT. Ooops. . Could lowering cholesterol not do good but harm? The entire cholesterol hypothesis began to wobble. Billions were at stake.
IMPROVE-IT delivers the goods?
Two imaginative moves saved the day. First came the suggestion that statins protected against CVD through mechanisms other than cholesterol lowering. Perhaps by reducing inflammation and ‘plaque stablisation’
Then, more recently, the American Heart Association/American College of Cardiology guidelines came out, suggesting a fire and forget approach to statins. No need to know what was happening to cholesterol levels, just prescribing a statin was enough. Interesting…wobble, wobble.
Which is where the IMPROVE-IT study comes in. It was set up to ‘prove’ that Ezetimibe, in addition to a statin, did indeed reduce the risk of CVD. It probably won’t come as a surprise to learn that IMPROVE-IT came up with the goods. The business magazine Forbes declared that it: ‘…demonstrates real but modest clinical benefit for Ezetimibe.’ Phew!
But it is hardly a ringing endorsement and there are enough shortcomings to the trial that you’d be forgiven for wondering if it proves anything at all.
Encouraging the trial to produce the right results
As the pharmaceutical columnist for Forbes pointed out the study had been going on for far longer than it was ever designed to. That was because it the researchers kept changing the way it was done to increase the chance it would produce a favourable result.
All trials have what are called primary end points – the things you want to show the drug can do. If Ezetimibe was an effective heart drug you’d expect it to reduce cardiovascular death, nonfatal MI, nonfatal stroke, rehospitalization for unstable angina (UA), and coronary revascularization. When it didn’t show an effect on enough of these, the researchers just increase the number of patients, from 10,000 to 18,000. The trial was designed so that it could spot a 9.3% reduction in these primary endpoints. In the end they only fell by 6%.
A trial that is not getting enough points to achieve statistical significance simply doubles the number of patients half way through? Then changed the statistical end-point they were looking for? Bong! Commenting on this study in Forbes, Professor Steven Nissen, whom I have criticized in the past, said that…
‘..the trial result was trustworthy and that the trial was well conducted, but he emphasized the modest effect and asked “when have you ever seen a risk reduction as small as 6%?” He noted that the trial was powered to detect a 9% difference but got only a 6% reduction.’ [Reference]
Time to start counting the spoons
Two comments here that are worthy of attention. Firstly, Niseen states that the result was ‘trustworthy.’ An interesting comment that. Why would he comment on the trustworthiness of the study? Well, as P.G. Wodehouse once said… ‘When an Englishman says ‘trust me’ it’s time to start counting the spoons.’ The other comment is that the trial was powered to detect a 9% different but got only a 6% reduction. Yet, this trial managed to meet its original end-point? Sorry, run me through that one again.
But even these figures give a too favourable an impression. There was just a 2% difference in the end point reduction among those who got the drug and those on a placebo. What that means is that 50 people would have to take the drug for six years to prevent just one of those primary endpoints. A modest benefit indeed.
But don’t worry about such trivial things about altering the number of patients in the trial and changing the end-points mid-way through, all is well, because Dr. Christopher P. Cannon, reporting on the trial stated that this is of no significance… ‘the trial “reaffirms the LDL hypothesis, that reducing LDL prevents cardiovascular events” and that the results should be incorporated in future guidelines.’ Which is code for…. The FDA should now approve PCSK9-inhbitors without delay
Cheers of relief from all those with their money riding on Alirocumab and Evolocumab. Two of the key companies about the launch PCSK9-inhbitors are Sanofi/Regeneron. So perhaps you would be interested to read Dr. Christopher Cannon’s conflict of interest statement:
Dr. Christopher P. Cannon discloses the following: Research grants/support from the following companies: Accumetrics, AstraZeneca, CSL Behring, Essentialis, GlaxoSmithKline, Merck, Regeneron, Sanofi, Takeda. Advisory Board (funds donated to charity): Alnylam, Bristol-Myers Squibb, Lipimedix, Pfizer. [Reference]
I cannot really say much more about the IMPROVE-IT study, because all we have to go on at present is an abstract that was presented at a conference. Normally the whole paper would be released. Forbes magazines asked one of the Chief investigators why it hadn’t been. ‘The reason is very simple: it was never our intention to submit it at the same time, even though that is what we usually do, because the time since unblinding was not long enough. We’ve been working on the presentation.’
Does that make you reassured that the cholesterol hypothesis is safe after all?