Zetia: still crazy after all these years

The role of cholesterol in coronary heart disease has been known for decades, and the discovery of drugs that lower cholesterol has prevented millions of heart attacks and strokes, and saved countless lives.  The primary class of drugs—HMG-CoA reductase inhibitors, or “statins”—have been particularly effective, and their benefit seems to extend beyond their effect on cholesterol levels.

This “added benefit” of statins has been a bit frustrating for drug researchers.  Isolating this non-cholesterol-lowering effect has been an elusive goal.  Rather than search for a biological mystery, drug companies have invested resources in finding other drugs that lower cholesterol.  Some of these early compounds, such as fibric acid derivatives, have been shown to be effective at lowering heart disease mortality, but they are not as potent as statins, and are more difficult to tolerate (some are powders or are available as very large pills, for example).

In the first decade of the 21st century, an new cholesterol-lowering compound was discovered.  Called ezetimibe (Zetia), it prevents absorption of cholesterol in the intestine, and is effective in lowering levels of LDL cholesterol.  But it doesn’t seem to prevent heart attacks or deaths.   If lowering cholesterol doesn’t improve important outcomes, then cholesterol is just a number.

We know that statins lower cholesterol, heart attacks, and death, and this reduction is proportional to cholesterol level.  Apparently, Zetia, which effectively lowers cholesterol, doesn’t have the sine qua non that gives statins the ability to save lives.

Zetia’s manufacturer, Merck, came up with a very clever marketing strategy.  In addition to marketing ezetimibeto lower cholesterol, they marketed another drug, Vytorin, which contains both ezetimibe and simvastatin, a statin with proven efficacy.  Vytorin was supposed to give that little extra something, that little extra bump needed to help patients achieve their cholesterol goals.  But Merck never showed that Zetia had the same magic as statins, and it wasn’t clear that Vytorin had any advantage over simvastatin alone (except to Merck: Vytorin costs about $130 per month, simvstatin only $10).

Merck has invested a lot in ezetimibe, and is still trying to find a clinical home for it.  This year The Lancet published a new ezetimibe study.  This one looked at patients with chronic kidney disease (CKD).  CKD patients have a higher risk for heart attack, although the cause isn’t completely clear.  Neither is this study.  Rather than picking the most important outcomes—heart attack and death—they chose “major athersclerotic events”, a much broader category.

Of course they found that Zetia reduces this outcome in these patients, but not by much.  They failed to find a significant reduction in cardiac death, heart attacks, or death (despite the linguistic acrobatics in the Discussion section).  If the failure to prevent important outcomes wasn’t enough, the study design had a basic, fundamental flaw:  rather than comparing Vytorin (or in this case two tablets, Zetia + simvastatin) to a statin alone, they compared it to placebo.  There is no way of separating out the effect of the Zetia and the simvastatin.  The only conclusion that can be drawn from the study is that using a statin + ezetimibe improved outcomes.  For all we know, these outcomes would have been just as good without the ezetimibe.

The study design seems to me to be completely dishonest, perhaps unethical.  Their justification for the design was curious:

To achieve an average reduction in LDL cholesterol of about 1 mmol/L without the use of high statin doses (which are associated with an increased risk of myopathy, especially in patients with impaired renal function), a low dose of a statin (simvastatin 20 mg daily) was combined with a cholesterol-absorption inhibitor (ezetimibe 10 mg daily).

So they are using the ezetimibe under the assumption that these patients would either not benefit from a low-dose statin, or would have side-effects on a higher dose statin.  Curious indeed.

More curious, though, is the fact that the study began with three groups: simvastatin, simva+ezetimibe, and placebo:

Patients entering SHARP were initially randomised three ways between simvastatin 20 mg plus ezetimibe 10 mg daily,simvastatin 20 mg daily, and placebo to assess the safety of adding ezetimibe to simvastatin during the first year (with no safety concerns identified and those initially allocated simvastatin alone were then rerandomised to simvastatin 20 mg plus ezetimibe 10 mg daily versus placebo after 1 year…

Wait…what?  What happened to the statin-only group?  Why did they drop it? Why don’t we get to see those results?

I can see no raison d’etre for this study other than as a marketing tool for ezetimibe.


Scandinavian simvastatin survival study group, . (1994). Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S) The Lancet, 344 (8934) DOI: 10.1016/S0140-6736(94)90566-5

Baigent, C., Landray, M., Reith, C., et al. (2011). The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial The Lancet, 377 (9784), 2181-2192 DOI: 10.1016/S0140-6736(11)60739-3

Previous Post


  1. D. C. Sessions

     /  November 16, 2011

    There are times you really wish there were a shock buzzer button to communicate with the authors of egregious papers. I may feel guilty contemplating the idea, but guilty pleasures are pleasures none the less.

  2. Dianne

     /  November 19, 2011

    A few thoughts:
    1. This sort of thing is why I’ve always heard people say that when attempting to collaborate with drug companies for clinical trials one must get everything in writing in advance and be willing to drop it all at any phase. Because the company will want to do only the study that makes their drug look good.

    2. What was Lancet thinking publishing this? It’s not a well designed or important trial. My conclusion from it is “there is no useful data there”.

    3. What was the IRB thinking? Given what we know about cholesterol, heart disease, and statins, is a placebo control really ethical in this situation? I don’t feel it is. Apparently someone disagreed.

    • The argument made was that statins haven’t been found to be as effective in CKD patients, therefore it was “ethical”.

      • Dianne

         /  November 19, 2011

        Hmm…maybe…but then it really should have been a 4 arm trial (placebo, each drug alone, combination). Also, there must be post-marketing data on the use of statins in people with CKD-a lot of people must be using it off label. Why not look at that for preliminary data, including safety data?

  3. There is an ongoing trial, the IMPROVE-IT trial, comparing 40 mg simvastatin/10 ezetimibe with 40 mg simva/placebo in patients with acute coronary syndrome, but the results won’t be available for a couple of years. Clearly, Merck/Schering Plough delayed starting an outcomes trial because they had such a good marketing thing going.

    Many people are saying that lipid-modifying drugs should not be approved solely on the basis of their effect on biomarkers. For example, see this commentary:

    Allison B. Goldfine, M.D., Sanjay Kaul, M.D., and William R. Hiatt, M.D. Fibrates in the Treatment of Dyslipidemias — Time for a Reassessment
    NEJM 2011; 365:481-484.

    http://clinicaltrials.gov/ct2/show/NCT00202878 (IMPROVE-IT trial record)

    There is a 2009 Cochrane Review on statins for people with chronic kidney disease not requiring dialysis. http://summaries.cochrane.org/CD007784/statins-cause-a-significant-reduction-in-the-risk-of-all-cause-and-cardiovascular-mortality-in-ckd-patients-who-are-not-requiring-dialysis
    This is based on post-hoc analyses of statin trials. At the time the SHARP trial was begun in 2003, I’m not sure if that kind of analysis had been done.

    There were two negative trials of statins in patients on dialysis, AURORA and 4D, and the benefit in the dialysis subgroup in SHARP was much smaller than in the pre-dialysis subgroup. The FDA advisory committee voted to approve the new indication in pre-dialysis patients with CKD, but not in patients on dialysis. We’ll see what the FDA does.

  4. Granted, Zetia has not been shown to reduce the risk of hard outcomes (MI, stroke)…

    But, statins, when meta-analyzed for primary prevention, also fail to show benefit for hard outcomes (even in ‘high risk’ patients).

    What’s your take on using Zetia and statins for primary prevention? If there’s a place for statins, then, surely there’s a place for Zetia, no?

  5. @Michael I believe you are referring to the meta-analysis that was published in Archives of Internal Medicine. That meta-analysis looked only at total mortality. In that meta-analysis, total mortality was 9% lower with statins but the difference was not statistically significant. However, it is very well established that statins reduce the risk of heart attacks and strokes in both primary and secondary prevention and the risk of total mortality in secondary prevention.

    Ezetimibe, in contrast, has never been shown to reduce the risk of heart attacks, strokes and death in any population whatsoever. That is being tested in the IMPROVE-IT trial.

    @PALMD My comment from yesterday is still awaiting moderation. Is it stuck in your spam filter?

%d bloggers like this: