The speech that killed off Schering-Plough

Below are the prepared remarks made by Dr. Krumholz at the ACC last Sunday that led to Schering-Plough stock dropping 25% on Monday and another 6% later in the week.

Fred Hassan, Schering-Plough CEO said “we were very disappointed with the way ACC unfolded. We hoped to see an open and balanced scientific discussion, and this did not happen.”

Dr. Krumholz responded to Hassan's criticism by making his speech available to the public, and also issued this comment:

“We were asked to comment on the Enhance study and discovered over the course of several weeks of discussion that we had a clear consensus about the recommendations for practice. My comments reflected that consensus - and were consistent with two editorials in the New England Journal of Medicine - and with the published guidelines. The Enhance study was not definitive, but increased uncertainty about the ability of ezetimibe to improve patient outcomes."

Here is the historic speech that killed off Schering-Plough:

Sometimes negative studies can provide very important insights. You have just seen a negative trial that should change practice, especially given the way that we, in this country, have been prescribing ezetimibe.

There are several things to notice about this study and Dr. Kastelein’s presentation. It seems to be a strong study for one focusing on a surrogate outcome of artery thickness. In the ASAP study, his earlier study of a similar population, such an approach showed that more intensive lipid lowering with atorvastatin was associated with less progression of atherosclerosis. In a pediatric population, another Kastelein study, this approach showed that pravastatin compared with placebo was associated with less progression of atherosclerosis. In this study, the addition of ezetimibe to a statin did not slow the progression of atherosclerosis. And this was true for various subgroups, including those who were statin naïve and those with stratified by baseline artery wall thickness.

Critics may opine about reasons for the findings including the possibility the measures were not precise enough or the population was not typical – but the most likely explanation is that the compound did not work. It lowered LDL, but did not retard the progression of atherosclerosis as we saw in prior studies where the use of statin therapy or intensive statin therapy had this effect. Now this is still just one study of ezetimibe and one that employed measurements of arteries and not clinical endpoints, but this study provides no new evidence to support the use of the drug. And it moves us to more uncertainty about the benefit of the drug.

What about the presumption that knowing that a drug lowers LDL is enough to know its effect on our patients. Dr. Kastelein’s presentation also has a message that is important in this respect. In his 5th slide Dr. Kastelein said that ezetimibe was known to reduce LDL-c when added to a statin but that the effect on the progression of atherosclerosis was not known. This is a very important slide. This study was developed by experts.

Whether the incremental lowering of LDL by this drug had an effect on the progression of atherosclerosis was an open question – worthy of study. As is the question of what effect the drug has on our patients – it is an open question – worthy of study. We do not know.

Remember that this is a new drug – with a novel mechanism – first in class – we do not have outcomes studies.

But you might say, but the drug lowers LDL, isn’t that a good thing? Shouldn’t that be associated with benefit? Wasn’t that enough for the FDA?

Well, although LDL is an important risk factor for cardiovascular disease, we have already refuted the assumption that just because a drug reduces LDL it must improve patient outcomes. Hormone replacement therapy reduces LDL and is not associated with cardiovascular benefit. Torceptrapib, the Pfizer drug, lowered LDL and raised HDL, but did not improve outcomes and never made it to FDA approval.

Drugs are complex compounds with an array of biological effects – knowing how they affect lipids does not tell us how they affect people. This observation does not unravel the lipid hypothesis, but says it may matter how we lower cholesterol. The drugs that we use make a difference in the net effect on people.

With knowledge of its effect on LDL but no information about its effect on the progression of atherosclerosis or patient outcomes, how have we used this drug? We, in the US, have embraced this medication. Amid an aggressive marketing campaign this drug has been rapidly adopted into practice and become one of our favorite options for patients with hypercholesterolemia. Within 6 years of its introduction, prescriptions for Zetia and Vytorin in this country skyrocketed. In a paper published online today by the NEJM, Jackevicius and colleagues show that by 2006, four years after its introduction, 15% of US prescriptions for LLAs included ezetimibe. In Canada, the rate was only 3%. In the US ezetimibe supplanted statins to some extent and led to use of lower doses of statins.

And the cost was great. In 2006 we in the US spent billions of dollars on this drug. Perhaps $1.5-2.0 billion more was spent than would have been had our pattern been more like the Canadians.

So what do we know now about the effects of this drug on people’s health?

There are three possibilities with this drug. Eventually – one day – when outcomes studies are finally done — we may recognize that it is an effective medication for reducing cardiovascular risk – the ENHANCE study makes that less likely – but it is not impossible.

Or it could be that ezetimibe is just an expensive placebo and its principal harm is that it drains precious resources from our health care system – and maybe leads people to use less of the drugs that have been shown to be beneficial. The ENHANCE study suggests that this may be true.

Third, it could be harmful. Honestly, we do not know enough about the clinical risks of this drug. It is well tolerated and there are no obvious safety problems, but we cannot say if there is an increased risk of AMI or death or another important health problem.

Important clinical risks like this can be imperceptible in clinical practice; we need large clinical studies to tell us about them. We learned that with encainide and flecanide – and hormone replacement therapy – and many others.

For this novel medication for which there are some plausible biological mechanisms that could link it with harm – and see the NEJM editorial by Taylor and Brown for some possibilities – it is important to know more about safety. No one can tell you with certainty that they know about the clinical risks of this drug. We just do not know. And ENHANCE does help us in this regard.

So where does this leave us? ENHANCE is an important negative study that provides no new support for a widely prescribed drug and whose surprising findings remind us how little we know about the overall risks and benefits of this drug – whether there is really a net clinical benefit to its lipid lowering effect.

For clinicians who may have employed this medication before exhausting options with statins, the strongest recommendation here is to turn back to statins, especially those with favorable outcomes data. Go back to what we know works. Let us stay with the evidence. Patients who need medication to treat cholesterol should be maximized on statin therapy – and different statins may need to be tried before a patient is considered to have failed statin therapy.

Then, the next options should be medications that have been shown to be associated with better clinical outcomes – niacin, fibrates and resins. We know that they are not tolerated as well, but they have evidence and are worth trying.

And then for those who have failed these therapies, and this should be a relative small group, the question of whether we should use ezetimibe will likely be unresolved until the outcomes studies are available. Until then we will not know the net effect of this drug on patients – and whether the reduction of LDL with this drug produces a clinical meaningful effect. And we will not really know the strategy that is in the best interests of our patients. It is an unfortunate predicament.

This study heralds the need for clinical research to guide us in decisions for our patients – and ideally this work needs to be done early in the drug’s development. It is not right that we are this far down the line with this drug and we have so much uncertainty about its balance of risks and benefits. We need to understand the effect of new drugs on people. And that relying on a drug’s effect on a set of lab tests may not tell the whole story. We have learned this lesson before – it appears that we need to learn it again.