Column: In the pipeline

Derek Lowe wonders what lessons we can learn from the Vytorin fiasco

The controversy over cholesterol drug Vytorin has been going on for months, and more fallout is surely on the way. Surveying the wreckage, my first thought is: what a terrible shame the whole affair is. After all, the drug is a very reasonable idea for a combination therapy. It makes perfect sense that curbing endogenous cholesterol synthesis with a statin (simvastatin) while reducing dietary cholesterol absorption (with ezetimibe) would do more to lower low-density lipoprotein (LDL) levels than either drug on its own. 

And so it does. Numerous studies have confirmed that Vytorin lowers LDL quite thoroughly, even when the patients take less than the stand-alone dose of the statin component. 

Based on those numbers, Merck and Schering-Plough have promoted their combination drug very aggressively, hoping to take market share in one of the most lucrative therapeutic areas in the world. They’re businesses, of course, and that’s what businesses do. Rationally, you’d expect that they would make money and their patients would, in return, get lower LDL levels. 

All the more reason for dismay, then, when the Enhance (Ezetimibe and Simvastatin in Hypercholesterolaemia Enhances Atherosclerosis Regression) trial (conducted in patients with genetically high cholesterol levels) apparently showed that Vytorin offered no benefit at all compared to the plain statin, when assessed by the generally accepted clinical marker of artery wall thickness. The drug did no harm, but it appeared to do no added good, either - which, given its price, is most certainly not ideal. The full results were released on 30 March at a meeting of the American College of Cardiology in Chicago, US, and also published by the New England Journal of Medicine. 

Makes no sense at all 

I should mention that ezetimibe was discovered right around the corner from my first lab in the drug industry, and I worked alongside its medicinal chemistry team every day during the compound’s early years. I have an emotional and intellectual connection to it, as it was the first drug I ever saw emerge in front of my own eyes, made by people I knew. 

So when I say that the Enhance result makes no sense, it might be tempting to put that down to my own biases. But the clinicians think the same thing, and the companies wouldn’t have run the trial at all in this patient population if they hadn’t seen good chances for a result to brag about. An explanation is needed: this result is telling us something important about human cholesterol handling and our approaches to testing it. It’s a pity the message is so unintelligible. 

That explanation won’t do much to help Schering-Plough and Merck, though, even if it ends up vindicating the drug. Its commercial prospects have surely been damaged beyond repair. There is a larger trial underway to evaluate outcomes with Vytorin in normal patients, but we won’t see that data until around 2011. Patients will benefit if the drug comes back into favour as a result, but the companies will have lost the compound’s most profitable years, and will have to watch it eventually turn generic while never having fulfilled its economic potential for them. 

Perfect example 

In hindsight, Vytorin illustrates the bind that the drug industry is in. Blockbuster drugs come along rarely, so companies need to make the most of them while they can. For most therapeutic areas, that means using surrogate markers, because the numbers that are most useful take the most time (and the most money) to obtain. LDL blood levels, for all the publicity surrounded them, are merely a surrogate. Artery wall thickness is another, albeit one step closer to the goal of slowing atherosclerosis. The real outcome is lowered rates of cardiac events, and the ultimate measures are lower mortality and lower morbidity. 

You’d think that all of these would follow in some sort of logical progression, but the truth is that we’re not completely sure about any of those links. Some of them are firmer than others; but the further you are from real mortality numbers (which take years, if not decades to obtain), the less sure you are. That fact means there is a terrible mismatch with the needs of drug development and the demands of the market which, in the end, are the demands of the general public and their physicians. Human beings will not put up with a 25 year drug trial, whether they are human patients, human cardiologists, or human stockholders. That means that the Vytorin fiasco may not be as much of a surprise as it seems. The odds, in fact, are very good that we’re going to do just this sort of thing again and again. 

Derek Lowe is a medicinal chemist working on preclinical drug discovery in the US.