Vytorin is a combination of two different types of cholesterol-lowering drugs: Zetia, which blocks cholesterol absorption in the gut, and Zocor, one of the statin drugs that slows cholesterol production. Other statins include Crestor, Lescol, Lipitor, Mevacor, and Pravachol.
As expected, patients who took Vytorin had lower levels of bad LDL cholesterol than did patients who took Zocor alone. But Vytorin patients did not have any less plaque in their arteries than those who took Zocor. In fact, they had slightly more.
That mystery is troubling to the FDA. Because lower LDL cholesterol is strongly linked to lower risk of heart attacks and stroke, the FDA approves drugs that can safely lower cholesterol. Even so, drugmakers must prove their products actually prevent heart disease before they can say this on the drugs' labels.
So why didn't Vytorin's superior ability to cut LDL cholesterol translate into superior ability to cut plaque? That's what the FDA wants to know, John Jenkins, MD, director of the FDA's Office of New Drugs, said in a news conference.
"We have not received the final study report and cannot explain why the lower LDL did not lead to lower amounts of plaque," Jenkins said. "Once we fully review the data, we may consider whether any further action is required, and whether this has any effect on our way of approving cholesterol-lowering drugs."
Jenkins quickly added that the FDA does not expect to change its longstanding policy of approving drugs that safely lower LDL cholesterol. All of the statin drugs were initially approved on this basis. All but Crestor -- a newer statin that is completing such studies -- eventually were shown to reduce the risk of heart attack and stroke.
Jenkins said it would take the drug companies Merck and Schering-Plough -- which jointly market Vytorin -- two or three months to get the study data to the FDA. After that, he said, it will take the FDA up to six months to review the data.
"When we talk about the full study report, we are talking about thousands of pages of documents and analyses. This is not a three- or four-page journal article. ... So if we get it in a couple of months, we'll be done in no more than six months, and it may well be sooner."
Reporters pressed Jenkins on why, since the study was completed in April 2006, it is taking so long to get the data to the FDA -- and why preliminary results were reported only last week.
"As a general rule, after the last patient visits the clinic and the trial period is over, a lot of work goes on to collect information," Jenkins said. "In this case, a central reading committee had to read the plaque images. That can take some time. Also, there is a question of the priority of how the company assigns its resources to get the work done. It is not unusual for it to take months, or even longer than months, to do all the data cleanup you need to say you are ready."