Healthy Skepticism Library item: 940

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: news

Herper M.
Fuzzy Drug Company Math
Forbes 2004 Oct 5

Full text:

There are lies, damn lies and statistics—and then there are drug companies.

The math involved in new medicines can sometimes be heavily spun—or simply misstated during fast-moving debates about what works and what doesn’t. Two recent examples of drug company math could stand with a little more clarity.

The first has to do with the recall of the $2.5 billion arthritis drug Vioxx by Merck. The second relates to exactly which data tiny biotech AtheroGenics should be presenting—and it bears direct relevance to whether the drug company acted properly in disclosing clinical trial data for its experimental heart drug.

Merck’s Messy Numbers
In the rush to get the word out about Vioxx—which was pulled from the market Sept. 30 because the drug doubles the risk of heart attack and stroke—Merck seems to have made some errors. The first was to say that 84 million people have taken Vioxx. On Oct. 1 it said that it had been wrong.
Vioxx had been prescribed 100 million times in the U.S., but no numbers for the total number of patients who have taken the drug are available. In the U.S., Merck says, 20 million people used Vioxx at one time or another.

An even more important question: How many of the people who took Vioxx had heart attacks? The controlled clinical study that led Merck to pull the drug could help here. But the way the company described the relevant data was more than a little confusing.

On a conference call with analysts, Merck research chief Peter Kim said that at the end of the study, 15 people out of every 1,000 taking Vioxx had heart attacks or strokes after three years, compared with 7.5 out of every thousand taking a placebo. However, those appear to be the number of heart attacks that occurred per year.

But according to Eric Topol, chairman of cardiovascular medicine at the Cleveland Clinic, some 45, or 3.5%, of the 1,300 patients taking Vioxx in the trial had heart attacks, strokes or other cardiovascular events. About 25, or 1.9%, of those taking sugar pills had heart attacks or strokes. He says the event rates were given to him by an executive at Merck. Merck could not comment in time for deadline.

Topol was one of the first to call attention to Vioxx’s potential for causing heart problems. Now he says, “Even when they’re withdrawing the drug, they’re spinning it.”

For investors trying to judge Merck’s legal risk, a clear idea of how many patients were affected is vital—and right now, it’s very difficult to come up with such a number.

Muddy Arteries For AtheroGenics
On Sept. 27, AtheroGenics released what it said was positive data for its heart drug, AGI-1067, the first pill designed to reduce artery inflammation.
One of the big reasons for confidence in the drug was the presence of Steven Nissen, a cardiologist at the Cleveland Clinic who pioneered the use of a technique that uses ultrasound to measure the amount of plaque in the arteries.

Today, Nissen says he wishes that he had not gone on the AtheroGenics conference call. He is surprised about how well AtheroGenics stock has done since, and worries that investors misunderstood him. “I didn’t think people would listen to those words and say this is the best thing since sliced bread.”

Topol, the clinic’s top cardiologist, is even more negative on the AtheroGenics study. “You can’t make anything out of these data,” he says.
“I’d love to see the drug work; that’s not the issue. I just cannot accept their methodology.”

Topol first criticized the AtheroGenics studies in Barron’s.

AtheroGenics never presented its original promised result, an analysis from the Montreal Heart Institute. Instead, the company enlisted Nissen to re-analyze the Montreal data, throwing out artery scans that his team at the Cleveland Clinic found unreadable. AtheroGenics shares surged after it presented the first half of Nissen’s data, which showed the pill may be decreasing the amount of plaque in arteries. However, Nissen’s interim analysis included only 133 patients enrolled in the trial, and will include only 260 when it is finished. Some 500 patients were enrolled in the trial.

Other cardiologists also criticized the study. “When you exclude half the population, you begin to have questions of any analysis that was done,” says Prediman K. Shah of Cedars-Sinai Medical Center. “It does not inspire confidence. It’s not entirely kosher.” Says Columbia University’s Alan Tall, “Even after reading it, I’m afraid I’m a bit confused.”

Mark Colonnese, AtheroGenics’ chief financial officer, said that the company stands behind its data. But he also emphasized that these were interim results, and that people should withhold final judgment until final data are in.

For investors hoping AtheroGenics will soon ink a deal with a big drugmaker such as Pfizer, AstraZeneca or Novartis, the doubts about the company’s studies can’t be encouraging.