WSJ:  March 9, 2004:  For Bristol-Myers, Challenging Pfizer Was a Big Mistake

For Bristol-Myers, Challenging Pfizer Was a Big Mistake

NEW ORLEANS -- Four and a half years ago, Bristol-Myers Squibb Co. placed a huge bet by announcing a 4,162-patient study pitting its cholesterol-lowering pill Pravachol against Pfizer Inc.'s Lipitor.

Though Lipitor was known to be better at reducing cholesterol, Bristol-Myers believed the drugs were essentially equivalent where it counts: in preventing deaths and heart attacks. A trial proving that point would give Pravachol a key marketing weapon in the drug industry's biggest category, the $22 billion market for cholesterol drugs known as statins. Bristol-Myers pugnaciously called its study "Prove-It."

The study didn't prove what Bristol-Myers hoped. Monday, researchers told an audience of several thousand heart doctors who jammed a lecture hall at the American College of Cardiology meeting here that Lipitor was significantly better than Pravachol at reducing the risk of death, heart attack or other serious complications within two years of treatment. The results suggest that for every 25 people taking Pravachol, one will die or suffer a bad event that would have been avoided by taking Lipitor.

The findings are the most decisive yet to show that lower is better when it comes to LDL, or "bad" cholesterol. They are likely to lead to more aggressive national treatment goals for lowering cholesterol, heart experts said. "Statins are going into a whole new high-gear mode," said Eric Topol, chairman of cardiovascular medicine at the Cleveland Clinic. "It's going to change medicine in a big way."

It may also change the outlook for Bristol-Myers, which has been slowly emerging from an accounting scandal that broke out in 2002. Pravachol was the company's top drug in 2003, with $2.8 billion in sales. While Bristol-Myers had expected Pravachol's sales to drop sharply in 2006 when its patent expires, that drop-off may now come sooner. Meanwhile, the study could help Pfizer extend Lipitor's run as the world's top-selling drug even after cheap generic versions of Pravachol and Merck & Co.'s Zocor become available in 2006.

It's too early to declare an end to the long-running debate about what the ideal level of cholesterol is and how people should get there. Bristol-Myers noted that the patients in the study were in a group that is especially susceptible to heart attacks and they may not reflect the experience of healthier people with high cholesterol. "Generalizing from this result without additional clinical trial evidence is not good for patients," said Andrew Bodnar, Bristol-Myers's senior vice president for strategy, medical and external affairs.

RELATED ARTICLE   • The Daily Scan: Stents Lead the Fight Against Heart Disease1

It's rare for pharmaceutical companies to mount head-to-head studies of their drug against a competitor's product because of the risk of unfavorable results. The Prove-It study is the first to demonstrate that one member of the statin class outperforms another in protecting patients from cardiovascular disease. Dr. Bodnar, a cardiologist by training, said the study reflects Bristol-Myers's "long tradition of asking the most important questions and designing studies to answer them."

The study examined patients with acute coronary syndrome, a term that includes conditions ranging from worsening chest pain called unstable angina to a major heart attack. Such patients are at highest risk of death and recurrent heart problems.

The study found that 26.3% of the patients on Pravachol had either died, suffered a heart attack or had other serious complications within two years of treatment compared with 22.4% of those on Lipitor, a relative reduction of 16% in the risk of a bad outcome. Patients in the study took either a 40-milligram dose of Pravachol, the highest available at the time, or 80 milligrams of Lipitor.

Of the patients on Pravachol, 3.2% died from any cause within two years, compared with 2.2% of those on Lipitor, a 28% reduction. That result fell barely short of being statistically significant but researchers said it was still impressive because the study tracked patients for only two years.

'Turning Point'

The findings represent a "turning point" in heart care, said Dr. Topol of the Cleveland Clinic. He noted that they back up a study completed last year, funded by Pfizer, showing that Lipitor is better than Pravachol in limiting the progression of disease in coronary arteries. Dr. Topol wasn't involved in the Prove-It trial and said he was surprised by the results. The study and an accompanying editorial by Dr. Topol were published online by the New England Journal of Medicine to coincide with the presentation at the cardiology meeting in New Orleans.

More data are coming on the issue of how low cholesterol should be. Both Pfizer and Merck have been carrying out long-term studies comparing high and low doses of their own drugs in broad populations and results are expected in the next year or so.

Bristol-Myers has long acknowledged that Pfizer's Lipitor reduces bad cholesterol more, but it theorized that the benefits of LDL lowering begin to fade once levels get to 120 to 125 milligrams per deciliter of blood. The Prove-It study was particularly striking because patients on Pravachol achieved an average LDL level of 95 milligrams, meeting the national guidelines that call for getting patients below 100. Thus they got what was considered the standard of care and still didn't fare as well as patients on Lipitor, who got their LDL down to an average of 62.

Moreover, the added benefits of the high-dose Lipitor started showing up within just 30 days after treatment began. In broader placebo-controlled trials of standard doses of statins, the benefits didn't begin to appear for 12 to 24 months. The rapid appearance of a benefit for the aggressive strategy was surprising, Dr. Topol said.

Only a fraction of people with high cholesterol are on statins, despite a barrage of drug-company advertising backed up by guidance from public-health officials. About 11 million Americans currently take one of the statins, while some public health experts say that at least 36 million should probably be on one. Globally, the discrepancy is even more dramatic: About 25 million are taking the pills while an estimated 200 million meet guidelines for treatment.

"This should be a wake-up call to everybody that treating cholesterol is serious business," said the Prove-It study's lead investigator, Christopher Cannon, a cardiologist at Brigham and Women's Hospital in Boston and Harvard Medical School. "If we see differences in super-low versus very-low LDL, then people with high LDL should no longer hesitate about getting appropriate treatment."

Statins do have side effects, including some at higher doses. That is the "one balancing thing that doctors have to remember" when they use the drugs more aggressively, Dr. Cannon said. In the study, 3.3% of patients on Lipitor had a high level of enzymes that are a predictor of liver problems, compared with 1.1% on Pravachol. The two medicines were roughly equivalent in many other measures of side effects. Some doctors believe muscle pain called myalgia is an underrecognized issue that leads many patients to go off statins.

This isn't the first time Pfizer, the world's largest drug company, has benefited from studies paid for by its rivals. In the early 1990s, many doctors were skeptical that lowering cholesterol could lead to a real reduction in deaths. But three studies involving a total of 15,000 patients -- two by Bristol-Myers and one by Merck -- demonstrated that the cholesterol-lowering statins saved lives and prevented heart attacks in a broad spectrum of patients.

Lipitor arrived on the market in 1997 without any studies showing that it saved lives or prevented heart attacks. But Pfizer did show from the start that Lipitor was a more-powerful cholesterol reducer. Pfizer's marketing machine persuaded many doctors that if lowering cholesterol was a good thing and Lipitor lowered cholesterol the most, it must be the best drug. It quickly became the best-selling drug in the world, while Merck's Zocor took the No. 2 spot in the statin market and Pravachol lagged behind as No. 3.

The turn of events left Bristol-Myers executives frustrated, said John LaRosa, a statin expert and president of State University of New York, Downstate Medical Center, Brooklyn. "The company felt they had done the right thing by investing in these very well-designed trials only to have [Lipitor], without any outcome trials, walk away with the market," Dr. LaRosa said.

Bristol-Myers tried to counteract Pfizer's marketing by pointing to analyses of two of its Pravachol studies. These showed that patients who got the biggest reduction in bad cholesterol from Pravachol didn't do markedly better than those with the smallest reduction. A further analysis of one of the studies in April 1998 by researchers at Harvard concluded that there wasn't any additional clinical benefit for patients when LDL was reduced below 125.

That conclusion, published in the American Heart Association journal Circulation, conflicted with national guidelines that then recommended a target of 100 or lower. One critic of the Circulation article, Steven Nissen of the Cleveland Clinic, persuaded Pfizer to fund a head-to-head study of Lipitor vs. Pravachol.

Dr. Nissen had been developing a new technique using ultrasound to look for the deposits of plaque that accumulate in the artery walls. Those deposits lead to heart disease. He believed the technique would show that aggressive Lipitor treatment was better than Pravachol at halting the progression of plaque deposits. The trial, dubbed Reversal, enrolled its first patient in June 1999.

In November of that year, Bristol-Myers announced the Prove-It study. It would be the first study to compare two statins on their ability to prevent heart attacks and death. The idea was to show that Pravachol is essentially equivalent to Lipitor.

Short Time Frame

The plan was to follow patients for just two years instead of four as in earlier landmark statin trials. Bristol-Myers says the short time frame was acceptable because very sick patients are likely to have more bad outcomes, which enables a faster analysis. Critics believed the limited duration of the study was a deliberate effort to make sure Pravachol wouldn't lose since based on the earlier trials it wasn't likely that two years was long enough to see distinct differences emerge.

In the end, both Dr. Nissen's Pfizer-funded trial and the Bristol-Myers Prove-It trial led to the same conclusion: Lipitor is better. The results of both trials could also boost AstraZeneca PLC's Crestor, a statin that is even more potent than Lipitor in lowering cholesterol. And it may lead more doctors to use Zetia, a pill sold by Schering-Plough Corp. and Merck that delivers extra reduction of bad cholesterol when combined with a statin.

Doctors aren't sure whether Lipitor's edge over Pravachol is solely due to its cholesterol-lowering power or involves other factors. A growing body of research suggests that inflammation plays a central role in the progression of coronary artery disease. In both the Prove-It and Reversal trials, Lipitor was significantly more effective than Pravachol in reducing levels of an inflammatory marker called C-reactive protein. It could be that lowering cholesterol leads directly to less inflammation, or Lipitor may be reducing inflammation via a separate unknown mechanism.

Bristol-Myers's stock fell slightly on news of the study, closing down 1.2% in New York Stock Exchange trading. Pfizer shares closed up 0.9%. Although Bristol-Myers's research labs have gone through a long dry spell, investors are hoping that several new drugs including the colon-cancer drug Erbitux, which it licensed from ImClone Systems Inc., will turn its fortunes around in the next few years.

Write to Ron Winslow at ron.winslow@wsj.com²

Updated March 9, 2004