In late 2006, Swedish researchers reported problems with drug-coated stents, the cardiac devices implanted in more than 6 million patients around the world over the past five years, and the product that transformed Natick's Boston Scientific Corp. into the region's largest life sciences company.
The Swedish study, one of several such alarming reports that year, found patients who received drug-coated stents were more likely to die within three years than those with uncoated implants. The news caused some doctors to stop using the devices, wounding manufacturers. Boston Scientific, the biggest maker of drug-coated stents, lost hundreds of millions of dollars in sales and is no longer the largest local life-sciences firm.
But after examining several additional years of data, the same researchers at the Uppsala Clinical Research Centre in Sweden reported last year that there was actually no difference in death rates and that drug-coated stents were safe after all.
Boston Scientific chief executive Jim Tobin dubbed the flip-flop the "oops from Uppsala." And more recent studies have indicated that drug-coated stents may, in fact, be safer than the old, bare-metal versions. At an investor meeting in San Francisco last month, the company's chief operating officer, Paul LaViolette, said three dozen studies over the past few years found they reduced deaths an average of 21 percent over two to three years compared with their bare-metal counterparts.
Stents are tiny wire-mesh tubes used to prop open coronary arteries after they are cleared through an angioplasty procedure. The drug coating was hailed as a major advancement because it blocks the growth of scar tissue, reducing by half the need for doctors to repeat the artery-clearing procedure in subsequent years.
The future of drug-coated stents is crucial to Boston Scientific, which relies on them for about one-quarter of its $8 billion in annual sales and hopes the latest research will revive sales.
The Swedish turnabout illustrates how billions of dollars in medical sales can hinge on studies. It also demonstrates how researchers can draw seemingly conflicting conclusions, based on small differences in the design or interpretation of studies, creating confusion for doctors and patients alike.
"It doesn't mean the science is bad or the studies were wrong," said Dr. Meir Stampfer, an epidemiology professor at Harvard University's School of Public Health. "It means the biology is very complex and can have really different effects on different people."
For example, Vioxx became a best-selling drug after federal regulators approved the painkiller in 1999 based on clinical studies showed it was effective and safe. Later - after the drug hit the market and was being used by millions of people - broader studies found it increased the risk of heart attacks and strokes, prompting Merck & Co. to stop selling it in 2004, and forcing the drug maker to spend nearly $5 billion settling patient lawsuits.
Another once-popular treatment, hormone-replacement therapy, was thought for decades to benefit older women until a study published in July 2002 suggested it increased the risk of breast cancer and did not reduce heart disease, leading to a dramatic falloff in its use. Scientists and doctors now recommend that women take hormone pills only to treat menopause symptoms and to stop taking them when they subside.
In the case of the Swedish stent study, Dr. Donald Baim, Boston Scientific's medical officer, said he believes the most likely explanation for the conflicting results is that when drug-coated stents were introduced in 2003, they were given to the sickest patients. Those patients were also more likely to die - making it appear as if drug-coated stents didn't work as well as bare-metal ones, he said. As drug-coated stents became more widely used, the mortality differences disappeared.
Dr. Stefan James, who helped conduct the Swedish study, said Baim could be correct. But there are other possible explanations for the improved outlook, he said. For instance, doctors may have started recommending patients take anticlotting drugs for a longer time to reduce the clotting risk.
The initial Swedish study was not the only one to raise concerns about drug-coated stents. A small European study in March 2006 found an increased mortality rate for patients with drug-coated stents after six months. Later that year, Swiss researchers, doing a fresh analysis of previous studies, also found evidence that drug-coated stents might carry a higher death rate.
But more recent data have been positive. A study published in the New England Journal of Medicine last week found that drug-coated stents were as safe as or safer than bare-metal stents when doctors prescribed them "off-label" - to treat conditions for which they have not been formally approved by the US Food and Drug Administration. LaViolette said he expects the analysis of the three dozen studies cited by Boston Scientific - showing drug-coated stents reduce deaths - to be formally presented at the American College of Cardiology medical meeting in March and in a major medical journal in the spring.
Still, some analysts and doctors are skeptical that drug-coated stents will regain their previous popularity soon. At their height, the devices accounted for 89 percent of the stent market; that dropped to 63 percent by fall 2007, according to Boston Scientific. The company plans to release updated figures Tuesday.
"I think we will see some rebound in 2008, but the recovery will be a cautious one," said J.P. Morgan analyst Michael Weinstein.
Many medical specialists now believe it takes the body longer to heal with drug-coated stents, requiring patients to take anticlotting drugs for at least six months or a year to reduce the risk of blood clots, compared with as little as one month for bare-metal stents. Some patients can't take anticlotting drugs that long, however, because they can't tolerate the side effects or might need additional surgery, interrupting the treatment. For them, cardiologists often recommend bare-metal stents.
Meanwhile, some doctors are holding off on stents altogether to see if drugs alone might work. A report published in the New England Journal of Medicine last March found bare-metal stents did not reduce the risk of death or stroke compared with drug treatment for patients with stable angina (chest pain that typically occurs during exercise or stress). As a result of the study, Bank of America analysts said that over the next several months there was a 10 percent drop in the number of angioplasties to clear clogged arteries.
But Baim said stents remain key to relieving many symptoms of heart problems, even if it turns out they don't always cut the risk of death. Lately, the company said, it has seen evidence that procedures using any kind of stent have started to inch upward again as doctors find that drugs alone aren't enough for many patients.
Another factor is cost. Drug-coated stents sell for an average of $2,150 apiece, more than double the price of bare-metal versions, according to Millennium Research Group, a market research firm in Toronto. A study in the British medical journal Lancet two months ago concluded that drug-coated stents were not worth the extra money for all patients, which could prompt some insurers to discourage the use of drug-coated stents in some cases.
After considering the flood of data, some local cardiologists say they are spending more time with patients to learn what makes sense for them, based on symptoms and lifestyles.
"I'm individualizing therapies much more than I did before," said Dr. Jeffrey Popma, director of Invasive Cardiovascular Services for the Caritas Christi Health Care System, the state's second-largest hospital chain. Popma added that as a result of the publicity about stents, patients are better informed and ask more questions. "It is a much longer discussion about the risks and benefits," he said.
Todd Wallack can be reached at twallack@globe.com.
