Whose responsibility is it to ensure a product sold to millions of Americans is safe? In the drug industry, it's no simple question.
Take, for example, the case of Amgen Inc. The Thousand Oaks biotech giant Thursday released the results of a highly anticipated safety study of the company's popular anemia drug Aranesp, which along with similar anemia medications is taken by more than a million people in the U.S. each year.
The Amgen-sponsored trial found that Aranesp showed no increased risk of death compared with a placebo in a subset of lung cancer patients, sending Amgen's shares up nearly 4%.
Industry analysts had called results of the study crucial to the future of Amgen's anemia products, which have been dogged by questions about their safety when used in some patients and dosages. The drug and its shorter-acting predecessor, Epogen, account for about half of Amgen's sales.
"These results contribute to the growing body of evidence on [the drug's] safety, reinforcing the neutral impact [of the drug] on survival in cancer patients suffering from chemotherapy-induced anemia," Roger M. Perlmutter, Amgen's executive vice president of research and development, said in a statement.
Yet there is one question that Thursday's announcement didn't answer: What took so long for the research to happen in the first place?
Although the study's results are reassuring, some wonder whether such questions should have been explored sooner.
"We prescribe this drug to a million people a year and only now are we seriously examining if how we are treating them is harmful?" said Dr. Ajay K. Singh, professor of medicine at Harvard Medical School. "That's very disconcerting."
Singh and other experts point out that anemia drugs have been on the market for nearly two decades. And questions were first raised about the safety of giving patients higher doses of the drugs in 1998, when a study published in the New England Journal of Medicine was closed early because of potential safety issues.
Although many doctors and researchers expected that additional trials would follow soon after to help clear up those fears, that didn't happen.
Meanwhile, sales of the drugs, classified under the name erythropoietin and also sold by Johnson & Johnson under the brand name Procrit, have grown exponentially over the last decade, reaching almost $11 billion last year. Erythropoietin is now the single biggest medication expense in the federal Medicare program.
"How often do we have to wait and see flashing yellow and red lights about a drug's risk before we demand trials are done that ensure they're safe?" asked Dr. Jerome Avorn, a Harvard Medical School professor and author of "Powerful Medicines," a book about the role of the drug industry in medical research.
Although government regulators and the drug industry strongly disagree, critics call it a frequent pattern: A drug that is already on the market shows possible safety problems, but years go by before anyone -- the company, federal regulators or independent scientists -- does more research to show whether a negative safety signal, as it is known, can be corroborated.
It's possible the trend could become more widespread. As drug companies' research pipelines have thinned in recent years, many are trying to find new uses for current medications to maintain revenue and profit growth.
To do that, drug companies are attempting to show that their drugs have benefits beyond what is approved on their labels. Some worry that such "off label" uses could come with risks that aren't explored for years.
Others fear that the government's push to approve drugs more quickly is adding to the problem.
For instance, a study in 2000 showed patients using Merck & Co.'s painkiller Vioxx had a higher rate of serious cardiovascular problems than those on a similar older drug. Yet 2 million Americans were taking Vioxx when it was pulled from the market four years later.
More recently, studies have raised questions about the use of nesiritide, a drug made by Johnson & Johnson's Scios unit to treat acute congestive heart failure. The intravenous drug has been given to more than 600,000 patients nationwide since it debuted in the U.S. in 1987.
Some studies in 2005 suggested that the drug, sold under the brand name Natrecor, could cause kidney problems and death.
That led the Food and Drug Administration to request that the manufacturer add additional safety information to the package's insert. A large trial looking at the drug's safety is underway.
Many drug companies deny that the system for monitoring safety problems of drugs already on the market is too lax.
"We take action as soon as there is the slightest indication action should be taken," said Dr. Craig Tendler, vice president of clinical affairs of Ortho Biotech, the division of Johnson & Johnson that sells Procrit.
Still, the volume of criticism about how drug safety is monitored is rising.