It didn't take long. Within days of the pharmaceutical company Merck announcing on Sept. 30, 2004, the voluntary withdrawal of the COX-2 anti-inflammatory painkiller Vioxx, lawyers began offering their services to anyone who had ever taken the drug. Vioxx, along with Celebrex and Bextra, the other COX-2 inhibitors, had enjoyed incredible sales and was touted as a great advancement in pain medication. Just as fast, the drug was off the market and vilified as a killer.
Dr. David Graham, associate director for science and medicine at the U.S. Food and Drug Administration's Office of Drug Safety, testified before the U.S. Senate Finance Committee in November that Vioxx may have caused between 88,000 and 139,000 heart attacks and 'of these, 30 to 40 percent probably died.' Graham further concluded that not only was Merck responsible, but the FDA as well for allowing Vioxx to stay on the market after studies indicated an increased risk for cardiovascular problems.
How did the FDA approve Vioxx and the other COX-2 inhibitors? Why were they kept on the market -- Bextra was only recently removed and Celebrex is still available -- despite the known health dangers? How reliable is the FDA in protecting the public from dangerous drugs, and should the COX-2s be completely unavailable?
Merck applied to the FDA for Vioxx approval in 1998. It received a six-month priority review because it was believed that Vioxx, like the other COX-2s, caused fewer gastrointestinal side effects than other non-steroidal anti-inflammatory drugs (NSAIDs) already on the market. The product was tested on approximately 5,000 patients and did not reveal any increased chance of heart attack or stroke. It did demonstrate a lower risk of gastrointestinal problems, and the FDA approved Vioxx in May 1999.
Dr. John Jenkins, the FDA's director for the Office of New Drugs, points out that approval didn't mean Vioxx was free of any problems or potential ones. 'When the FDA approves a drug it doesn't mean that the drug has no risk,' Jenkins says. 'It just means that we believe that, when used appropriately and according to the instructions in the labeling, the drug's overall benefits outweigh the risks. That doesn't mean in each individual patient the benefits outweigh the risks, but that's the assessment we make on a population basis based on the data we have in front of us.'
That data came from the 5,000 patients tested with Vioxx who didn't show any increased risk of heart attacks. Once Vioxx hit the market, however, there were millions more people taking the drug. In the first year alone, more than 4.8 million prescriptions were written for Vioxx. Celebrex, already on the market, was prescribed more than 17.5 million times.
Dr. Stephen Lindsey, section head of rheumatology at the Ochsner Clinic Foundation in Baton Rouge, who recommended the drugs to his arthritis patients, thinks there are limits to testing. 'So a drug does very well with 3,000 people, but when it goes to market, it goes from 3,000 people to 500,000 people,' he says. 'So if something happens in 1 in 10,000 people, you're never going to see it until it gets out there to 500,000 people. You see five or 10 cases of it and you ask how did this drug ever get approved? That's the way it is -- I don't think you can completely obviate that. You have to float with the punches and be pragmatic.'
Many doctors thought that the COX-2 inhibitors worked well. As opposed to other NSAIDs, which often require multiple daily dosages, patients took the COX-2s only once a day and were highly functional. Dr. Tom Lyons, a New Orleans orthopedic surgeon, says they were his first choice for anti-inflammatory drugs. 'There's dozens of NSAIDs out there, but over the last few years the COX-2s became really popular. Generally, they were more effective. Basically that's what I used, and I think many orthopedists used the COX-2s pretty much exclusively.'
If it were up to Geraldine Harris, she would still be taking Vioxx. A local librarian in her 50s, Harris suffers from osteoarthritis, which she describes as 'severe and constant pain.' Her doctor prescribed Vioxx because Harris had developed ulcers from other anti-inflammatories. Not only was Vioxx easier on her gastrointestinal tract, but Harris even began working out for the first time in years. When Harris heard that Vioxx was being taken off the market, she was devastated.
'I was very upset because the Vioxx was working,' says Harris. 'It was one of the best medications I had taken. I really wanted to continue.'
As word about the success of the COX-2s spread, more people asked their doctors for them. Because the drugs were effective and there was no reason to refuse them, Lyons says it was a matter of satisfying his patients.
'A lot of people came in and wanted Celebrex, Bextra or Vioxx because their neighbor was on it, or they saw the TV ad, so we gave it to them,' he says. 'With the Internet and advertising, many times our prescriptions are dictated by what the patient wants.'
Patient demand translated into huge sums of money for Merck and Pfizer, the producer of both Celebrex and Bextra. In 2003, the last full year all three COX-2 inhibitors were available, COX-2s accounted for $5.3 billion in national sales.
Not everyone was happy with these developments, however. In fact, some doctors like Sidney Wolfe began warning people as early as April 2001 to stop taking Celebrex or Vioxx (Bextra wasn't on the market yet).
In 1971, Wolfe, a longtime critic of the FDA, founded with Ralph Nader the Health Research Group division of Public Citizen, a consumer advocacy organization. Since that time, Wolfe has petitioned the FDA to withdraw 27 medicines -- 16 of which were eventually removed, with others severely restricted. In early 2001, he testified before a FDA advisory committee that a black-box warning -- the most serious labeling for prescription medicine, which forbids the drug from being advertised -- should be put on Vioxx and Celebrex. His testimony was based on the findings of the VIGOR (Vioxx Gastrointestinal Outcomes Research) study, which was conducted by Merck to examine the gastrointestinal effects of the drug.
As Wolfe explains, the VIGOR study resulted in both good news and lethal news. 'The VIGOR study showed that there was some improvement gastrointestinally with Vioxx,' he says. 'It also showed a four-fold increase in heart attacks. So that completely swamps out any benefit with Vioxx.'
The FDA didn't completely agree with Wolfe. It didn't issue a black-box warning, but a year later it approved labeling changes to reflect the cardiovascular risk. Doctors were advised to use caution in prescribing Vioxx for patients with heart disease. (According to the FDA's '2003 Report to the Nation,' the most current report available, the FDA has approved 303 drugs since 1994, with seven then being withdrawn for safety reasons.) Vioxx continued to be advertised on television and elsewhere.
When Vioxx was finally removed from the market in September 2004, a number of studies had come out against the drug, and it appeared that the risks of Vioxx outweighed the benefits. The controversy continued in April 2005, when the FDA announced it would carefully review any proposal from Merck for the re-introduction of Vioxx to the market. Additionally, it announced that it would ask Pfizer to withdraw Bextra but not Celebrex.
The FDA's Jenkins says that because all of the NSAIDs are potentially dangerous, it became a matter of risk/benefit analysis. 'All of the drugs in this class, including the older agents that are not considered COX-2 selective, may increase the risk of serious cardiovascular events,' he says. 'If they all have that risk and we can't differentiate in terms of rank order of one being worse than the other, then we looked if there were any other mitigating factors between the drugs. Did they have any unique risks, or benefits?'
In the case of Vioxx, it appears that the risk/benefit ratio is still open for discussion. Celebrex received a passing grade because it was determined that it didn't have a greater risk of adverse cardiovascular events than the non-selective NSAIDs. Bextra didn't either, but it was associated with an increased risk of serious skin reactions. This tipped the scales against Bextra, but Lindsey, who had an 81-year-old patient beg him to remain on Bextra following the drug's withdrawal, thinks there was another factor involved: legal liability.
'I think the United States is fraught with legal problems, so I think this whole issue here was based on the worries about legal implications -- people worried about getting sued and all the things that could happen from that. It drives a lot of these decisions to take something off the market. There's no 100 percent in medicine -- you have to individualize to the person. That's why I don't like this ruling. It doesn't allow a doctor to individualize to the patient.'
At Ochsner, Lindsey will continue to give his patients Celebrex as long as they don't have a risk for heart attack or stroke. As for FDA's role in drug safety, Jenkins believes Americans can rely on the FDA, but physicians and patients are responsible as well. "The FDA's role is to make sure those high-level decisions about the overall benefits and risks of a drug are favorable, so that they can be approved and available for marketing," Jenkins says. "Beyond that, we have a role in trying to inform and educate patients and physicians in how the drugs can be used safely and effectively. We can't be there with every prescribing decision about how they use the drug. It clearly has to be an entire network effort of the healthcare system to impact the safe use of drugs."