WASHINGTON -- Over 15 years, the Food and Drug Administration has swung from taking too long to get medicine to dying AIDS patients to drawing fire for rushing drugs to market that wound up killing people.
The agency's troubles were highlighted last week, when it asked Pfizer Inc. to suspend sales of Bextra because the painkiller can cause fatal heart and skin problems. Bextra was the latest casualty in a drug safety controversy that began last fall with Vioxx, an arthritis drug quickly approved by the FDA and then taken off the market. The multiple sclerosis drug Tysabri followed the same script: speedy approval followed by withdrawal, because it may cause a rare brain disease.
The profound change within the agency came from several directions. AIDS activists in the late 1980s besieged the agency to demand it make more drugs available to combat what, at the time, looked like an unstoppable epidemic. Threats by major drug companies to move overseas spurred Congress to pass laws that created a speedier approval process funded by drug makers themselves. The speed movement took hold with people at the top ranks of the FDA, who pressured staffers charged with studying the effectiveness and safety of drugs to hurry up.
''There is no doubt that there has been a cultural change," said John Jenkins, director of the FDA's Office of New Drugs. ''I don't think it needs correction."
Jenkins said the agency did not err in giving faster reviews for problematic drugs and said the FDA did not compromise safety. ''There is a distinction between the speed at which we complete our review vs. whether we've changed the evidentiary standard for approval," Jenkins said. ''I don't think we've altered the standards for approval."
A closer look at the internal and external pressures on the FDA shows that the safety problems did not happen overnight. From 1993 to 1996, 1.6 percent of new drugs approved by the FDA were taken off the market because of safety problems. Of the drugs approved between 1997 and 2000, 5.3 percent were later withdrawn.
Many were top sellers: About one in 10 adults had been taking drugs that were withdrawn from 1997 to 2000, said John Abramson, a clinical instructor at Harvard Medical School and author of the book ''Overdosed America."
By the mid-1990s, when the agency's efforts to speed drug review were going full throttle, the number of drug approvals became part of employees' performance evaluations. Meanwhile, Congress expanded the kinds of drugs the FDA could fast-track. Once reserved for cancer and AIDS drugs, faster reviews were launched for medicines not aimed at life-threatening diseases. Viagra for male impotence, Celebrex for pain relief, and treatments for itchy eyes and irritable bowels were all granted priority approvals.
''There was nobody out there saying, `I'm dying of erectile dysfunction; you must approve this drug quickly,' " said Abbey Meyers, president of the National Organization for Rare Disorders. ''Speedy reviews were supposed to be for serious illnesses and life-saving drugs, and now it's for everything."
An outcry over AIDSThe climate was dramatically different in 1988. On Oct. 11, Sharon Smith Holston, a senior FDA manager, drove past uniformed police officers, flashed an ID badge to enter the agency's garage -- an unheard-of security measure at that time -- and peered out the window at a restless crowd massing at the agency's headquarters in Rockville, Md.
Gay activists, united in fear of the deadly AIDS plague ravaging their ranks, had organized a demonstration meant to pressure FDA leaders to speed access to life-saving drugs. It was taking the FDA 30 months to approve a typical drug. Many AIDS patients would die waiting, the activists said. So protesters stood atop buses and stopped traffic. Some broke into the FDA and smashed computers. They hung then-FDA commissioner Frank Young in effigy.
''It felt like we were under siege," said Holston. ''We felt as if we were doing good. The signs were hurtful." Protesters were chanting, ''Hey, hey, FDA how many people have you killed today,' " she said.
Matthew Sharp, now a 14-year AIDS survivor, took part in the protest with pride. ''It was the first time patients had mobilized to any direct action toward a federal agency for their disease cause."
They succeeded. After other AIDS protests and lobbying, the agency responded by zipping through more than a dozen experimental AIDS remedies. Those drugs, many say, changed the course of an epidemic.
''At the time, there was only one drug available for HIV, and it was mediocre," recalls former FDA commissioner David Kessler. ''People were dying."
Kessler, now dean of the school of medicine at University of California-San Francisco, said, ''When it comes to serious and life-threatening diseases, you have to make a decision. Are you willing to take risks? And, if you're willing to take risks, you're going to be wrong sometimes. But, I think in the case of HIV, it worked."
Frustrated drug companies also wanted speed. In Washington, industry lobbyists told legislators and regulators that drug companies would move research, development, and clinical studies abroad, where it was easier to initiate patient trials and get a drug through the development and approval process, said Kenneth I. Kaitin, director of the Tufts Center for the Study of Drug Development. The White House was worried that the pharmaceutical industry would follow the high-tech industry to Asia.
''There was a concern that the industry would move overseas," Kaitin said.
Within the FDA, a feeling was growing that something needed to change. When Richard Kapit, a former FDA clinical reviewer, got Prozac to review in late 1984, he was told the application for the new antidepressant had sat untouched for two years.
''Everyone was erring on the side of caution," Kapit said. ''There was general recognition that something needed to be done."
In a budget-cutting atmosphere, Congress passed the Prescription Drug User Fee Act of 1992. It introduced new goals for the FDA: Most ordinary drug reviews would take no longer than 12 months, instead of the 30 months it took in the late 1980s. Important drugs promising significant medical benefits could earn speedier six-month reviews. In exchange for speed, the industry would pay hefty fees that helped the FDA boost resources, including the number of reviewers examining new drug applications.
Industry funding, called user fees, has grown substantially. By 1997, user fees were adding $87.5 million to the FDA's budget each year. In fiscal 2006, industry dollars flowing to the FDA are expected to hit $382 million.
The sums reflect a sizable portion of the FDA's budget for reviewing drugs. Last year, the agency spent $575.9 million for what it calls pre-market activities, including reviews of new drug, medical device, and biologic applications. Industry user fees accounted for $222 million of that total.
Despite the cost, the drug industry likes the system. Shaving six months from a drug review gives a company ''a fantastic amount of money," said William Vodra, a former FDA attorney now at Arnold & Porter, a Washington, D.C., law firm. The Tufts Center estimates that half of the $802 million expense of developing a new drug reflects revenue a company forgoes on unrealized investments because its money is locked up in the experimental drug.
With industry dollars, the FDA started to approve drugs much faster. In 1993, the median time for speedier drug review was 16.3 months. A standard drug review, by contrast, took 20.8 months. By 1995, priority reviews took 7.9 months while standard reviews took 16.2 months.
Signs of troubleA troubling pattern quickly emerged. Nine drugs approved after passage of the user-fee law were withdrawn for safety reasons by 2000.
''With each successive one, the question was raised again: Is this something that was the result of accelerated review?" said Dr. Jerry Avorn, a professor at Harvard Medical School and author of ''Powerful Medicines." Avorn said speed was one problem, along with what he called the FDA's sloppiness and culture of denial.
''That's about the purposeful suppression of individuals within FDA who are concerned about problems and are told, 'Don't make waves. We need to get this drug approved,' " Avorn said.
An agency bathed in the afterglow of protecting Americans from thalidomide dangers, which for decades rewarded caution, was undergoing a seismic cultural shift, said Kapit, who worked on new drug reviews beginning in 1984 through the '90s and left the FDA in 2002.
In the '80s, ''if I had a problem with some data that a company submitted, I checked that out. And I checked it out until I was satisfied," Kapit said. ''Now, if you're going to check that out -- and because you're checking it out you may go past the deadline -- well, you have to get permission."
Kapit said, ''Many supervisors really counted approvals as the major factor in deciding whether you were doing a good job."
Dr. Erick Turner reviewed psychotropic drugs from 1998 to 2001 at a time when the FDA's goal shrank to 10 months from 12 months. He said reviewers who recommended a clinical hold -- which stops action on a drug application until the company responds to the agency's concerns -- were summoned before ''tribunals" to defend their decision. Turner, now an assistant psychiatry professor at the Oregon Health and Science University school of medicine, said the tribunals added to the pressure to approve drugs too fast.
''The expectation is you're supposed to let them go ahead," he said.
The FDA's Jenkins discounts those assessments. He said Congress wanted the agency to clear backlogged drug applications quickly but wasn't ordering the agency to approve everything. ''All of the goals relate to review times, not any standard for X number have to be approved," Jenkins said.
Criticism of tribunals, he said, mischaracterizes an oversight committee now used as an education tool so employees in different divisions can learn from others' experiences.
''There were some concerns, at first, about people feeling like they were being put under the microscope. I think we've addressed those concerns," Jenkins said.
Kessler, commissioner from 1990 to 1997, also disputes that FDA leaders then were consumed exclusively with speed.
''If there was a single-minded focus, it was 'find drugs that work, find drugs that work, find drugs that work.' OK?" he said. Especially, drugs treating serious conditions like AIDS, multiple sclerosis, and cancer. ''I am proud of what the agency did in the 1990s. It was an historic period of drug development. But it's not generally applicable to every drug or every condition. And it was never designed to be that."
Among the first to cast doubt about speedier drug reviews, however, were some of the AIDS activists who were the catalysts behind the movement in the first place.
''Advocates who were, basically, initially scaling the walls of the FDA and burning FDA commissioners in effigy, six years later came into my office -- the same advocates -- saying, 'Don't give me another mediocre drug where I don't know the dose. FDA is being too lenient,' " Kessler recalled.
'An unacceptable risk'One of the first warnings the agency was going too fast came from Warner-Lambert's efforts to get the diabetes drug Rezulin approved. The FDA's medical reviewer assigned the drug, John L. Gueriguian, spotted signs the drug could cause liver problems. He and others predicted side effects could worsen to include acute liver failure, the need for transplants, and deaths.
Gueriguian and others at the FDA repeatedly told company officials about those concerns. Still, in 1996, the company responded with a new drug application that showed few liver problems.
''What they were telling me is people who were treated with Rezulin, as compared to people treated with placebo, were doing better as far as their liver was concerned."
Outraged, he swore at a Warner-Lambert official. The company complained to his bosses. They took him off the case and did not present his draft report to an advisory committee that ended up recommending the FDA approve Rezulin.
With the drug on the market, the FDA began receiving reports of liver problems. The agency sent letters to doctors explaining ways to reduce the number of serious side effects. Warnings on the drug's label were strengthened several times. Doctors closely monitored liver function in people taking Rezulin.
British drug regulators took Rezulin off the market in 1997. Three years later, the FDA agreed with Gueriguian's prediction of dire side effects and did the same. By that time, the agency had confirmed the drug caused 58 deaths. The drug posed ''an unacceptable risk to patients," the FDA said.
''That's the part where I really sensed a change," said Kapit, the Prozac reviewer. ''Back in the '80s, boy, they would have pulled that drug right away."
Pfizer, which has since acquired Warner-Lambert, declined to comment.
The drug industry says successes of faster drug reviews have eclipsed failures. Even Rezulin was not a total disaster, because it heralded an important new class of diabetes treatments, said Alan Goldhammer, vice president for regulatory affairs at the Pharmaceutical Research and Manufacturers of America, the industry's lobbying arm in Washington.
But the FDA also has granted priority status for drugs that treat non-life-threatening diseases, sometimes with fatal consequences. Despite evidence of serious gastrointestinal side effects that were flagged by one of its doctors during a review that took less than eight months, the FDA approved GlaxoSmithKline's drug Lotronex for irritable bowel syndrome in 2000.
After at least four fatalities and 45 bowel surgeries required to treat severe side effects, GlaxoSmithKline withdrew the drug. The company reintroduced it in 2002 with stricter guidelines for its use, but some doctors still won't prescribe it. Dr. David Cave, a gastroenterologist at Caritas St. Elizabeth's Medical Center in Brighton who oversaw several patients in a Lotronex clinical trial, says the FDA erred by giving it rapid approval.
''I know it was the first drug that came out for irritable bowel syndrome, but irritable bowel syndrome never killed anybody," Cave said. ''So what's the hurry?"
To the FDA's growing number of critics, the problems with speedier reviews culminated with Vioxx, a painkiller taken by millions of Americans, most unnecessarily.
During the review process for Vioxx, an FDA medical officer looking at clinical trials noted a threefold increase in cardiovascular problems by people taking it, but could not definitively blame Vioxx. A similar drug, Celebrex, was already on the market, yet the FDA gave Vioxx priority treatment with a goal of making a decision within six months. It hit that goal and approved Vioxx for sale in 1999 to ease acute pain, osteoarthritis, and painful menstrual cramps.
But through extensive marketing, the drug's maker, Merck & Co., sold $2.5 billion of Vioxx worldwide in 2003, well beyond that relatively small group of patients.
Within six months of the speedy FDA approval, another Vioxx clinical trial pointed to a doubling of heart attacks and strokes. Last Sept. 30, Merck took Vioxx off the market. Had the drug gotten a standard review and the FDA pushed for better studies earlier, the extra evidence could have led the agency to restrict who used the drug or delay its arrival on the market, critics said.
The FDA said Vioxx's cardiovascular risks were not known until after the drug was approved and defended its speedy review because of the drug's potential to reduce the possibly fatal gastrointestinal side effects caused by many other painkillers. The FDA's Jenkins said Vioxx had a lower rate of stomach ulcers than other painkillers but the agency decided it needed stronger evidence before adding that to the drug's label.
''Based on the information we had at the time we made those decisions, I think those decisions were appropriate," Jenkins said.
A Merck spokeswoman said Vioxx qualified for the agency's speedy review process, and both the FDA and its advisers agreed the company's studies of more than 5,000 osteoarthritis patients showed the painkiller's benefits outweighed its risks.
''With regard to whether additional pre-marketing studies should have been performed, neither the FDA nor the advisory committee believed this to be the case," the spokeswoman said.
David Graham, an FDA senior reviewer who blasted his own agency in testimony before Congress, estimated that up to 139,000 people suffered a heart attack or stroke, and 40 percent of them died, because of the agency's haste to get Vioxx on the market.
''Safety, as I said before, is at the back of the bus," Graham said.