FDA WALKS A FINE LINE
Agency tries to find the right balance between thoroughness and efficiency in drug approval
Congress and the Food & Drug Administration have striven mightily over the past several years to make the agency more efficient and responsive to the concerns of drugmakers and consumers alike. But whether those efforts have been a success or are actually a threat to the health and well-being of the U.S. population is a hotly debated question.
There is no doubt that FDA is approving new drugs at a much faster rate now than it did a decade ago. The time the agency takes to review New Drug Applications (NDAs) has been shortened. The time pharmaceutical firms spend in developing new drugs is also shorter than it would have been without changes that have taken place at FDA.
However, there is a downside to this push to get drugs to market faster by speeding up drug reviews. According to Congress' General Accounting Office, the rate of drug withdrawal from the market because of adverse reactions has increased over the past eight years, and the morale of professionals in FDA's Center for Drug Evaluation & Research (CDER) is low, due in part to an increased workload.
And FDA is at the center of a number of other contentious debates over such issues as postmarket surveillance of drugs for serious adverse side effects, testing in children of drugs potentially useful for childhood illnesses, how to ensure the develop-ment of drugs for rare diseases, generic drug approval, approval of biologics, and the quality of information pharmaceutical companies provide to doctors about their products.
As CDER is taking steps to address these issues, an overarching concern is whether the Prescription Drug User Fee Act (PDUFA), first enacted in 1992, has made FDA too close to the pharmaceutical industry. The industry now provides about half the funds for drug reviews.
THE USER FEE ACT has had many positive effects, both in terms of speeding up FDA's drug approval times and in minimizing the time it takes for pharmaceutical firms to develop drugs. When PDUFA was enacted in 1992, the median drug approval time at FDA for standard drugs was 27 months. Now the median is 14 months. For priority drugs--which are used to treat life-threatening diseases--approval time has dropped more than two-thirds, from 21 months in 1992 to six months today. As a result, valuable medications have been made available to Americans more quickly.
In both 1996 and 1997, CDER approved 120 new drugs, but in 2001 that number dropped to 66. This decline can be attributed to the agency now receiving fewer NDAs than it did before, not to a backlog in the system, CDER Director Janet Woodstock says.
One added plus is that the U.S. economy as a whole benefits from the fact that, today, 80% of new drugs appear for the first time in the U.S. rather than in Europe, Japan, or any other country, according to CDER Deputy Director Steven K. Galson.
PDUFA has gone through three distinct incarnations. The 1992 law focused primarily on speeding up drug approval times. User fees from the drug industry were used to hire 600 additional FDA reviewers. This alone was responsible for a large decrease in approval times, says Jane Axelrad, CDER's associate director for policy.
The first reauthorization of PDUFA was in 1997. PDUFA-2 focused on reducing development times at drug firms. The changes were made in response to complaints that companies needing input on critical drug decisions from FDA had to wait months just to meet with agency officials, Axelrad explains. Also, companies were complaining of difficulties in resolving disputes with FDA staff.
PDUFA-2 set up a number of performance goals for FDA, requiring timely meetings with the drug's "sponsor" and prompt responses to questions. With the staff increases funded by user fees, FDA is holding more frequent meetings with industry. "This enables the level of communication between our reviewers and industry sponsors to be a lot higher than it was before PDUFA," Galson says.
An overarching concern is whether the Prescription Drug User Fee Act, first enacted in 1992, has made FDA too close to the pharmaceutical industry.
HOWEVER, the extra funding made available to FDA under PDUFA-2 did not match the increased workload imposed by the legislation, Axelrad says. So PDUFA-2 was somewhat less successful than the original act. "There were some resources being spent to meet the metrics of PDUFA-2 that probably took resources from the review process, particularly for standard drugs," says Alan Goldhammer, associate vice president for regulatory affairs at the Pharmaceutical Research & Manufacturers of America (PhRMA).
"PDUFA-2 may have overregulated FDA," says Daniel Carpenter, a professor of government at Harvard University who is leading an extensive research effort on FDA. "There is a sense that it resulted in a great deal of micromanagement," he explains. His research is supported in part by the National Science Foundation and the Robert Wood Johnson Foundation.
In 2002, the third reauthorization of PDUFA focused on making up for the resource shortfall created by PDUFA-2, Axelrad says. And, for the first time, it allowed FDA to spend part of the user fees it collects on risk management activities, such as postmarket surveillance of drugs for adverse side effects. PDUFA-3 also authorized an increase in the number of safety officers involved in postmarket surveillance from 100 to 200 over five years.
Under PDUFA-3, FDA will study drug utilization patterns, looking at what types of patients are getting particular drugs and what types of physicians are prescribing them. In addition, it will use a large database to link adverse events with drug utilization, says Victor Raczkowski, director of CDER's Office of Drug Safety.
However, it is difficult to track adverse side effects in the U.S. "The fact is, our country doesn't have active surveillance for adverse events from drugs," FDA's Woodcock says. "We have active surveillance for a lot of things--sexually transmitted diseases, smoking, tuberculosis--but drug adverse events is not one of them," she explains. Consequently, the U.S. doesn't have baseline and trend information for adverse events, she adds. Several countries with national health care systems, such as France, which has regional pharmacovigilance centers, have more complete surveillance, she says.
Nevertheless, FDA plans to try to use data from health maintenance organizations (HMOs) to track adverse events from drugs, Woodcock says. "They often know which patients got what drugs and what happened to them, and everything is computerized," she explains.
ANOTHER OF THE RISK management activities that PDUFA-3 will fund is an effort to improve the drug informational bro-chures--called labels--that drug firms provide as package inserts in all medicines. Critics claim the information conveyed in the brochures is inadequate, confusing, and often misleading, and that the information leaflets that pharmacies provide to patients are difficult to understand and are sometimes biased.
Woodcock agrees. "Currently, the information is not presented in the way most useful to doctors," she says. A couple of years ago, FDA developed a proposal to reformat the brochures, and this will be published in the near term, she notes. "We are engaged in many activities" that could lead to further changes to the brochures and are getting involved in the electronic transmission of information, she says, because "doctors in the future are going to use handheld computers and other electronic devices to get their information."
Peter Lurie, deputy director of Public Citizen's Health Research Group, applauds this effort. "Currently, the information supplied to doctors is very badly organized. It doesn't give them the information they need in a simple way," he says.
PhRMA is addressing another part of the information problem. It is involved in a pilot project with several pharmacies to improve the brochures provided to pharmacies in medicine packages. The drug brochures are provided in an electronic format, so doctors can quickly scroll through the label to access the material they need. "The system won't be ready for prime time for at least a couple of years," Goldhammer says.
FDA is engaged in a parallel attempt to improve the drug information leaflets that pharmacies provide patients. "For years, we have been upset about the lack of clear information for patients," Lurie says. "We have always argued that there should be mandatory, FDA-regulated patient information provided with every prescription," he explains.
"The agency has proposed repeatedly over the years that it regulate the leaflets that are given to patients, but Congress has always voted it down," Woodcock says. "So we are engaged in a process of evaluating how well a voluntary effort by pharmacies to give patients understandable leaflets is going. We set up an advisory committee on that last summer," she says. "But we are not there yet."
One problem CDER is struggling with that doesn't seem to have a straightforward solution is a high rate of turnover among its professional staff. With the exception of chemists, whose attrition rate is on average about 6%, CDER's other scientists--biologists, statisticians, toxicologists, medical officers, and microbiologists--are spending less time at CDER than they do at comparable agencies, such as the National Institutes of Health. Woodcock claims that the only reason professionals quickly leave CDER is that they acquire skills there that are very useful in pharmaceutical firms, and the firms offer them jobs that pay at least twice as much. She maintains that "people like working here. CDER's quality of life is rated high."
The GAO report, in contrast, claims that morale is low at CDER because the professionals are overworked, struggling to meet the many performance goals set by the PDUFAs, and consequently unable to attend seminars and conferences that enhance their professional skills. It says that low morale as well as higher industry salaries are the reasons scientists stay only a few years.
Lurie claims that "unwise drug approvals, due in part to increased workloads caused by PDUFA, have led to poor morale among drug reviewers." A survey conducted by watchdog group Public Citizen found that at least 27 medical officers whose job it is to review drugs were sometimes "made to shade their reviews to be more favorable toward a drug or to not present data adverse to a drug before an advisory committee."
Woodcock says this is not likely. "Under our regulations and in our procedures, if a medical officer at any level does not think a drug should be approved, they have to sign a statement to that effect," she says. "We pay them a lot of money to give us their honest scientific opinion. We checked out our records, and we could find only a couple of instances" where a drug had been approved over the objections of a medical officer, she explains. "We want people to tell us what they think," she adds.
A new series of articles in the medical journal Lancet published in November claims that the multinational pharmaceutical industry has too much influence over the production of information related to clinical practice and too much influence over drug regulation in both Europe and the U.S. In one article, Joe Collier, a pharmacologist at St. George's Hospital Medical School, London, writes that most of the information generated by multinational drug companies remains secret and that "few clinical trials of new medicines are without important shortcomings, especially with respect to trial design and data interpretation" [Lancet, 360, 1405 (2002)].
Woodcock says that the Lancet articles reflect "a debate that has been going on for a long time." She concedes that the investigators who do clinical trials for the drug companies do "put a very rosy spin on the data when they appear in the medical literature. Sometimes, we are surprised it is the same study that we reviewed, because we came to different conclusions about the data. That is why the FDA review is important, because we have an unbiased assessment."
FDA is attempting to publicize a more balanced assessment of clinical trials by posting its reviews on the Web, Woodcock says. It recently adopted a template for reviews of clinical trials that will have a table of contents and all of the information in the same order. "We will be sending these out to professional societies that are interested in certain diseases. They can then report on the reviews to member physicians," she says.
The GAO report also claims that the drug withdrawal rate rose from 3.10% in the eight-year period before PDUFA's implementation to 3.47% in the eight years following implementation.
FDA, WHICH TREATS the data somewhat differently, disputes the GAO analysis. "The drug withdrawal rate is the same in the 10 years before as it was in the 10 years after PDUFA was enacted," Woodcock says.
Harvard's Carpenter cautions "against making inferences from a reported small increase in the rate of drug withdrawals, because drug withdrawals are so rare." He does say, however, that evidence suggests that "many people are now taking drugs for the wrong reasons." For example, some women are taking birth control pills because "they got the impression from an advertisement that the pills would help them lose weight." Misuse of drugs is a real problem, he says.
Some critics see a problem in that the vast majority of NDAs received by FDA are for new formulations of pharmaceuticals that are already on the market or for drugs in the same class as an existing approved drug. In other words, a fairly small percentage--about 17%--of NDAs are for priority drugs, or drugs viewed by FDA as offering an important new therapeutic advantage.
"Therapeutic classes are crowded because there is a market," Lurie says. "And companies keep going back to the same market." For example, they keep producing drugs for depression, even though the new drugs are no better than the old ones, and for pain, even though the new drugs are no better, he explains.
Woodcock, in contrast, does not consider this a drawback. Having several drugs in the same class may benefit patients, Woodcock says. "We don't think that having choices is necessarily a bad thing," she says. For example, many people who need to take a statin drug to lower cholesterol find they can't tolerate the drug that they are prescribed. But they are able to tolerate another statin--a drug in the same class, she notes.
Furthermore, even new preparations of existing drugs can benefit some patients, Woodcock says. For example, many very young and very old people can't swallow pills. A liquid or slurry preparation of an existing drug may allow them to take the medicines they need. In addition, sustained release medicines help those who have difficulty remembering to take medications several times a day, she says.
"Actually, we are seeing fewer drugs that are very similar to existing drugs than we saw before the passage of PDUFA," Woodstock says. One reason is that companies often have difficulty getting a pharmaceutical on a formulary--an HMO's list of acceptable drugs--if it is very similar to an existing drug, she explains. Also, mergers and acquisitions in the industry have had an effect. "When two companies merge that are both developing the same kind of drug, they will suppress one of them," she explains.
One measure that has helped the development of priority drugs is the Orphan Drug Act, passed in 1983. Drug firms were given several kinds of incentives to develop orphan drugs, which treat rare diseases. Not only do the firms qualify for special tax credits, they can also obtain grants from FDA totaling about $12 million annually to help defray some of the costs of clinical trials. Also, orphan drugs have several years of market exclusivity beyond the normal 17-year patent. FDA usually gives orphan drugs expedited review because most of them are for serious or life-threatening diseases.
As a result, more than 200 drugs and biological products for rare disorders--defined as diseases afflicting fewer than 200,000 people--have been approved and marketed since 1983, in contrast to 10 in all the years before 1983.
Another success story at FDA, according to the agency, is that the generic portion of the marketplace has risen dramatically, growing from 19% in 1984 to about 50% today. FDA has approved about 8,000 generic drugs since 1984, when the Hatch-Waxman amendments concerning patent drug law passed, easing the approval of generic drugs (C&EN, Sept. 23, page 53).
"Our country doesn't have active surveillance for adverse events from drugs."
FURTHER INCREASING the availability of generic drugs--and consequently lowering prescription drug costs--is the aim of a new rule the Bush Administration announced this past October. Under the current Hatch-Waxman law, a brand-name manufacturer is allowed to file for multiple new patents on a drug, and each new filing triggers a 30-month delay in a generic drug's approval. The new rule would close that loophole and allow only one automatic 30-month stay. According to a study by the Federal Trade Commission, new patents on brand-name drugs are often frivolous, concerning the color of the pill bottle or the shape of the pill. The Administration estimates that the new rule will cut $3 billion off the nation's drug bill.
But attempts to increase the availability and safety of drugs used to treat children were dealt a blow recently. In October, a federal judge invalidated a 1998 rule that required drug companies to test some of their products on children. The judge ruled FDA does not have statutory authority to require pharmaceutical companies to test their products on children, even though the pediatric rule "might reflect the most thoughtful, reasoned, balanced solution to a vexing public health problem." Most observers believe that drugs that are potentially useful for treating childhood illnesses should be tested on children to determine proper dosages and safety. Before 1998, very few drugs were tested in children because of concerns that children could not give informed consent.
"THE TRUTH IS, there will always have to be some testing on children if we wish to know whether drugs work on children," Lurie says. Woodcock refuses to comment on the court ruling because the issue is still under litigation.
Tackling a relatively new issue, FDA recently began an organizational move that is expected to speed up the approval of biologics--drugs that are produced through biotechnology--and make review of these drugs more consistent with the approval of standard drugs under CDER. In September, the agency decided to move the approval process, but not the research on biologics, into CDER from the Center for Biologics Evaluation & Research (C&EN, Sept. 16, page 8). "We face a transition period that will be challenging, but I think there are advantages," Woodcock says.
Carl B. Feldbaum, president of the Biotechnology Industry Organization, praises FDA's decision. "We are pleased with the Bush Administration's efforts to bring lifesaving drugs to patients faster, alongside greater consistency in the drug developing and review process," he says.
Despite that praise, FDA will undoubtedly continue to find itself embattled for a long time to come. As has happened for decades, Harvard's Carpenter says, the agency will be bashed from one side by those who want drug approvals speeded up, and bashed from the other by those who believe the agency is not taking enough care in approving new drugs and not using enough vigilance in monitoring adverse drug reactions. There is no easy way to find the right balance between taking time enough to make very careful drug decisions and keeping the drug approval process efficient, which helps get needed drugs to patients quickly and increases the profits of the pharmaceutical industry.
The pharmaceutical industry has recently suffered a downturn in productivity and investor goodwill, but a number of factors remain in its favor
DRUGMAKERS PLAY IN CONTRACT FIELD
Several big-name drug companies have a lesser known business serving their competitors
FDA WALKS A FINE LINE
Agency tries to find the right balance between thoroughness and efficiency in drug approval
STEP BY STEP
Long, Laborious Path To Drug Approval
NEW 'OME' IN TOWN
The "metabonome" provides real-world information about drug toxicity, gene function