Article in Journal of National Institutes of Health on Impact of FDA Reform
Current FDA "Reform" Would Endanger Americans
This article, by Health Research Group Director Sidney Wolfe, M.D., appeared in the August 1996 edition of The Journal of NIH Research.
The organization I direct, the Public Citizen's Health Research Group, has been for 25 years the most outspoken critic of the Food and Drug Administration (FDA). On more than 50 occasions, we have submitted petitions (occasionally resulting in lawsuits against the agency) to ban or force warning labels on dangerous drugs or medical devices or to require the agency to uphold its legal obligations concerning other regulated products such as food dyes and raw (unpasteurized) milk.
Although we have not been able to persuade FDA to act on all of these products, we have in many cases succeeded. Examples of our successes include the banning of the arthritis drugs Oraflex and Tandearil, the painkiller Suprol, the diabetes drug phenformin, and the Bjork-Shiley heart valve, as well as the requiring of warning labels on aspirin for Reye's syndrome and on tampons for toxic shock syndrome. Although some in industry would not agree, our work, is constructive criticism of the agency in the form of petitions and, when necessary, lawsuits. In sharp contrast is the latest FDA "reform" proposed by industry supporters, which falls into the category of destructive criticism of FDA and which would cripple the agency's ability to protect Americans adequately from dangerous drugs, medical devices, and other products that FDA regulates. Whatever the outcome of this destructive criticism in Congress this year, extreme proposals within the House and Senate bills likely will resurface in future Congresses.
In this article, I will concentrate on some of the most destructive concepts now circulating, those embodied in the House bill H.R. 3199 concerning drugs and biological products, which is still pending in Congress. The need for the changes called for in this legislation-intended to speed up the drug-approval process-is highly questionable in light of three facts. First, user fees paid to FDA by manufacturers submitting new drug applications (NDAs) under the Prescription Drug User Fee Act of 1992 have already safely speeded up the approval process-especially for important drugs-by providing more funds to hire additional FDA reviewers of NDAs. Second, as a result of user fees and a much faster track for those few drugs that are important therapeutic advances over existing drugs, FDA has in recent years been approving important breakthrough drugs as quickly as, or more quickly than, other countries. Third, since these important drugs-the ones patients and their doctors are most concerned about-are already getting approved and marketed in the United States in an expedited manner, the main "beneficiaries" of any legislative changes to force further speedup of the drug-approval process would be "me-too" drugs, which make up 70 percent to 80 percent of drug-approval applications. Me-too drugs are those that are not important therapeutic advances over existing drugs. Manufacturers, nevertheless, are eager to get quick approval of these drugs so as to cash in on multibillion dollar U.S. markets established by breakthrough drugs.
In a statement to The Washington Post reported in the May 1, 1996, issue, Rep. James Greenwood, R-Pa., the floor manager of this legislation, referred to the purported benefits of this bill that would allow drug companies to choose third-party (outside of the FDA) review of NDAs instead of FDA review and would set new limits on the time of review. He said the proposed charges would eliminate "nit-picking" by federal bureaucrats "saying 'not yet, not yet, not yet' for years and years and years" to drug-approval applicants. I believe his remarks apply almost exclusively to me-too drugs because for important drugs, as mentioned above, FDA has already expedited review as safely as possible.
I would like to briefly review the rewards to Americans of such "nit-picking" by summarizing a study I did last year entitled, "Differences in the Number of Drug Safety Withdrawals: United States, United Kingdom, Germany, France 1970-1992." In this study, we looked at all drugs approved for the first time in one of these four countries from 1970 on and later withdrawn (before the end of 1992) because of dangers to patients.
There were 56 drugs that fell into this category. Of the 56, 31 were withdrawn in France, 30 in Germany, 23 in the United Kingdom, and only nine in the United States. In the cases of three of the U.S. drugs-Oraflex, Selacryn, and Merital-their manufacturers later pleaded guilty to criminal charges in withholding safety information from FDA, information that would likely have prevented approval of the drugs had FDA been aware of it.
Not one of these 56 drugs was an important breakthrough drug. Thus, Americans were spared exposure to dozens of drugs that caused thousands of serious injuries or deaths in Europe, where weaker drug regulatory procedures allowed the drugs on the market that were kept off the U.S. market.
What has been referred to as nit-picking-FDA's asking more questions about questionable drugs that do not remotely offer a breakthrough in treatment-turns out to be quite healthful for the American public. Replacing this process with outside review by third parties chosen by the drug companies-likely based on the belief that the chosen third party's review will get a drug to market more certainly and more quickly poses significant dangers.
The United States still has more rigorous drug-approval standards than the rest of the world, mainly because it has had rigorous standards for efficacy of drugs since 1962 in addition to standards of safety. As far as the use of third parties-outside of the government-to review NDAs, European countries have been using such groups for some time. It is this practice and European countries' lower efficacy standards that most differentiate these countries' drug-approval procedures from those of the United States.
H.R. 3199 has two provisions that seek to weaken our laws and regulations so as to make them more like those in Europe. One, Section 5 of the bill, would allow effectiveness to be established by results of "one or more" clinical investigations, which could be interpreted to set the default at one, as opposed to the current system in which two or more clinical investigations is the norm but a single study is allowed in unusual circumstances.
Other parts of Section 5 also seriously undermine FDA's standards for drug effectiveness. One allows FDA to waive the requirement of well-controlled investigations, thereby creating a situation in which known risks would have to be measured against unknown benefits. Another allows approval of new uses of previously approved drugs without evidence from well-controlled studies if the new use is common among clinicians experienced in the field and is based on reliable clinical experience and confirmatory information. This might be called the "DES (diethylstilbestrol) provision," because, under its terms, the cancer-causing drug DES (which ultimately turned out to be ineffective for the unapproved use of preventing miscarriage) would have been approved.
Another part of Section 5 that seriously weakens the efficacy standard is that effectiveness determination shall not include relative effectiveness. Thus, if there are data indicating that one drug is clearly less effective than another drug that is equal in safety, FDA would be barred from using this information as a basis for saying no to the approval of the less effective drug.
A second provision of the bill, Section 7, would allow an applicant to submit an NDA to FDA or to an accredited third-party reviewer. The third-party reviewer must submit its report and recommendation to FDA prior to 60 days before the end of the statutory deadline for NDA review (usually 180 days, making the report due by day 120).
The outside reviewer's approval recommendation would have to be followed by FDA unless FDA finds reasonable probability that the drug is not safe or effective. And FDA must evaluate the external reviewer's recommendation within 60 days of receiving it. If FDA does not accept a recommendation of approval, it must provide a detailed explanation of the basis for its disagreement.
A whole new private-industry bureaucracy, probably quite profitable, will be created by this legislation. Most drug companies will want to choose this "fast" route, and it will be in the third-party reviewers' interests to recommend approval in order to get future business from the drug company. There is little question that unsafe and/or ineffective products that current FDA review keeps off the market would reach American patients under this dangerously expedited private pathway, which is likely fraught with conflict of interest.
Passage of FDA "reform" of the type now being promoted by industry and its supporters would be a dangerous step backward for our health and safety, although it might represent a bonanza for those companies whose me-too products would thereby get approved. Some of the legislation introduced to date would cripple FDA's ability to carefully scrutinize new drug applications. In countries such as the United Kingdom, France, Germany, and others whose drug-approval process is weaker and faster than ours for these less-than-important drugs, the toll on health and lives is clear.
Were the provisions in H.R. 3199 to become law, they would unleash the approval of a torrent of me-too drugs now being kept off the U.S. market, many of which would eventually have to be banned because they are too dangerous for Americans now protected from such drugs.