A recent article in the Wall Street Journal described “a growing backlash against the pharmaceutical industry that is already affecting the development and marketing of drugs in the U.S.” The article also cites increasing emphasis by the FDA on drug safety, “which probably will inhibit the arrival of new treatments to the market [because] drug makers are rethinking what sorts of drugs to pursue and develop, based on their anticipation of potential safety and cost problems.” These are ominous prospects for an aging population that increasingly needs innovative new medicines for their degenerative diseases.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />
Many issues that surround the development of new drugs are far more complex than they seem. Although it is true, for example, that several major medical journals recently introduced a requirement for companies to register clinical trials in a publicly available database before they are launched in order for the results subsequently to be considered for publication, that initiative has been widely misunderstood, and its usefulness exaggerated. The rationale is that “negative” results from clinical trials could no longer be obscured and simply not reported. In other words, companies should not be able to “cherry pick” studies, making public only those that give favorable results and suppressing the rest.
This unilateral action by journal editors was arguably little more than a power trip. Nothing in our society is more stringently regulated and monitored than drug development. During each phase of clinical testing, the FDA reviews and must grant permission for every clinical trial and has access to all of the proprietary information about the drug (as well as related drugs). When the manufacturer has accumulated evidence that the drug is safe and effective—as part of the application for marketing approval, the results of every trial and everything else that is known about the drug – all, must be reported to the FDA. That means in the United States and abroad. And, of course, trials must be performed with appropriate treatment of statistics. Moreover, the FDA serves as a repository for data on similar drugs made by other manufacturers. All of this prevents statistical “cherry-picking” or “data mining” that could mislead regulators.
It is true that the control that FDA exerts over drug developers would not preclude some selection bias when researchers submit a paper to a medical journal. However, when a manuscript reporting a clinical trial is submitted, all of the relevant data related to that experiment is included therein, and the referees and journal editors can ask for additional information that they feel is necessary to ensure the integrity of the conclusions. If they’re not satisfied, the paper doesn’t get published.
The editors have overreached by demanding the disclosure of background material related to sensitive, proprietary details of study design, timing, manufacturing methods, etc. Competition creates incentives for efficiency and progress, whether in a neighborhood restaurant, academic research lab or corporate boardroom. In an environment as competitive as the drug industry, why should companies tell their competitors what they’re doing, how, when, where, and with whom?
Who gave journal editors the right to appoint themselves omniscient arbiters of medical progress? Companies and the academics who perform their clinical trials should simply refuse to comply with the editors’ demands. If necessary, perhaps through a non-profit foundation, the industry can endow new, less imperious, less hostile, peer-reviewed journals. Perhaps the first could be Nature Clinical Trials.
Finally, a clinical trials database might be useful to ascertain whether an individual patient would be eligible for a clinical trial that is under way—a function already served by www.ClinicalTrials.gov . But—except for offering a bonanza to plaintiffs’ attorneys trolling for business—the benefit of a publicly available database of clinical trial results would be minimal. Few practicing physicians currently take the time to read carefully even the existing labeling for drugs, newly-approved or otherwise. Why should we expect that they’ll sit down at a computer and surf through a trial results database for data on an as-yet unapproved drug that they cannot prescribe, or to try to make sense of “negative” clinical trials that may have been performed with an approved drug?
That brings up the question of what we mean by “negative.” In the context of scientific and clinical experiments, the term has a meaning very different from the common usage.
At last count, on average the results of more than 70 clinical trials are submitted by a corporate sponsor to support an application to FDA for marketing approval of a drug, but generally only two or three are "pivotal" trials that provide the required definitive evidence of safety and efficacy. The remainders are sometimes referred to as "negative" trials, a tag that carries an unnecessarily and inappropriately pejorative connotation.
These trials are seldom “negative” in the sense of revealing that the drug being tested inflicts harm, but for a variety of reasons, they may not be useful or applicable to the indications (uses) for which approval is being sought. The reasons can include: insufficient statistical power (that is, number of patients) in the study; inappropriate choice of route of administration, dose or frequency of administration, or in the stratification of subjects; or simply a failure of the drug to be effective for the indication for which it was tried
The bottom line: “negative” does not imply “harmful.” “Irrelevant” might be a more appropriate and less judgmental term.
The Wall Street Journal article notes that FDA “has become more aggressive in making its concerns public even before the link between a drug and a side effect is clear, and plans to do more in this area through a proposed "Drug Watch" Web site. This initiative is a poorly conceived, precipitate response to political pressure on FDA to seem more concerned about side effects. According to the agency, Drug Watch “is intended to identify drugs for which FDA is actively evaluating early safety signals. The Drug Watch is not intended to be a list of drugs that are particularly risky or dangerous for use; listing of a drug on Drug Watch should not be construed as a statement by FDA that the drug is dangerous or that it is inappropriate for use. Rather, inclusion on the Drug Watch signifies that FDA is attempting to assess the meaning and potential consequences of emerging safety information.”
It is hard to imagine how access to such preliminary data—which will be made available on the agency’s website—will benefit physicians and other healthcare providers, let alone members of the public. There is a difference between indiscriminate data and useful information, and Drug Watch seems destined to provide far more of the former than the latter.
We have already had a hint of the mischief that can be wrought by the indiscriminate release of data on drug safety. By filing a Freedom of Information request, an analyst at the stock brokerage firm Morgan Stanley obtained a list of adverse-events reports made to the Food and Drug Administration related to Tysabri, a multiple sclerosis drug withdrawn from the market by its manufacturers in February. (Following an FDA priority review of new data, the drug should again become available in 2006.) After analyzing that data, the analyst released a brief report noting that a number of Tysabri patients had died of rare infections. Are “published” opinions about the safety of drugs destined to have only the level of reliability as the average blog?
Moreover, given the current fervor at FDA for ensuring drug safety—and the difficulty of proving a negative—how will a “suspect” drug ever be able to clear its name and get off the Drug Watch list? It would be far more constructive, surely, to update product labeling continually and rapidly—which can now be accomplished using the FDA’s website and e-mails to healthcare providers and consumers—once regulators are past the stage of merely “attempting to assess the meaning and potential consequences of emerging safety information.”
There are important lessons in all of this. First, research, development and commerce thrive when restrictions and constraints are kept, insofar as possible, to the level that is necessary and sufficient.
Second, society offers civil servants like those at the FDA lifetime tenure in return for resisting political pressure and the regulatory fad du jour, and for considering only the public interest in its decisions about policies and individual products. FDA officials must uphold their end of the contract.
Third, public policy toward science, technology and medicine should be made—insofar as it’s possible – without regard to the prevailing public biases and perceptions of the time. Instead, our leaders should lead… Now there’s a novel concept.