|Among 10 medications pulled off the market recently, 8 had more side effects for women.
The ultimate reason for conducting clinical studies on new pharmaceutical products is to prove their safety and effectiveness; without such proof, the FDA will not grant licenses for marketing and sale in the United States. Occasionally, however, when approved drugs enter the marketplace, unexpected, adverse results occur. In these cases, it is not uncommon for the FDA to rescind its approval and demand the withdrawal of the products. In fact, between January 1997 and January 2001, this happened on 10 occasions (see Table 1).
|PHOTO: TONY FERNANDEZ
Only concerns about safety or efficacy can result in an approved drugs removal from the market, and in the case of these 10 therapies, safety was the overriding issue. In particular, according to a General Accounting Office (GAO) report dated January 19, 2001, most of the drugseightwere withdrawn because they posed greater health risks to women than to men. (The report is available online at www.gao.gov/new.items/d01286r.pdf.)
Based on this report, several questions about approved drugs and clinical research follow: Why would the safety of medications not be considered fully for women during trials? Isnt there a requirement that researchers conduct tests on men and women?Does an increase in the number of side effects reported after a drug is approved indicate that the trials used to demonstrate its safety were flawed? And finally, does a drugs withdrawal for health risks indicate that the drug is fundamentally dangerous?
There are essentially three facts about the ways in which clinical trials are conducted that can cause difficulties when a product is introduced into the general population. First, most research data is weighted in favor of men; fewer women participate in trials. Second, the sample size of research studies, although statistically significant, is small compared with the general population that will use the drug. Third, the therapeutic use of a medication is controlled only during its trial; real world physicians commonly prescribe approved drugs for alternative (off-label) uses, and patients do not always follow regimented dosing schedules outside of research studies.
It is important to recognize that the U.S. Code of Federal Regulations does not address the question of sex in clinical trials; there is no formal requirement that women make up a significant portion of a trials patient population or even participate at all. The reality is that unless a trial specifically mandates female participation (e.g., a test for the safety and efficacy of a new birth control pill), men overwhelmingly predominate. For pharmaceutical companies, there is little motivation to recruit women. The first reason for this lack of motivation has to do with attempts to limit potential liabilities. Other concerns stem from this issue and focus on economic and legal realities that are stumbling blocks to expanded inclusion of women in research.
First, though, how do concerns about company liability limit the recruitment of female patients? Two issues explain this: pregnancy and infant care. No drug developer wants to be responsible if the product it is testing turns out to have an adverse effect on fetal development or harms a breast-feeding infant through milk contamination. The result, therefore, is that most trials require participating women to be either postmenopausal or nonlactating and actively using an acceptable form of birth control. Some investigators, uncertain of what constitutes an acceptable control method and fearful of patient noncompliance, simply refuse to enroll female patients of childbearing agein their trials. This reduces the number of women eligible to enter any given study, making recruitment that much harder.
The financial situation regarding equal testing of women, though, is based on reality and a male-oriented cost-benefit analysis. Specifically, because of the eligibility restrictions just mentioned, entering women in trials is often more expensive than entering men. Consider that an average Phase III trial may take up to eight months to reach full enrollment. If women must be enrolled in equal numbers with their male counterparts, there are only two possibilities: the time it takes to reach full enrollment must lengthen, or increased funds must be made available to advertise for and recruit female patients quickly. The latter option is clearly cost-related, but in fact, so is the former.
Pharmaceutical companies must file patents on experimental drugs before conducting animal studies and then must recoup development costs before the approved drugs patent life expires. As a result, anything that increases development time, necessarily shortens the period in which costs can be recouped and thus impacts the bottom line. (This issue represents the legal reality that further impinges on the recruitment of women.)
Finally, from a pharmaceutical companys standpoint, the costs of increasing female participation in trials are significantly greater than the odds that an approved drug, tested primarily on male subjects, will be withdrawn by the FDA because of unexpected health effects in women. Subsequently, although pharmaceutical companies do not actively dissuade female participation from trials, their recruitment efforts rarely focus on women, and the lack of female participation in most research is not a point of concern.
Once a drug reaches the market, additional factors often become apparent. For example, the general users of a medication may differ from those who originally tested the product and, among other things, may include a significantly higher percentage of women. Further, nonresearch patients generally consume pharmaceutical products more haphazardly (e.g., skipping doses) than do their research counterparts. Further, in the real world, interactions with concomitant medications that were disallowed during trials sometimes produce unpredictable effects.
The single greatest challenge to approved medications comes in the form of physicians prescribing off-label uses. When the FDA approves a drug, it is for the treatment of a specific indication and not simply an approval of the drug in general. (Pharmaceutical companies must conduct separate clinical trials for each therapeutic use they wish to market.) In reality, physicians often discover, through their own experiences, that a drug approved for one use can be helpful in treating another disease. For example, a headache remedy might be beneficial for the treatment of arthritis. Other misapplications also occur, such as physicians allowing the long-term use by patients of a drug approved solely as a short-term treatment option. The obvious problem is that when doctors prescribe approved drugs to treat patients in ways that have not been researched, health risks can result.
The conclusions of the GAO report indicate that, of the 10 medications withdrawn during the last 4 years, 2 posed no greater health risk to women than to men. The remaining eight medications that did appear to pose greater risks for women, however, were considered in two groups: those with questionable sex connections and those with definite sex-dictated consequences.
The former set included four drugs that produced greater health risks for women but were also used more extensively by women (two, for example, were appetite suppressants). As a result, it is not clear whether those four drugs were definitively more dangerous to women or whether the data was simply skewed by the fact that few men used them (a gender roles issue). In other words, regardless of the percentage of men and women who suffered severe events related to those medications, the total number of reported events was markedly greater in women and represented a significant health risk for them.
The remaining four drugs were prescribed widely to both men and women, but women reported significantly more adverse events while using them. For these drugs, the sex-related effects are clear and demonstrably caused by physiological differences between men and women that were not addressed in trials.
A final accounting
The FDA, in considering withdrawals, takes into account the presence or absence of other approved medications in the same therapeutic class. For example, removing a dangerous antibiotic from the market would not impact human health because other approved treatment options exist. However, the drug Propulsid, despite conclusive evidence of its danger to women, was conditionally left on the market to treat patients with debilitating conditions for which there is no other approved therapy. Thus, there is a balance between safety, efficacy, and therapeutic need.
The GAO report also cannot be taken to mean that the withdrawn drugs were necessarily unsafe or ineffective. The consequences of off-label prescribing of approved medications can provoke FDA withdrawals. In these cases, it is not always true that the offending drug is unsafe, but rather that the manner in which it is used commercially is such that allowing it to remain on the market poses a significant threat.
But what emerges from the report and the discussion thus presented is that women are not considered to an appropriate degree in the clinical testing of pharmaceutical products and, subsequently, are sometimes at risk when using recently approved medications. Ultimately, as the number of new drugs and approvals increases each year, this research flaw has the potential to become an even greater concern.
Cullen T. Vogelson is an assistant editor of Modern Drug Discovery. Send your comments or questions regarding this article to firstname.lastname@example.org or the Editorial Office by fax at 202-776-8166 or by post at 1155 16th Street, NW; Washington, DC 20036.
Return to Top || Table of Contents