Participation of Women in Clinical Trials: Ethically Achieving Representative Enrollment

Introduction

Clinical trial safety and efficacy results are integral components of the marketing applications for new drugs, medical devices and biologics. The trials signal the possibility of new and improved therapies for the greater patient population and, very importantly, provide individuals suffering from a disease with an opportunity to receive potentially effective new therapy, particularly in Phase 3 trials. Twenty years have passed since progressive regulatory agencies first attempted to reverse discriminatory attitudes against inclusion of women in studies, yet today they remain underrepresented in medical areas responsible for the highest risk to the female population. The sub-group of pregnant women continues to be actively excluded from all study participation. This article explores the lingering issues behind the slow adoption of representative enrollment and addresses these concerns from the feminist bioethical perspective, which supports the moral imperative of responsibly including women and their unborn children in all areas of research that affect them.

Traditional Barriers to Women's Participation in Clinical Trials

Female sex characteristics and female anatomy covariates such as hormonal variations and body size produce different responses to drug and medical device therapies compared to those of male subjects. Therefore, the exclusion of women from clinical trials negatively affects the medical community's understanding of these responses. In the absence of clinical data on women, there is no evidence or impetus to support marketing drug products with modified dosages or medical devices tailored for women's anatomy. This dearth of basic knowledge regarding prescription and over-the-counter drugs and medical devices is juxtaposed against the industry's progress toward biotechnology solutions in the field of personalized medicine. Healthcare customization assumes biotechnology firms' investment in tailored product development for the female anatomy, which is dependent upon female enrollment in clinical studies.

Prior to 1970, regulatory agencies and federal legislation prohibited inclusion of women of childbearing age in clinical trials on the basis that their exclusion protected the potential fetus from investigational drug exposure.¹ Due to this practice, in 1992 the US Government Accountability Office (GAO) found through investigation of clinical trial conduct that women were dramatically underrepresented and sex difference data were analyzed in less than half of drug studies.² In response, the US Food and Drug Administration (FDA) published a guidance document, Guideline for the Study and Evaluation of Gender Differences in the Clinical Evaluation of Drugs, to promote the study of women in dosing trials and increase data collection on sex response differences.³ The National Institutes of Health (NIH) was mandated by Congress the following year to monitor study inclusion of women as part of the NIH Revitalization Act of 1993.⁴ Other countries followed FDA's lead; for example, in 1997, Health Canada issued its own Inclusion of Women in Clinical Trials policy.⁵ Recent assessment of these NIH monitoring reports showed that overall average enrollment in Phase 3 trials increased between 1995-2010, but that only 28% of the trial results included gender-specific response data.⁶ Therefore, while enrollment has increased due to regulatory engagement, the availability of substantive data regarding the female patient population remains limited.

In response to the enduring need for representative enrollment and data, particularly within the medical device field, FDA's Center for Devices and Radiological Health (CDRH) gathered a number of interested parties to provide input into the persistent barriers to women's enrollment and explore potential recommendations. These parties included medical professional societies, patient advocacy groups and other stakeholders. The result was the 2011 draft guidance, Evaluation of Sex Differences in Medical Device Clinical Studies.⁷ This document identifies several enrollment barriers: a pervasive fear of harming a potential fetus; lack of information regarding disease causation and effect within the female body; and study personnel's avoidance of female patients due to the notion that recruitment of women requires more study resources.⁸ Each identified inclusion barrier represents a risk-based perspective that ignores the associated risk of excluding women. While CDRH focused on providing recommendations for improved enrollment through study design changes and data reporting solutions, this article explores the rationale behind the barriers and attempts to weigh these perspectives against bioethical responses.

Fear of Fetal Harm

To date, avoiding harm to a potential fetus has been the most common rationale for excluding women from participating in clinical trials and presents the most difficult social stigma to overcome. Proponents of protecting the fetus through study exclusion of all women of childbearing age arrive at their viewpoint from both legal and moral avenues. Clinical trials are inherently associated with some level of risk due to the investigative nature of their research questions. The task of all research sponsors and investigators is to weigh this potential risk to the trial subjects against the perceived benefits for the subjects as well as the overall associated patient population. This careful assessment of risk is mandated through the established regulatory process for drugs and devices, a phased approach that gradually exposes increasing numbers of subjects to an investigational therapy as the early trial phases prove successful. Early trial phases demonstrate safety in humans prior to the addition of study design complexity and statistical significance to prove the efficacy of the therapy. In this environment of calculated risk, FDA is charged with approving the research sponsor's benefit-risk approach and study design, while research sponsors and clinicians are ultimately responsible for thorough communication of these risks to the subjects during the informed consent process. Despite these requirements and safeguards, litigation and liability for adverse events are a real concern for research sponsors.

The early feminist response to these legal concerns is exemplified in a collection of essays edited by Susan Wolf, entitled Feminism and Bioethics, Beyond Reproduction.⁹ One of the essayists, Vanessa Merton, acknowledges that if a female subject conceives a child during the course of receiving investigational therapy and subsequent adverse effects are observed in her offspring, these effects could be causally linked to the therapy and the sponsor held financially liable.¹⁰ Legally proving fetal harm through this causal link would present a relatively novel variety of lawsuit. While quantifying and mitigating the associated risk is certainly important from both a clinical and business perspective, the presence of liability concerns alone does not justify de facto exclusion of women.

An example of a marketed product in the field of contraceptive medical devices concerns the American Home Products Institute (AHP), maker of an implantable contraceptive. AHP settled a $50 million five-year class action lawsuit on behalf of more than 36,000 women who suffered minor side effects, such as headache, all of which were included in the product labeling.¹¹ Although AHP won several jury trials and more than 14,000 lawsuit dismissals, the cost of litigation simply became too high for the company to sustain and it settled.¹²

In the face of current legal risk and the tendency of some juries to award large settlements for adverse events claims, many sponsors feel they cannot afford to undertake the corporate responsibility associated with a potential birth defect lawsuit.¹³ However, this risk exists despite the exclusion of women. Unless the study patient population is limited to sterilized men, there is still a risk that birth defects could be connected to a study medication.¹⁴ In fact, scientific findings on a variety of birth defects such as brain cancer, tumors and other health conditions show evidence linking male subject exposure to transmission of prenatal deformities.¹⁵

This argument highlights the weakness of a zero-risk approach to legal consequences. Clinical research will never be devoid of liability; rather than avoiding women subjects, jurisprudence should be addressed by carefully crafting the informed consent document. If a disclosed and consented prenatal risk for a study manifests as an adverse event, should liability for the event extend past the consenting parent?

The fact that litigation risk works as an obstacle to women's enrollment underlies a greater issue with the modern system of risk and liability. Sponsors currently insure the risk associated with their clinical trials through purchased insurance policies, and subjects harmed in the course of testing must use the legal system to pursue their claims. A more ethical approach to mitigating risk may be implementation of mandatory compensation for research injuries.¹⁶ Since 1993, governmental committees such as FDA's Recombinant DNA Advisory Committee and the National Bioethics Advisory Commission have recommended investigation of a program that would provide a proactive mechanism for compensating research subjects for injury.¹⁷ While these recommendations did produce improvements in the informed consent regulations, no government agency has pursued the kind of mandatory compensation program that would address the inadequacies of the current legal and insurance systems. Until the compensation system addresses sponsors' current reluctance to accept prenatal risk and liability, researchers will continue to avoid enrolling women of childbearing age.

The concern over potential fetal harm is a moral issue for many investigators, who see the unknown consequences of investigational treatment as a violation of the Hippocratic Oath. This appears to be a selective point of view; however, as most women require medication during their pregnancies, the lack of data regarding fetal safety creates an inexcusable moral situation where pregnancy itself is essentially an "off-label condition."¹⁸ As a result, obstetricians must regularly prescribe based on results of limited, nonclinical fetal studies in animals. The risk associated with this practice was seen with ACE inhibitors, which were widely used for hypertension in the first trimester. A 2006 study revealed a previously unknown, statistically significant risk of fetal cardiovascular defects resulting from use of ACE inhibitors during the first 12 weeks of pregnancy.¹⁹ The moral rationale of preventing fetal harm through exclusion of women in research does not stand up to the realities of three decades worth of prenatal abnormalities caused by ACE inhibitor prescription. Fetal safety for the greater patient population is best assured through prescription of treatments with known fetal consequences. This is best achieved through established, responsible and controlled research rather than the inevitable off-label use of products that have not been studied in childbearing women.

The well-established informed consent process provides assurance that individuals are provided understandable information to decide whether to participate as subjects in clinical research. It seems reasonable that a fetus be legally protected in the same manner as a child according to the 21 CFR 50 definition, which requires obtaining informed consent for the legally underage person from the child's parent or guardian.

Lack of Information Regarding Disease Causation and Effect on the Female Body

The 2011 draft CDRH guidance indicates that a lack of understanding regarding sex differences directly leads to clinicians' under-diagnosis and under-referral of women.²⁰ This is a troubling argument because it presents the lack of understanding as both an enrollment obstacle and cause for discrimination. It follows that the solution to overcoming the barrier is resolving the lack of medical knowledge. Researchers normally leap at such knowledge gaps and craft protocols specifically to address them. Why then does the medical community continue to accept this fundamental lack of information? The answer may be rooted in the longstanding scientific principle that homogeneity of gender in research is essential to generate reliable data about therapies more rapidly and inexpensively, which is to the ultimate benefit of public health.²¹ Due to understood differences in the female body, such as hormonal variation, study data that include women subjects can skew the otherwise "perfect" all-male dataset.²² This desire for homogeneity of gender underlies the existing research paradigm where men are the ideal subject and women are the variable, undesirable "other."²³ FDA addressed this flawed rationale by reiterating that the purpose of effectiveness trials is to extrapolate data as applicable to the greater patient population. Therefore, if the patient population includes women, the data must be derived from a proportional female subject pool. However, FDA currently promotes this reasoning through guidance documents, which only reflect the agency's current thinking on a topic and are not legally binding.

In the absence of enforceable equality regulation, substantive equality should be pursued through a pluralistic view of healthcare. A recent feminist bioethics response to traditional principle-based reasoning, wherein subjects are viewed as individual, rational beings, is an ethics-of-care perspective that views subjects as representative members of the patient population.²⁴ This viewpoint allows researchers to go beyond the protocol's inclusion and exclusion criteria to identify a subject pool that is a statistically significant representation.²⁵ A predisposition for homogeneity of data or underlying preference for male study subjects could result in data outcomes that only apply to a subset of the patient population. By seeing subjects as members of their respective communities and representatives of larger ethnic and gender groups, researchers are more likely to account for the applicable social framework in their study designs and avoid a preference for the male subject.

Perception That Female Patients Require More Study Resources

Even when study protocols do not explicitly discriminate against women, researchers avoid female patients due to the perception of increased complexity and cost involved in recruiting them. This perception is not unfounded. Women are often the primary childcare providers, affecting their ability or willingness to participate in research. Primary childcare providers and pregnant women have lower tolerance for the lengthy appointments and rigorous follow-up schedules associated with research.²⁶ CDRH's draft guidance outlines study design approaches such as recruiting from women's clinics and providing flexibility in follow-ups that include childcare services.²⁷These recommendations certainly encourage participation from the patient's perspective but fail to address the hesitancy of researchers to design these provisions into their protocols.

The cardiovascular field is among the medical areas with the least representative study enrollment. Improvement is critical to women's health, as cardiovascular disease is the number one cause of death in women in the US.²⁸ Heart disease is, in fact, the cause of most deaths in both men and women, yet women are consistently under-diagnosed and under-referred for optimal healthcare. The answer to this disparity lies with the entrenched perception among providers that women are more difficult patients. Studies on the treatment of heart disease reveal that providers overwhelmingly respond to male patients' description of symptoms with follow-up referrals, while women's reports of the same symptoms are taken as whining.²⁹ The result is that the same heart disease condition is more likely to be fatal in women.³⁰ Allowing this perception of female patients to persist clearly undermines the standard of care for all women. The fact that recruiting and enrolling female patients may require more study resources does not excuse the practice of exclusion any more than it excuses the exclusion of any traditionally hard-to-reach group. Access to research is as much an ethical mandate as protections from its risks.³¹ Researchers must account for social justice and perform an honest benefit-risk assessment that factors in the potential consequences of failing to prove safety and efficacy in women (pregnant or otherwise).

Conclusion

Progress toward achieving representative study enrollment is slow due to persistent barriers. An analysis of these obstacles reveals opportunities for government intervention in the areas of mandatory patient compensation for adverse effects and revision of the US Code of Federal Regulations to include enforceable gender difference study requirements. The identified barriers also expose significant, lingering prejudice against female study subjects. These prejudices ensure that a gap in substantive data regarding the female population will remain, undermining equitable standards of care. Responsible clinical trial conduct remains the best approach for improving therapies; social justice requires healthcare and research professionals to adopt a more pluralistic concept of the study subject, assigning value to their data based on its proportional relevance to the disease population.

About the Author:

Katherine St. Martin, MS, RAC (US) is the regulatory and compliance project manager for the National Marrow Donor Program. She has more than eight years of experience in regulated industries and received a graduate degree in Bioscience Regulatory Affairs from Johns Hopkins University. St. Martin is also a graduate of West Point and a veteran. She can be reached at [email protected]. 

References

1. US Food and Drug Administration Center for Devices and Radiological Health. Draft guidance for industry and Food and Drug Administration staff: Evaluation of sex differences in medical device clinical studies. Food and Drug Administration. FDA website. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm283846.htm. Accessed 21 November 2013.
2. Ibid.
3. Ibid.
4. Foulkes M. After inclusion, information and inference: Reporting on clinical trials results after 15 years of monitoring inclusion of women. Journal of Women's Health. 20(6); 2011:829.
5. Inclusion of Women in Clinical Trials. Therapeutic Products Directorate. Health Canada website. http://www.hc-sc.gc.ca/dhp-mps/alt_formats/hpfb-dgpsa/pdf/prodpharma/womct_femec-eng.pdf; 1997. Accessed 21 November 2013.
6. Op cit 4.
7. Op cit 1.
8. Ibid.
9. Wolf S. Feminism and bioethics: Beyond reproduction. New York, New York: Oxford University Press; 1996.
10. Merton V. Ethical obstacles to the participation of women in clinical trials. Feminism and bioethics: Beyond reproduction. New York, New York: Oxford University Press; 1996.
11. Eaton ML. Ethics and the business of bioscience. Stanford, California: Stanford Business Books; 2004.
12. Ibid.
13. Ibid.
14. Op cit 10.
15. Op cit 11.
16. Ibid.
17. Ibid.
18. Lyerly A, Little M, et al. The second wave: Toward responsible inclusion of pregnant women in research. International Journal of Feminist Approaches to Bioethics. 1(2); 2008.
19. Ibid.
20. Op cit 1.
21. Op cit 10.
22. Ibid.
23. Ibid.
24. Scully JL, Baldwin-Ragaven L, Fitzpatrick P. Feminist bioethics: At the center, on the margins. Baltimore, Maryland: Johns Hopkins University Press; 2010.
25. Ibid.
26. Op cit 4.
27. Op cit 1.
28. Women and heart disease. American Heart Association website. http://www.heart.org/HEARTORG/Advocate/IssuesandCampaigns/QualityCare/Women-and-Heart-Disease_UCM_430484_Article.jsp; 2012. Accessed 22 November 2013.
29. Op cit 24.
30. Ibid.
31. Op cit 18.

Cite as: St. Martin, K. "Participation of Women in Clinical Trials: Ethically Achieving Representative Enrollment." Regulatory Focus. November 2013. Regulatory Affairs Professionals Society.