Gender-specific Regulatory Challenges to Product Approval: A Panel Discussion

Abstract

On May 13, 2014, a 1-hour panel discussion session titled “Gender-Specific Regulatory Challenges to Product Approval” was held during the Academic Emergency Medicine consensus conference, “Gender-Specific Research in Emergency Medicine: Investigate, Understand, and Translate How Gender Affects Patient Outcomes.” The session sought to bring together leaders in emergency medicine (EM) research, authors, and reviewers in EM research publications, as well as faculty, fellows, residents, and students engaged in research and clinical practice. A panel was convened involving a representative from the Office of Women’s Health of the U.S. Food and Drug Administration, two pharmaceutical executives, and a clinical EM researcher. The moderated discussion also involved audience members who contributed significantly to the dialogue. Historical background leading up to the session along with the main themes of the discussion are reproduced in this article. These revolve around sex- and gender-specific research, statistical analysis of sex and gender, clinical practice, financial costs associated with pharmaceutical development, adaptive design, and specific recommendations on the regulatory process as it affects the specialty of EM.

Scientific advances in understanding the variables that relate to health, disease, and response to treatment are becoming the focus of medical product development. These changes are becoming more relevant as customized treatments geared to the individual and emphasis on personalized medicine are approaching standard of care.

Sex and gender are powerful modulators of human development that influence disease and injury epidemiology, diagnosis, response to therapy, safety, and outcomes. Hence, sex and gender play important roles in pharmaceutical regulation, from the design of clinical trials and the approval of new drugs to postmarketing surveillance.¹ Because the United States is the leading market for the consumption of prescription drugs worldwide, the pharmaceutical industry often prioritizes its drug development and marketing strategies according to the requirements and guidelines outlined by the U.S. Food and Drug Administration (FDA).¹ The FDA evaluates clinical trials in its review of drug applications, approves drugs for marketing, and assesses how pharmaceuticals are marketed to the public.

The FDA 1977 Guidance of General Considerations for the Clinical Evaluation of Drugs essentially had the effect of industry excluding women of child-bearing potential from participation in clinical trials.² Because of the lack of participation of women in clinical trials that resulted, Congress mandated the formation of the FDA’s Office of Women’s Health, which was established in 1994.Then in 1998 the FDA issued the so-called “demographic rule” requiring all drug and biological manufacturers to provide analyses of effectiveness and safety data for women and other key demographic subgroups in marketing applications and tabulation of enrollment of subjects into clinical studies in investigational new drug annual reports.¹ Moreover, these requirements apply to marketing as well as to product labeling, which must include any sex-based differences that might influence the prescription decisions physicians and patients make.³

The significance of including and reporting sex as a critical variable in clinical studies is highlighted by the recent (2013) FDA notification to the manufacturer of a widely prescribed insomnia medication, zolpidem, to lower the recommended bedtime dose in women due to new data showing that blood levels in female patients may remain high enough in the morning after administration to impair activities such as driving. Women appear to be more susceptible to this risk due to eliminating zolpidem more slowly than men.⁴ This FDA decision is an important example of the significance of sex-specific research and highlights potential dangers of excluding sex and gender as a critical study variable.

A 1-hour discussion session titled “Gender-Specific Regulatory Challenges to Product Approval” was held during the 2014 Academic Emergency Medicine consensus conference, “Gender-Specific Research in Emergency Medicine: Investigate, Understand, and Translate How Gender Affects Patient Outcomes.” A multidisciplinary panel was assembled to address the sex- and gender-specific challenges to product development and approval. Each member of the panel brought his or her unique perspective to this discussion.

Members were as follows:

Moderators: Judd E. Hollander, MD, Associate Dean for Strategic Health Initiatives, Thomas Jefferson University, Philadelphia, PA; and Alyson J. McGregor, MD, MA, Division Director, Women’s Health in Emergency Care, Alpert Medical School at Brown University, Providence, RI

Panelists: Helen Barr, MD, Deputy Director and Senior Medical Officer, FDA Office of Women’s Health, Washington, DC; David Wright, MD, Director, Emergency Neurosciences, Emory University School of Medicine, Atlanta, GA; and Peter Wildgoose, PhD, Senior Director of Clinical Development, Janssen Pharmaceuticals, New Brunswick, NJ.

In addition, clinical and academic perspectives were sought through an active discussion with the audience that comprised academic emergency physicians, nurses, patients, PhD researchers, and trainees. A full description of the audience is described in the executive summary.⁵ The following are adapted summaries from the session:

SEX- AND GENDER-SPECIFIC RESEARCH

Currently there is an unmet need in the area of sex- and gender-based research.⁶ One of the responsibilities of academia within emergency medicine (EM) is to raise the level of interest of a particular research area or focus that is considered significant and essential to the practice of emergency care. If the specialty as a whole concentrated on discovering novel sex- and gender- specific mechanisms, it could attract pharmaceutical companies and lead to investment in bringing novel products forward to the public. This is the evolution of how academia interacts with pharmaceutical companies. The primordial basic science research that needs to be conducted to discover sex and gender differences is very costly and often low yield, thereby leaving the early research up to academia. Pharmaceutical companies would advance novel treatments for both men and women, or for one particular sex, if the potential to generate income and improve health outcomes were demonstrated.

The FDA’s focus is to approve safe and effective medical products. To do this, the different demographic groups that will use the product need to be represented in the clinical trials that are submitted to the FDA as part of the application for product approval. As part of the FDA Safety and Innovation Act (FDASIA), Congress recently required the FDA to publish a report on the inclusion and analyses of demographic subgroups in FDA applications and the public availability of such information.⁷ The report concluded that, in general, sponsors are describing the demographic profiles of their clinical trial participants and the majority of applications submitted to FDA include demographic subset analyses.

The public availability of this information varied across the different centers within the FDA. Whites represented a high percentage of clinical trial study participants for biological, drug, and medical device applications, in some cases underrepresenting other racial groups. The FDA opened a docket for comments and held a public meeting to receive comments about the report and is developing a follow-on action plan. The primary message from the FDA was that it alone cannot increase demographic diversity in trials, employ novel clinical trial designs and analyses for subpopulation differences, and advance sex and gender research. The FDA’s partners in the solution are industry, advocacy, academia, and sister agencies like the National Institutes of Health (NIH). Patients and prescribers are demanding confidence that the medical products they are using are safe and effective in all demographics of the population who will use the product.

STATISTICAL ANALYSIS OF SEX- AND GENDER-BASED STUDIES

The FDA requires clinical trial analysis to be performed by sex, age, and race. Sometimes these are secondary analyses and may not include descriptive statistics. At times, the testing and analysis may not reveal sex and gender differences for a specific indication. Sometimes, after a drug has been on the market, the manufacturer will request a new indication for the drug and the information for a new indication brings to light a sex difference previously unrecognized or raises the threshold for a difference previously thought to be of no or little clinical significance. For instance, analysis of clinical trial results of the drug zolpidem was initially performed many years ago; however, newer extended release formulations revealed higher serum levels in women of the active drug the following morning. This example highlights the importance of not only performing the correct analysis to determine if a sex and gender difference exists, but also reflecting that information in appropriate labeling recommendations. Another important caveat to investigating previously accepted data analysis is that thresholds for serum drug concentrations and the determinacy of safety and efficacy can change, rendering a previously acceptable drug outside of either the therapeutic or safety window.

CHANGING CLINICAL PRACTICE IN THE ED

One of the duties of the FDA is to make sure that the manufacturer’s proposed indication(s) for medications are safe and effective in the intended use population. There are examples of drugs that have been studied in both sexes but in the end are labeled as indicated for only one sex. One example is Lotronex (alosetron HCl), a medication indicated for some women with irritable bowel syndrome. Another example is with Oxytrol (oxybutynin transdermal), a medication for women to treat overactive bladder that is available without a prescription. Physicians need to be aware of possible sex differences in drug labeling. E-mails, smart phone applications, computer screen reminders, and hospital-based protocols are possible methods that can assist in disseminating this information.

Another area of possible sex differences for physicians to consider is determining sex-specific normal ranges for diagnostic tests. For instance, determining baseline levels of the cardiac biomarker troponin by sex will improve the specificity of this diagnostic tool, avoiding misclassification of patients.⁸ Additional reference ranges that differ by sex include those for calcium, creatinine, uric acid, hemoglobin and hematocrit, and cholesterol.⁹ However, separate reference ranges by sex are often not used in clinical practice despite statistically significant differences reported. Having these differences incorporated as part of clinical protocols or highlighted in reported results would help clinicians better apply the findings.

The clinical practice that seems to be most currently adaptive to the inclusion of sex and gender is within risk factor profiling. Most emergency physicians present during this consensus conference agreed that they consider sex and gender when ordering certain tests, such as erythrocyte sedimentation rate in women, as they most often suffer from autoimmune conditions. Many consider sex and gender as a significant factor in the presentation of disease but currently not in the interpretation of test results or treatment. Other areas for sex-based research could also be around control parameters. For example, should women have the same blood pressure or diabetes control parameters as men?

The examples brought forth suggest that work is being done on the part of biomedical organizations with respect to inclusion of men and women to determine differences in serum concentrations and reference ranges and by the FDA around some sex-specific drug indications and dosing recommendations. The question remains as to what is considered clinically significant and how physicians can use this information to change the practice of EM.

COST OF SEX AND GENDER RESEARCH IN PHARMACEUTICAL DEVELOPMENT

Despite the desire and need for sex- and gender-specific research in pharmaceutical development, there is a substantial financial cost that must be considered. Most of the expenditure is not necessarily in laboratory drug development but in the cost of the clinical trials. For instance, a pharmaceutical company secures a patent for drug development which expires on a particular date. Powering clinical trials for safety differences would require massive studies that would not be feasible and would stifle development and innovation due to increased time and money spent in enrollment and conducting the trial itself. This results in less time prior to the patent expiration for development, promotion, and financial gain for the company. Ultimately this cost is relayed to the consumer. Therefore, in some cases, novel ways of designing and analyzing clinical studies are needed to detect subpopulation differences. Additionally, initiatives to increase the enrollment of women and other demographic groups in clinical trials are needed. Pharmaceutical Research and Manufactures of America (PhRMA) recently announced an initiative to increase the enrollment of racial and ethnic minorities in clinical trials. Although in general women are included in numbers sufficient for the FDA to make decisions about safety and efficacy of products, the previously described FDASIA report did find some underrepresentation of racial and ethnic groups in trials submitted to the FDA as part of product applications. Also, even larger numbers are needed when looking for effect modification by sex in efficacy and adverse events of a drug. It also stands to reason that the more women and the more diversity in general in clinical trials, the more likely it is that significant differences will be discovered or that the confidence level will rise that there are not significant differences.

ADAPTIVE DESIGN

In 2010, the FDA issued guidance on adaptive trial design to modify the way drugs travel from the laboratory to the market.¹⁰ This initiative evaluates patients’ reactions to a drug early in a clinical trial to modify the trial in accordance with those findings. This adaptation process includes dosage, sample size, and patient selection criteria. The aim is to quickly identify drugs that have a therapeutic effect and to define patient populations for whom the drug is appropriate.

Adaptive design can aid in a smooth transition from Phase 2 to Phase 3, going from a largely proof-of-concept phase to safety and efficacy determination. If needed, there can be adjustments in Phase 3, which are usually done around studying efficacy. A case in point is the discovery phase of neuroscience translational work. Often sex differences are not intended as a result of working with male animals or animals whose sex is not identified in publications. There is a move toward having standards for incorporating sex differences throughout the entire translational continuum. The acceptance of an evolving study design is crucial to preventing the current issue of drugs having to be withdrawn from the market due to side effects against women.

Guidance for FDA reviewers issued in December 2013 has them engaging with the sponsors at each phase of the application and review process, including trial design and analysis to help ensure that people are not arbitrarily excluded from clinical trials around age cutoffs or coexisting conditions, and that the trials contain representation of the different demographics who will be using the drug. There is a balance that the FDA is trying to achieve by ensuring demographic diversity without stifling innovation.

HOW FAR DO WE GO?

After this consensus conference on inclusion of sex and gender research into EM, how far should we as a specialty go? This issue of AEM is dedicated to the publication of the proceedings from the breakout group sessions, panel discussions, executive summaries, and a call to researchers for manuscripts that have significance within sex- and gender-specific emergency care research. What happens after the dust settles? Should editors mandate that each study design, clinical trial, or preclinical research manuscript comment on sex and gender differences that may or may not have pertained to the results of that study? This would have more of an effect than simply listing in a sample design table how many participants were men and women. Although the discussants agreed that this has the potential to affect study design, it may also result in underpowered analyses. The challenge lies in promoting the value of conducting studies to adequately determine if sex and gender differences exist.

Recommendations put forth by the group include the following:

Clinical Trial Methodology

It is clear that to determine drug effect modifications regarding sex and gender, studies will require more advanced analyses than simple secondary gender comparisons. To determine the magnitude of the effect that a drug may have on men or women, there needs to be large enough clinical trial enrollment to detect these differences.

Data from 2006 for the 15 largest pharmaceutical companies show that the majority of patient visits associated with an industry-sponsored clinical trial occur outside of the United States and speaks to the cost and difficulty associated with conducting clinical research in the United States.¹¹ Investigators from the United States enroll two-thirds as many subjects into clinical trials as investigators in the rest of the world. Among U.S. investigators participating in a clinical trial, 27% fail to enroll any subjects, compared with 19% of investigators elsewhere. Seventy-five percent of investigators fail to enroll the target number of subjects worldwide.¹¹ Enrollment is one of the largest issues faced by the pharmaceutical industry today. The effort and time spent by the principal investigator has a significant effect on educating each patient and securing the enrollment; however, due to the changing landscape within the health care industry, less time is available for this type of patient interaction. The FDA is also trying to engage the public by informing them that participation in clinical trials is the mechanism by which safe and effective medical products reach the marketplace.

One example is with the ProTECT Trial for the evaluation of progesterone for treatment of traumatic brain injury. This research study, spanning 50 sites, was able to garner successful enrollment due to the commitment of the individuals within that research network. Another reason for the success is due to the Neurological Emergencies Treatment Trials (NETT) Network, created by the NIH to assist in conducting large trials for the treatment of neurological emergencies with a narrow treatment window. Increasing clinical trial enrollment is the business of everyone—academia, industry, government, and advocacy. Employing best practices around recruitment and retention using known successes as models is needed.

Postmarket Surveillance

There are numerous ways to conduct postmarket surveillance, the practice of monitoring the safety and effectiveness of a pharmaceutical drug or medical device after it has been approved for marketing. Clinical trials often involve relatively small numbers of men and women in relation to the general public. Postmarket surveillance can further refine the safety of a drug in people with a wide variety of medical conditions. Postmarket studies can also be done voluntarily by industry to examine such things as sex or other demographic differences, and the FDA can also require postmarket studies to be done. The FDA also has a voluntary reporting system called MedWatch where physicians and patients can report adverse drug reactions. Both the FDA and the pharmaceutical companies use other surveillance mechanisms, existing disease and pregnancy registries, and social media to detect safety signals.

EM Registries

Emergency medicine is looking for ways to establish registries that provide information on reportable quality measures adhered to by our specialty. For instance, the American College of Cardiology maintains the comprehensive National Cardiovascular Data Registry with more than 2,400 hospitals and nearly 1,000 outpatient providers participating.¹² This is a high benchmark to achieve, with potential for high financial cost that would likely be deferred to the members within the EM professional society. Financing registries is also debated by the members of the National Quality Forum as they consider potential costs of mandatory reporting. Currently, government agencies do not provide funding and costs are deferred to those requesting the data, whether that be the academic researcher, hospitals, or even independent outpatient physicians who inquire about their own quality data. The challenge for us would be to ensure that the information collected is robust enough to be useful.

CONCLUSIONS

It is our hope that we have shed some light regarding the role and performance of sex- and gender-specific research as it relates to the development, testing, and approval of drugs for marketing. Although the sex of trial participants must be reported in FDA applications and analyses performed for possible sex differences, more women in trials will hopefully garner more information about sex differences in response to drugs. Additional initiatives are needed to increase the diversity in clinical trials so that trial populations resemble those who will be taking the drug, without causing undue cost or delay in new drug development. Similarly, issues pertaining to dissemination of the sex-specific dosages and laboratory references ranges were raised, which are areas ripe for research. Examining sex differences by using cells and animals of both sexes is needed to increase the knowledge around sex differences in therapeutic responses and disease manifestation, and peer-reviewed journals should consider requiring the disclosure of this information from its authors. In the end, we identified more questions than we answered, but only through discussion of these important issues are we likely to ultimately find the answers we seek to improve the care of both men and women.