3

The Legal Landscape

NAVIGATING LEGAL BARRIERS AND LIABILITY RISKS

“There is a deeply entrenched myth that it is legally impermissible to include pregnant and lactating people in research,” asserted Leslie Meltzer Henry, professor of law at the University of Maryland Carey School of Law and core faculty at Johns Hopkins Berman Institute of Bioethics. There are legal and ethical ways to conduct this research, but she said that outreach and education are needed to dispel this myth. The legal landscape is nuanced and complex, and legal barriers—real and perceived—still impede the inclusion of pregnant and lactating persons in clinical trials. Henry said that human subjects research regulations and liability concerns are often cited as reasons why researchers avoid including pregnant and lactating persons in clinical studies.

There are several legal regimes that shape the conduct of research with pregnant and lactating persons. Henry highlighted three areas that demonstrate some of the complexities that clinical researchers must navigate. First, pregnancy and lactation are each addressed differently in the law. The regulations for the protection of human subjects in research address research involving pregnant persons, fetuses, and neonates—as addressed

in 45 Code of Federal Regulations (CFR) 46 Subpart B¹—but do not address research involving lactating persons. Liability for harm varies relative to whether the exposure resulting in harm occurred prenatally or postnatally. Next, small molecule drugs and vaccines are regulated under different legal regimes.² Liability schemes are also different. Henry explained that a private party is generally liable for issues associated with a drug product, while there is a federal compensation program for injuries associated with vaccine use.^3,4 Lastly, regulation of the research, development, and approval processes varies according to whether products are intended for routine use versus use in a public health emergency. For example, the U.S. Food and Drug Administration (FDA) can issue an emergency use authorization to allow the use of an unapproved product during a public health emergency, as was done during the COVID-19 pandemic.⁵ Liability schemes are also different for routine versus public health emergency use. Henry explained that there is immunity from liability associated with medical countermeasures used in response to a public health emergency.

Based on a series of consultations with legal experts, Henry and colleagues found that “lawyers and regulators have the potential to be involved in decision-making that influences the conduct of research at every point throughout the research pathway,” including decisions around the inclusion of pregnant persons in studies. They identified five legal factors that can be obstacles to the inclusion of pregnant persons in clinical research (Mastroianni et al., 2017):

• Regulatory ambiguities. There is no regulatory language in 45 CFR 46 Subpart B that prohibits the conduct of research with pregnant persons, Henry said. There is also no stated requirement to include pregnant persons, and no penalty for excluding them. There is ambiguity surrounding what constitutes “minimal risk” to the fetus, so this leaves the regulatory language open to interpretation. Those interpretations are usually cautious owing to liability concerns, she said.

___________________

¹ 45 CFR 46 Subpart B, Additional Protections for Pregnant Women, Human Fetuses and Neonates Involved in Research.

² Products such as biologics and devices are also regulated under different legal regimes.

³ See https://www.hrsa.gov/vaccine-compensation (accessed September 12, 2022).

⁴ A new Rule took effect in January 2022 that included vaccines recommended for routine administration in pregnant persons to the vaccines covered by the National Vaccine Injury Compensation Program. For more information, see https://www.federalregister.gov/documents/2021/12/02/2021-26197/national-vaccine-injury-compensation-program-adding-the-category-of-vaccines-recommended-for (accessed November 6, 2022).

⁵ See https://nap.nationalacademies.org/read/26441/chapter/1 (accessed September 22, 2022).

• Perceptions of liability. “Liability is frequently cited as the single most common reason that decision makers exclude pregnant people from research,” Henry said. However, based on the available evidence, “stakeholders’ fear of liability seems to be outsized compared to the actual risk.” While there is a risk of harm to pregnant trial participants and their fetuses in the course of clinical research, adverse events are relatively infrequent and limited by the size of the study population, remarked Henry. There are strategies to mitigate the risk of liability associated with these events (e.g., the informed consent process, indemnity or trial insurance, a plan for how research-related injuries will be compensated). Conversely, the risk of liability may increase considerably once a drug is approved and used in the general population if adverse events occur in subpopulations who might have been excluded from studies. Henry said fear that including pregnant persons in clinical trials will lead to “another thalidomide” is misplaced. She pointed out that the thalidomide crisis occurred because pregnant people were not included in clinical studies of the drug before it was widely used in pregnant persons. “We might actually be engaging higher risk levels when we do not include pregnant people in research than if we were to include them and have a smaller pool of risk to manage and mitigate,” she said.
• Risk management challenges. “Lawyers are…experts at risk management [and] should thoughtfully engage in risk mitigation,” Henry said. The Task Force on Research Specific to Pregnant Women and Lactating Women (PRGLAC Task Force) Implementation Plan discusses a range of potential strategies for mitigating the liability risk associated with research-related injuries. For example, there are different options for indemnity programs (e.g., trial insurance, self-indemnity), and she noted that compensation for research-related injuries is done in most other countries.
• Risk shifting and the learned intermediary doctrine. Fear of liability is stoked by the application of the “learned intermediary doctrine” in the courts, which Henry explained “shifts the [legal] risk from the pharmaceutical company downstream to the health care provider.” This doctrine asserts that it is the clinician’s responsibility to advise the patient of the potential health risks of taking a drug, and therefore the clinician is liable if the patient experiences an adverse event they were not warned about. Henry said this “is a heavy burden to bear when you do not have the requisite evidence to make that kind of disclosure to your patient.”
• Venue-specific laws. Each state has laws addressing a range of harms to the fetus (e.g., fetal homicide charges if pregnant person

• miscarries because of a car accident). Henry said that concerns about how these laws might be interpreted and applied in the context of a clinical trial can affect decisions to include pregnant persons in studies.

Regulatory Barriers to Inclusion

The Common Rule no longer includes pregnant persons as a potentially vulnerable population with regard to consenting to clinical trial participation. However, Maggie Little, Georgetown University, said there are concerns that 45 CFR 46 Subpart B, Additional Protections for Pregnant Women, Human Fetuses, and Neonates Involved in Research, still presents regulatory barriers to conducting clinical trials in pregnant and lactating individuals.⁶ One of these, as mentioned by Henry, is the ambiguity around the term minimal risk to the fetus. Little said that the PRGLAC Task Force report includes a recommendation to modify 45 CFR 46 Subpart B to reduce these barriers as they relate to the minimal risk standard for clinical trials and the requirements for parental consent. She cautioned against focusing exclusively on regulatory solutions, which usually require legislation to effect. Regulations are not generally significant barriers for the conduct of research, Little continued, and 45 CFR 46 Subpart B, while not perfect, is based in common sense. The most significant barriers to research with pregnant and lactating persons are related to culture, liability, and capacity, she said.

Lessons Learned from Pediatric Research

To inform the discussion of legislative solutions to increasing the inclusion of pregnant and lactating persons in clinical trials, a brief history of the Best Pharmaceuticals for Children Act (BPCA) and the Pediatric Research Equity Act (PREA) was provided by Lynne Yao, director of the Division of Pediatrics and Maternal Health at FDA (summarized in Box 3-1).⁷

BPCA was first introduced as the Better Pharmaceuticals for Children Act and was passed with bipartisan support as part of the Food and Drug Administration Modernization Act (FDAMA) in 1997. An updated version of BPCA—the Best Pharmaceuticals for Children Act—was introduced in 2001 and signed into law in 2002. Yao pointed out that the 2001 BPCA was passed by unanimous consent in the Senate and by voice

___________________

⁶ See https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/common-rule-subpart-b/index.html (accessed September 6, 2022).

⁷ Yao noted that the opinions expressed are her own and do not necessarily represent an official position of FDA.

vote in the House, which she said is an indication of the broad bipartisan agreement on the importance of the measure. PREA was introduced in the Senate in 2002, and again in 2003 with bipartisan sponsorship in both the Senate and the House. PREA was also passed by unanimous consent in the Senate and by voice vote in the House. Yao emphasized the role of bipartisan support and collaboration in passing both BPCA and PREA, and she noted that the environment in Congress is currently quite different, suggesting that similar legislative remedies for pregnant and lactating populations may be more challenging and require sustained collaboration.

The key difference between PREA and BPCA, Yao explained, is that PREA gives FDA the authority to require pediatric studies for certain products, while BPCA is voluntary, and offers incentives for sponsors to conduct pediatric studies. She shared data on pediatric labeling changes from 1998 through 2020 which indicate that just over half of the 854 labeling changes stemmed from PREA (i.e., studies were required), 22 percent were associated with BPCA incentives, and 14 percent were associated with both.

PERSPECTIVES ON LIABILITY

Clinician Perspective

Carmen Zorrilla, professor of obstetrics and gynecology in the University of Puerto Rico School of Medicine, shared her perspective as a practicing clinician and researcher. She highlighted three common sources of drug-related risk during pregnancy and the patient perceptions behind them. In her experience, she said that newly pregnant persons often discontinue a drug they are taking for an existing condition because they perceive it to be harmful during pregnancy. However, as discussed, there can also be risks associated with not continuing to treat the existing condition. Another source of risk is when a pregnant person does not take a new drug prescribed to them during pregnancy, again because of perceived potential harms. This risk may be compounded if the provider attempts to manage the patient’s complications while under the impression they are taking the drug as prescribed. Taking over-the-counter (OTC) drugs or supplements is also a source of risk during pregnancy, both directly and as a result of drug interactions. Many patients do not consider OTC products to be drugs and do not perceive any risk. The underlying issue that needs to be addressed for each of these sources of risk is a lack of patient knowledge about the safety of these products and their effect during pregnancy. Zorrilla said that providers need to have conversations with their patients about what is known about a drug and the provider’s

recommendation based on their experience and expertise (e.g., the low likelihood that a larger molecule drug could cross the placenta).

Cost-containment strategies that affect access to treatments during pregnancy present a risk. Zorrilla said that, in Puerto Rico, insulin (for diabetes) and labetalol (for hypertension) must be prescribed by an endocrinologist or a cardiologist, respectively, to be covered by insurance companies. She noted that these specialists are part of the pregnant person’s care team, but it is the obstetrician/gynecologist (OBGYN) who has the expertise on treating these conditions during pregnancy.

There can also be liability concerns associated with pregnancy drug registries and drug information databases. Zorrilla said these resources can be “extremely useful,” especially when providers enroll their patients prospectively (e.g., the Antiretroviral Pregnancy Registry). However, when patients enter their own data, Zorrilla cautioned that there may be biases and misinterpretation of the information introduced to the database. She mentioned the Vaccine Adverse Event Reporting System (VAERS)⁸ as an example of a database in which anyone can report any outcome they experience after receiving a vaccine.

Jeff Roberts, associate vice president of vaccine clinical development at Merck Research Laboratories and former associate director for scientific affairs in the Office of Vaccine Research and Review at FDA, reflected on a trend in clinicians’ liability perceptions—that other providers are often uncomfortable treating a pregnant person without engaging an OBGYN (e.g., when a pregnant person presents in the emergency department for a nonpregnancy-related issue, they call for an OBGYN). He suggested that, similarly, clinical investigators who are not OBGYNs might be uncomfortable enrolling and managing pregnant persons in a clinical trial. Roberts suggested that involving OBGYNs in clinical trials, even when not specifically designed to evaluate pregnant individuals, could begin to remove barriers to inclusion of this population.

Research Perspective

Jessica Cohen, director of the Office of Research Affairs at PATH, discussed a range of questions and concerns associated with liability risk for researchers across the clinical research continuum.

• Satisfying regulatory requirements. Questions about regulatory requirements include what evidence regulatory bodies will consider to be sufficient to advance the development of a drug for use in pregnant persons, and what data from nonpregnant popu

___________________

⁸ See https://vaers.hhs.gov (accessed September 12, 2022).

• lations are needed before studies can include pregnant persons, Cohen said.
• Securing funding. There are also questions about the extent to which funders will support the costs and timelines associated with the expanded regulatory pathway for including pregnant and lactating persons in studies.
• Participant safety. Is there sufficient evidence of safety (e.g., from preclinical developmental and reproductive toxicology studies) for clinical investigators to be willing to enroll pregnant or lactating persons in a trial?
• Managing injury and adverse events. Managing liability claims will vary based on where the research with pregnant persons is being conducted. Cohen said it is important to understand “the statutory requirements and timeframes for bringing claims in a specific setting…and how this will impact the investigator and the institutional responsibilities, budgets, [and] insurance requirements.”
• Partnership complexity. Cohen said there are many complexities to managing large, multistakeholder research partnerships, including the need to build stakeholder alignment on values, norms, and risk thresholds.
• Reputational risk. Cohen observed that there can be a “heightened sensitivity” to certain research involving pregnant or lactating persons (even though they are no longer designated as “vulnerable” in regulations), and there can be reputational risks to researchers and institutions when problems occur in a study involving pregnant or lactating persons.
• Perceptions of the ethical challenges and complexities. Cohen added that the “years of labeling pregnant [persons] as vulnerable populations and the perception that these studies will intrinsically be more complex” has had a “chilling effect” on research. These perceptions persist and can result in barriers such as challenges or delays from institutional review boards (IRBs) reviewing proposed studies in pregnant and lactating persons.
• Timeline for sponsor or researcher liability. There are questions surrounding when liability ends. Research including pregnant or lactating persons involves both the pregnant or lactating person and child, and can also affect other infants, for example if the lactating person is a breast milk donor. From a liability perspective, “Understanding the possible long-term safety impacts of research interventions is critical,” Cohen said.
• Safety surveillance systems. Tracking and analyzing adverse outcomes associated with the use of vaccines and therapeutics by

• pregnant and lactating persons around the world is challenging because of variability across surveillance systems (e.g., lack of standardized data) and limited availability of surveillance systems in many low- and middle-income countries. Cohen said this limits the ability to compare data across regions and to analyze rare events relative to baseline data from before a product was introduced. Uncertainties about the coverage of a vaccine under the vaccine injury compensation program, and the determination of eligibility for compensation, also create liability concerns for researchers.

Industry Perspective

Aviva Wein, assistant general counsel at Johnson & Johnson, discussed product liability risk from the perspective of a medical product sponsor.⁹ “Pharmaceutical companies develop, manufacture, [and] sell products all with the hope that they will help change the course of a disease, prevent another disease, change a life, and maybe even save a life,” Wein said. The primary concern for a company conducting a clinical trial is protecting trial participants while taking “informed and necessary risks” to gather sufficient evidence to bring a product to market and enable providers to make informed decisions for their patients, she said. Wein emphasized the importance of getting products to those who need them without delay and pointed out that “adding complex populations to those trials is a consideration.” While the risk of liability is a consideration pre- and postapproval, she said that any hesitancy to include pregnant and lactating people in trials is not attributable solely to liability risk.

Real and perceived liability associated with research involving pregnant and lactating persons exists within the larger context of the U.S. liability landscape. Wein said that, prior to 2007, most liability litigation was prompted by media attention to a safety-related action by FDA, such as a product recall or a label change. In more recent years, lawsuits have been filed before any action on the part of FDA, which were based on safety concerns detected by mining big data sources or shared on social media. In some cases, it is litigation that has led to FDA analysis of an adverse event. This litigation trend in the United States has led to “an avalanche of claims,” which she said are based on weak evidence before FDA has had an opportunity to investigate the safety concerns and take action on a particular drug product. Wein explained that mass tort liability litigation in the United States has become a “multibillion-dollar business” that advertises ongoing liability lawsuits to the public to increase the number

___________________

⁹ Wein noted that the opinions expressed are her own and do not necessarily represent an official position of Johnson & Johnson.

of claims and extract huge settlements from companies. In this litigious ecosystem, product claims involving pregnant and lactating persons in clinical trials will be “another fertile area for plaintiff lawyers to expand their business model,” she said.

Wein acknowledged that the emotions surrounding research with pregnant and lactating persons can lead to a greater perception of risk. However, “We live in a world of risk,” she said, and research with pregnant and lactating persons is simply another area of risk for which mitigation strategies are needed. Different stakeholders in clinical research have different levels of risk tolerance and ability to manage liability, and she said that expanding inclusion of pregnant and lactating persons in clinical trials “will require all parties to confront their fears of liability for injuries.”

INSTITUTIONAL POLICIES AND PRACTICES THAT PROMOTE INCLUSION IN RESEARCH

Anna Mastroianni, Charles I. Stone professor of law at the University of Washington (UW) School of Law, described some of the findings of a qualitative study to identify institutional strategies that can facilitate research with pregnant persons. The study was based on interviews with investigators, IRB staff members, university administrators, and ethicists at UW who had been involved in clinical studies that included pregnant participants. She noted that UW is a major academic research institution with a track record of including pregnant participants in clinical research, which stems from an institutional commitment that is based on a presumption of inclusion rather than exclusion. Mastroianni quoted from the UW Human Subjects Division Standard Operating Procedures, which states that UW

Mastroianni shared several insights from the study (Mastroianni et al., 2020). From a legal perspective, “Ambiguity offers opportunities for bounded interpretation,” Mastroianni said. Recognizing the flexibility that the regulatory ambiguities in 45 CFR 46 Subpart B afforded, UW developed formal policies and practices to promote clarity and consistent implementation of the institution’s interpretation of the regulations. Among these were:

• Readily accessible web-based standard operating procedures (SOPs) and worksheets;

• Training of IRB staff and committee members on regulatory definitions and risk assessment (including risks associated with not conducting the research); and
• Accessibility of IRB staff for consultation prior to protocol review and throughout the trial.

Mastroianni noted that the UW IRB review of research involving pregnant participants routinely includes an obstetrician, and that the transparency and consistency of the IRB process has helped build a trusting relationship between the investigators and the IRB.

Another enabling factor identified by the study is the UW Human Subjects Assistance Program, a self-funded compensation program that provides medical and other assistance to participants who incur injuries that are “more likely than not caused by UW-conducted research.” Mastroianni said study respondents felt the compensation program also protected researchers, and information provided indicates that “UW has fewer research-related lawsuits and tort claims than comparable institutions.” Mastroianni added that a multipronged approach is required to overcome the legal hurdles impeding the inclusion of pregnant and lactating persons across the clinical trial continuum, she said.

Cohen described some of the elements of PATH’s institutional research oversight framework that help mitigate risks associated with human subjects research. First, the scientific merit review process requires that protocols include sound scientific rationale for the inclusion or exclusion of specific populations in the study, and PATH is developing new guidance on equity and justice in research. PATH has a policy on study-related injuries, which requires studies to secure appropriate liability insurance and have a plan for how treatment of study-related injuries will be delivered and paid for. There is guidance for developing consent forms, which discusses the inclusion of readable and understandable information about how to get help for a study-related injury. All studies in which participants might incur greater than minimal risk are required to have a crisis communication plan with strategies to address real or perceived concerns about a study. PATH requires that there be IRB review in the country where the study is being conducted. Cohen explained that this helps ensure compliance with local regulatory requirements and that local contextual factors guide risk–benefit analyses. Cohen added that PATH has a master insurance policy that covers all of its human subjects research worldwide. The master global policy does not have any limitations regarding birth control or pregnancy during a study and includes no-fault medical payment coverage for clinical trials in the event of an injury. PATH will also incorporate locally admitted policies where needed and/or required, and the master global policy will be excess over any locally placed policy.

REGULATORY EFFORTS TO PROMOTE INCLUSION IN RESEARCH

“FDA is committed to advancing research in pregnant and lactating people,” stated Catherine Sewell, a board-certified OBGYN and acting deputy director and deputy director for Safety in the Division of Urology, Obstetrics, and Gynecology at FDA.¹⁰ She cited examples of the agency’s involvement in recent national efforts, including the PRGLAC Task Force; a workshop on scientific and ethical considerations for including pregnant people in clinical trials that was coconvened with Duke-Margolis; and a conference on using real-world data for maternal health coconvened by the FDA Office of Women’s Health and the Johns Hopkins University (JHU) Center of Excellence in Regulatory Science and Innovation (CERSI).

Regulatory Considerations for Inclusion of Pregnant and Lactating Persons

In recent years, FDA has issued several Guidance for Industry documents addressing regulatory issues pertinent to drug research and development involving pregnant and lactating persons. The guidance, Pregnant Women: Scientific and Ethical Considerations for Inclusion in Clinical Trials, discusses issues such as balancing the risks and benefits of continued trial participation for a person who becomes pregnant during a trial.¹¹ The Postapproval Pregnancy Safety Studies guidance addresses the design of required postmarketing studies such as prospective cohorts (i.e., registries) and complementary studies to close gaps in data.¹² Other relevant guidances mentioned by Sewell address the design of clinical lactation studies and increasing diversity in clinical trials.¹³ Additionally, FDA has worked to better harmonize FDA regulations with the Common Rule by removing references to pregnant persons as a vulnerable population in this context.¹⁴

___________________

¹⁰ Sewell noted that the opinions expressed are her own and do not necessarily represent an official position of FDA.

¹¹ See https://www.fda.gov/regulatory-information/search-fda-guidance-documents/pregnant-women-scientific-and-ethical-considerations-inclusion-clinical-trials (accessed September 6, 2022).

¹² See https://www.fda.gov/regulatory-information/search-fda-guidance-documents/postapproval-pregnancy-safety-studies-guidance-industry (accessed September 6, 2022).

¹³ See https://www.fda.gov/regulatory-information/search-fda-guidance-documents/clinical-lactation-studies-considerations-study-design, and https://www.fda.gov/regulatory-information/search-fda-guidance-documents/enhancing-diversity-clinical-trial-populations-eligibility-criteria-enrollment-practices-and-trial (accessed September 6, 2022), respectively.

¹⁴ After the workshop, FDA released a Proposed Rule that would remove pregnant people as a vulnerable population. See 87 FR 58733 (August 28, 2022).

FDA Efforts to Advance Data Collection from Pregnant and Lactating Persons

Sewell discussed FDA’s proposed commitments regarding pregnancy safety for the forthcoming reauthorization of the Prescription Drug User Fee Act. Specifically, she said that by September 30, 2023, FDA will host a “public workshop on postmarketing safety studies in pregnant people,” by September 30, 2024, the agency will “publish a workshop report describing the proposed framework,” and by September 30, 2027, FDA will “update the proposed framework and develop a guidance for a standardized process in determining the necessity and type of pregnancy postmarketing studies.” The framework will incorporate the use of Sentinel, FDA’s medical product safety surveillance system.

Another effort involves expanding CURE ID,¹⁵ which is a data repository for information on how clinicians have repurposed existing drugs to treat infectious diseases. The CURE Pregnancy Treatment Repository will capture data on the use of repurposed drugs for the treatment of infectious diseases in pregnant persons. This expansion will help identify areas where clinical studies are needed to address treatment needs during pregnancy, Sewell said.

Roberts described FDA’s efforts to increase data collection from pregnant persons during the development of COVID-19 vaccines.¹⁶ The first FDA guidances addressing the development of COVID-19 vaccines encouraged sponsors to remove the requirement for contraception (after conducting reproductive toxicology studies to provide adequate support for the safe initiation of studies in pregnant persons) from their phase 3 clinical trial protocols because, he said, once the products are broadly marketed, “Women are going to be vaccinated before they are aware that they are pregnant.” The agency also encouraged sponsors to conduct reproductive toxicity studies early in development and to conduct specific studies in pregnancy; however, these were not done prior to phase 3 studies. He pointed out that this is not specific to the development of the COVID-19 vaccines. Reproductive toxicity studies are conducted late in the product development pathway, and studies specific to pregnancy are generally conducted after a product is approved. In vaccine development in general, “Pregnancy is an afterthought,” he said.

___________________

¹⁵ See https://cure.ncats.io/home (accessed September 12, 2022).

¹⁶ Roberts spoke from his regulatory experience, and the opinions expressed do not represent official positions of Merck & Co., Inc.

Opportunities for Collaboration with Regulators

Sewell noted that FDA can engage in discussions with IRBs, researchers, and product sponsors about “making the inclusion of pregnant people the default where appropriate” and justifying exclusion. The agency can also publicize priority therapeutic areas where data in pregnant and lactating persons are most needed. Discussions are also ongoing about creating targeted incentive programs and requiring sponsors to develop a study plan for inclusion of pregnant and lactating persons.

As a result of the failure to collect data on pregnant and lactating persons during the COVID-19 vaccine trials, Roberts said that the Office of Vaccines at the FDA Center for Biologics Evaluation and Research (CBER) has been considering regulatory options, such as asking sponsors to include a pregnancy and lactation plan in the product development plan. For example, in a pre-investigational new drug meeting, a sponsor might discuss how it anticipates its product would be used during pregnancy or lactation and, accordingly, how its product development plan will enable the collection of safety and effectiveness data to support use in these populations. Or, as discussed by Sewell, a sponsor might justify why pregnant or lactating persons would not be included in a particular study (rather than exclusion as the default development plan).

ADDRESSING LEGAL AND REGULATORY CONCERNS

Taking the First Steps

Panelists discussed some of the practical first steps to addressing liability concerns that impede the inclusion of pregnant and lactating people in trials. Mastroianni emphasized that lawyers manage risk on a daily basis. A first step could be to engage lawyers as partners in the study development process and prioritize focusing on ways to address the liability risks of including pregnant and lactating persons in clinical trials, Mastroianni said.

Mastroianni highlighted the need to learn from institutions that are already conducting research that includes pregnant and lactating persons. The University of Southern California, for example, established the Flexibility Coalition, a group of research organizations that share approaches on how to effectively use the flexibility within federal human subjects research regulations and how to mitigate risk. She emphasized that researchers, participants, and other stakeholders throughout the research pathway should be engaged in discussions of addressing liability risk for including pregnant and lactating persons in clinical trials, not just senior administrators and lawyers.

Cohen said that if a research institution wants to support clinical research that includes pregnant and lactating persons, then the necessary supportive infrastructure, policy framework, insurance framework, and legal framework to achieve this goal can be developed. She said that at PATH, the legal affairs team, product development teams, ethics committee, and office of research work collaboratively across the institution.

Zorrilla noted the effect of public awareness campaigns, and she cited the successful campaigns to raise awareness about the effects of alcohol and cigarette use during pregnancy as an example. She suggested that expanded messaging about preparing for pregnancy is needed, covering issues such as optimizing prepregnancy health, folic acid, and discussing any medications being taken for chronic conditions with a health care provider before making any changes. She also emphasized the need to combat the increasing prevalence of misinformation, disinformation, misinterpretation, and misunderstanding about health and health care.

Designing Studies to Include Complex Populations

Roberts emphasized the need to design trials for inclusion of pregnant and lactating persons from the start, engaging obstetricians and maternal–fetal medicine specialists on study design committees and IRBs, even when trials are not pregnancy specific. He noted that, in vaccine trials in healthy populations, serious adverse events are rare. Adverse outcomes in pregnancy are fairly common, however, so it is important to have an expert perspective when establishing how adverse events in pregnant clinical trial participants will be evaluated (e.g., determining expected baseline rates of miscarriage or preterm labor). Zorrilla described the frustration on the part of trial participants and their clinicians when a participant inadvertently becomes pregnant and is dismissed from the trial. Zorrilla suggested that in many cases, based on the mechanism of action for a given drug, the risk to the pregnant person or the fetus can be predicted. Cohen said that medical product development should engage researchers, clinicians, legal colleagues, funders, regulators, and other stakeholders from the outset of a given study. Sewell encouraged sponsors to engage with FDA early on in trial development through a pre-investigational new drug application meeting with the agency.

Ruth Karron, professor of international health at the JHU Bloomberg School of Public Health, said there is a perception that a product sponsor’s vested interest in bringing a product to market quickly dissuades them from including pregnant persons in trials. Wein emphasized that a sponsor’s goal is bringing a product to market as safely and efficaciously as possible. The more complex the trial, the more that time to market can be delayed, and the longer patients must wait for innovations to address

a disease state. The time needed to assess whether a particular adverse event is specifically associated with a complex population or attributable to something else can cause significant delays. She added that postmarket studies in select populations is an option if they are not included in premarket studies.

Wein suggested that, if multiple companies were working on developing a new class of drug and were including pregnant and lactating persons in the trials, there could be a common database for the timely sharing of data on adverse events (while still protecting competitive information as needed). Henry countered that there is conflicting evidence on whether excluding pregnant people from clinical trials necessarily leads to faster times to market. She agreed that postmarket studies are an option but observed these are not often done. The sooner a product reaches the market the sooner there are financial returns, and she said that during the consultations with legal experts (described above) it was emphasized that financial factors play a role in study design and “to not mistake something as legal when what is actually going on is financial.”

Brittany Bettendorf, University of Iowa, asked if the learned intermediary doctrine creates a disincentive for pharmaceutical companies to include pregnant persons in clinical studies, as shifting liability risk downstream to the clinician would seem to be in the best interest of the company. Henry said that including pregnant persons in studies would provide clinicians with information to support shared decision-making with the patient. Although this is a win-win for patients and providers, she agreed that sponsors could potentially face increased liability. Henry suggested that regulations or statutory requirements might be needed to ensure inclusion of pregnant and lactating persons in trials (similar to how studies are required to include children if the product is to be used in a pediatric population). Mastroianni noted that one of the original goals of the learned intermediary doctrine was to allow clinicians to “practice their art” and make informed, risk/benefit-based decisions about how best to address the specific needs of their patient. It was not designed as a liability protection.

Products for the Specific Benefit of the Pregnant/Lactating Person or Fetus/Child

Aaron Pawlyk of the National Institute of Child Health and Human Development raised the topic of liability risk related to the development of drugs specifically for obstetric conditions (e.g., preeclampsia, preterm birth). Henry said there is little empirical evidence on how the law is applied in these cases, and she encouraged researchers to engage lawyers sooner in the drug development process. Wein noted that trials

are planned and designed specifically for complex populations, knowing there will be some level of risk for trial participants, risk mitigations are therefore put in place.

Roberts pointed out that clinical trials are under way for a vaccine that is administered to the pregnant person to protect the offspring from disease. This is different from other products FDA has reviewed, and will raise new legal liability considerations, he said. Henry suggested that one of the reasons this particular research is being done is because there is benefit to the fetus. She speculated that, under 45 CFR 46 Subpart B as written, research with a direct benefit to the fetus might be more acceptable than including pregnant persons in studies of products with no pregnancy-specific benefit.

Potential Effect of Fetal Personhood on Clinical Trials with Pregnant Persons

Panelists discussed how potential federal and state laws governing the personhood status of a fetus might affect the liability associated with clinical trials involving pregnant persons.¹⁷ Henry noted that some states have existing statutes regarding the personhood of a fetus and said, “There is no evidence to date that any of these statutes have been used to prevent research with pregnant or lactating people.” However, misuses of these statutes remain a concern. For example, there have been hundreds of cases in which a pregnant person was prosecuted for child neglect or child abuse because it was argued that their late miscarriage or other fetal injury was a result of their actions (e.g., not following doctor’s orders for bed rest). Henry suggested that fetal personhood statutes create additional challenges for interpreting 45 CFR 46 Subpart B, such as defining the extent of the role of the father in the informed consent process. She also raised a concern that, under these statutes, a guardian ad litem could be appointed to the fetus and would determine whether research involving the fetus could be done under 45 CFR 46 Subpart B. In contrast, several states, including Maryland, have codified that a fetus will not be awarded personhood. Henry noted that Maryland and a few other states have also established their own human subjects research regulations. She predicted, that as more states recognize the potential effect of fetal personhood on clinical research, there might be more state-level human subjects research regulations.

___________________

¹⁷ One week after this workshop, on June 24, 2022, the U.S. Supreme Court issued a decision overturning the Roe v. Wade and Casey v. Planned Parenthood decisions. These workshop discussions took place in the context of the anticipated decision based on the leaked draft opinion by Justice Samuel Alito.

Mastroianni pointed out that personhood statutes could inadvertently lead to clinical research involving pregnant persons only being conducted in certain states (i.e., those that do not recognize personhood of the fetus). Henry suggested that some states might become supportive of this research, and there could even be a specialized IRB for research involving pregnant persons. Sewell raised concerns about how limiting where trials are done would affect demographic representation and whether the data would then be relevant to the entire intended population.

Although a pregnant person and their fetus/child are separate entities from an ethics perspective, Ruth Faden, Johns Hopkins University, said, “In almost all cases, the interests of the pregnant person and the baby that will be born align, in research and in clinical practice.” The health of the pregnant person affects the health of the baby, and it is also of “tremendous importance” to pregnant people that their babies are born healthy and are not at risk for serious illness, she said. Awarding the fetus legal personhood creates two separate entities and could lead to two competing sets of interests.

Sabra Anckner of the Association of Maternal and Child Health Programs suggested that pregnant persons have been considered a vulnerable population with regard to clinical trial participation because of concerns about the fetus, not the pregnant person. Current and impending changes to laws are raising significant concerns about the potential criminalizing of pregnant persons and their care providers should any harm come to the fetus during the pregnancy. If a pregnant person miscarries while enrolled in a clinical trial, will the provider who referred the patient to the trial be held criminally liable? Will the threat of criminal liability for fetal adverse events deter providers from referring their pregnant patients for clinical trials? Veronica Gillispie-Bell, Oschner Health System and Louisiana Department of Health, agreed that clinicians should not have to be concerned about going to jail for practicing medicine in the rapidly changing legal environment for reproductive health. In addition, there has been so much focus on the fetus that the pregnant or lactating persons are nearly forgotten. “When you realize that babies come from mothers and you take care of the moms, that will take care of the babies,” she said.

It was also discussed that many patients, ethnic minorities in particular, do not want to share information about having had a miscarriage or abortion during the course of a clinical trial owing to possible legal action. Gillispie-Bell said these are legitimate concerns in both trial enrollment and routine care. The first step is to establish trust with the patient, and then assure them that their data is protected and explain what that protection is. Present them with the risks and benefits of participation and then support their decision, she said.

OPPORTUNITIES TO ADDRESS LIABILITY CONCERNS: BREAKOUT DISCUSSION HIGHLIGHTS

Following the panel discussion, participants divided into breakout groups to consider opportunities to address liability concerns from the standpoint of clinical investigators or trial sponsors.¹⁸ A summary of the points made by individual breakout group participants was provided in plenary session by a designated leader for each of the breakout groups.¹⁹

Opportunities to Address Liability Concerns for Clinical Investigators

Participants in two groups considered liability concerns from the perspective of clinical investigators. Highlights of their discussions were provided by William Cooper, professor of pediatrics and health policy at Vanderbilt University, and Leyla Sahin, FDA.

The following is a list of potential solutions to address liability risk that were discussed:

• Establishing requirements for the inclusion of pregnant and lactating persons in clinical trials. Participants discussed that the lack of any requirement to include pregnant or lactating persons in studies is a barrier for clinical investigators. Cooper reported the suggestion that there are lessons from the implementation of BPCA and PREA that could be applied to developing a balance of incentives and mandates to promote inclusion of pregnant and lactating persons in trials.
• Including obstetric, pharmacokinetic, and pediatric expertise on IRBs. A barrier to inclusion that was highlighted was a concern about the potential negative effects on the developing fetus or the child (e.g., teratogenicity). Several participants suggested that including relevant expertise on IRBs would better inform decision-making around risk.
• Using a range of methodologies to collect data. Another approach to addressing the concerns about adverse effects on the developing fetus or child that several participants discussed is compiling evidence about drug product uses through registries and large cohort studies.
• Designing and funding trials for inclusion from the start. Cooper relayed discussions that there are costs associated with the inclusion

___________________

¹⁸ Breakout groups were convened both in person and via Zoom for online attendees for each of the topic areas.

¹⁹ This section is the rapporteurs’ summary of the breakout group reports by the group leaders and should not be construed as reflecting any group consensus.

• of complex populations in trials, and when inclusion is an afterthought, the funding needed is not available. Applying lessons from BCPA and PREA, participants discussed that FDA, the National Institutes of Health (NIH), and other funders, sponsors, and investigators need to work together and commit to designing and funding studies with the intent to include pregnant and lactating persons. There was also a discussion about the need to extend funding cycles as “academic investigators feel like they are constantly reapplying for funding and are unable to complete their clinical research.”
• Training OBGYNs in clinical research. Participants discussed that OBGYNs are generally not trained in clinical research, which presents a significant barrier to the inclusion of pregnant and lactating persons in studies.
• Developing the infrastructure to conduct trials beyond the academic medical center. Several participants observed that clinical studies are often conducted at academic medical centers, and participants highlighted the importance of reaching out to communities and rural areas and enrolling populations that are often underrepresented.
• Countering misinformation. Participants discussed that clinicians often need to address the misinformation that patients are exposed to before they can interest them in enrolling in a clinical trial. It was noted that this misinformation is spread over social media, often by celebrities and other influencers.
• Improving the accessibility of clinical trial participation. A range of practical barriers to patients’ participation in studies were discussed, such as the need to go to a specific trial site for laboratory testing or imaging. Solutions suggested by participants (drawn in part from lessons learned from COVID-19 clinical trials experiences) included allowing more flexibility in how samples for laboratory testing can be collected (e.g., by a local laboratory, by home health care workers, self-collection), and more flexibility for follow-up visits (e.g., telemedicine, follow-up texts).
• Improving the informed consent process. The consent process was also discussed as another practical barrier. Participant suggestions included consent via telemedicine, Spanish language consent forms, and electronic consent. It was also discussed that the informed consent process should be thought of as a process, not a brief event, and a longer time window should be allowed. The provider would document when the conversation began and when the participant signed off, which might span several days. It was suggested that running through the consent process too quickly could be coercive.

• Instituting policies designed to mitigate liability risk. Participants discussed the need to implement institution-wide policies for risk mitigation such as has been done by the University of Washington (discussed by Mastroianni, above). The need to develop an injury compensation fund for drug trials was also discussed.

Opportunities to Address Liability Concerns for Trial Sponsors

Two breakout groups focused on the liability concerns of trial sponsors. Highlights of the points made by individual participants were provided by Henry and Anthony Beardsworth, senior medical director for patient safety at Eli Lilly. They noted that much of their groups’ discussions aligned with the points made by the groups that focused on issues for clinical investigators. The following additional points were made:

• Addressing regulatory ambiguity. Several participants discussed the need for more clarity around issues such as what constitutes minimal risk and what data are sufficient to justify inclusion of pregnant or lactating persons in a trial. It was noted that what constitutes minimal risk might vary over the course of pregnancy, and that this is not addressed by 45 CFR 46 Subpart B. Participants also noted the need for predictable and consistent guidance from regulators across different divisions.
• Breaking down liability risk into manageable segments. Several participants discussed that liability is often discussed as one huge, nonspecific risk. However, participants noted that liability risk varies across the stages of pregnancy and the phases of research. Liability risk is affected by the availability and usefulness of data from preclinical nonhuman studies, and by whether the drug is intended specifically for use in pregnancy. Risk also differs depending on whether the harms are incurred during the course of the study or postmarket. Several participants suggested that research is needed to understand the different sources and levels of risk across the clinical research continuum.
• Providing sponsors with relevant legal information. Henry reported that sponsors often lack the relevant legal information and expertise to shift from presumed exclusion to routine inclusion of pregnant and lactating persons on trials. Their perspectives are influenced by high-profile, traumatic events (e.g., the thalidomide crisis) and the risk of long-tail liability across potentially generations of claimants. Several participants discussed that it would be helpful for sponsors to (1) have information such as how frequently long-tail liability becomes an issue for companies,

• and (2) develop a greater understanding of relevant risks associated with different types of studies and different points during pregnancy or lactation.
• Establishing requirements for inclusion. Incentives and requirements for including pregnant and lactating persons in clinical trials were also discussed by participants considering the liability issues for sponsors. It was suggested by some that sponsor concerns about liability might outweigh any incentive for inclusion, and that without a mandate, studies including women and lactating persons might not be undertaken, though other participants were more skeptical of a mandate. It was discussed that incentives might be successful in motivating studies for conditions that are specific to pregnancy (e.g., preeclampsia, preterm labor), but that incentives may not succeed in encouraging sponsors to conduct studies for conditions not specific to pregnancy. Some participants suggested that a regulatory requirement for studies to include pregnant and lactating persons could reduce sponsor risk.
• Committing to inclusion as part of the company’s core mission. Overcoming inertia from leadership was noted as a barrier to the inclusion of pregnant and lactating persons in trials by sponsors. Several participants discussed that corporate leadership needs to declare that inclusion is important and embed it in the core mission of the company. It was suggested that would then shift focus from avoiding liability risk to managing that risk.
• Improving protection against liability claims. Participants discussed the challenges of securing clinical trial insurance and said that sponsors are generally self-insured. Improved options for liability protection for sponsors is needed.

This page intentionally left blank.