Conducting randomized controlled pharmaceutical trials in the pregnant population: challenges and solutions


Introduction

The development of a new drug for use in humans classically follows a strict sequence, beginning with in-vitro and animal experiments to identify promising novel molecules, their sites of action, and biological effects. After efficacy and safety in animals is confirmed, three phases of human clinical trials need to be completed prior obtaining approval for use in clinical practice. Phase I studies are conducted with healthy volunteers and serve to establish the safety of the drug, its therapeutic range, and appropriate dosage. Phase II trials involve a limited number of participants and are conducted in precisely defined conditions to demonstrate drug efficacy in a carefully selected population. Phase III trials are large, usually multicenter trials, which include a broader population and serve to establish the effectiveness and safety of drugs, simulating usual clinical use. After commercialization, phase IV trials are used to document drug effectiveness and adverse events at the population level, or to investigate new dosages or indications [ ]. The randomized controlled trial (RCT) is the gold standard for the evaluation of a new drug or technology, and is the design usually used for phase II and III trials.

This stepwise process is essential to ensuring the development and approval of new drugs which are both effective and safe. However, this process is difficult to apply to the development of drugs used in pregnant women. Indeed, due to the potential teratogenic effects of new drugs on the fetus, the ethical dilemma of knowing how much evidence is required to justify conducting a clinical trial including pregnant women has often led to the complete exclusion of this population from drug trials. Apart from data available in registries, there is little high-quality evidence on the innocuity and effectiveness of common medications in the pregnant population. Furthermore, few drugs are developed specifically to target obstetrical diseases, and the ones which are tested or repurposed for these indications often fail at the phase III trial level.

This chapter will explore the specific considerations and challenges of conducting randomized controlled pharmaceutical trials in pregnant women and provide potential solutions to help safely include this population in future trials.

Ethical considerations

Maternal and fetal risks in pharmaceutical research

The most sensitive and controversial ethical aspect of pharmaceutical trials in the pregnant population is the risk of harm to the embryo or fetus. Historically, this risk has led to the systematic exclusion of pregnant women from pharmaceutical research [ ]. Although the AIDS/HIV epidemic led to the recognition that all groups of the population should be offered inclusion in the studies of new medications which might benefit them [ ], pregnant women are still vastly underrepresented in pharmaceutical trials [ ]. The ethical considerations of clinical research in the pregnant population are discussed in detail in Chapter 8 . A key point to consider in this endeavor is that pregnant women have agency. The National Institute of Health and the American College of Obstetrician and Gynecologists advocate that pregnant women should no longer be considered a vulnerable population, but rather a scientifically and ethically complex one [ , ]. Indeed, women who are pregnant retain their full agency and can decide independently whether or not they wish to take part in research. Furthermore, consent of the male partner is not necessary to the participation of women in a research project [ ].

In 2018, the Food and Drug Administration (FDA) published a draft guidance document which may help direct research efforts in the pregnant population. This document outlines that pregnant women should be included in pharmaceutical trials of any medication that might benefit them, as long as there is sufficient evidence derived from animal studies and nonpregnant women to provide an estimate of the potential risks to the mother and fetus. Research in the pregnant population is deemed acceptable if: “The risk to the fetus is caused solely by interventions or procedures that hold out the prospect of direct benefit for the woman or the fetus; or, if there is no such prospect of benefit, the risk to the fetus is not greater than minimal and the purpose of the research is the development of important biomedical knowledge which cannot be obtained by any other means” [ ].

The evolution of ethical considerations concerning the integration of pregnant women in research calls for their broader inclusion in pharmaceutical trials. Recent recommendations by the FDA provide a key impetus to both industry and the academic community to pursue clinical trials involving pregnant women.

Equipoise

Equipoise is an ethical standard that needs to be met to justify the conduct of any clinical trial. A more precise definition of equipoise has been proposed as “a state of genuine agnosticism or conflict in the expert medical community about the net preferred medically established procedure for the condition under study” [ ]. A trial is warranted when there is “equipoise,” that is when there is a reasonable uncertainty concerning the benefit of an experimental treatment over placebo or an existing treatment. This is a delicate balance between the presence of insufficient evidence to justify a trial (in which case more preliminary/elementary research is needed) and a state of overabundance where an additional trial would not be ethically acceptable. Physician-researchers and members of institutional review boards (IRBs) have the task of determining whether a trial will help resolve equipoise [ ]. Different IRBs can disagree markedly on the same protocol as to their opinion on the presence or absence of equipoise [ ]. This may reflect variations in the expertise of the members on the IRB, as well as the absence of guidelines regarding what is evidence of sufficient quality to justify a trial.

What is sufficient evidence to conduct a trial?

General considerations

In fact, the evidence required to justify the decision to conduct a phase III RCT in the area of maternal-fetal medicine has not been adequately defined. The hypothesis for a trial is normally based on an understanding of the impact of a drug on a specific disease process. The hypothesis should be supported both by in-vitro and animal studies [ ]. There must be sufficient preliminary evidence from phase II trials to indicate that the treatment under investigation might improve patient outcomes in a clinically significant way, and that it is not harmful [ ]. A phase III clinical trial is “warranted if there is sufficient but not definitive evidence that the intervention to be assessed would have a favorable risk-benefit ratio in the population to be enrolled” [ ].

To determine this, the available evidence needs to be reviewed and assessed, usually through a systematic review. The proposed RCT must add to the existing knowledge base; thus, it is important to establish what information is missing from the current knowledge base so that the trial can be designed to fill these knowledge gaps. A thorough understanding of the available evidence is key to formulating the appropriate research question, in defining the target population, in estimating the effect size, and in assessing feasibility.

If available evidence is reliable and already provides a definitive answer, there is no need for a study, although there may be a need for a study in a specific target group or a larger study to refine therapeutic strategies or define optimal dosage [ ].

A methodological question arises as to what constitutes sufficient evidence to consider that a treatment is efficacious. If the effect estimate derived from a number of small trials shows statistical evidence of an effect, what conclusion should be drawn? Is an additional large RCT justified? This question is even more difficult to address given that studies have underlined the discrepancy between the findings of meta-analyses and the largest trials of the same therapy [ , ]. One important factor to consider when examining the results of a systematic review is the potential for publication bias. Indeed, clinical trials with significant findings are more likely to be published and are published faster than negative trials [ ], and phase I and II trials are more likely to suffer from publication bias than phase III and IV trials [ ]. This has the potential to lead to an overestimation of the effect of an intervention in meta-analyses of small randomized clinical trials. Therefore, when the efficacy of a treatment is established based on several small randomized clinical trials, a robust phase III RCT may still be warranted to confirm the treatment's effectiveness in a large population and to evaluate the presence of adverse events. Again, at least one large RCT showing positive results is generally required to obtain regulatory approval [ ].

Level of evidence to justify a trial during public health crises

In the context of a public health emergency, special consideration must be given to the level of evidence required to justify a clinical trial that includes pregnant women. Indeed, when outbreaks of diseases threaten the lives of pregnant women and their unborn children, the level of evidence required to justify a clinical trial can be lowered. In this context, the potential benefits of each molecule targeting the disease must be carefully weighed against its potential risks. When sound biological rationale and animal studies indicate a potential benefit of a molecule against a lethal disease, consideration may be given to “short-circuiting” the drug development pipeline, and in such cases, it may be unethical to deprive any patient population from a potentially life-saving drug [ ].

The best example of this approach is the rapid approval of zidovudine to prevent vertical transmission of HIV. In 1991, a multicenter RCT was launched in the United States and in France to evaluate the efficiency and safety of zidovudine. The study was based only on animal studies and Phase I evidence. The results showed that zidovudine lowered vertical transmission of HIV by two thirds when compared with placebo [ ]. Based on the interim results of this study, the CDC recommended the use of zidovudine in pregnant women to prevent vertical transmission of HIV in August 1994, three months before the publication of the trial results [ ].

The importance of including pregnant women in pharmaceutical trials of agents targeting lethal diseases is further supported by the example of the Ebola epidemic. During the 2013–2016 Ebola outbreak, pregnant women were excluded from all 16 randomized clinical trials of antiviral treatments or vaccines. Trial investigators justified their decision based on potential embryotoxicity and teratogenicity of the therapeutic agents. The case fatality rate of Ebola in pregnant women is 55%, and only one live birth from a pregnant woman infected with Ebola has ever been reported [ ]. Excluding pregnant woman from pharmaceutical trials for a disease whose natural course is maternal and fetal death, based on the fear of causing fetal harm, is both absurd and unacceptable. Investigators have the ethical duty to ensure that pregnant women are given fair access to any trial of a pharmaceutical agent targeting a disease from which they might suffer [ ]. In the case of deadly diseases, teratogenic potential alone is not a sufficient reason to preclude the inclusion of pregnant women in research.

Despite serious concerns raised by the scientific community [ ], pregnant and lactating women have been almost systematically excluded from pharmaceutical research in the current covid-19 pandemic [ ]. Pregnant women were excluded from phase III RCTs for all the commercially available vaccines [ ]. With recent evidence showing  increased risks of morbidity and mortality from covid-19 infection in pregnancy [ ], this population has become eligible for vaccination in several countries (such as the United States of America, Canada and United Kingdom), but definitive evidence on efficacy and safety in pregnancy is lacking. Most strikingly, pregnant and lactating patients were even excluded from trials of agents known to be innocuous in pregnancy such as hydroxychloroquine and colchicine [ ]. This situation highlights the need for public policies to mandate the inclusion of pregnant and lactating women in pharmaceutical trials.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here