Most pharmacokinetics (PK) clinical studies, and most clinical studies in general, exclude pregnant women in order to protect the safety of the mother and fetus. Furthermore, if a study participant becomes pregnant during a trial, the common practice is to immediately halt administration of study drug and discontinue her from the study, with ongoing monitoring of safety outcomes and developmental complications at least through the end of pregnancy.
While these policies are good in that they reduce the risk of developmental and other pregnancy-related complications during the investigational phase of drug development, they can also be problematic once these products are on the market. Because recommended dosing is almost always based upon PK studies in nonpregnant individuals, risks posed to pregnant women who may take the drug are not well-established. This is especially concerning given the large number of women who take at least one drug during their pregnancy. This includes prescription and over-the-counter medications. The numerous physiological changes that occur during pregnancy that can impact drug exposure include:
- Decreased gastrointestinal motility, altered gastric emptying, and increased gastric pH
- Increased body water, plasma volume, glomerular filtration rate, and blood flow to organs
- Decreased concentration of drug-binding proteins in the blood
- Changes in drug metabolizing enzyme activity in the liver
PK Assessments in Pregnant Women
The goal of a PK assessment in pregnant women should be to determine if drug doses need to be adjusted during pregnancy and to make recommendations for those adjustments based on gestation or trimester. Any dose adjustments (or lack thereof) should be reflected in the product label (i.e., the package insert).
The FDA Guidance for Industry, “Pharmacokinetics in Pregnancy – Study Design, Data Analysis, and Impact on Dosing and Labeling,” recommends that PK studies be conducted if the drug is prescribed (or is expected to be prescribed) during pregnancy, especially during the second or third trimester. PK studies should also be conducted if changes in the dose can have a substantial impact on the patient, and if the PK characteristics of the drug are likely to be significantly changed.
Examples of drugs that are of particular concern include those with a narrow therapeutic range (such as many oncology drugs) or those that are predominately renally excreted. This recommendation applies whether the drugs are intended for acute symptom relief or for chronic use.
However, careful consideration must be given to the potential safety implications of drug administration in pregnant women. The key is to minimize risk. By statute (45 CFR Subpart B 46.204), pregnant women may be involved in PK studies if 1) there are already sufficient data from preclinical studies and clinical studies (including nonpregnant women) to assess potential risk to pregnant women and fetuses and 2) if the potential risk is minimal and the purpose of the research is to gather important information that cannot be otherwise obtained.
Phase of Drug Development
PK studies can be considered in Phase 3, prior to submission of a marketing application (i.e., NDA, BLA). Typically, however, most information regarding the use of a drug during pregnancy is gathered during the post-marketing period. This information may be obtained from clinical studies or from real world experiences with the drug (positive and negative). Most often, post-marketing clinical studies in pregnant women will enroll women who have already been prescribed the drug by their own physicians as part of their clinical care.
Information regarding product use during pregnancy should be reported as part of the sponsor’s Periodic Safety Update Report. This report, along with other sources of safety/efficacy information in published literature, can be useful for determining the need for a dedicated PK study in pregnant individuals.
Recommended Study Designs
If clinical studies are performed, the FDA recommends one of the following study designs:
- Longitudinal Design. A longitudinal design would be the most applicable for a chronically administered drug, such as an HIV medication. The FDA recommends that each woman serve as her own control for PK parameters and that samples be taken at specified time points (“windows”) in each trimester.
- Population PK Design. A population PK (popPK) approach with nonlinear mixed effects modeling can be used to assess the impact of various covariates on PK, such as age, race, trimester, etc. This approach allows sparse PK sampling that reduces the number of blood draws required from the mother.
In addition, a number of other aspects of the study design should be considered, including:
Participants: Participants should be representative of the typical patient population. Consideration should be given to intrinsic and extrinsic factors that may impact the PK of the drug (e.g., race, ethnicity, stage of pregnancy, genetic polymorphisms in drug metabolizing enzymes, weight, diet and renal function). The study protocol should include specified metrics for dating the pregnancy as well as uniform diagnostic measures.
Pre-pregnancy and Post-partum Assessments: Depending upon the intended use of a drug, an individual subject may act as her own control. If a drug is to be given chronically, a pre-pregnancy PK/PD assessment can serve as an appropriate comparator to the pregnant state. For drugs that are administered both during and after pregnancy, post-partum PK/PD assessments can serve as the control. Consider longitudinal sampling to capture potential PK impacts of a woman’s changing physiology after childbirth (which can include very rapid changes and changes that take much longer). For certain pregnancy-related medical conditions that resolve rapidly following childbirth, a single-dose PK assessment in the post-partum period may be most appropriate. Safety precautions should be taken to avoid adverse effects on the infant from transfer of drug into the breast milk. If warranted, a lactation study in parallel with post-partum PK sampling might be an appropriate addition to the study design.
Sample Size: The appropriate sample size will depend upon a number of factors, including: study objective, PK/PD variability, study design (e.g., dosing regimen, sampling schedule), and stage of pregnancy.
Drug Administration: The amount of drug administered will depend upon the characteristics of the drug, its intended use, the findings of previous PK studies in non-pregnant individuals, and the expected risk to the mother and fetus. Lower doses may be appropriate in volunteer studies, whereas dose adjustments (higher or lower) may be required when a mother needs the drug to achieve a particular health outcome. Alternative designs for drug administration may also be appropriate.
Sample Collection and Analysis: Samples should include plasma or whole blood and urine for assessing the concentration of the parent drug and active metabolites. Since plasma protein binding is often reduced during pregnancy, consider calculating unbound drug and metabolites, especially if the extent of plasma protein binding of the drug is high (>80%).
Pharmacodynamic Assessments: Pharmacodynamic endpoints, including relevant biomarkers and potentially even fetal PD endpoints, can be important. Informative study design considerations should be discussed with the FDA prior to study initiation.
Women who are pregnant often take prescription and/or over-the-counter medications. However, dosing recommendations almost always rely on PK parameters obtained from non-pregnant individuals, and the impact of pregnancy on drug PK is often unknown. This can pose a significant risk to the mother and fetus, especially given the large number of physiological changes that occur during pregnancy that can impact the PK behavior of a drug. For drugs that are likely to be prescribed or otherwise used during pregnancy, studies in pregnant women may be warranted to inform dosing recommendations and ensure safe use.