Antibody Transfer From Mother to Fetus/Infant: An Opportunity for Model‐Informed Precision Medicine

Therapeutic antibodies have become increasingly important for the treatment of autoimmune disorders, such as inflammatory bowel disease (IBD), in pregnant patients and mothers.1 For example, a retrospective study assessing the pattern of biologics use from pre-pregnancy to postpartum has reported 71.6% have continued biologics use at least once during pregnancy. Compared to patients with rheumatoid arthritis, patients with Crohn's disease and ulcerative colitis were more likely to continue the biologic during pregnancy, while patients with psoriasis or psoriatic arthritis were less likely to continue biologic use. The percentage of pregnant subjects with live birth that continued biologics declined from 68.6% in trimester 1 to 58.8% in trimester 2 to 48.6% in trimester 3.2 While transplacental transfer of endogenous antibodies supports passive immunity against infectious diseases to the fetus and newborn, the long-term safety of therapeutic antibodies that cross the placenta remains poorly understood. The impact of transplacental transfer of therapeutic antibodies and their impact on fetal health represents a substantial knowledge gap in maternal pharmacotherapy with consequences for clinical decision-making.3, 4 The management of chronic disease during pregnancy and lactation has undergone a paradigm shift in recent years with increasing focus on maintaining maternal disease control while ensuring healthy outcomes for the fetus or infant. More robust data is needed to support dosing strategies that optimize the health of both expecting mothers and their children. Currently, most evidence supporting antibody dosing strategies in pregnant patients is based on observational studies, attributable to the ethical challenges of conducting clinical trials during pregnancy. Reliance on observational, or clinically convenient data, has led to inconsistent clinical recommendations for the use and dosing of antibody drugs during pregnancy. Our compilation of selected societal guidelines highlights the variability in recommendations for antibody drug use in selected autoimmune disorders, particularly with respect to dosing schedules (Table S1). Post-approval pregnancy studies, when conducted, tend to prioritize severe congenital malformations as safety endpoints, often challenged with the detection of milder outcomes due to limitations of sample size and/or follow-up time, which could still affect infant health. Meanwhile, hormonal, immunological, and microbial fluctuations during pregnancy can alter the course of autoimmune diseases, influencing risk of relapse, attack severity, and postpartum recovery.5 Drug labeling typically adopts a cautious stance, prioritizing fetal safety over maternal well-being, despite risk signals that are often ambiguous. Data gaps persist, particularly with respect to the mechanisms that govern dynamics of antibody pharmacokinetics and transplacental transfer in pregnant individuals, as well as the implications for drug efficacy and safety. These data gaps underscore the need for advanced tools to balance maternal therapeutic needs and fetal safety. To address these challenges, researchers have increasingly turned to model-based approaches, such as physiologically based pharmacokinetic (PBPK) modeling, to study the pharmacokinetics and transplacental transfer of antibody drugs.6, 7 PBPK modeling is increasingly used as a method for predicting the transfer of small molecules from mother to fetus or infant8-12 that also shows promise for therapeutic antibodies. These models can account for physiological changes during pregnancy in both the mother and in neonates, enabling predictions of antibody transfer and resulting exposures. These predictions are critical to evaluating and optimizing dosing regimens with respect to both maternal efficacy and fetal safety. For instance, Chen et al developed a PBPK model to evaluate the placental transfer of anti-tumor necrosis factor-alpha (anti-TNFα) biologics in pregnant women with IBD during the third trimester.13 Their model sought to facilitate timely infant vaccination by determining a treatment discontinuation time that maintained maternal efficacy while minimizing fetal exposure at birth. We commend the rigor of Chen et al's approach, however, a key limitation merits attention: the model overlooks the ontogeny and functional maturation of the neonatal Fc receptor (FcRn), a critical factor in antibody transfer exhibiting a strong pregnancy time-dependent behavior. FcRn is ubiquitously expressed, including in the syncytiotrophoblast layer of the placenta, and facilitates antibody transfer from maternal to fetal blood. The ontogeny of placental FcRn encompasses both increasing expression and function maturation with gestational age. FcRn expression in the syncytiotrophoblast and endothelial layers of the placenta exponentially increases after around the second trimester.14, 15 The efficiency of IgG transfer (measured as fetal:maternal IgG ratio) is a proxy of both increase FcRn expression and function, promoting an exponential rise in endogenous IgG transfer to the fetus, particularly in the second and third trimesters.16 This marked increase in FcRn expression likely contributes to increased fetal exposure to therapeutic antibodies as gestation progresses, with a more pronounced effect later in pregnancy (Figure 1). Accounting for FcRn ontogeny and functional maturation in PBPK models is thus essential for accurately predicting antibody exposure and determining optimal discontinuation times, especially since discontinuation, if pursued, typically occurs in the third trimester—a period of heightened FcRn expression and functional activity. Societal guidelines often recommend discontinuing antibody drugs (if at all) around the third trimester, however this must be an individualized decision that considers maternal disease severity and treatment response. Active IBD in expecting mothers has been associated with an increased rate of adverse pregnancy outcomes, prompting some experts to advocate for continuation of anti-TNF therapy throughout pregnancy to prevent maternal flares and associated adverse fetal outcomes.17 Notably, discontinuation in the third trimester has not been associated with reduced infant infection risks, as active placental transfer exposes the infant to anti-TNF agents regardless.18 Consequently, live vaccinations are often delayed for these infants to avoid potentially severe outcomes.3, 19 Moreover, placental FcRn expression can be altered by gestational disorders. For example, type 2 diabetes has been shown to significantly decrease FcRn expression20 which can add complexity to the optimal timing of treatment and treatment discontinuation. Given this variability, a one-size-fits-all approach may fall short, underscoring the need for more individualized strategies that carefully weigh the benefits of therapy against potential risks. For these patients, abruptly halting therapy could exacerbate their condition, jeopardizing both maternal and fetal health. Rather than pursuing complete treatment cessation, tailored dosing strategies that meet the specific needs of individual patients may be a more practical and effective solution. Beyond pregnancy, FcRn remains relevant in newborns, where it facilitates antibody transfer via breastfeeding, contributing to neonatal immunity. FcRn is expressed in the gastrointestinal epithelium of newborns, where it contributes to the transfer of maternal antibodies to the infant via breastfeeding.21 This process, though partial, is vital for neonatal immunity and highlights the physiological significance of this receptor to child development after birth (Figure 1). An improved understanding of FcRn postnatal ontogeny could further inform dosing and discontinuation schedules for biologic therapies in new mothers. Integrating PBPK models with FcRn ontogeny offers a powerful framework for precision medicine, enabling clinicians to adapt treatment to individual patient profiles. Model-informed approaches offer an opportunity to optimize maternal efficacy while mitigating fetal/infant risk both during and after pregnancy. By integrating PBPK models with insights into FcRn ontogeny and functional maturation, clinicians gain a powerful tool for precision medicine. We have developed a PBPK model that accounts for FcRn ontogeny and functional maturation throughout pregnancy to predict optimal dosing regimens that maintain maternal efficacy while minimizing fetal exposure.22 Furthermore, since the efficacy threshold of therapeutic antibodies can vary by patient factors such as disease history, status, or response to treatment, individualized dosing strategies may be more appropriate than a one-size-fits all approach. We incorporated our minimal PBPK model in an R Shiny application (www.github.com/mkardouh/mPBPK_antibodies_pregnancy) as a user-friendly web tool to produce simulations based on a given scenario. This approach enables a more nuanced, model-informed strategy that is adaptable to individual patient scenarios. Currently, our model, as presented in the R Shiny app, incorporates gestation-related changes in physiology and can accommodate different dosing schedules prior to and throughout pregnancy. Future development of PBPK models should focus on incorporating patient-specific characteristics collected from real-world data. Key factors such as disease history, severity, response sensitivity to antibody therapies, and pharmacokinetics can be integrated to determine the optimal dose and dosing regimen for each patient.23-26 Additionally, this information can be further tailored to account for gestational age, physiological shifts during pregnancy, and required therapeutic efficacy, ensuring effective treatment while minimizing fetal risks. Antibody transport from mother to fetus is also highly relevant to drug development. For instance, we often rely heavily on non-human primates to assess potential developmental and reproductive toxicities (DART), as most clinical trials exclude pregnant subjects. However, there may be opportunities to waive certain DART studies by providing a bottom-up assessment of antibody exposure in both the mother and fetus, along with an evaluation of the corresponding safety margins. Ultimately, combining advanced modeling techniques with a deeper understanding of FcRn dynamics provides a promising pathway for optimizing maternal and fetal care in this complex clinical setting. This integrated approach has the potential to refine dosing regimens, improve maternal and fetal outcomes, and address the unique challenges of managing chronic diseases during pregnancy. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Miramar Sami Kardouh, Tyler Dunlap, and Rui Zhong conducted this research under affiliation with the University of North Carolina at Chapel Hill. Miramar Sami Kardouh is currently at Inncelerex, Jersey City, New Jersey. Tyler Dunlap is currently at Metrum Research Group. Rui Zhong is currently at Bristol Myers Squibb. All authors declared no competing interests for this work. This work was supported by grants from the National Institutes of Health (NIH R35 GM152449 to Yanguang Cao and UC2HD113039 to Jacqueline B. Tiley). Miramar Sami Kardouh was a PK/PD fellow sponsored by Certara Inc. Rui Zhong was a PK/PD fellow sponsored by Bristol Myers Squibb. Tyler Dunlap was supported by the National Institute of General Medical Sciences of the National Institutes of Health under Award Number T32GM086330. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This manuscript does not generate any new primary data. All analyses presented are based on previously published or publicly available datasets, and no proprietary or unpublished data were developed for this work. Consequently, there is no additional data to be shared beyond what is contained in the manuscript and its supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.