Laure-Hélène Préta (1), Frantz Foissac (1,3), Léo Froelicher (1,2), Victoria Buth (4), Alisha Dinnall (1), Annouk Dauvilliers (1), Faheemah Padavia (1), Saïk Urien (1,3), Jean-Marc Tréluyer (1,2,3,4), Sihem Benaboud (1,2), Naïm Bouazza (1,3), Gabrielle Lui (1,4)
(1) URP 7323 Pharmacologie et évaluation des thérapeutiques chez l'enfant et la femme enceinte - Université Paris Cité Paris France, (2) Service de Pharmacologie - Assistance Publique Hôpitaux de Paris Hôpital Cochin Paris France, (3) Unité de Recherche Clinique - Assistance Publique Hôpitaux de Paris Hôpital Tarnier Paris France, (4) CIC-1419 Mère-enfant Cochin-Necker Paris France
Introduction: Depression is common in pregnant women [1]. Perinatal depression affects around 6-12% of pregnant women at different trimesters [2-3]. Between 2.5 and 10% of pregnant women are treated with antidepressants [4], with an overall incidence rising in recent years. Of the different classes of antidepressants available, selective serotonin reuptake inhibitors (SSRIs) are the most widely prescribed in pregnant women [5-6]. While some data on SSRIs use in pregnant women are reassuring, concerns about malformations, prematurity or neurodevelopmental disorders in children prenatally exposed to these drugs have been raised [7]. Currently, little is known about the transplacental transfer of these drugs and whether there are any differences between them. For ethical reasons, pregnant women are not included in clinical trials, but ex vivo perfusion of human placenta cotyledon method combined with modelling have shown their suitability for predicting fetal exposure [8].
Objectives: To model ex vivo placenta perfusion data for escitalopram and sertraline; To identify the covariates that may influence transplacental transfer for each molecule; To compare ex vivo transfer ratio for these two SSRIs.
Methods: First, ex-vivo human placenta perfusion experiments in double closed circuit were performed for escitalopram and sertraline. Medications were added at therapeutic concentrations (i.e. 80 ng/mL for escitalopram and 100 ng/mL for sertraline). Antipyrine was added as a control (20 mg/L). The perfusions were carried out over 3 hours, with samples taken on both maternal and fetal sides. Drugs were then quantified following a validated method. Fetal transfer rates (FTR), defined as the ratio of fetal-side medication amount to total medication amount, were calculated at the end of each perfusion. Second, a non-linear mixed-effects model (NLME) was used to describe the concentration time course on both sides using Monolix® (version 2023). The following covariates were tested: presence/absence of albumin, origin and age of the mother, smoking, gestational age, delivery mode, weight and sex of the newborn and weight of the cotyledon. Individual placental cotyledon time courses were derived from the final model using empirical Bayesian estimates. Areas under the curves AUC(0-3h) on the maternal and fetal sides for both molecules and AUC fetal to maternal ratios (AUCf/m) were calculated. Ratios for the two molecules were compared using a non-parametric Wilcoxon test.
Results: Of the 31 cotyledons perfused, 26 were validated by antipyrine (13 for each molecule). Ex vivo experiments yielded 230 concentrations for escitalopram i.e. 115 for each side (maternal and fetal) and 232 concentrations for sertraline i.e. 116 for each side, over 3 hours of perfusion. Median (IQR) FTR was 46.9% (43.8-49.5) for escitalopram and 30.7% (17.2-33.6) for sertraline. The most suitable model for describing our data was a five-compartment model: 1 compartment for the maternal reservoir, 1 compartment for the fetal reservoir and 3 compartments for the cotyledon (maternal part of the cotyledon, placental barrier and fetal part of the cotyledon). Regarding covariates, a significant effect of albumin on the transfer of these two molecules was observed. Median (IQR) AUCf/m ratios for escitalopram were 0.36 (0.34-0.37) without albumin and 0.42 (0.37-0.47) with albumin. For sertraline, median AUCf/m ratios were 0.029 (0.026-0.03) without albumin and 0.21 (0.12-0.27) with albumin. For escitalopram, the model found a median AUCf/m ratio 1.2 times higher in presence of albumin compared to without albumin. For sertraline, this ratio was 7.2 times higher in presence of albumin compared to without albumin. Finally, AUCf/m ratio in presence of albumin were doubled for escitalopram compared with sertraline (p < 0.0001).
Conclusions: Data from ex vivo experiments integrated into the NLME model showed that escitalopram and sertraline cross the placental barrier. An albumin effect was observed for both compounds, but appeared to be particularly significant for sertraline, which seems consistent with their protein binding rate (escitalopram: 55-56%, sertraline: 98-99%). If we compare the ex vivo transplacental transfer of these two medications, escitalopram appears to significantly more cross the placental barrier than sertraline. These initial results will be used to carry out PBPK modelling for predicting fetal exposure.
References:
[1]: Ko JY et al. Journal of Women’s Health. 2012;21(8):830-836
[2]: Gavin NI et al. Obstetrics & Gynecology. 2005;106(5, Part 1):1071-1083
[3]: Bennett HA et al. Obstet Gynecol. 2004;103(4):698-709
[4]: Bénard‐Laribière A et al. Brit J Clinical Pharma. 2018;84(8):1764-1775
[5]: Hurault-Delarue C et al. Eur Arch Psychiatry Clin Neurosci. 2019;269(7):841-849
[6]: Bérard A et al. BMJ Open. 2017;7(1):e013372
[7]: Becker M et al. Curr Psychiatry Rep. 2016;18(3):32
[8]: Eke AC, Olagunju A et al. Clin Pharmacokinet. 2020 Oct;59(10):1185-1194
Reference: PAGE 32 (2024) Abstr 10979 [www.page-meeting.org/?abstract=10979]
Poster: Drug/Disease Modelling - Paediatrics