Optimized protocol for non-clinical data generation regarding drug milk excretion and breastfed infant drug exposure (D3.10)

Le résumé fourni par la source

This Deliverable summarizes the results obtained by WP3 on the development of a platform of non-clinical models (‘tools’) to predict human milk drug concentrations and subsequent exposure of nursing infants to maternal medication. The non-clinical model systems developed and evaluated all employ methodologies to quantitatively assess the passage of small molecule medicines across the blood milk barrier. The underlying hypothesis is indeed that the ability of a compound to cross this barrier will ultimately determine the milk concentrations of that compound (Ventrella et al., 2019; Nauwelaerts et al., 2021). The complete non-clinical platform could be applied both during early and late phases of drug development, but also after a marketing authorization has been granted. Currently, there is no well-developed non-clinical model that is accepted by the health authorities that can be used by the pharmaceutical companies and researchers to predict medicine secretion into human milk. Of the standard reproductive toxicology studies that are conducted during the development of a new drug candidate, the pre- and postnatal developmental toxicity study is the only one that includes a lactation phase. However, this toxicity study provides only limited information regarding lactation, since no quantitative data are collected for the potential medicine transfer into milk. If a dedicated milk transfer study is conducted in rats, it can be helpful in determining the “presence” or “absence” of a medicine into the milk, but due to species-specific differences in lactation physiology, animal lactation data typically do not reliably predict quantitative levels in human milk. Therefore, an estimation of the relevant infant dose (RID) is not possible. Consequently, at the time of new medicine approval, limited or no information is available that can be used to adequately inform women on the safety and the proper use of the medicine while nursing their infant. In most of the cases, the medicine’s label will recommend to the woman not to breastfeed while taking the medication due to the absence of data. This situation poses significant challenges for both the mother and the infant. The lack of appropriate information could deprive the infant of important benefits of breastfeeding, such as optimal nutrition and antibodies for protection from illnesses and diseases. Additionally, for many women, breastfeeding increases both physical and emotional bonding with their infant. The proposed non-clinical platform would allow a more reliable prediction of medicine concentrations in human milk along with systemic exposure in breastfed infants during drug development, especially since human data on breastfeeding is typically not available at marketing authorization. In addition, the methods developed in this platform may be taken into consideration towards adoption into regulatory guidelines for the use of medicines during lactation. Results obtained in this non-clinical platform/workflow are expected to facilitate a risk assessment and labeling of a medicine candidate, as well as the creation of a basis to advise women who would like to breastfeed. Importantly, pharmacokinetic information obtained in the proposed non-clinical platform will need to be merged with toxicodynamic information for the specific compound. n the context of this work package, small molecule drugs have been selected to be in-scope. Amoxicillin (antibiotic), cetirizine (antihistamine), metformin (anti-diabetic), and venlafaxine (antidepressant) are the four medicines selected for investigation in both WP3 and WP4 (see Section 2, Tables 1 and 2, for more information on compound selection). In addition, WP3 will generate non-clinical data in the in vivo minipig model for valaciclovir, levetiracetam, escitalopram and atenolol, as well as investigate several additional model compounds in the in vitro model and PBPK models (see below). Future work could expand the use of this non-clinical platform to other therapeutic modalities. The development of this non-clinical platform encompasses three distinct methodologies that may be used either independently or together depending on the development stage of the medicine in question. Note: Graphical abstract modified from https://doi.org/10.1016/j.biopha.2020.111038 and created using Biorender.com The three methodologies are: State-of-the-art in vitro models that are representative of the blood milk barrier and can be used for accurate and rapid determination of plasma-milk transfer rates (permeability coefficients) of medicines and drug candidates across this blood-milk barrier. The in vitro model can be used early in drug development to calculate the extent to which a drug will cross the mammary epithelium, enter the milk, and be available to a nursing infant. These permeability coefficients will be converted to clearance values for milk secretion, which can then be used as data input for drug-specific lactation PBPK models. In addition, the utility of implementing these permeability coefficients in in vitro to in vivo extrapolation (IVIVE) algorithms for the M/P ratio will be explored. An animal lactation model in lactating Göttingen Minipig sows to determine maternal systemic exposure, milk drug concentration, and infant systemic medicine concentrations, which can be used as a prediction of the human maternal and infant exposure to a medicine. The in vivo model will also enable the generation of maternal systemic PK data and maternal milk drug concentrations. Furthermore, infant systemic drug concentrations resulting from exposure via breast milk can be determined. Together these data can allow for the calculation of a relative infant dose. Combined with known toxicity data from the development of the drug, recommendations for safe use during lactation could be provided in the label. 3. Physiologically-Based PharmacoKinetic (PBPK) models for predicting drug concentration time profiles in the blood and milk of lactating women, as well as pediatric PBPK models to predict systemic exposure of breastfed infants to the maternal medication. These predictions will better inform the label on safe use of medicines during lactation. The method(s) chosen will be dependent on the stage of medicine development, and/or the amount of (pre)clinical data available for the medicine (candidate). Consequently, the workflow will be customized for each medicine. The outcomes would enable risk assessment and could be used to inform labelling recommendations for the safe use of the medicine during lactation. Since this non-clinical platform allows to generate medicine-specific quantitative data, an estimated relative infant dose may also be calculated. IMI ConcePTION is also seeking EMA Qualification Advice of this innovative non-clinical platform for the reliable prediction of medicine concentrations in human milk and an estimation of the systemic exposure in breastfed infants.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.