Grace Fraczkiewicz1, Revathi Chapa1, Angele Fleury2
1Simulations Plus, Inc., 2Boehringer Ingelheim Pharma
Introduction: Glucagon-like peptide-1 (GLP-1) receptor agonists are a class of medications intended to manage Type 2 diabetes, obesity and Metabolic dysfunction-associated steatohepatitis (MASH). However, one of their side effects is a delay in gastric emptying, which can affect the absorption and pharmacokinetics (PK) of comedications taken by the patients. For this reason, pharmaceutical companies developing GLP-1 receptor agonists are required to conduct clinical DDI studies with potential comedications. Objectives: The objective of this analysis was to reproduce the impact of delayed gastric emptying induced by the GLP-1 receptor agonist liraglutide [1, 2] on the PK of several oral drugs belonging to different classes of the biopharmaceutical classification system (atorvastatin, digoxin, ethinylestradiol, griseofulvin, levonorgestrel, and lisinopril) [3-6] using a Physiologically-Based Pharmacokinetic (PBPK) modeling approach. Methods: GastroPlus® version 9.8 (Simulations Plus, Inc.) was used to build PBPK models for paracetamol, atorvastatin, digoxin, ethinylestradiol, griseofulvin, levonorgestrel, and lisinopril to describe their in vivo absorption and PK. The published in vitro and clinical data describing physicochemical properties and PK of these compounds were used to develop and validate their PBPK models. Gastric emptying was manually deconvolved from paracetamol plasma concentration vs time profiles obtained in absence or in presence of liraglutide at a steady state. Two types of deconvolutions were conducted: simple and complex. During the simple deconvolution, one value of gastric emptying time (GET) was fitted over the entire length of paracetamol absorption to approximately reproduce its observed plasma concentration vs time profiles. Whereas, during the complex deconvolution, multiple GETs were fitted over the time of paracetamol absorption to match its plasma concentration vs time profiles as closely as possible. The deconvolved GET using the complex deconvolution was then applied to simulate the PK of the other comedications in absence and in presence of liraglutide. The simulated impact of liraglutide-induced prolonged gastric emptying was compared against the observed data. Results: Overall, the simulated results were in line with observed values and showed that while slower gastric emptying tends to prolong Tmax and lower the Cmax due to the slower absorption, the AUC was hardly affected. The results show that most PK parameters were predicted within bioequivalence (BE) limits except for ethinylestradiol AUCs, which were slightly below the BE limit of 0.8-fold, and digoxin Cmax and AUC0 inf, which slightly exceeded the upper BE limit of 1.25-fold. Low solubility drugs and drugs that have a high first-pass extraction can be more affected than the other types of drugs by the gastric emptying prolongation. Longer gastric emptying can enhance the absorption of the low solubility and high permeability drugs either by slowing down the intestinal precipitation process (weak bases) or by increasing intestinal dissolution by reaching solubility limits at later times (lipophilic and not ionizable drugs). Griseofulvin belongs to the later class of drugs and displayed slightly positive absorption and exposure effects with longer gastric emptying. However, this effect was low (~20%). Ethinylestradiol, which is a low solubility and high permeability drug but is also highly extracted pre-systemically, was affected to a low extent by longer emptying time in terms of exposure, even though the model predicted a slightly higher gut extraction ratio for this compound under slower gastric emptying conditions. Atorvastatin, which is highly metabolized and transported, was almost unaffected by the longer gastric emptying as well. Model predictions were aligned with a very small exposure decrease observed for atorvastatin during liraglutide coadministration. Conclusions: This analysis was intended as a proof-of-concept to show that PBPK models can reliably project the DDIs caused by prolonged gastric emptying utilizing limited clinical studies with paracetamol and the perpetrator compound. The results above support the validity of this methodology and show that such an approach can reduce the number of clinical studies with comedications.
[1] https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/022341s027lbl.pdf. Revised: August 2017 [2] European Medicines Agency, E.M., Saxenda, INN-liraglutide. 2015. [3] Malm-Erjefalt M et al. Mol Pharm. (2015), 12(11): 4166-73. [4] https://ClinicalTrials.gov/show/NCT01508858 [5] https://ClinicalTrials.gov/show/NCT00978393 [6] Glerup H et al. Scand J Gastroenterol. (2007), 42(10): 1182-6.
Reference: PAGE 33 (2025) Abstr 11433 [www.page-meeting.org/?abstract=11433]
Poster: Drug/Disease Modelling - Absorption & PBPK