Camille Massaux 1, Flora Musuamba 1,2,3, Jean-Michel Dogné 1,2, Happy Djokoto 1, Lisa Hanquet 1, Lisa Wellin 1, Grace Shalom Govere 1
1 University of Namur (Namur, Belgium), 2 Federal Agency for Medicines and Health Products (Brussels, Belgium), 3 University of Lubumbashi (Lubumbashi, Democratic Republic of the Congo)
Objectives: Physiologically based pharmacokinetic (PBPK) models are widely used to characterize oral drug absorption, yet important structural differences exist between commonly used platforms. The Advanced Dissolution, Absorption, and Metabolism (ADAM) model implemented in Simcyp relies on a segmental representation of the gastrointestinal tract, whereas PK-Sim is based on a whole-body PBPK framework integrating detailed anatomical compartments. Despite widespread use, the impact of these structural differences on pharmacokinetic predictions remains insufficiently characterized.
The objective of this work was to reconstruct the ADAM absorption model in R to make its equations, parameters, and relationships explicit, thereby improving mechanistic understanding of its structure. This reconstruction was then used to assess its structural consistency with the Simcyp implementation and evaluate how these differences may influence pharmacokinetic predictions. In addition, the gastrointestinal representation of PK-Sim was analyzed to provide a mechanistic comparison across platforms.
Methods: The ADAM model was reconstructed in R using a system of ordinary differential equations describing gastrointestinal transit, dissolution, absorption, and first-pass processes across nine intestinal segments [1]. Intestinal transit was modeled using first-order rate constants derived from regional mean residence times. A diffusion layer model (DLM) was implemented but not activated. Dissolution was instead described using a time-dependent profile, consistent with the “dissolution profile” option in Simcyp for immediate-release formulations.
Simulations were performed in a virtual population of 100 fasted female subjects aged 18–45 years following a single oral dose of 0.1 mg of levonorgestrel, with profiles evaluated over 24 hours. Model comparison was conducted at two distinct levels. First, a structural comparison was performed, examining the underlying equations and parameter definitions between the R implementation and Simcyp. Second, a quantitative comparison was carried out using key pharmacokinetic metrics, including fraction absorbed (Fa), maximum concentration (Cmax), time to maximum concentration (Tmax), and area under the curve (AUC). Additionally, the PK-Sim gastrointestinal model was analyzed based on published descriptions, with particular attention to its anatomical structure and governing equations [2-4].
Results: The explicit reconstruction of the ADAM model in R clarified its segmental structure and enabled evaluation of its impact on pharmacokinetic predictions. The model predicted near-complete absorption, with fraction absorbed (Fa) of 0.998 and intestinal availability (Fg) close to unity. Absorption was rapid, with a T50 of 0.72 h and a T90 of 2.54 h. The mean predicted Cmax was 1.20 ng/mL with a Tmax of 0.75 h, and the mean AUC was 11.24 ng·h/mL. Regional analysis showed that absorption was predominantly localized in proximal segments, with approximately 93% of the dose absorbed in the duodenum and jejunum, while distal segments contributed about 7%.
In Simcyp, absorption was lower, with a mean Fa of 0.89 (median 0.91) and moderate variability (CV ≈ 7–8%, 5th–95th percentile: 0.78–0.98), despite identical formulation inputs. Intestinal availability remained close to unity. Systemic exposure was comparable, with a slightly higher Cmax (1.32 ng/mL, + 9.6%) and a lower AUC (10.64 ng·h/mL, – 5.3%) compared to the R model. These findings indicate that similar pharmacokinetic metrics may arise from different underlying absorption dynamics.
Analysis of PK-Sim highlighted a fundamentally different representation of gastrointestinal absorption. The model describes the gastrointestinal tract using multiple anatomical segments within a whole-body PBPK framework, with explicit representation of luminal, interstitial, and cellular compartments [2-4]. Drug absorption is governed by permeability-driven transport across membranes, while dissolution, transit, and absorption occur simultaneously. In contrast, the ADAM model relies on a segment-based representation with predefined first-order kinetic processes [1].
These structural differences result in distinct absorption behaviors. ADAM-based models concentrate absorption in proximal segments and generate rapid drug input into systemic circulation. In contrast, PK-Sim distributes absorption more gradually and is more sensitive to physiological determinants such as gastric emptying and membrane permeability [2-4]. These differences are reflected in pharmacokinetic profiles, with a lower Cmax in PK-Sim (0.59 ng/mL) compared to the R model (1.20 ng/mL) and Simcyp (1.32 ng/mL), together with a delayed Tmax.
Conclusion: This analysis shows that models based on similar physiological foundations may generate different dynamic behaviors depending on their mathematical structure. The explicit reconstruction of the ADAM model in R provides a transparent framework to examine these structural choices and better characterize how they may influence interpretation of oral absorption processes and their contribution to pharmacokinetic predictions.
References:
[1] M. Jamei et al., “Population-based mechanistic prediction of oral drug absorption,” AAPS J, vol. 11, no. 2, pp. 225-37, Jun 2009, doi: 10.1208/s12248-009-9099-y.
[2] S. Willmann, J. Lippert, M. Sevestre, J. Solodenko, F. Fois, and W. Schmitt, “PK-Sim®: a physiologically based pharmacokinetic ‘whole-body’ model,” Biosilico, vol. 1, no. 4, pp. 121-124, 2003, doi: 10.1016/s1478-5382(03)02342-4.
[3] K. Thelen, K. Coboeken, S. Willmann, J. B. Dressman, and J. Lippert, “Evolution of a detailed physiological model to simulate the gastrointestinal transit and absorption process in humans, part II: extension to describe performance of solid dosage forms,” J Pharm Sci, vol. 101, no. 3, pp. 1267-80, Mar 2012, doi: 10.1002/jps.22825.
[4] K. Thelen, K. Coboeken, S. Willmann, R. Burghaus, J. B. Dressman, and J. Lippert, “Evolution of a detailed physiological model to simulate the gastrointestinal transit and absorption process in humans, part 1: oral solutions,” J Pharm Sci, vol. 100, no. 12, pp. 5324-45, Dec 2011, doi: 10.1002/jps.22726.
Reference: PAGE 34 (2026) Abstr 12247 [www.page-meeting.org/?abstract=12247]
Poster: Drug/Disease Modelling - Absorption & PBPK