Marina Navas Bachiller 1, Nikoletta Fotaki 1
1 Department of Life Sciences, University Of Bath (Bath, United Kingdom)
INTRODUCTION
Successful treatment of uncomplicated malaria relies on oral therapies, yet the absorption of antimalarial agents such as artesunate remains poorly characterised, contributing to variable exposure and treatment outcomes.
Reduced artesunate exposure was reported for a paediatric granule formulation compared to a reference tablet formulation in Korean subjects, with dissolution related mechanisms proposed as a potential cause [1]. To mechanistically investigate the key formulation and physiological factors underlying this disparity, physiologically based biopharmaceutics modeling (PBBM) can be employed.
In this work, the first AS PBBM was developed and applied to investigate key determinants of oral absorption for two formulations of artesunate in two populations.
METHODS
BIORELEVANT DISSOLUTION
Dissolution studies were conducted for one tablet (60 mg artesunate) or one granule sachet (20 mg artesunate) in three set-ups (n = 3, 37oC): (a) USP II apparatus in 500 mL of FaSSIF-V2 at 75 rpm for 4 h, (b) USP II apparatus in 250 mL of FaSSGF for 30 min at 75 rpm followed by FaSSIF (250 mL, 75 rpm) for 210 min (two-stage) and (c) USP IV apparatus with Ø 22.6 mm cells using FaSSGF for 30 min at 12 mL/min followed by FaSSIF for 240 min at 4 mL/min (open-mode). 2 mL samples were taken at pre-specified timepoints, filtered and analysed by HPLC.
PBBM DEVELOPMENT
A PBBM for artesunate was developed in Simcyp® v22 using drug physicochemical parameters [2, 3]. Distribution was described with a perfusion-limited minimal model. The model was validated against clinical studies [4, 5].
An Advanced Distribution and Metabolism Model (ADAM) oral absorption model was then developed using two approaches. Firstly, a Diffusion Layer Model (DLM) was developed and optimized using clinical data from [6]. Key inputs included solubility in FaSSGF/FaSSIF [2], luminal degradation rate constants representing pH-dependent hydrolysis [7], particle size and effective permeability.
Subsequently, physiologically relevant dissolution profiles from each set up were input as tabulated data or fitted to a Weibull equation (Equation 1) and integrated into the model. The healthy Korean population from SimCYP repository was utilised. Predicted profiles were compared to the clinical ones in Korean healthy subjects [1]. Additional exploratory simulations were conducted with the healthy Caucasian population to assess the impact of ethnicity on artesunate pharmacokinetic profiles. Model performance was evaluated through fold error metrics.
F=F_max (1-e^(-((t^β))⁄α) ) Equation 1
RESULTS
IMPACT OF GASTROINTESTINAL PHYSIOLOGY ON ARTESUNATE DISSOLUTION
Artesunate dissolution in gastric media was slower for the tablet than the granule (3% vs 21% dissolved at 30 min), consistent with the low gastric solubility of artesunate and the higher drug load of the tablet.
In the USP IV apparatus, the tablet dissolved only slightly slower than the granule (7% vs 16% dissolved at 30 min), supporting saturation-limited dissolution in the USP II apparatus.
IMPACT OF GASTROINTESTINAL PHYSIOLOGY ON PREDICTED PHARMACOKINETIC PROFILES
The ADAM DLM model, with solubility and luminal degradation inputs, overpredicted the plasma concentration profiles for both formulations, indicating that solubility alone could not capture the in vivo behaviour of these artesunate formulations (AAFE = 2.35-3.55). The integration of formulation-specific data in biorelevant conditions mimicking the gastrointestinal tract environment led to successful predictions.
When biorelevant dissolution data generated in the USP II apparatus in intestinal media were integrated in the model, an overprediction was observed for both formulations. No overprediction was observed when the full dissolution profile in gastric and intestinal media was integrated, highlighting the importance of considering dissolution in gastric environments even for weakly acidic compounds like artesunate.
IMPACT OF ETHNICITY-ASSOCIATED PHYSIOLOGY ON PREDICTED PHARMACOKINETIC PROFILES
Exploratory simulations with a Caucasian population indicated increased exposure compared to the Korean population and overall decreased predictive power. The difference can be attributed to the higher haematocrit in the Korean population which results in higher artesunate partitioning into red blood cells and the higher liver volume, which results in higher hepatic clearance and reduced exposure.
Although ethnic differences in CYP and UGT were integrated in the model, artesunate clearance is not driven by these pathways, highlighting the importance of characterizing and including non enzymatic physiological differences in PBBM.
CONCLUSION
The current work demonstrated that incorporating gastrointestinal and population specific physiology is essential for accurately predicting artesunate absorption, highlighting PBBM as a valuable tool to characterise key determinants of antimalarial absorption.
References:
REFERENCES
1. European Medicines Agency (EMA), Pyramax. Assessment report. EMA/813257/2015. 2015.
2. Charman SA, Andreu A, Barker H, Blundell S, Campbell A, Campbell M, Chen G, Chiu FCK, Crighton E, Katneni K, Morizzi J, Patil R, Pham T, Ryan E, Saunders J, Shackleford DM, White KL, Almond L, Dickins M, Smith DA, Moehrle JJ, Burrows JN, Abla N. An in vitro toolbox to accelerate anti-malarial drug discovery and development. Malaria Journal. 2020;19(1):1. 10.1186/s12936-019-3075-5
3. Arey R, Reisfeld B. Predicting the disposition of the antimalarial drug artesunate and its active metabolite dihydroartemisinin using physiologically based pharmacokinetic modeling. Antimicrobial Agents and Chemotherapy. 2021;65(3). 10.1128/AAC.02280-20
4. Tarning J, Hanboonkunupakarn B, Hoglund RM, Chotivanich K, Mukaka M, Pukrittayakamee S, Day NPJ, White NJ, Dondorp AM, Jittamala P. Safety and pharmacokinetic properties of a new formulation of parenteral artesunate in healthy Thai volunteers. Malar J. 2024;23(1):296. 10.1186/s12936-024-05085-9
5. Li Q, Cantilena LR, Leary KJ, Saviolakis GA, Miller RS, Melendez V, Weina PJ. Pharmacokinetic profiles of artesunate after single intravenous doses at 0.5, 1, 2, 4, and 8 mg/kg in healthy volunteers: A phase i study. Am J Trop Med Hyg. 2009;81(4):615-21. 10.4269/ajtmh.2009.09-0150
6. Navaratnam V, Mansor SM, Mordi MN, Akbar A, Abdullah MN. Comparative pharmacokinetic study of oral and rectal formulations of artesunic acid in healthy volunteers. Eur J Clin Pharmacol. 1998;54(5):411-4. 10.1007/s002280050484
7. Haydar MA, Degradation of artesunate in aqueous solution. 2011, Curtin University.
ACKNOWLEDGEMENTS
This publication is based on research funded in part by the Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect positions or policies of the Gates Foundation.
Reference: PAGE 34 (2026) Abstr 12309 [www.page-meeting.org/?abstract=12309]
Poster: Drug/Disease Modelling - Other Topics