(595g) Digital Twin Modeling of Liver-on-a-Chip Systems: Integrating Computational Fluid Dynamics and Biochemical Kinetics for Drug Metabolism Studies

Accurate modeling of liver function is essential in drug discovery, providing critical insights into hepatic metabolism, toxicity, and drug interactions [1]. As the primary site for drug biotransformation, the liver converts pharmaceuticals into active or inactive metabolites, influencing therapeutic efficacy and safety. Traditional in vivo and in vitro approaches for studying hepatic metabolism are constrained by ethical concerns, species-specific physiological differences, and high operational costs [2]. Emerging platforms, such as organ-on-a-chip technologies and digital twin models, address these limitations by enabling high-fidelity simulation of liver function in a controlled environment [3-5].

This study focuses on developing a digital twin framework for a liver-on-a-chip system designed to emulate the tissue-vessel interface of hepatic microenvironments. The device comprises two fluidic channels embedded within a solid scaffold, separated by a central porous hydrogel supporting functional hepatic cell culture [6]. Nutrients such as glucose and amino acids diffuse across a semi-permeable membrane into the hydrogel, where they are consumed by liver cells. Concurrently, metabolic byproducts diffuse out of the hydrogel and are transported downstream through the outlet channel. Species transport in the hydrogen domain is governed by diffusive gradients, with effective diffusivities estimated using the Stokes-Einstein relation and porous media models. Fluid flow within the microchannels is assumed to be laminar and described by the Navier-Stokes equations, with advective transport incorporated into the species balance equations. Biochemical reactions, including amino acid catabolism and carbohydrate metabolism, were modeled using kinetic parameters calibrated from experimental studies involving primary human hepatocytes [7]. The resulting reaction network captured key proteolytic trends, particularly the release of amino acids into the extracellular medium.

Simulation results demonstrated distinct spatial and temporal concentration profiles during the first 48 hours, with nutrient gradients and metabolite levels stabilizing as the system approached pseudo-steady state conditions. These outcomes underscore the digital twin’s ability to mimic physiological liver responses, offering predictive capabilities for in vitro drug metabolism and toxicity assessments. In addition to simulating local hepatic activity, this work incorporated a systemic perspective by examining the liver’s role in whole-body energy homeostasis. An integrated computational model was developed using experimental data [8] to quantify metabolic fluxes associated with energy production, amino acid turnover, and urea cycle activity. A computational fluid dynamics modeling approach was employed to represent hepatic perfusion dynamics, capturing the interplay between molecular transport and biochemical reaction kinetics under varying physiological conditions. The digital twin model was further validated under conditions simulating glucose-enriched perfusion, highlighting its adaptability to diverse experimental setups.

References:

[1] Kratochwil, N. A., Meille, C., Fowler, S., Klammers, F., Ekiciler, A., Molitor, B., ... & Parrott, N. J. (2017). Metabolic profiling of human long-term liver models and hepatic clearance predictions from in vitro data using nonlinear mixed-effects modeling. The AAPS journal, 19(2), 534-550.

[2] Godoy, P., Hewitt, N. J., Albrecht, U., Andersen, M. E., Ansari, N., Bhattacharya, S., ... & Hengstler, J. G. (2013). Recent advances in 2D and 3D in vitro systems using primary hepatocytes, alternative hepatocyte sources and non-parenchymal liver cells and their use in investigating mechanisms of hepatotoxicity, cell signaling and ADME. Archives of Toxicology, 87, 1315–1530.

[3] Yang, J. W., Khorsandi, D., Trabucco, L., Ahmed, M., Khademhosseini, A., Dokmeci, M. R., ... & Jucaud, V. (2024). Liver-on-a-chip integrated with label-free optical biosensors for rapid and continuous monitoring of drug-induced toxicity. Small, 20(48), 2403560.

[4] Lin, C., & Khetani, S. R. (2016). Advances in engineered liver models for investigating drug-induced liver injury. BioMed Research International, 2016, 1829148.

[5] Kuepfer, L., Niederalt, C., Wendl, T., Schlender, J. F., Willmann, S., Lippert, J., Block, M., Eissing, T., & Teutonico, D. (2016). Applied concepts in PBPK modeling: How to build a PBPK/PD model. CPT: Pharmacometrics & Systems Pharmacology, 5(10), 516–531.

[6] Khorsandi, D., Yang, J., Jenson, S., Kajino, T., Maity, S., Salih, A. R. C., ... & Dokmeci, M. R. (2024). In Lab-on-a-chip devices for advanced biomedicines: Laboratory scale engineering to clinical ecosystem (Vol. 25, Ch. 17, pp. 560–614), edited by A. Parihar & P. Pradeep Mehta. Royal Society of Chemistry.

[7] Guthke, R., Schmidt-Heck, W., Pless, G., Gebhardt, R., Pfaff, M., Gerlach, J. C., & Zeilinger, K. (2006). Dynamic model of amino acid and carbohydrate metabolism in primary human liver cells. In Biological and Medical Data Analysis: 7th International Symposium, ISBMDA 2006, Thessaloniki, Greece, December 7–8, 2006. Proceedings 7 (pp. 137–149). Springer.

[8] Ashworth, W. B., Davies, N. A., & Bogle, I. D. L. (2016). A computational model of hepatic energy metabolism: Understanding zonated damage and steatosis in NAFLD. PLoS Computational Biology, 12(9), e1005105.