Background: Andrographolide has a potent antiviral effect in the treatment of coronavirus disease (COVID-19). However, there are no in vivo studies of andrographolide as an anti-COVID-19 treatment.

Objective: The study aims to develop a physiologically based pharmacokinetic (PBPK) animal model and scale it up to a human model to predict andrographolide concentrations in the lungs.

Methods: ADAPT5 (version 5.0.58) was used to establish the PBPK model based on 24 enrolled pharmacokinetic studies.

Results: The perfusion-limited PBPK model was developed in mice and extrapolated to rats, dogs, and humans. The metabolism of andrographolide in humans was described by the Michaelis-Menten equation. The saturation of the metabolism occurred at a high dose (12 g), which could not be used therapeutically. The optimized oral bioavailability in humans was 6.3%. Due to the limit of solubility, the dose-dependent absorption between 20-1000 mg was predicted by GastroPlus®. Using the extrapolated human PBPK model together with the predicted dose-dependent fraction of the dose absorbed that enters the enterocytes by GastroPlus®, the oral dosage of 200 mg q8h of andrographolide would provide a trough level of free andrographolide at a steady state over the reported IC₅₀ value against SARS-CoV-2 in the lungs for the majority of healthy humans. Based on the reported CC₅₀ value, toxicity might not occur at the therapeutic dosage.

Conclusion: The PBPK model of andrographolide in animals and humans was successfully constructed. Once additional data is available, the model would be needed to recalibrate to gain an understanding of a dose-response relationship and optimization of dosage regimens of andrographolide.