A semi-physiologically based pharmacokinetic pharmacodynamic model for glycyrrhizin-induced pseudoaldosteronism and prediction of the dose limit causing hypokalemia in a virtual elderly population.

Glycyrrhizin (GL) is a widely used food additive which can cause severe pseudoaldosteronism at high doses or after a long period of consumption. The aim of the present study was to develop a physiologically based pharmacokinetic (PBPK) pharmacodynamic (PD) model for GL-induced pseudoaldosteronism to improve the safe use of GL. Since the major metabolite of GL, glycyrrhetic acid (GA), is largely responsible for pseudoaldosteronism via inhibition of the kidney enzyme 11β-hydroxysteroiddehydrogenase 2 (11β-HSD 2), a semi-PBPK model was first developed in rat to predict the systemic pharmacokinetics of and the kidney exposure to GA. A human PBPK model was then developed using parameters either from the rat model or from in vitro studies in combination with essential scaling factors. Kidney exposure to GA was further linked to an Imax model in the 11β-HSD 2 module of the PD model to predict the urinary excretion of cortisol and cortisone. Subsequently, activation of the mineralocorticoid receptor in the renin-angiotensin-aldosterone-electrolyte system was associated with an increased cortisol level. Experimental data for various scenarios were used to optimize and validate the model which was finally able to predict the plasma levels of angiotensin II, aldosterone, potassium and sodium. The Monte Carlo method was applied to predict the probability distribution of the individual dose limits of GL causing pseudoaldosteronism in the elderly according to the distribution of sensitivity factors using serum potassium as an indicator. The critical value of the dose limit was found to be 101 mg with a probability of 3.07%.