Successful Prediction of Human Pharmacokinetics by Improving Calculation Processes of Physiologically Based Pharmacokinetic Approach

Application of Minimal Physiologically-Based Pharmacokinetic Model to Simulate Lung and Trachea Exposure of Pyronaridine and Artesunate in Hamsters

A fixed-dose combination of pyronaridine and artesunate, one of the artemisinin-based combination therapies, has been used as a potent antimalarial treatment regimen. Recently, several studies have reported the antiviral effects of both drugs against severe acute respiratory syndrome coronavirus two (SARS-CoV-2). However, there are limited data on the pharmacokinetics (PKs), lung, and trachea exposures that could be correlated with the antiviral effects of pyronaridine and artesunate. The purpose of this study was to evaluate the pharmacokinetics, lung, and trachea distribution of pyronaridine, artesunate, and dihydroartemisinin (an active metabolite of artesunate) using a minimal physiologically-based pharmacokinetic (PBPK) model. The major target tissues for evaluating dose metrics are blood, lung, and trachea, and the nontarget tissues were lumped together into the rest of the body. The predictive performance of the minimal PBPK model was evaluated using visual inspection between observations and model predictions, (average) fold error, and sensitivity analysis. The developed PBPK models were applied for the multiple-dosing simulation of daily oral pyronaridine and artesunate. A steady state was reached about three to four days after the first dosing of pyronaridine and an accumulation ratio was calculated to be 1.8. However, the accumulation ratio of artesunate and dihydroartemisinin could not be calculated since the steady state of both compounds was not achieved by daily multiple dosing. The elimination half-life of pyronaridine and artesunate was estimated to be 19.8 and 0.4 h, respectively. Pyronaridine was extensively distributed to the lung and trachea with the lung-to-blood and trachea-to-blood concentration ratios (=Cavg,tissue/Cavg,blood) of 25.83 and 12.41 at the steady state, respectively. Also, the lung-to-blood and trachea-to-blood AUC ratios for artesunate (dihydroartemisinin) were calculated to be 3.34 (1.51) and 0.34 (0.15). The results of this study could provide a scientific basis for interpreting the dose–exposure–response relationship of pyronaridine and artesunate for COVID-19 drug repurposing.

A Novel Experimental and Theoretical Method for Estimating Albumin-Mediated Hepatic Uptake Based on the Albumin Binding Fraction in Plasma and Human PK Prediction Using a Physiologically-Based Pharmacokinetic Approach

Recently, protein-facilitated uptake has been suggested to be an important factor in the precise prediction of the pharmacokinetic (PK) profiles of drugs. In our previous study, a physiologically-based pharmacokinetic (PBPK) approach considering the mechanism of albumin-mediated hepatic uptake was developed for predicting human PK profiles. It was assumed that drugs affected by albumin-mediated hepatic uptake would bind only to albumin, which means that there would be over-estimation of the contribution of protein-facilitated uptake for a drug that could bind to multiple proteins. In this study, we developed a method that can evaluate the albumin binding fraction in plasma considering the affinity for other proteins. Based on the albumin binding fraction, the contribution of albumin-mediated hepatic uptake was theoretically estimated, and then the human PK profiles were predicted by our proposed PBPK approach incorporating this mechanism. As a result, the predicted human PK profiles agreed well with the observed ones, and the absolute average fold error of PK parameters was almost within a 1.5-fold error on average. These findings show the importance of considering protein-facilitated uptake and also suggest that our proposed PBPK approach can be useful in scientific discussions with regulatory authorities.

The Novel In Vitro Method to Calculate Tissue-to-Plasma Partition Coefficient in Humans for Predicting Pharmacokinetic Profiles by Physiologically-Based Pharmacokinetic Model With High Predictability

Proper prediction of human pharmacokinetic (PK) profiles can accelerate the compound selection in drug discovery. Recently, we reported a robust bottom-up physiologically-based pharmacokinetic (PBPK) approach (J Pharm Sci. 2019 Aug; 108(8):2718-2727), which uses the in vivo rat distribution volume at the steady state (V_ss) to determine human tissue-to-plasma partition coefficients (Kp_tissue). Here, we report on a bottom-up PBPK approach that can simulate the PK profile with both high-throughput and high-predictive accuracy only using in vitro data. In this study, as an alternative parameter of in vivo rat V_ss which was used for the correction of human Kp_tissue, V_{ss, in vitro} was obtained from protein binding data in rats, and the values of V_{ss, in vitro} for 31 reference compounds showed good correlation with the observed rat V_ss (R² = 0.859). Next, rat and human PK profiles of reference compounds were predicted by the bottom-up PBPK approach using Kp_tissue corrected by rat V_{ss, in vitro}. As a result, the absolute average fold errors for pharmacokinetic parameters were almost less than 2, showing that these PK profiles could be accurately predicted using in vitro data. This method enables the screening of promising compounds with good PK profiles in humans at an early stage of drug discovery.

A Physiologically Based in Silico Tool to Assess the Risk of Drug-Related Crystalluria

Drug precipitation in the nephrons of the kidney can cause drug-induced crystal nephropathy (DICN). To aid mitigation of this risk in early drug discovery, we developed a physiologically based in silico model to predict DICN in rats, dogs, and humans. At a minimum, the likelihood of DICN is determined by the level of systemic exposure to the molecule, the molecule's physicochemical properties and the unique physiology of the kidney. Accordingly, the proposed model accounts for these properties in order to predict drug exposure relative to solubility along the nephron. Key physiological parameters of the kidney were codified in a manner consistent with previous reports. Quantitative structure-activity relationship models and in vitro assays were used to estimate drug-specific physicochemical inputs to the model. The proposed model was calibrated against urinary excretion data for 42 drugs, and the utility for DICN prediction is demonstrated through application to 20 additional drugs.