Application Of Gastroplus^TM Software In The Field Of Drug Modeling Establishment, Human Clearance Prediction And Clinical Formulation Design

During the early stage of drug discovery, a report pointed that approximately 40% compounds of the candidates failed due to the undesirable pharmacokinetics. Usually, it takes about 0.8 billion dollars and 15 years for a new drug from the early phase to the commercial stage. So, it is very important to predict and evaluate human pharmacokinetics performance by using animal trial data and appropriate simulation tools.Currently, there exist much commercial software to predict compound pharmacokinetics in animal and human, including GastroPlusTM(Simulation Plus Inc.,Lancaster,California,USA), SimCYP(SimCYP Ltd,Sheffield,UK), PKSIM(Bayer Technology Services,Germany) etc. From another point of view, many international pharmacy companies have used these computer tools to achieve rational experimental design, shorten research time and save research money.This thesis took tolbutamide, omeprazole and diltiazem hydrochloride as modeling drugs and studied how to establish rat and human model with the aid of GastroPlusTM software systematically. Meanwhile, in this thesis, parameter sensitivity analysis was also performed for searching the factors that impacted the simulated concentration-time profile. In addition, we also concluded the application of GastroPlusTM in the field of human clearance prediction and compared different clearane prediction methods. The influence of dosage form, dose and food effect on the simulated result was investigated further.To sum up, some results were obtained through our research. Firstly, the predictive of animal model had a huge influence on the predictive accuracy of human model. Secondly, the accuracy of human clearance values was very important for the accuracy of human model. Thirdly, for the compounds which were easily degraded in physiological media, it’s better to use.cdd file for obtaining a better prediction. Finally, appropriate model could increase the confidence of dosage form and dose prediction. In addition, parameter sensitivity analysis could provide the factors that influenced the concentration-time profile.With the help of GastroPlusTM model, the in vivo rat clearance values predicted via the models from rat hepatocyte and rat liver microsomes were within acceptable range compared with the data reported in literature. However, the prediction of human in vivo clearance using the same approach resulted in predicted values outside acceptable range for most drugs studies. In contrast, human in vivo clearance values predicted from rat in vivo clearance using allometric scaling method gave values within acceptable range compared with human in vivo clearance reported in literature.During the early stage of drug discovery, the human clinical data was not acquired. So the establishment and validation of animals had a huge influence on the accuracy human model. Based on this situation, multi-species models validation is significantly important for increasing the accuracy of human model. However, due to time and data derivation, we didn’t perform multi-species models validation. Hopefully we look forward to completing this section further. In summary, this thesis provided some suggestions and references for usage of Gastro PlusTM in the field of drug discovery.