Analysis of Risk Factors in Human Bioequivalence Study That Incur Bioinequivalence of Oral Drug Products

Publication: Mol Pharm
Software: ADMET Predictor®

Abstract

In the study of human bioequivalence (BE), newly developed oral products sometimes fail to prove BE with a reference product due to the high variability in pharmacokinetic (PK) parameters after oral absorption. In this study, risk factors that incur bioinequivalence in BE study were analyzed by applying the Biopharmaceutics Classification System (BCS). Forty-four generic products were selected from a database of BE studies in the past 10 years at Towa Pharmaceutical Co., Ltd. (Osaka, Japan), and 90% confidence interval (CI) of AUC and C(max) in human BE study for all products were converted into coefficient of variation (CV(90)). Then, the required number of subjects to confirm BE was estimated from the 90% CI in human BE study of new products. It was found that both the permeability of drugs to human intestinal membrane (P(eff)) and the dose number calculated from their water solubility did not correlate well to CV(90) and the estimated subject number in human BE study, suggesting the contribution of other factors to cause the variability in oral drug absorption. As the PK parameter of drugs, the value of AUC/dose was calculated and plotted against CV(90) and the estimated subject number by classifying drugs into 4 BCS classes. For drugs in classes 1 and 3, AUC/dose gave a clear criterion to distinguish the drugs with a high risk of bioinequivalence, where drugs with low AUC/dose showed high CV(90) and large number of subjects. It was suggested that the high metabolic clearance (for class 1 drug) and low oral absorption (for class 3 drug) could be significant factors to incur bioinequivalence in human BE study, although for drugs in classes 2 and 4, clear factors were not defined. Consequently, for drugs in BCS classes 1 and 3, risks in human BE study to incur bioinequivalence could be predicted by calculating the AUC/dose. In the case of generic drugs, since the parameter of AUC/dose is available before initiating human BE study, this finding is expected to promote an efficient and cost-saving strategy for the development of oral drug products.