Abstract

In this study, the data of 113 human bioequivalence (BE) studies of immediate release (IR) formulations of 74 active pharmaceutical ingredients (APIs) conducted at Sawai Pharmaceutical Co., Ltd., was analyzed to understand the factors affecting intra- and intersubject variabilities in oral drug absorption. The ANOVA CV (%) calculated from area under the time–concentration curve (AUC) in each BE study was used as an index of intrasubject variability (Vintra), and the relative standard deviation (%) in AUC was used as that of intersubject variability (Vinter). Although no significant correlation was observed between Vintra and Vinter of all drugs, Vintra of class 3 drugs was found to increase in association with a decrease in drug permeability (P_eff). Since the absorption of class 3 drugs was rate-limited by the permeability, it was suggested that, for such drugs, the low P_eff might be a risk factor to cause a large intrasubject variability. To consider the impact of poor water solubility on the variability in BE study, a parameter of P_eff/Do (Do; dose number) was defined to discriminate the solubility-limited and dissolution-rate-limited absorption of class 2 drugs. It was found that the class 2 drugs with a solubility-limited absorption (P_eff/Do < 0.149 × 10^–4 cm/s) showed high intrasubject variability. Furthermore, as a reason for high intra- or intersubject variability in AUC for class 1 drugs, effects of drug metabolizing enzymes were investigated. It was demonstrated that intrasubject variability was high for drugs metabolized by CYP3A4 while intersubject variability was high for drugs metabolized by CYP2D6. For CYP3A4 substrate drugs, the K_m value showed the significant relation with Vintra, indicating that the affinity to the enzyme can be a parameter to predict the risk of high intrasubject variability. In conclusion, by analyzing the in house data of human BE study, low permeability, solubility-limited absorption, and high affinity to CYP3A4 are identified as risk factors for high intrasubject variability in oral drug absorption. This information is of importance to design the human BE study for oral drug products containing APIs with a risk of large intrasubject variability in oral absorption.