Comparison of Fraction Unbound Between Liver Homogenate and Hepatocytes at 4°C

Fraction unbound (f_u) values obtained from liver or hepatocyte homogenate with equilibrium dialysis (f_u,homo) or the hepatocyte partition coefficient method at 4°C (f_u,c) are both frequently used to estimate unbound drug concentrations (C_u) and unbound partition coefficient (Kp_uu) of the liver. Literature data are somewhat controversial on this topic: some reported that the two methods gave comparable f_u values, while others showed that they had no correlation. To better understand the two approaches, 44 structurally diverse compounds with wide ranges of f_u values were used for the comparison study. The results indicate that f_u values from the two methods are comparable with an average fold error of 2.9-fold and a bias of 1.0. Although some outliers were observed, the reasons are not entirely clear and further investigations are needed. As the f_u data from both methods are correlated, f_u,homo measurement using tissue homogenate is recommended over cells at 4°C (f_u,c) in early drug discovery. This is because f_u,homo method is more reliable, has good in vivo predictability, and feasibility for any tissue types where representative cells may not be readily available. The approach can be used to estimate C_u and Kp_uu of tissues in order to develop pharmacokinetic/pharmacodynamic relationships, and to estimate therapeutic indices, as well as to predict drug-drug interactions.