Comparison of the Predictability of Human Hepatic Clearance for Organic Anion Transporting Polypeptide Substrate Drugs Between Different In Vitro–In Vivo Extrapolation Approaches

Abstract

Prediction of human pharmacokinetic profiles of drug candidates is an essential step toward first-in-human studies. However, it remains difficult to quantitatively predict hepatic clearance, particularly when hepatic uptake is mediated by transporter(s). Using 15 organic anion transporting polypeptide (OATP) substrate drugs, we tested 3 in vitro-in vivo extrapolation (IVIVE) approaches to predict overall hepatic intrinsic clearance in vivo (CL). IVIVE approaches involved metabolic intrinsic clearance in human liver microsomes (CL) with or without hepatocyte-to-buffer maximum unbound concentration ratio (K) correction and uptake intrinsic clearance at 37°C (PS) in human hepatocyte suspensions. K and PS values were determined in 2 hepatocyte batches, and all tested compounds showed temperature-dependent uptake, consistent with the fact of transporter-mediated uptake. CL substantially underestimated CL. By multiplying CL by K values, the prediction performance was much improved; however, in vitro-in vivo correlation was poor. Among the IVIVE approaches, PS showed the most robust correlation with CL, and one of the hepatocyte batches could predict CL with a minimal empirical scaling factor. These results suggested IVIVE with hepatic uptake clearance determined in hepatocyte suspensions as one of the most practical approaches for predicting CL of OATP substrate drugs.

Current In Vitro Methods to Determine Hepatic Kp uu : A Comparison of Their Usefulness and Limitations

“…Only for cyclosporine A and erythromycin, temperature-dependent deviations of >4-fold were observed, likely resulting from alterations in membrane binding and lysosomal trapping processes, respectively. Izumi et al 22 recently applied the temperature method to determine Kp uu in human hepatocytes and obtained highly comparable data for an overlapping subset of compounds, thus confirming robustness and reproducibility of the data obtained via the temperature method. Because Kp uu from the temperature method correlated well with ECM-based data for drugs in ECM classes 2/4 and 1/3 (with low intrinsic clearance), we conclude that fu hep from the temperature method matches with the corresponding Kp data and provides realistic estimates of Kp uu in the suspended human hepatocytes (Fig.…”

“…For all compounds, Kp uu was calculated using Kp from steady-state hepatocyte uptake experiments and fu hep either based on the temperature method (Kp uu,temp ), the homogenization method (Kp uu,hom ), or the logD 7.4 method (Kp uu,logD ) according to Equations 7-9: The ECM method has been proposed for the prediction of overall intrinsic hepatic clearance (CL int,h ) using in vitro process clearances as follows 8,17,18,22 (10) where PS inf is the sum of active (PS inf,act ) and passive uptake membrane permeability (PS inf,pas ), PS eff is the sum of active (PS eff,act ) and passive sinusoidal efflux membrane permeability (PS eff,pas ) and CL int is the sum of intrinsic metabolic (CL int,met ) and biliary clearances (CL int,sec ) ( Table 3). Accordingly, ECM-based Kp uu values (Kp uu,ECM ) can be calculated according to Equation 11:…”

The Importance of Incorporating OCT2 Plasma Membrane Expression and Membrane Potential in IVIVE of Metformin Renal Secretory Clearance

“…Many groups have used primary cells (e.g., human/rat hepatocytes) and empirical scaling factors (Jones et al, 2012;Zou et al, 2013;Izumi et al, 2017) to predict hepatobiliary clearance of drugs. Such methods lack mechanistic insight and therefore may not necessarily apply to a new molecular entity.…”

Prediction of Clinical Transporter‐Mediated Drug–Drug Interactions via Comeasurement of Pitavastatin and Eltrombopag in Human Hepatocyte Models

“…The clinical pharmacokinetic data following an oral 1 mg pitavastatin dose in healthy volunteers 2 were used to validate the pitavastatin PBPK model. As the PBPK model was developed for pitavastatin in the blood, the clinical data were converted to blood (ng/mL) by dividing them by the blood:plasma ratio obtained from the literature (0.425 24 ).…”