PWII-002 - DEVELOPMENT OF ADEFOVIR PBPK MODEL TO SUPPORT PROSPECTIVE EVALUATION OF BIOMARKER-INFORMED OAT1 INHIBITION AND EFFECT OF RENAL IMPAIRMENT

S. Tan¹, H. Wu², A. Rostami-Hodjegan^1,3, D. Scotcher¹, A. Galetin¹; ¹The University of Manchester, Manchester, United Kingdom, ²University of Manchester, England, United Kingdom, ³Certara Predictive Technologies, Sheffield, United Kingdom.

Background: Combining endogenous biomarker clinical data with physiologically-based pharmacokinetic (PBPK) modelling is envisioned to inform or replace dedicated transporter-mediated drug-drug interaction (DDI) clinical trials with clinical probe substrates. Similarly, PBPK modelling may be leveraged upon to predict the effect of renal impairment. The aim of this study was to develop a PBPK model of adefovir, a renal organic anion transporter 1 (OAT1) clinical probe, to predict OAT1-mediated DDI and effect of chronic kidney disease (CKD).

Methods: PBPK model of adefovir was developed in Simcyp V23 by incorporating experimental measurements of adefovir OAT1-mediated transport and passive diffusion for in vitro-in vivo extrapolation of adefovir renal secretion clearance. A previously verified probenecid model incorporating in vivo OAT1/3 inhibitory constant estimated from OAT1/3 endogenous biomarker (4-pyridoxic acid) clinical data was applied [1]. Application of adefovir model to patients with CKD accounted for our previously recommended additional 50% decrease in OAT1 activity beyond the decline of glomerular filtration rate in severe CKD.

Results: Adefovir model was verified with a large clinical dataset investigating a range of adefovir intravenous and adefovir-dipivoxil oral doses (n = 9 studies, 188 subjects). Application of the biomarker-informed probenecid model accurately predicted the increase in adefovir area under the curve (AUC) and maximum plasma concentration (Cmax) after 0.75g to 1.5g single oral probenecid dose within the stringent Guest criterion [2]. Interaction with the lowest probenecid dose of 0.5g was slightly overpredicted, but AUC and Cmax ratio were predicted within 1.5-fold error. Extension of PBPK models to severe CKD populations successfully predicted the 6.2-fold increase to AUC and 6.1-fold decrease to renal clearance of adefovir.

Conclusion: This study has reinforced the role of biomarker-informed PBPK modelling to predict OAT1-mediated DDI in lieu of dedicated clinical study. Moreover, PBPK modelling can be applied to predict the effect of renal impairment on OAT1 renally eliminated drugs and potentially reduce need for CKD clinical studies.

[1] Tan S.P.F et al. Clin Pharmacol Ther. 114, 1243-53 (2023)
[2] Guest et al. Drug Metab Dispos. 39:170-3 (2011)