(S-052) Update on Current Status of PBPK Modeling Application in Hepatic Impairment Populations: Recent Successes & Knowledge Gaps
Sunday, October 19, 2025
7:00 AM - 5:00 PM MDT
Location: Colorado A
Laura Al-Amiry Santos – Certara; Zoe Barter – Certara; Kuan-Fu Chen – Certara; Oliver Hatley – Certara; Eleanor Howgate – Certara; Hannah Jones – Certara; Alice Ban Ke – Certara
SVP Head of Simcyp PBPK Modelling Services Certara Concord, , Massachusetts, United States
Disclosure(s):
Hannah M. Jones, PhD: No financial relationships to disclose
Objectives: Hepatic impairment studies are resource intensive and it can take several years to recruit enough patients with different levels of disease severity. Physiologically based pharmacokinetic (PBPK) models are increasingly being used within a quantitative translational framework to predict the exposures of drugs in subjects with differing ages, genetics, ethnicities and disease states including organ impairment, and can be used to provide dose recommendations in clinical studies and labeling. The purpose of this work was to demonstrate the successes and/or limitations of PBPK modeling in supporting the dosing or labeling of recently approved drugs in hepatic impairment (HI) populations.
Methods: We extracted several case studies of recently approved drugs from publicly available FDA OCP reviews and publications. Relevant information were reviewed to understand the predictability of PBPK models for HI populations, the types of applications commonly used and the acceptance by regulators.
Results: Based on this analysis it was determined that prior to performing a HI study, PBPK modelling is often performed to rationalize the timing or need for a study; to broaden eligibility criteria in phase II/III trial; to help design a leaner HI study and select doses for the study. In addition, once a HI study has been performed, PBPK modelling is used to extrapolate from a single dose to multiple dose scenario for drugs with time dependent PK; extrapolate from one or two HI categories to all categories; supplement limited PK data if enrollment is incomplete; or contribute to the totality of evidence to support labelling for HI. Example drugs where PBPK modelling was used for these purposes include adagrasib, olanzapine, samidorphan and avapritinib. There are several areas where knowledge gaps exist that warrant further research, including the lack of mechanistic understanding of the impact of HI on drug absorption.
Conclusions: We envision a broad application for PBPK modeling approaches in supporting exposure and dose predictions in HI populations throughout clinical development of drugs and believe it can contribute to a totality of evidence to support labeling for HI with a reduced clinical trial burden.
