Online Program

Return to main conference page

All Times EDT

Friday, September 25
Fri, Sep 25, 11:45 AM - 12:45 PM
Virtual
Poster Session

PS19-Joint Analysis of the Relationships Between Drug Exposure and Liver Enzyme Elevations for Assessing Drug-Induced Liver Injury (301121)

View Presentation

Ling Cheng, AbbVie Inc 
Xiu Huang, AbbVie Inc. 
Wei Liu, AbbVie Inc 
*Weihan Zhao, AbbVie Inc. 

Keywords: Exposure-Response Analysis, Joint Modeling, Drug-Induced Liver Injury

Drug-induced liver injuries (DILI) caused by therapeutic drugs are serious and sometimes even fatal in some patients and have resulted in disapproval of new drugs and removal of approved drugs from the market. To evaluate DILI, both the widely accepted Hy’s law and the FDA’s evaluation of drug-induced serious hepatotoxicity (eDISH) program focus on two key liver chemistry tests, namely alanine aminotransferase (ALT) and total bilirubin (TBL). An association of elevated ALT and TBL with drug exposure would indicate a high DILI risk to subjects exposed, especially to those with high exposures. Despite the fact that ALT and TBL elevations are generally correlated and may be associated with different risk factors, in practice exposure-response analyses are usually performed individually for these two measures with the same set of covariates considered in each analysis.

In this work, simulated continuous and categorical (binary and ordinal) data are used to compare statistical methods for correlated bivariate data for assessing the relationship between drug exposure and ALT/TBL elevations. Model recommendations are provided for different data assumptions, although it was found that the appropriate method to use is largely dependent on the objective and available data. These statistical models for correlated continuous and categorical data are applied in the exposure-safety analyses of an AbbVie dataset to predict the safety impact under scenarios of increased exposures. The results are compared with those from the univariate models to demonstrate the advantages of the joint analysis.