Mechanistic modeling of biologics-induced liver injury (BILI) predicts hepatotoxicity of Tocilizumab through both on- and off-target effects
Biologics address a range of unmet medical needs. However, there are increasing numbers of BILI cases which slow therapeutic development or require frequent monitoring of liver function.
To assess the potential clinical BILI liability of biologics, a novel
quantitative systems toxicology (QST) platform, BIOLOGXsym™, was developed incorporating relevant liver biochemistry, mechanistic representations of key pathophysiologic pathways, and assay data from a human biomimetic liver microphysiology system.
Tocilizumab (TCZ), a human anti-interleukin (IL)-6 receptor antagonist biologic used for treating inflammatory diseases, can lead to transient elevations in alanine aminotransferase (ALT), a biomarker of liver injury. Within BIOLOGXsym, a mechanistic model of TCZ was developed including direct representation of IL-6, IL-6 receptor, and specific on-target (CYP expression, hepatocyte regeneration) and offtarget
(TCZ-induced reactive oxygen species, [ROS]) effects.
Exposure of human (vascularized) liver acinus microphysiology system ([v]LAMPS) to 1.6 μM TCZ demonstrated TCZ-induced ROS and recovery of IL-6 inhibited CYP activity, offering insights into potential hepatotoxic effects of TCZ. These data, along with data from the literature and TCZ exposure predictions from GastroPlus®, were used to parameterize the model and run proof-of-concept simulations.
Simulated TCZ administration (8 mg/kg Q4W for 12 weeks) to a small cohort (n=4) of individuals with elevated IL-6 demonstrates ALT above three times the upper limit of normal in one individual. When TCZ is co-administered with repeat therapeutic doses of acetaminophen, a medication with CYP-dependent hepatotoxicity, all individuals in the cohort show significant ALT elevations. Mechanistic analysis of these
responses indicates persistent ALT elevations during co-medication when either off-target or on-target effects are excluded from the simulation.
These results demonstrate the potential of BIOLOGXsym to predict BILI, identify key toxicity mechanisms, and evaluate drug-drug interactions for developing biologics.
- Schiff 2011
- Nishimoto 2003
- Nishimoto 2008
- Long 2016
By: Lara Clemens, James J. Beaudoin, Lawrence A. Vernetti, D. Lansing Taylor, Albert Gough, Christina Battista
Presented at Society of Toxicology (SOT) 61st Annual Meeting and ToxExpo, March 27-31, 2022