Related Experiment Video
Updated: Feb 4, 2026

Human Liver Microphysiological System for Assessing Drug-Induced Liver Toxicity In Vitro
Published on: January 31, 2022
A liver-immune coculture array for predicting systemic drug-induced skin sensitization
Lor Huai Chong1, Huan Li, Isaac Wetzel
1Department of Biomedical Engineering, National University of Singapore, Singapore. biety@nus.edu.sg.
Abstract:
Drug-induced skin sensitization is prevalent worldwide and can trigger life-threatening health conditions, such as Stevens Johnson Syndrome. However, existing in vitro skin models cannot adequately predict the skin sensitization effects of drugs administered into the systemic circulation because dermal inflammation and injury are preceded by conversion of parent drugs into antigenic reactive metabolites in the liver and subsequent activation of the immune system. Here, we demonstrate that recapitulation of these early tandem cellular processes in a compartmentalized liver-immune coculture array is sufficient to predict the skin sensitization potential of systemic drugs. Human progenitor cell (HepaRG)-derived hepatocyte spheroids and U937 myeloid cells, a representative antigen presenting cell (APC), can maintain their respective functions in 2 concentric micro-chambers, which are connected by a diffusion microchannel network. Paradigm drugs that are reported to cause severe cutaneous drug reactions (i.e. carbamazepine, phenytoin and allopurinol) can be metabolized into their reactive metabolites, which diffuse efficiently into the adjoining immune compartment within a 48 hour period. By measuring the extent of U937 activation as indicated by IL8, IL1β and CD86 upregulation upon drug administration, we show that the liver-immune coculture array more consistently and reliably distinguish all 3-paradigm skin sensitizing drugs from a non-skin sensitizer than conventional bulk Transwell coculture. Given its miniaturized format, design simplicity and prediction capability, this novel in vitro system can be readily scaled into a screenable platform to identify the skin sensitization potential of systemically-administered drugs.
Insights
A novel liver-immune coculture array effectively predicts drug-induced skin sensitization by mimicking early metabolic and immune responses. This advanced in vitro model offers reliable identification of potential skin sensitizers for systemic drugs.
Area of Science:
- Toxicology
- Immunology
- Drug Development
Background:
- Drug-induced skin sensitization is a significant global health concern, potentially leading to severe conditions like Stevens Johnson Syndrome.
- Current in vitro models struggle to predict systemic drug sensitization due to the complex interplay of liver metabolism and immune activation preceding skin reactions.
Purpose of the Study:
- To develop and validate a novel compartmentalized liver-immune coculture array capable of predicting skin sensitization potential of systemically administered drugs.
- To recapitulate the early tandem cellular processes involving drug metabolism in the liver and subsequent immune system activation.
Main Methods:
- A microfluidic coculture system was designed with human hepatocyte spheroids (HepaRG) and U937 myeloid cells (antigen-presenting cells) in concentric micro-chambers connected by diffusion channels.
- Paradigm sensitizing drugs (carbamazepine, phenytoin, allopurinol) and a non-sensitizer were administered to assess metabolite diffusion and U937 cell activation over 48 hours.
- U937 activation was quantified by measuring upregulation of IL8, IL1β, and CD86.
Main Results:
- The liver-immune coculture array successfully metabolized paradigm sensitizing drugs into reactive metabolites that diffused into the immune compartment.
- The system demonstrated higher consistency and reliability in distinguishing sensitizing drugs from non-sensitizers compared to conventional Transwell cocultures.
- U937 activation markers (IL8, IL1β, CD86) effectively indicated the sensitizing potential of the tested drugs.
Conclusions:
- The compartmentalized liver-immune coculture array accurately predicts skin sensitization potential by simulating early systemic drug reaction mechanisms.
- This miniaturized, simple, and predictive in vitro system is suitable for scaling into a high-throughput platform for identifying systemic drug sensitizers.
Related Concept Videos
Sensitivity, Specificity, and Predicted Value
Sensitivity is the...
What is the Immune System?
Predicting Molecular Geometry
Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
Humoral Immune Responses
Second Order systems II

