Table 1.

Criteria for evidence-based selection of training compounds for the MIP-DILi project

DILI category	Model hepatotoxins that selectively target specific pathways/systems in the hepatocyte
	Model hepatotoxins that cause specific forms of DILI in preclinical model systems
	Drugs that have a well-defined association (clinical phenotype, frequency, severity) with particular forms of DILI in man and in non-clinical models
	Drugs that cause DILI in man but did not in available non-clinical test systems
	Compounds that do not show liver damage either in pre-clinical tests or in man, but which are chemically related to drugs that are clearly associated with DILI, to act as negative controls
Mechanism known	Molecular target
	Reactive metabolites
	CYP independent cell injury
	Mitochondrial impairment
	Inhibition of BSEP
	Innate/adaptive immune activation
	Other
DILI initiating primary event	Evidence for primary event (in vitro/in vivo)
	Evidence for mechanism (in vitro/in vivo)
	Drug or metabolite involved
	Dose–response
ADME data available?	Characterization of drug exposure and metabolite profiles (Phase I–III)
DILI frequency in humans?	Clinical evidence of liver injures reported
Human-specific DILI?	Evidence of human only mechanism(s) of liability and liver injury
Human selective DILI?	Sensitivity as it relates to in vivo or in vitro test systems