Development of a reference and proficiency chemical list for human steatosis endpoints in vitro

“

TL;DR

This study proposes a list of 18 high-priority reference and proficiency chemicals (9 steatosis-inducing, 9 negative) for developing and validating human in vitro test methods to detect chemical-induced hepatic steatosis, a key event in non-alcoholic fatty liver disease (NAFLD). The chemicals were selected through systematic literature review and weight-of-evidence assessment, prioritizing structurally diverse compounds with moderate-to-strong evidence and global availability for regulatory use. This chemical set is intended to support OECD test guideline development and integrated approaches to testing and assessment (IATA) for metabolic disrupting chemicals.

Full summary

4,279 CHARS

**Background:** Non-alcoholic fatty liver disease (NAFLD), characterized by hepatic steatosis (>5% hepatocyte fat accumulation), affects approximately 25% of the global population and is linked to both lifestyle factors and chemical exposures. Metabolic disrupting chemicals (MDCs) can promote steatosis, but standardized regulatory in vitro test methods for this endpoint are lacking. The EU Horizon 2020 GOLIATH project aims to develop and pre-validate such methods. A critical first step is establishing a set of reference and proficiency chemicals with well-characterized activity toward hepatic steatosis.

**Methods:** A tiered literature search strategy was employed using the Scopus database. An initial broad search for publications on human hepatic steatosis was conducted, followed by chemical-specific sub-searches for 36 prioritized chemicals. Searches were performed between 6-8 January 2021 and 15 February 2022. Inclusion criteria required structurally diverse chemicals from multiple sectors (pharmaceuticals, industrial chemicals, pesticides, food additives, natural compounds) with reasonable global availability and cost. Chemicals subject to international restrictions (e.g., Stockholm Convention) or undefined mixtures were avoided where possible. Highest priority was given to human-relevant evidence (epidemiological studies, clinical trials, human in vitro studies using primary human hepatocytes, HepaRG, or HepG2 cells). Rodent and other animal studies were used as supportive evidence. Full-text assessment and data extraction were conducted by two team members with independent review by a third. Chemicals were classified as 'positive' (inducing steatosis/lipid accumulation) or 'negative' (no induction or reduction of steatosis). The target was 25-50% negative chemicals to assess test method discriminatory ability.

**Key Results:** From an initial 160 chemicals identified, 36 underwent detailed chemical-specific database searches. After full-text review and weight-of-evidence assessment, 18 chemicals were identified as high-priority tentative proficiency chemicals. The 9 steatosis-inducing chemicals are: amiodarone, benzo[a]pyrene, mono-ethylhexyl phthalate (MEHP), oleic acid, tributyltin chloride (TBT), tebuconazole, tetracycline, triphenyl phosphate (TPP), and valproic acid. The 9 negative chemicals are: acetaminophen, ascorbic acid (vitamin C), caffeine, docosahexaenoic acid (DHA), fenofibrate, metformin, resveratrol, rosiglitazone, and rotenone. Four lower-priority candidates (DDE, PFOA, chlorpyrifos, thiacloprid) were identified but have limitations such as international use restrictions or weaker evidence. Nine chemicals were not prioritized due to redundancy with higher-priority chemicals (e.g., cyproconazole, ketoconazole, pemafibrate, pioglitazone, rifampicin, thiamethoxam) or technical limitations (fructose, glucose, DEHP, BPA, niacin). The authors note that a potency range (weak/moderate/strong) could not be robustly established due to the lack of a gold standard in vitro method and variability in lowest observed effect concentrations across studies.

**Clinical Implications:** This chemical set provides a critical foundation for developing, optimizing, and validating human-relevant in vitro test methods for hepatic steatosis, which could ultimately support regulatory hazard assessment of metabolic disrupting chemicals under EU chemical regulations (e.g., REACH). The authors emphasize that steatosis is a reversible key event in NAFLD progression, and reliable in vitro methods could reduce reliance on rodent in vivo testing while improving human relevance. Integration of these methods into IATA frameworks, as demonstrated in a case study with valproic acid, could enable tiered testing strategies combining transcriptomics, reporter gene assays, mitochondrial stress tests, and cellular triglyceride accumulation assays. The authors note limitations including the inability of single-tissue liver models to capture secondary steatosis mechanisms (e.g., BPA acting via pancreatic estrogen receptor signaling), highlighting the need for multi-organ approaches and IATA frameworks. The study will contribute to an OECD detailed review paper on metabolism disrupting chemicals (workplan project 4.147).