For fasiglifam (TAK875) and its metabolites the substance‐specific mechanisms of liver toxicity were studied. Metabolism studies were run to identify a putatively reactive acyl glucuronide metabolite. In vitro cytotoxicity and caspase 3/7 activation were assessed in primary human and dog hepatocytes in 2D and 3D cell culture. Involvement of glutathione (GSH) detoxication system in mediating cytotoxicity was determined by assessing potentiation of cytotoxicity in a GSH depleted in vitro system. In addition, potential mitochondrial liabilities of the compounds were assessed in a whole‐cell mitochondrial functional assay. Fasiglifam showed moderate cytotoxicity in human primary hepatocytes in the classical 2D cytotoxicity assays and also in the complex 3D human liver microtissue (hLiMT) after short‐term treatment (24 hours or 48 hours) with TC50 values of 56 to 68 µM (adenosine triphosphate endpoint). The long‐term treatment for 14 days in the hLiMT resulted in a slight TC50 shift over time of 2.7/3.6 fold lower vs 24‐hour treatment indicating possibly a higher risk for cytotoxicity during long‐term treatment. Cellular GSH depletion and impairment of mitochondrial function by TAK875 and its metabolites evaluated by Seahorse assay could not be found being involved in DILI reported for TAK875. The acyl glucuronide metabolites of TAK875 have been finally identified to be the dominant reason for liver toxicity.

The toxicological profile of insulins is exclusively due to exaggerated pharmacology resulting in hypoglycemic findings. Insulin analogues displaying modifications and aimed at improving pharmacokinetics do not induce different toxicity. The main target is the brain displaying neuronal necrosis. Wallerian degeneration of nerves occurs rarely after severe hypoglycemia. These findings are of potential human relevance; nevertheless, these changes are induced in normoglycemic animals whereas diabetic patients suffer from hyperglycemia. Therefore, it is usually not difficult to achieve a therapeutic window for subsequent use in patients. Based upon this and in the absence of classical toxicity, there has been no scientific need for diabetic animal models. A greater challenge is the mitogenicity already inherent with regular insulin. Thus, the focus for preclinical safety evaluation of analogues is to demonstrate that modifications in regular insulin do not result in enhanced mitogenicity. The approaches used to assess the mitogenic potential of insulin analogues have changed over time driven by scientific progression and changes within the regulatory environment. Therefore, in vitro and in vivo evaluation of cell proliferation has become common practice, and to date there has been no evidence that the mitogenic potential of insulin analogues may be increased compared to regular insulin.

Fibroblast growth factor 21 (FGF21) and FGF15/FGF19 belong to the same subgroup of FGFs and are believed to have therapeutic potential in the treatment of type 2 diabetes and associated metabolic dysfunctionalities and pathological conditions. FGF19 has been proposed to induce hyperplasia and liver tumors in FVB mice (named after its susceptibility to Friend leukemia virus B), mediated by the FGF receptor 4 (FGFR4). The goal of this work was to investigate whether FGF21 might also have a potential proliferative effect mediated via FGFR4 using liver-specific Fgfr4 knockout (KO) mice. We conducted a mechanistic 7-day study involving female Fgfr4 fl/fl and Fgfr4 KO mice with a treatment regimen of twice daily or daily subcutaneous injections of FGF21 or FGF19 (positive control), respectively. The Ki-67 liver labeling index (LI) was evaluated by a semi-automated bioimaging analysis. The results showed a statistically significant increase in FGF21- and FGF19-treated Fgfr4 fl/fl mice. Interestingly, in Fgfr4 KO mice, this effect was absent following both treatments of FGF19 and FGF21, indicating that not only the FGFR4 receptor is pivotal for the mediation of hepatocellular proliferation by FGF19 leading finally to liver tumors but it seems also that FGFR4/FGF21 signaling has an impact on the hepatocellular proliferative activity, which does not promote the formation of hepatocellular liver tumors based on the current knowledge.