Highlight report: hepatotoxicity prediction with Hep3B cells

Recently, Yu et al. (2018) have contributed a study with 3D-cultivated Hep3B cells tested with 22 model compounds. LC₅₀ values were compared to rat LD₅₀ values and human C_max values. At a first glance, one may ask, whether it is really justified to introduce another liver tumor cell line besides the already well-established HepG2 and HepaRG cells. However, the obtained LC₅₀ values were remarkably low. For example, acetaminophen resulted in an LC₅₀ of 300 µM, which is lower compared to many previously reported cell systems, such as HepG2. Finally, a remarkably good overall predictively of human hepatotoxicity was obtained. Currently, human hepatocytes are still considered a gold standard for in vitro testing of human hepatotoxicity (Pfeiffer et al. 2015; Leist et al. 2017; Grinberg et al. 2014; Ghallab 2017; Hammad and Ahmed 2014; Hammad et al. 2015). Moreover, hepatotoxicity represents a complex process that may involve interaction of several cell types (Jansen et al. 2017; Vartak et al. 2016; Albrecht 2017; Schenk et al. 2017). If the high predictively of the Hep3B-based test system (Yu et al. 2018) would be confirmed by further compounds, it might indeed facilitate prediction of hepatotoxicity. However, whether prediction of idiosyncratic hepatotoxicity will be possible based on these still remains to be shown.