Regulatory approval rates of new chem- ical entities (NCEs) are currently lower than at any previous time. Recent data indicate that the average success rate for all therapeutic areas is approximately 11% – implying that only one in nine compounds survives the drug devel- opment process – and that the major causes of attrition in the clinic are safety issues [1]. Between 1975-2007, furthermore, 47 drugs were withdrawn from the market due to adverse events. In order to increase the success rate of drug development and to minimise the risk to patients, the prediction of drug safety must improve significantly. Recent data indicate three toxicities as the most common clinical adverse events: hepatotoxicity (14%), car- diotoxicity (16%) and neurotoxicity (22%). Cardiac and hepatic toxicity also contributed disproportionately to drug withdrawals, with 21 of the 47 drugs withdrawn between 1975- 2007 linked to hepatotoxicity and 21 to cardiotoxicity. For these reasons, eliminating potentially toxic com- pounds as early as possible in the drug discovery process – particularly in the case of hepatotoxicity and cardiotoxic- ity – would clearly improve the overall efficiency of drug development. As cell-based toxicity assays address only cell-autonomous toxicity, and in vivo toxicity often involves drug absorption, distribution, metabolism and excretion, in vivo models should be re-considered for early-stage toxicity screening in drug discovery. As large-scale compound screens are not practical in mammalian models due to ethical, financial, and throughput issues, embryos and larvae of the zebrafish (Danio rerio) represent an attractive alternative. Zebrafish have several advantages for screening large numbers of compounds, both for drug discovery and for toxicity applications. Adult zebrafish are small, cost-effective to maintain, and breed easily, producing large numbers of offspring. Zebrafish embryos and larvae, with which most bioassays are performed, are microscopi- cally small – between 1 and 4 mm in length, and therefore compatible with 96-well microtitre plates. Importantly, zebrafish embryos and larvae are opti- cally transparent, readily allowing the visualisation of internal organs such as the liver. The organisation of the genome and the genetic pathways con- trolling signal transduction and devel- opment appear to be highly conserved between zebrafish and humans. In addition, the cardiovascular, nervous and digestive systems of zebrafish have Pharmaceutical companies are under increasing pressure to develop safe and effective drugs, while at the same time controlling R&D expenditures. One partial solution to this challenge may be new technologies capable of generating in vivo data on the safety and efficacy of large numbers of compounds as early as possible in the drug discovery process, thereby reducing the number of drug candidates that fail in clinical development. Zebrafish offer the unique possibility to carry out high-throughput in vivo screens for compounds at the hit to lead stage, to detect both therapeutically relevant bioactivities and to assess potential toxicities. Here, we briefly highlight zebrafish assays for cardiotoxicity and hepatotoxicity. IN VIVO DISEASE MODELS in vivo in vivo