Xenograft models, and in particular the mouse xenograft model, where human cancer cells are transplanted into immunocompromised mice, have been used extensively in cancer studies. Although these models have contributed enormously to our understanding of cancer biology, the zebrafish xenograft model offers several advantages over the mouse model. Zebrafish embryos can be easily cultured in large quantities, are small and easy to handle, making it possible to use a high number of embryos for each experimental condition. Young embryos lack an efficient immune system. Therefore the injected cancer cells are not rejected, and the formation of primary tumors and micrometastases is rapid. Transparency of the embryos enables imaging of primary tumors and metastases in an intact and living embryo. Here we describe a method where GFP expressing tumor cells are injected into pericardial space of zebrafish embryos. At four days post-injection, the embryos are imaged and the formation of primary tumor and distant micrometastases are analyzed.